The Structure of China's Domestic Consumption Analyses and Preliminary Forecasts Li Xuezeng Yang Shengming SWP755 He Juhuang Introduction by Jacques van der Gaag WORLD BANK STAFF WORKING PAPERS Number 755 W#t WVS WORLD BANK STAFF WORKING PAPERS Number 755 cU67 lkiD e zo The Stucture of China's Domestic Consumption Analyses and Preliminary Forecasts jlNTEF.NATIONAL MONETP. AIY FUND J01,4T LlBItARY Li Xuezeng Yang Shengmning He Juhuang DEC O Introduction by Jacques van der Gaag INA : NT RECONSTRUCTIO:4 AN1D DDVELOe"'N WASHINGTON. D.C. 20431 The World Bank Washington, D.C., U.S.A. Copyright ©) 1985 The International Bank for Reconstruction and Development/THE WORLD BANK 1818 H Street, N.W. Washington, D.C. 20433, U.S.A. All rights reserved Manufactured in the United States of America First printing November 1985 This is a working document published informally by the World Bank. To present the results of research with the least possible delay, the typescript has not been prepared in accordance with the procedures appropriate to formal printed texts, and the World Bank accepts no responsibility for errors. The publication is supplied at a token charge to defray part of the cost of manufacture and distribution. The World Bank does not accept responsibility for the views expressed herein, which are those of the authors and should not be attributed to the World Bank or to its affiliated organizations. The findings, interpretations, and conclusions are the results of research supported by the Bank; they do not necessarily represent official policy of the Bank. The designations employed, the presentation of material, and any maps used in this document are solely for the convenience of the reader and do not imply the expression of any opinion whatsoever on the part of the World Bank or its affiliates conceming the legal status of any country, territory, city, area, or of its authorities, or concerning the delimitation of its boundaries, or national affiliation. The most recent World Bank publications are described in the annual spring and fall lists; the continuing research program is described in the annual Abstracts of Current Studies. The latest edition of each is available free of charge from the Publications Sales Unit, Department T, The World Bank, 1818 H Street, N.W., Washington, D.C. 20433, U.S.A., or from the European Office of the Bank, 66 avenue d'Iena, 75116 Paris, France. Li Xuezeng, Yang Shengming, and He Juhuang are research associates at the Chinese Social Sciences Academy. The views, opinions, and theories that they present here are theirs alone and do not represent the Chinese Social Sciences Academy or its dependent agencies. The statistical data used are from statistical offices in China. Anyone who wishes to cite any of the data should check them against the original sources. Jacques van der Gaag is an economist in the Development Research Department of the World Bank. Library of Congress Cataloging-in-Publication Data Li, Xuezeng. The structure of China's domestic consumption. (World Bank staff working papers ; no. 755) Includes bibliographical references. 1. Consumption (Economics)--China--Peking. 2. Consumption (Economics)--China--Hupeh Province. 3. Consumption (Economics)--China--Surveys. 4. Consumption (Economics)--China. I. Yang, Sheng-ming. II. He, Juhuang. III. Title. IV. Series. HC430.C6L5 1985 339.4'1'0951 85-22768 ISBN 0-8213-0640-5 - iii - ABSTRACT This paper is the final product of a cooperative research project conducted by staff of the Chinese Academy of Social Sciences in Beijing and the Development Research Department of The World Bank. Mainly based on previously unpublished data, household consumption patterns in China are analyzed, emphasising changes over time and differences between urban and rural areas and among households at different points of the welfare distribution. Though detailed data on the structure of both income and consumption is presented, the analytical focus is on household consumption and its likely development up to the year 2000. The report contains formal statistical analyses of consumption patterns as well as policy oriented sections that discuss existing problems in China's market for consumption goods, and include proposals for solutions. The authors hope that the total is more than just the sum of its parts, but each section is self-contained and can be read as an independent paper on one major aspect of China's Domestic Consumption. - iv - ACKNOWLEDGEMENTS We are extremely grateful to our World Bank colleague Jacques van der Gaag for discussing with us China's consumption structure. We also thank Edwin Lim, Benjamin King, Adrian Wood, and Shujiro Urata for their assistance in this research. We thank Dr. Liu Guoguang of the Chinese Social Sciences Academy for his support; we wish to thank in particular Mr. Dong Furen for his support of this work, from start to finish. We also wish to thank all those at the Chinese National Bureau of Statistics, the Beijing Statistics Office, and the Hubei Statistics Office for the great deal of data and assistance they furnished. Sections II-VI are translations from the Chinese original papers. Many thanks to Brenda Holt and Wendy Shinn for the typing of the various drafts of this report and expert editorial assistance. TABLE OF CONTENTS Page I. Introduction (Jacques van der Gaag) 1 t. Some Analyses of Beijing Staff-and-Worker Household Consumption 6 Structure A. Structure and Population Structure of Survey Households 6 1. Staff-and-Worker Living Condition Survey Households: 6 Basic Conditions 2. Structure of Survey Households 8 3. Population Structure of Survey Households 10 4. Employed Population Structure of Survey Households 11 B. Staff-and-Worker Cash Income, Expenditure, and Consumption 12 Structure 1. Average Per Capita Monthly Cash Income Structure of 13 Staff-and-Worker Households 2. Changes in Average Per Capita Monthly Cash Income 15 Structure 3. Average Per Capita Monthly Cash Income of Employed 16 Staff-and-Workers 4. Staff-and-Worker Income Structure by Sector of 19 National Economy 5. Average Per Capita Monthly Cash Expenditure Structure 19 in Staff-and-Worker Households 6. Primary Commodity Consumption Structure of Staff-and- 22 Worker Households with Different Income Levels 7. Changes Over Time in Consumption of Major Commodities 28 by Beijing Staff-and-Worker Households (Changes in average Per Capita Annual Consumption) C. Some Suggestions 32 1. Low Wages and High Employment are Unfavorable for 32 Family Planning and Population Control 2. Income from Mental and Physical Labor is Essentially 33 the Same; Such Egalitarianism Is Detrimental to Achieving Modernization 3. Whether the Supply Quota System for Cereals, Cotton 33 and Oils Can Be Eliminated Should Be Discussed 4. Low Rents Are Detrimental to Resolving the Tight 34 Housing Situation 5. Some Characteristics Expressed by Engel's Coefficient 35 in China 6. We Must Vigorously Develop the Service Sector 36 - vi - III. Some Analyses of Consumption Structure in Hubei Peasant 37 Households A. Structure of Hubei Peasant Survey Households 37 1. Basic Conditions in Peasant Survey Households 2. Structure of Survey Households, Population, and 39 Labor Force by Net Income B. Income Structure of Hubei Peasant Survey Households 42 1. Hubei Peasant Household Total Income and Its Sources 43 2. Sources of Net Per Capita Income in Hubei Peasant 44 Households 3. Relationship Between Hubei Peasant Household Size and 44 Per Capita Total and Net Income 4. Dynamic Analysis of Per Capita Net Income Levels in 47 Hubein Peasant Survey Households 5. Effect of Household Size on Per Capita Consumption 50 and Consumption Structure 6. Preliminary Analysis of Consumption Structure at 54 Different Net Income Levels in Hubei Peasant Households C. The Income-Expenditure Situation and Structure Among Peasant 54 Survey Households and Their Position Nationwide D. Some Summary Assessments and Conclusions. 64 IV. Using a Statistical Data Approach to Forecast Consumption 68 Structure for a Comfortable Standard of Living (Yang Shengming) A. Forecasting People's Consumption Levels 72 B. Forecasting Urban Consumption Structure 72 1. Consumption Structure in 1990 72 2. Consumption Structure in 2000 75 C. Forecasting Peasant Consumption Structure 78 1. Peasant Consumption Structure in 1980 79 2. Peasant Consumption Structure in 2000 79 D. Forecasting National Domestic Consumption Structure 82 - vii - V. Demand System Analysis and Consumption Structure Forecasting 84 (He Juhuang, with the assistence of Cong Yi) A. Some Problems Requiring Attention in Studying China's 84 Consumption Structure 1. Collective Consumption 84 2. Free Health Care 85 3. Price Subsidies 86 4. Supply Controls 86 5. Irrational Relative Pricing 87 B. Extended Linear Expenditure System 88 1. Model 88 2. Demand Elasticity 91 C. Analysis of Beijing Cross-section Data 93 D. Analysis of Hubei Cross-section Data 101 E. Forecasting Consumption Structure 110 VI. Some Policy Questions Regarding China's Consumption Strategy 116 and Consumption Structure (Li Xuezeng) A. The Problem of the Ratio of Social Consumption 118 Funds to Individual Consumption Funds B. The Problem of Price Subsidy Policy 120 C. The Problem of Staff-and-Worker Labor Insurance 122 Benefits Policy D. The Problem of Consumption Structure Policy 127 E. Policy Questions Regarding the Industrial Structure 136 Corresponding to Consumption Structure - viii - LIST OF TABLES Table Number Title Page 1 Basic Conditions of Survey Households, Beijing 6 Staff and Workers 2 Basic Conditions of Survey Households, Nation- 7 wide Staff and Workers 3 Beijing Staff-and-Worker Survey Household 8 Structure, by Income Class 4 Structure by Income of Staff-and-Worker Survey 9 Households Nationwide 5 Population Structure of Beijing Households Broken 11 Down by Income 6 Employed Population Structure of Beijing Staff-and- 12 Worker Survey Households 7 Average Per Capita Monthly Cash Income of Staff-and- 13 Worker Households 8 Average Per Capita Monthly Cash Income of Staff-and- 14 Worker Households 9 Per Capita Monthly Cash Income of Staff-and-Worker 15 Households 10 Changes in Average Per Capita Monthly Income Struc- 15 ture (percent) 11 Average Per Capita Monthly Cash Income of Employed 16 Staff and Workers 12 Staff-and-Worker Cash Income Structure, Beijing, 18 1982 (percent) 13 Monthly Per Capita Cahs Income by Sector, Beijing, 20 1982 14 Staff-and-Worker Cash Income Structure by Economic 21 Sector (percent) 15 Average Per Capita Monthly Cash Expenditures 22 16 Changes in Average Per Capita Monthly Expenditures 22 - ix - 17 Average Monthly Primary Commodity Purchases Per 23 Capita in Beijing Staff-and-Worker Households 18 Changes in Retail Prices of Major Commodities and 26 Impact on Living Standards, Beijing, 1982 19 Changes Over Time in Household Consumption, Beijing 29 20 Total Purchases of Durable Consumer Goods by 1,200 31 Staff-and-Worker Households, Beijing, 1982 21 Relationship Between Living-Expense Income and 32 Employment in Beijing Staff-and-Worker Households, 1982 22 Basic Conditions: Hubei Peasant Survey Households 38 23 Structure of Hubei Peasant Survey by Net Income Level 39 24 Structure of Hubein Peasant Household Population 41 by Net Income 25 Structure of Hubei Peasant Household Labor Force 41 by Net Income 26 Total Income Structure of Hubei Peasant Survey 43 Households, 1982 27 Structure of Hubei Peasant Household Income Received 44 From the Collective 28 Net Per Capita Income and Its Sources in Hubei 45 Peasant Survey Households, 1982 29 Structure of Net Per Capita Income from the Collective, 45 1982 30 Relationship Between Survey Household Size and Net 46 and Total Income 31 Labor Force Structure of Survey Households 46 32 Hubei Peasant Survey Households by Per Capita 48 Total and Net Income *33 Structure of Sources of Per Capita Total and Net 49 Income in Hubei Peasant Survey Households (%) 34 Relationship Between Survey Household Size and 51 Consumption Expenditure, 1982 35 Relationship Between Household Size and Consumption 52 Structure, 1982 (%) 36 Basic Situation of Peasant Households in Hubei and 55 Nationwide 37 Average Per Capita Net Income Level and Structure of 56 Peasant Survey Households, Hubei and Nationwide 38 Average Per Capita Consumption Expenditure Levels 60 and Structure in Peasant Survey Households, in Hubei and Nationwide 39 Proportion of Commodity Expenditures in Peasant 61 Consumer-Goods Expenditures 40 Consumption Structure of Staff-and-Worker Households 73 with Average Per Capita Monthly Living-Expense Incomes Over 60, Nationwide, 1982 41 Consumption Structure of Staff-and-Worker Households 74 with Average Per Capita Monthly Living-Expense Incomes Over 601, Beijing, 1982 42 Retail Sales Structure of Social Consumer Goods 76 43 Changes in Staff-and-worker Household Consumption 77 Structure Nationwide (%) 44 1982 Peasant Consumption Structure (%) in Shanghai, 78 Beijing, Guangdong and Hubei 45 1982 Consumption Structure of Urban Staff-and-Worker 80 Households with Per Capita Monthly Living Expenses Over 451 46 Changes in China's Peasant Consumption Structure 80 47 National Domestic Consumption Structure (%) 83 48 Application of National Income, 1982 85 49 Abstract of Staff-and-Worker Household Income Survey 95 Data, Beijing, 1982 50 Parameters of Demand Functions Calculated from 1982 97 Beijing Data 51 Untitled 97 - xi - 52 Total Expenditure Elasticity of Demand Among Beijing 99 Staff-and-Workers 53 Price Elasticity of Demand Among Beijing Staff- 100 and-Workers 54 Price Elasticity of Demand Among Beijing Staff- 100 and-Workers (income compensation) 55 Nationwide State-operated Industry and Commerce List 101 Price Index 56 Summary of Sample Survey Data 1981 Hubei Peasant 103 Household Income 57 Summary of Sample Survey Data: 1982 Hubei 104 Peasant Household Income 58 Demand Function Parameters Calculated According to 105 1981 and 1982 Combined Data 59 Parameters of Engel's Double Logarithmic Function 106 Calculated According to Hubei Cross-Section Data 60 Total Expenditure Elasticity of Demand Among 108 Hubei Peasants 61 Price Elasticity of Demand Among Hubei Peasants 109 62 Staff-and-Worker Household Consumption Structure 114 63 Peasant Household Consumption Structure 115 64 Nutrients Absorbed from Food Per Person Per Day, 129 Urban and Rural AReas, 1982 65 Comparison of Average Per Capita Major Food Consumption 129 among Peasant and Staff-and-Worker Households, 1982 66 Per Capita Dialy Nutrient Intake in Several Countries 131 67 Proportions of Nutrients Supplied by Animal and 133 Vegetable Products I. DNTRODUCTION This report is the result of a cooperative research project of the Chinese Academy of Social Sciences (CASS) and the World Bank. Issues studied include Enterprise Management, Structural Change and Private Consumption. This is one of two reports on Private Consumption l/ As one of the authors of this report wrote elsewhere, the objective of expanding production and all other economic activities is "to ensure the highest degree of satisfaction of growing material and cultural needs of the laboring people." (Yang Shenming, "The Structure of People's Consumption", Chapter 18 in: China's Economic Structure, People's Publishing Company, December 1981.) Thus, to a large extent, the success of economic policies can be measured by their ability to increase the people's living standard by providing sufficient means to satisfy consumption needs. Furthermore, when the consequences of various policies on future economic development are analysed, heavy emphasis should be placed on the magnitude and composition of private consumption. This study highlights these two important aspects: consumption as an indicator of the people's level of economic well-being, and consumption as a key variable in discussions on growth and economic development. The following sections are all translations of the original Chinese contributions from the staff of CASS. Each section is self-contained and can be read as a separate paper on one aspect of household consumption. Sections 1/ The first report on Private Consumption is "Private Household Consumption in China; a Study of People's Livelihood", World Bank Staff Working Paper Number 701, Jacques van der Gaag, 1984. - 2 - II and III present detailed descriptions of the structure of both household income and household consumption in the People's Republic of China. Though historical data are incomplete, an attempt is made to trace the development from 1955 to 1982. In Section II this is done for staff and workers living in urban areas. Very detailed information is available for the years 1981 and 1982, including information on price changes that took place between these two years. Though the section is primarily descriptive in nature, it ends with a number of policy recommendations on a wide variety of topics, ranging from family planning to supply quotas to rent subsidies. Section III parallels the previous section, with a focus on peasant household consumption. Using selected data for the period 1954-1982 the development of total household income and consumption, and changes in their structure are analysed. The section includes a detailed analyses of year-by- year changes for the period 1978-1982. Not surprisingly, the introduction of the Economic Responsibility System shows a large impact on the structure of household income during this period. The changes in household consumption patterns, especially for the latter years, provide indications about future developments that can be expected if economic growth continues at its current pace. In Section IV an explicit attempt is made to make such projections, based on past development, the current level and composition of household consumption, and assumptions about the development of the economy until the year 2000. Under the assumption of a 6.5-7.0 percent annual growth rate, with a capital accumulation rate of 29 percent (leaving 71 percent for consump- tion), and a population policy that results in two children per couple (or 1.22 billion people in 2000), the average Chinese household in the year 2000 is expected to have 701w available for private and collective consumption, plus about 200w in collective "social" accumulation funds and other "non- productive" accumulation funds. The consequences of this development are traced for the structure of consumption in urban and in rural areas. Section V is the most technical. After a discussion of several features of the market for consumption goods in China that deserve special attention (collective consumption, price subsidies, supply controls, etc.), a formal model of consumer demand is presented and estimated, using grouped cross-section data. The model is the Extended Linear Expenditure System (ELES) that allows for the calculation of own- and cross-price elasticities in the absence of direct observations on prices..1! The estimation results are used to project private consumption for the year 2000, based on an economic growth scenario and changes in relative prices, in particular an increase in the price of food. The final section is the most policy oriented. It discusses the balance between accumulation and consumption, and between social and private consumption. It provides a critical assessment of price subsidy policies, housing subsidies and insurance benefits for staff and workers. Where feasible solutions to existing problems are proposed. The section concludes by discussing the structure of household consumption (emphasizing nutrition and cultural services) and its relation to China's industrial structure. During the course of this cooperative research effort, we were fortunate to be able to benefit from China's recent "open door" policy. 1/ See Lluch, C., A.A. Powell, R.A. Williams; Patterns in Household Demand and Saving, Oxford University Press, 1977. - 4 - Beginning in 1982 China publishes a Statistical Yearbook.11/ Our study made extensive use of data on income and consumption published in these Year- books. Furthermore, very detailed data on Beijing city and Hubei province were made available especially for this study by the Statistical Bureaus. Still, as the reader will notice, information on consumption in China is rela- tively scarce. Complete time series data do not exist over a sufficiently long period, while individual cross-section data are not available for analyses. On top of that, the reader is reminded in various parts of this report that a number of specific features of the Chinese market for consump- tion goods need special attention if one wants to make reliable forecasts of the structure of consumption. Consequently all efforts to make projections for the future development of private household consumption should be viewed as tentative. Still, with these warnings in mind, we hope that this report2/ will serve as an information base about current consumption patterns in China. We also expect that the results of our cooperative research project will be used as a "baseline" study for future research projects that can further benefit 1/ Statistical Yearbook of China 1981; 1983. Compiled by the State Statistical Bureau, PRC. Published by the Economic Information Agency, Hong Kong, 342 Hennessy Road, Hong Kong. 2/ And its companion Van der Gaag, 1984, op. cit., which can be read as a complement to the current report. - 5 - from improved information and better data that are likely to result from the continuation of China's "open door" policy. The continued cooperation between Chinese scholars and researchers in other countries will, no doubt, lead to improved insight in the economic development of the People's Republic, and to further understanding of the development process in general. -6- II. SOME ANALYSES OF BEIJING STAFF-AND-WORREK HDUSEHOLD CONSUMPTION STRUCTURE We have analyzed of the consumption structure of staff-and-worker households in Beijing, using survey data on Beijing staff-and-worker household living conditions and taking into consideration other relevant data we have gathered, in an attempt to find some regular patterns. A. Structure and Population Structure of Survey Households 1. Staff-and-Worker Living Condition Survey Households: Basic Conditions In Table 1 we present summary statistics on selected household surveys for staff and workers in Beijing, conducted during the period 1955- 1982. Table 1: Basic Conditions of Survey Households, Beijing Staff and Workers Item and Unit 1955 1957 1965 1981 1982 Number of households 180 475 100 1,200 1,200 surveyed (households) Population in survey 877 1,960 524 4,961 4,667 households (persons) Average population 4.87 4.13 5.24 4.13 3.89 per household Number of employed persons 223 598 156 2,886 2,815 in survey households Average number of employed 1.24 1.26 1.56 2.41 2.35 persons in household Average number of dependents 3.93 3.27 3.35 1.68 1.62 per employed person Average monthly living- 17.87 20.91 21.00 42.85 46.75 expense income per capita (V) - 7 - The data in the above table show that the size of staff-and-worker families is declining, with a 20% drop from 1955 to 1982; the average number of jobholders per household is on the rise, with an 89.5% increase from 1955 to 1982; the average monthly living-expense income per capita increased 1.6 times from 1955. The following Table compares these results for Beijing with data for the whole country. Table 2: Basic Conditions of Survey Households, Nationwide Staff and Workers Item 1964 1981 1982 Number of households 3,537 8,715 9,020 surveyed (households) Population in survey 18,746 36,952 37,343 households (persons) Average population 5.30 4.24 4.14 per household Number of employed persons 5,518 20,829 21,558 in survey households Average number of employed 1.56 2.39 2.39 persons in household Average number of dependents 3.40 1.77 1.73 per employed person Average monthly living- 18.92 38.64 41.69 expense income per capita (V) The 1982 survey data on living conditions in 9,020 staff-and-worker households nationwide show an average of 4.14 persons per household, slightly higher than the 3.89 in Beijing; the average number of jobholders per house- hold is 2.39, versus 2.35 in Beijing; the average number of dependents per - 8 - jobholder is 1.73 for the country as a whole, and 1.62 in Beijing; and the average monthly living-expense income per capita is 41.69Y, slightly lower than the 46.75Y in Beijing. These data show that the 1982 survey data on staff-and-worker household living conditions in Beijing to a large extent also reflect the situation in the country as a whole for 1982; the average family size is on the decline, while the average number of jobholders per household is rising, as is the average monthly living-expense income per capita. 2. Structure of Survey Households In the fifties and sixties, there was no breakdown by income of Beijing's survey households. The 1981 and 1982 survey households were broken down according to average monthly living-expense income per capita. Table 3: Beijing Staff-and-Worker Survey Household Structure, by Income Class 1981 1982 Average monthly living- Number of Percent Number of Percent expense income per capita households of total households of total Total 1,200 100.00 1,200 100.00 Under 20 V 7 0.58 3 0.25 20-25 V 17 1.42 11 0.92 25-30 V 61 5.08 40 3.33 30-35 V 156 13.00 85 7.08 35-40 V 232 19.33 187 15.58 40-45 V 236 19.67 236 19.67 45-50 V 167 13.92 198 16.50 50-60 V 202 16.83 249 20.75 Over 60 V 122 10.17 191 15.92 The data in Table 3 demonstrate the uneven distribution of the various groupings in the survey households. The survey households account for - 9 - a very small proportion of the low-income groups; conversely, there is a high proportion of survey households in the high-income groups. In 1982 in particular the highest proportion of survey households is in the 50-60Y group. Why this should occur must be clarified by further surveys. This situation may cause the average survey results to be biased upwards. In Table 4 we present similar data for the nation as a whole. Thble 4: Structure by Il=e of Staff-and-Worker Survey Hixelds Natiomwide 1964 1981 1982 Absolute Relative Absolute Relative Absolute Relative Living-IERDense figure figure figure figure figure figure ine grwp (Owlnlds) (percent) (hxwleolds) (percent) (housolds) (percent) Total 3,537 100.00 8,715 100.00 9,020 100.00 20 3 and below 2,113 59.74 179 2.05 83 0.92 2D-25 a 608 17.19 476 5.46 332 3.68 25-35 V 567 16.03 2,772 31.81 2,312 25.63 35-50 E 204 5.77 3,686 42.29 4,095 45.40 5D-60 ? 45 1.27 1,037 11.90 1,281 14.2D Over 60 1 565 6.49 917 10.17 Source: "ChLna StatisticAl Abstracts" (1983), Statistics Press, 1983, page 81. The data in Table 4 illustrate that the structure of staff-and-worker survey households nationwide is undergoing major changes. The proportion of the group with an average monthly living-expense income per capita of less than 20Y declined from 59.74 percent in 1982 to 0.92 percent; conversely, the proportion of the 50-60Y and 60+Y groups rose from 1.27 percent in 1964 to 24.37 percent. Such structural changes are a major factor in the rise in - 10 - monthly per capita living-expense income in staff-and-worker households nationwide from 18.921 in 1964 to 41.69Y in 1982, reflecting the rise in the staff-and-worker standard of living. Looking at the 1982 nationwide survey data for 9,020 households, we find the proportion of survey households with average monthly living-expense incomes in the 201 and 25-35Y groups is higher than in Beijing; the proportion of survey households in the 35-50Y, 50-60Y and 60+Y groups is lower than in Beijing. This shows that the living-expense income levels for Beijing staff- and-worker households is higher than the overall nationwide level; one of the reasons for this is that the low-income groups account for a small proportion of the survey household structure, while the high-income groups account for a large proportion. 3. Population Structure of Survey Households The population of 1981 and 1982 survey households in Beijing has been broken down by average monthly living-expense income per capita. This break- down reveals the income breakdown of the survey households. Table 5 illustrates basically the same issue as Table 3. It shows that the proportion of survey households in lower-income groups is rather low, while the proportion in high-income groups is rather high. This may cause the survey results to be somewhat skewed upwards. - 11 - Table 5: Population Structure of Beijing Households Broken Down by Income 1981 1982 Average monthly living-expense Absolute Relative Absolute Relative income figure figure figure figure per capita (people) (percent) (people) (percent) Total 4,961 100.00 4,667 100.00 Below 20 i 49 0.99 18 0.39 20-25 V 98 1.98 65 1.39 25-30 V 313 6.31 200 4.29 30-35 V 716 14.43 375 8.04 35-40 V 1,037 20.90 809 17.33 40-45 V 938 18.91 936 20.06 45-50 V 643 12.96 711 15.23 50-60 V 743 14.98 904 19.37 Over 60 V 424 8.55 619 13.91 4. Employed Population Structure of Survey Households The data in Table 6 show that the proportion of the jobholding population in the low-income groups is consistently very low, while the proportion of the jobholding population in the high-income groups is higher. That is, the number of jobholders has a significant impact on household income and on staff-and-worker household consumption patterns. More jobholders promotes social stability; the more jobholders, the higher the income, and the higher the standard of living. - 12 - Table 6: Employed Population Structure of Beijing Staff-and-Worker Survey Households 1981 1982 Average monthly living expense Absolute Relative Absolute Relative income figure figure figure figure per capita (people) (percent) (people) (percent) Total 2,886 100.00 2,815 100.00 Below 20 V 9 0.31 3 0.11 20-25 e 35 1.21 19 0.67 25-30 M 125 4.33 82 2.91 30-35 M 341 11.82 170 6.04 35-40 M 526 18.23 411 14.60 40-45 1 550 19.06 531 18.86 45-50 e 417 14.45 442 15.70 50-60 V 534 18.50 632 22.45 Over 60 V 349 12.09 525 18.65 B. Staff-and-Worker Cash Income and Expenditure Structure For urban staff and workers, wages, bonuses and other forms of cash income are the basis of their livelihoods. Hence before studying the consumption structure of staff and workers we must analyze their cash income structure and cash expenditure structure. - 13 - 1. Average Per Capita Monthly Cash Income Structure of Staff-and-Worker Households Following the data in the statistical tables in "Average Per Capita Monthly Cash Income of Staff-and-Worker Households" of the Beijing Statistics Office, the cash income situation looks as follows (Table 7): Table 7: Average Per Capita Ntxthly Cash Iucoe of Staff-end-Worker Households 1981 1982 Average Mnthly Actual Incoe Actual Income livig-expense Sipport Savings Support Savings inrcm Tbtal Labor and other and loan ibtal Labor and other and loan per capita ine ircxa incin incomm i inc income income Below 20 Y 22.34 18.09 0.31 3.94 18.95 18.09 0.48 0.38 20-25 V 26.86 23.75 0.83 2.28 27.38 23.63 1.78 1.97 25-30 V 33.35 28.67 1.56 3.12 34.22 29.65 1.21 3.36 30-35 V 40.12 35.17 1.00 3.95 39.67 34.67 1.46 3.54 35-40 V 44.43 38.79 1.02 4.62 44.69 39.29 1.31 4.09 40-45 V 51.47 44.55 1.56 5.36 51.11 45.00 1.46 4.65 45-50 V 56.77 49.31 1.72 5.74 56.77 49.13 1.99 5.65 50-60 V 64.40 55.56 2.37 6.37 64.61 55.99 2.03 6.59 Over 60 V 81.48 68.02 3.72 9.74 82.01 69.57 4.02 8.42 Average 51.63 44.60 1.67 5.36 55.67 48.30 1.97 5.40 The above data show that labor income accounts for more than 80 percent of monthly cash income, for both high-income and low-income households (see Table 8). This means that since China did away with the old oppressive system the masses now earn their living by their own labor. It further means - 14 - Table 8: Average Per Capita Itbthly Cash Imane of Staff-end-Worker Households 1981 1982 Average Mnthly Actual Incoam Actual Incame living-exnse Support Savings Support Savings irmge Total Labor and other and loan Total Labor amd other and loan per capita inox.e income imcome inane i income incam incne income Below 20 V 100 81.0 1.4 17.6 100 95.5 2.5 2.0 20-25 V 100 88.4 3.1 8.5 100 86.3 6.5 7.2 25-30 V 100 86.0 4.7 9.3 100 86.6 3.6 9.8 30-35 i 100 87.7 2.5 9.8 100 87.4 3.7 8.9 35-40 V 100 87.3 2.3 10.4 100 87.9 2.9 9.2 40-45 V 100 86.6 3.0 10.4 100 88.0 2.9 9.1 45-50 V 100 86.9 3.0 10.1 100 86.5 3.5 10.0 50-60 V 100 86.3 3.7 10.0 100 86.7 3.0 10.2 Over 60 V 100 83.5 4.6 11.9 100 84.8 4.9 10.3 Average 100 86.4 3.2 10.4 100 86.8 3.5 9.7 that the laboring masses are able to live on their own labor income, so that their lives are stable and relatively undisturbed by other factors. It is only the families with material difficulties in the under-20M bracket that have a relatively high level of annual loans. The State should provide greater care and attention to these people in the form of social relief and other actions. - 15 - 2. Changes in Average Per Capita Monthly Cash Income Structure In Table 9 we show how the composition of household income has changed over time. Table 9: Per Capita Monthly Cash Income of Staff-and-Worker Households (unit: M) Item 1955 1957 1965 1981 1982 Total cash income 21.58 27.09 26.31 51.63 55.67 I. Actual income 18.53 23.02 22.29 46.27 50.27 1. Labor income 18.53 23.02 21.87 44.60 48.30 2. Support and -- - 0.42 1.67 1.97 gift income II. Savings and 3.05 4.07 4.02 5.36 5.40 loan income Table 10 gives the same information in percentages. Table 10: Changes in Average Per Capita Monthly Income Structure (percent) Item 1955 1957 1965 1981 1982 Total cash income 100.0 100.0 100.0 100.0 100.0 I. Actual income 85.9 85.0 84.7 89.6 90.3 1. Labor income 85.9 85.0 83.1 86.4 86.8 2. Support and - - 1.6 3.2 3.5 gift income II. Savings and 14.1 15.0 15.3 10.4 9.7 loan income - 16 - The data show that in 20-plus years the proportion of loan income in average monthly staff-and-worker income has declined; the proportion of labor income remains basically unchanged; and the proportion of other income has risen. 3. Average Per Capita Monthly Cash Income of Employed Staff and Workers Table 11: Average Per Capita Mxithly Cash Ilme of Bndoyed Staff and Workers, Beijing, 1982 Average per capita monthly livilr*xpense incaie lbthly Below 2D- 25- 30- 35- 40- 45- 50- Over Iten average 20 V 25 30Y 35Y 4CM 45{ 50Y 601 60Y Total 78.10 108.56 77.04 72.32 75.75 75.77 78.14 77.67 77.27 82.81 I. 1tge i xm 71.88 90.07 70.70 67.06 70.30 70.27 71.78 71.40 71.02 75.91 1. Iburly and 53.44 62.38 54.83 50.90 53.86 53.74 52.14 52.94 52.16 56.67 piecework wages 2. AU bonuses 9.07 13.16 7.88 7.32 7.87 8.11 9.88 9.25 9.07 9.52 3. Supplemental 0.85 2.47 1.47 1.07 0.97 0.94 0.99 0.73 0.77 0.74 wages 4. Overtime wpges 0.80 1.09 0.44 0.62 0.46 0.50 0.90 0.77 0.88 0.98 5. All subsidles 7.72 10.97 6.08 7.15 7.14 6.98 7.87 7.71 8.14 8.00 U. labor insuraxnce and 5.27 17.82 5.87 4.75 4.81 4.73 5.48 5.37 5.57 5.19 welfare inc 1. Bath and haircut 2.78 1.61 1.54 2.15 2.61 2.58 3.00 2.79 2.93 2.70 experises 2. Difficulty 0.15 13.19 2.68 0.98 0.31 0.15 0.06 O.Od 0.04 0.05 supplemnts 3. Otber labor 2.34 3.02 1.65 1.62 1.89 2.00 2.42 2.49 2.60 2.44 insurance ard wlfare III. Other incone 0.95 0.67 0.47 0.51 0.64 0.77 0.88 0.90 0.68 1.71 NDte: The staff-and-wrker cash Lnccuu total does not include the inrme of lndependent laborers or other enployed persons. - 17 - We see that in Beijing staff-and-worker cash income is mostly average; the gaps between the various groups are not very large. It should be noted that the under 20M group is not small, and it is even larger than some of the other groups. The reason why their average monthly living-expense income per capita tends to be low is mainly that the household size is large, with fewer employed persons. Once the number of employed people rises, house- hold income rises quickly. It is clear that employment has a major impact on income and consumption. Wage and labor insurance and welfare income account for approximately 98 percent of staff-and-worker monthly cash income (Table 12). That is, China's staff and workers have virtually no other sources of income. Income from hourly and piecework wages accounts for roughly two-thirds of all income. Therefore, in examining staff-and-worker living conditions we cannot look only at their wage levels; we must take into account total income. The proportion of bonuses in the wages of the various staff-and-worker categories is roughly the same, which means that egalitarianism in the area of bonuses is quite strong. Difficulty supplements account for a relatively greater propor- tion of the income of the below-20V group, while it is very small in the other groups. This means that the vast majority of staff and workers are able to maintain themselves on their income and do not require social relief or diffi- culty supplements. As staff-and-worker income increases, the proportion of difficulty supplements will tend to decline, and this corresponds to general laws. - 18 - lable 12: Staff-end-Wbrker Cash Income Structure Beijing, 1982 (percent) Per capita Average per capita mothly livirg-expnse ince mxothly Blelow 2D- 25- 30- 35- 40- 45- 50- Over Itan average 20 V 25Y 30Y 35Y 44X 45Y 50Y 601 61X Tbtal staff-and-wrker 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 cash inc I. Wge inei 92.0 83.0 91.8 92.7 92.8 92.7 91.9 91.9 91.9 91.7 1. bwrly and 68.4 57.5 71.2 70.4 71.1 70.9 66.7 68.2 67.5 68.4 piecework 2. AU bomnses 11.6 12.1 10.2 10.1 10.4 10.7 12.6 11.9 11.7 11.5 3. Supplemntal 1.1 2.3 1.9 1.5 1.3 1.2 1.3 0.9 1.0 0.9 weges 4. Overtime wges 1.0 1.0 0.6 0.9 0.6 0.7 1.2 1.0 1.1 1.2 5. All subsidie 9.9 10.1 7.9 9.8 9.4 9.2 10.1 9.9 10.5 9.7 II. labor insurance arnl 6.8 16.4 7.6 6.6 6.3 6.3 7.0 6.9 7.2 6.3 welfare incame 1. Bath ard haircut 3.6 1.5 2.0 3.0 3.4 3.4 3.8 3.6 3.8 3.3 eqpenses 2. DLfficulty 0.2 12.1 3.5 1.4 0.4 0.3 0.1 0.1 0.1 0.1 supplents 3. Other labor 3.0 2.8 2.1 2.2 2.5 2.6 3.1 3.2 3.4 2.9 insurance and welfare III. Other irxane 1.2 0.6 0.6 0.7 0.9 1.0 1.1 1.2 0.9 2.1 - 19 - 4. Staff-and-Worker Income Structure by Sector of the National Economy A comparison of staff-and-worker cash income over various sectors of the national economy is presented in Table 13. We can see that average staff-and-worker cash income is highest in the communications, posts and telecommunications sector; it is lowest in government offices, trade and food service, and agriculture and forestry. Cash income in government offices is 18.5 percent lower than in communica- tions, posts and telecommunications. Average wages in those sectors where the intelligentsia are concentrated -- research, education and health - are essentially the same as in industry and capital construction. This means that the remuneration for mental labor is basically the same as for physical labor. Wage income in the communications, posts and telecommunications sector is so large primarily because it has the highest level of subsidies, corresponding to more than twice the level in the lowest sector. On the other hand, the reason wage income is so low in the agricultural sector is, first, that the hourly and piecework wages are low here, and, second, that supplemen- tal wages, overtime and subsidies are all quite low. 5. Average Per Capita Monthly Cash Expenditure Structure in Staff-and- Worker Households Tables 15 and 16 shQw that expenditures on commodities were 40.50Y in 1982, up from 14.26V in 1955. The proportion of expenditures spent on noncommodity purchases is declining. Ibbkl 13: FMxthly Per Cgpita 0h Inxcoe by Sector, Being, 1982 Erxwvic Sector Ariulture Omzmicatiss, FHxatixi Q - lmthly capital anmi posts ardi Trade, Public al ment iten aver Inustry coEtrwtim forestry ta1eaumzdcations food service utilities Besearch balth offices 88.46 89.36 92.71 87.34 96.30 86.00 87.24 89.65 92.02 78.51 LI e frr1TI 81.29 83.11 83.04 76.38 90.34 77.40 78.88 83.36 83.68 73.24 1. turrly and 58-70 57.33 57.43 51.93 61.03 53.15 54.37 67.15 67.06 61.87 2. All bxunius 11.44 13.31 15.74 17.26 13.15 14.29 11.08 9.21 5.73 5.26 3. Suiluita1 1.19 1.68 1.57 0.84 1.87 1.28 1.47 0.02 0.13 0.23 4. oerxtiwe waes 1.04 0.79 2.89 0.02 1.30 1.90 1.41 0.10 1.01 0.11 5. All sAmidies 8.92 10.00 5.41 6.33 12.99 6.78 10.55 6.88 9.75 5.77 II. labor fnsuranEe and 5.86 5.46 9.58 8.10 5.52 7.16 8.09 4.83 4.12 3.96 welIare inrm 1. Claning 2.96 2.60 4.51 5.00 2.66 4.41 5.01 2.27 1.77 1.31 2. DLfficulty 0.27 0.26 0.38 0.29 0.41 0.07 0.41 0.25 0.17 0.43 3. OCier labor 2.63 2.60 4.69 2.81 2.45 2.68 2.67 2.31 2.18 2.22 insurance ald welfare III. Otier Inse 1.31 0.79 0.09 2.86 0.44 1.44 0.27 1.46 4.22 1.31 Tab] 14: Staff-arWirker C3sh Inrme Strwture by Fccuic Setor (Percent) - --xmc Sector -Acu1ture 0mnmications, Nertio- Capital an posts and Tade, Public and ment Itan hMrage lxhstry acrutncxtio forestry te1cmmlmtions food service utities Rearrh hBalth offices at.al Staff-_ -fler 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 100.0 cah i I. Ihge l 92.0 93.0 89.6 87.5 93.8 90.0 90.4 93.0 90.9 93.3 1. Ihrly and 66.4 64.2 61.9 59.5 63.4 61.8 62.3 74.9 72.9 78.8 2. An bcCxmsB 12.9 14.9 17.0 19.8 13.7 16.5 12.7 10.3 6.2 6.7 3. Savplaxental 1.3 1.9 1.7 0.9 1.9 1.5 1.7 - 0.1 0.3 4. overtime meges 1.2 0.9 3.1 - 1.3 2.2 1.6 0.1 1.1 0.1 5. All subs1dies 10.2 11.2 5.8 7.2 13.5 7.9 12.1 7.7 10.6 7.3 Ia. labor insuraore and 6.6 6.1 10.3 9.2 5.7 8.3 9.3 5.4 4.5 5.0 wel iaor e 1. Bath and baircLut 3.3 2.9 4.8 5.7 2.8 5.1 5.7 2.5 1.9 1.7 e .ns 2. IkLfflculty 0.3 0.3 0.4 0.2 0.4 0.1 0.5 0.3 0.2 0.5 3. Otber labor 3.0 2.9 5.1 3.3 2.5 3.1 3.1 2.6 2.4 2.8 insurance ad welfare III. Otber irrame 1.4 0.9 0.1 3.3 0.5 1.7 0.3 1.6 4.6 1.7 - 22 - Table 15: Average Per Capita Monthly Cash Expenditures (Unit: M) Item 1955 1957 1965 1981 1982 Total cash expenditures 21.04 27.27 26.26 51.50 55.17 Expenditures to purchase commodities 14.26 16.53 17.36 38.65 40.50 Noncommodity expenditures 2.86 3.35 3.19 3.98 4.07 Support and gift expenditures 0.66 2.11 1.29 3.42 3.51 Savings and loan expenditures 3.26 5.28 4.41 5.45 7.07 Other expenditures - - - - 1.46 Table 16: Changes in Average Per Capita Monthly Cash Expenditures (Percent) Item 1955 1957 1965 1981 1982 Total cash expenditures 100.0 100.0 100.0 100.0 100.0 Expenditures to purchase commodities 67.8 60.6 66.1 75.0 73.4 Noncommodity expenditures 13.6 12.3 12.2 7.8 7.4 Support and gift expenditures 3.1 7.7 4.9 6.6 6.4 Savings and loan expenditures 15.5 19.4 16.8 10.6 12.8 Other expenditures - - - - 2.6 6. Primary Commodity Consumption Structure of Staff-and-Worker Households with Different Income Levels Detailed data on expenditures from commodities are presented in Table 17. - 23 - 1hble 17: Average Monxthly Prilmary COmmdity Purchases Per Capita in Beijing Staff-and-Worker Hkuseholds, 1982 (1) Iten Average per capita mxaithly living-exqpase incme (all in Jin unless Mbxthly Below 20- 25- 30- 35- 40- 45- 50- over otherwise specified) average 20 V 25Y 30Y 35Y 40Y 45Y 50Y 60Y 6C0 Cereals 25.80 32.71 25.32 23.57 24.89 23.85 24.93 26.12 25.94 29.99 Rice 7.16 6.51 7.06 5.78 7.39 6.96 6.94 7.16 6.86 8.51 Flour 14.73 2D.61 14.28 14.16 13.93 13.34 14.25 15.20 15.05 16.69 !4iYMsceUain3s cereals 0.95 2.90 1.35 1.23 0.95 0.83 0.99 0.90 0.85 1.08 Raw and prepared staples 2.96 2.69 2.63 2.40 2.62 2.72 2.75 2.86 3.18 3.71 Pork 2.80 0.84 1.44 1.56 2.05 2.32 2.55 2.89 3.31 3.94 Beef, lamb, goat 0.63 0.15 0.29 0.40 0.48 0.47 0.64 0.63 0.78 0.82 Animal oils 0.29 0.08 0.09 0.15 0.18 0.23 0.26 0.31 0.36 0.43 Pwltry 0.27 0.02 0.03 0.08 0.15 0.19 0.20 0.28 0.34 0.50 EBg 1.28 0.38 0.51 0.76 0.88 1.11 1.24 1.38 1.48 1.68 Fish, suriMp 1.11 0.30 0.45 0.65 0.81 0.94 1.04 1.10 1.26 1.59 Fresh vegetables 30.01 31.46 28.03 25.96 24.85 27.09 28.54 29.74 32.60 36.82 Vegetable oils 1.13 0.68 0.79 0.82 0.94 1.01 1.04 1.12 1.23 1.55 U8gar 0.70 0.28 0.51 0.54 0.62 0.65 0.67 0.69 0.76 0.85 Cigarettes (paCks) 2.48 0.79 1.93 1.48 1.63 1.52 2.32 2.56 3.23 3.70 leaf tobacco, cut tobacco 0.02 0.04 0.03 0.02 0.01 0.02 0.01 0.02 0.03 0.04 Distilled spirits 0.30 0.14 0.17 0.23 0.22 0.19 0.31 0.28 0.37 0.43 Other alcohol 0.55 0.04 0.11 0.25 0.32 0.38 0.52 0.56 0.69 0.90 Tea (liang) 0.51 0.30 0.45 0.32 0.38 0.40 0.46 0.52 0.60 0.72 Pastries 0.92 0.17 0.22 0.49 0.67 0.76 0.93 0.99 1.07 1.18 Candy 0.27 0.03 0.19 0.19 0.24 0.21 0.23 0.26 0.31 0.39 Fresh frults 3.54 0.56 1.06 2.06 2.36 2.85 3.46 3.72 3.99 4.98 Fresh melons 2.36 0.22 0.78 1.23 1.66 1.79 2.37 2.37 2.78 3.46 Fresh milk 1.89 - 0.84 1.15 1.37 1.70 2.2D 1.88 2.25 1.90 FOWdered milk 0.06 0.01 0.01 0.02 0.04 0.04 0.05 0.07 0.07 0.08 COtton cloth (Chi) 0.83 0.71 0.96 0.75 0.69 0.72 0.80 0.81 0.86 1.10 Polyester-cotton cloth 0.09 0.08 0.06 0.06 0.07 0.10 0.08 0.08 0.10 0.11 (meters) Other chemical fiber 0.07 0.01 0.02 0.03 0.05 0.07 0.06 0.06 0.07 0.12 cloth (meters) WODlens (meters) 0.02 - 0.01 0.01 0.01 0.01 0.01 0.02 0.02 0.04 Silks and satins (meters) 0.03 - 0.01 0.01 0.02 0.02 0.02 0.03 0.04 0.06 CoDtton clothing 0.05 0.03 0.04 0.05 0.05 0.06 0.06 0.05 0.06 0.05 (artiCleS) amical-fiber clothing 0.09 0.03 0.02 0.06 0.07 0.08 0.09 0.10 0.10 0.11 (articleS) Nylon clothing (articles) 0.02 - 0.01 - 0.01 0.01 0.01 0.02 0.02 0.03 Silk-satin clothing _ _ _ - - - - - 0.01 0.01 (articles) - 24 - Ta1 - 17 (cAntinid) Iten Average per capita iMitbhly livirg-ecpese incme (all in Jin unless Mxthly Below 2)- 25- 3D- 35- 40- 45- 50- Oer otherwise specified) average 20 V 25Y 3OY 359 40M 45Y 5CY 60V 60Y Wml yarn and woolen 0.03 0.01 0.01 0.01 0.03 0.02 0.03 0.03 0.04 0.05 clothirg lditwear (articles) 0.13 0.06 0.06 0.11 0.12 0.12 0.13 0.14 0.13 0.15 Socks and stockIngs 0.10 0.06 0.03 0.08 0.19 0.10 0.10 0.10 0.10 0.11 (pairs) RIibber shoes (pairs) 0.02 0.03 0.01 0.02 0.02 0.02 0.02 0.02 0.01 0.01 Cloth shoes (pairs) 0.07 0.13 0.06 0.07 0.07 0.07 O.0B 0.07 0.06 0.06 AU-plastic shoes (pairs) 0.03 0.05 0.02 0.02 0.03 0.03 0.03 0.03 0.03 0.03 latber shoes (pairs) 0.04 - 0.01 0.02 0.02 0.03 0.03 0.04 0.05 0.05 ool. blarkets (articles) - - - - - - - - - 0.01 Wool scarves (articles) 0.04 0.03 0.04 0.04 0.03 0.04 0.04 0.04 0.04 0.05 Bath soap (bars) 0.62 0.71 0.78 0.75 0.65 0.64 0.56 0.63 0.56 0.68 Scented or medicinal soap 0.19 0.08 0.14 0.11 0.15 0.16 0.19 0.23 0.21 0.27 (bars) Iardry detergent 0.21 0.27 0.17 0.17 0.19 0.19 0.18 0.22 0.24 0.25 lbothpaste (tubes) 0.20 0.16 0.13 0.13 0.17 0.18 0.18 0.22 0.21 0.25 TIermos bottles 0.01 - - - - 0.01 0.01 0.01 0.01 0.01 (articles) Aluminum pots (articles) 0.01 - - - 0.01 0.01 0.01 0.01 0.01 0.01 Coal 30.51 61.73 33.20 33.45 29.86 26.45 30.78 29.85 33.36 30.25 1G (kg) 1.85 0.56 0.61 1.48 1.87 1.70 1.75 1.83 1.96 2.35 The above data show that people's income levels have a clear impact on their consumption structure. As income level rises, consumption of all consumer goods (with the exception of cereals, coal, cloth shoes, plastic shoes and bath soap) rises in varying degrees. The gap in consumption of major consumer goods between the lowest and the highest income groups is gen- erally about 100 percent but can reach 300 percent, 400 percent, and in the maximum case 2,000 percent. For example, for beef and lamb consumption, the - 25 - gap is 500 percent; for poultry, 2,500 percent; for candy, 1,300 percent; for fresh melons, 1,500 percent, and so on. The standard of living of people in different income groups still varies considerably. In terms of foodstuffs, consumption of rice, flour, and miscellaneous cereals does not change with income levels but remains basically stable at a fixed level. This may be related to fixed-quantity supply. Consumption of raw and prepared staples increases with income, which means that if income increases the amount of prepared foods people buy in dining halls and restaurants will increase. Consumption of pork, beef and lamb, animal oils, meat products, poultry, eggs, fish and shrimp, fresh vegetables, soybean products, vegetable oils, sugar, cigarettes, distilled spirits, tea, pastry, candy, fresh fruits, milk powder, and canned goods increases with income. The gap between high and low income levels is generally around 100 percent, but may reach 2,000 percent in extreme cases. As for clothing, consumption of cotton cloth, polyester-cotton cloth, cotton clothing, cloth shoes, etc., does not change with income; but consumption of woolens, silks and satins, chemical fiber clothing, woolen clothing, silk-satin clothing, wool yarn, woolen clothing, knitwear, leather shoes, blankets, and so on does increase with income. The gap between high and low income levels may be several hundred percent, as much as 1,000+ percent. As for daily-use articles, consumption of bath soap, laundry deter- gent, thermos bottles and so on remains relatively stable and basically the same regardless of income level. Consumption of scented and medicinal soaps, cosmetics, and athletic gear increases with staff-and-worker income. In the area of fuels, coal consumption remains relatively stable in all income groups, but consumption of LPG increases with income. - 26 - Given these monthly expenditures, we calculated the direct impact of 1981-1982 price changes on the levels of living of staff and workers (Table 18). Table 18: ChEnges in Retail Prices of Major CamDdities and Tteir Imipact on Living Standards, Beijing, 1982 1982 average Iten and Unit If 1981 per capita Effect of (all in Jin unless 1981 1982 is 100, innthly piurchase price change on otherwise specified) price price 1982 is: eycediture (V) living standard (V) Cereals 0.2D 0.2D 100.0 5.20 Rice 0.19 0.19 100.0 1.34 Wheat flour 0.19 0.19 100.0 2.86 MYscllaneous cereals 0.16 0.16 100.0 0.15 Raw and prepared staples 0.28 0.29 103.57 0.85 -0.03 Pork 1.18 1.22 103.39 3.41 40.11 Beef and lamb 1.00 0.98 98.00 0.62 -0.01 A-inn, oils 1.08 1.00 92.60 0.29 -0.02 Poultry 1.42 1.26 88.73 0.34 -0.04 FEgs 1.15 1.16 100.86 1.48 40.01 Fish and shrimp 0.57 0.58 101.75 0.64 -0.01 Fresh vegetables 0.09 0.09 100.0 2.78 Vegetable oils 1.05 1.04 99.05 1.18 -0.01 Sugar 0.83 0.86 103.61 0.60 40.02 Cigarettes (packs) 0.37 0.46 124.32 1.13 40.22 Leaf tobacco, cut tobacco 1.00 1.50 150.0 0.03 -$0.01 Alcobal 1.24 2.0 161.29 0.60 40.23 'Ia (hang) 0.56 0.59 105.36 0.30 -0.02 Pastry 0.85 0.96 112.94 0.88 40.10 Candy 1.19 1.52 127.73 0.41 40.09 Melons and fruits 0.27 0.26 96.29 1.54 -0.06 Fresh milk 0.29 0.30 103.45 0.57 40.02 Milk powder 2.50 2.67 106.8 0.16 40.01 Cotton cloth (chi) 0.50 0.51 102.0 0.43 40.01 Polyester-cotton cloth 4.73 3.22 73.68 0.29 -0.10 (meters) Wbolens (meters) 15.50 15.50 100.0 0.31 _ Silks and satins (meters) 3.75 3.67 97.87 0.11 -0.01 CDtton clothing 5.80 5.20 89.66 0.26 -0.03 (articles) Chemical-fiber clothing 14.00 12.33 88.07 1.11 -0.15 (articles) - 27 - Table 18 (continied) 1982 average Item and Unit If 1981 per capita Effect of (all in jin unless 1981 1982 is 100, monthly purchase price change cm otherwise specified) price price 1982 is: expenditure (Y) living standard (V) Nylon clothing (articles) 31.50 30.0 95.23 0.60 -0.03 Silk-satin clothing 7.00 - - 0.06 - (articles) Wbolen clothing (articles) 17.00 16.33 96.06 0.49 -0.02 Knitwear (articles) 3.O8 3.0 97.40 0.39 -0.01 Wbol scarves (articles) 1.50 1.25 83.33 0.05 -0.01 Socks and stockixgs 1.78 1.50 84.27 0.15 -0.03 (pairs) Rbbter shoes (pairs) 5.0 3.50 70.0 0.07 -0.03 Cloth shoes (pairs) 4.0 3.57 89.25 0.25 -0.03 AU-plastic shoes (pairs) 3.0 3.0 100.0 0.09 _ iLather shoes (pairs) 16.50 14.50 87.88 0.58 -0.08 Bath soap (bars) 0.23 0.23 100.0 0.14 - Scented and medicinal 0.47 0.47 100.0 0.09 soap (bars) landry detergent 0.60 0.62 103.33 0.13 40.01 Toothpaste (tubes) 0.58 0.60 103.45 0.12 -0.01 lbermos bottles 3.0 3.0 100.0 0.03 (articles) Alu1iium pots (articles) 6.0 5.0 83.33 0.03 -0.01 Pens (articles) 1.0 - - _ _ Oml, coal briquets 0.01 0.015 - 0.46 Brwyccmb briquets 0.02 - - LPG (kg) 0.18 0.18 100 0.33 Of the 44 different types of commodities above, 10 underwent no price change, 16 rose in price, and 18 dropped in price. The rise in 16 commodity prices caused an increase of approximately 0.91Y in per capita monthly expenditure. The drop in 18 commodity prices caused a decrease of approxi- mately 0.68V in per capita monthly expenditure. This leaves a net effect of an increase of approximately 0.23V in per capita monthly expenditure due to - 28 - the price effect. This expenditure accounts for 0.49 percent of living- expense income. Per capita monthly living-expense income rose from 42.851 in 1981 to 46.751 in 1982, a net increase of 3.90M. This shows that a rise in retail prices has not affected improvements in people's standard of living. 7. Changes Over Time in Consumpton of Major Commodities by Beijing Staff- and-Worker Households (Changes in Average Per Capita Annual Consumption) In Table 19 we show how consumption patterns changed over time. The data show that in the 20-plus years between 1955 and 1982 the amount of cereals consumed in Beijing staff-and-worker households declined by 11 per- cent, with wheat flour consumption declining by 19.5 percent, rice consumption by 7 percent, and miscellaneous cereals consumption by about 90 percent. At the same time as consumption of staples was declining, there was a marked rise in all non-staples; those rising most include: alcohol, 1,117 percent, poultry 1,080 percent, fresh fruits 520 percent, pork 460 percent, fish 250 percent, eggs 230 percent, candy 250 percent, pastry 230 percent. This shows a marked improvement in the people's standard of living. In the area of clothing, cotton clothing, rubber shoes, and leather shoes showed small increases, while woolens, wool yarn, woolen clothing and other top-grade consumer goods showed marked increases, which demonstrates an improvement in the clothing quality of Beijing staff-and-worker households. In the area of articles for everyday use, all except bath soap showed a marked increase. Bath soap declined because of the increase in laundry detergent, which not only did not affect the people's standard of living but on the contrary saved on oils, fats and other raw materials for the State. - 29 - Mble 19: Qauges Oyer Time in Houisehld CQxwumtiin, BeiJing Iten and Unit (all in jin unless If 1955 is otherwise specified) 1955 1957 1965 1981 1982 100, 1982 is: Cereasls 347.3 384.9 336.5 316.4 309.6 89.1 Rice 92.5 121.9 99.3 90.1 86.0 93.0 WIhat flour 147.9 130.6 192.3 178.6 176.7 119.5 Miselr;lan Eo cereals 106.9 132.3 44.9 11.8 11.5 10.8 Raw and prepared staples - - - 36.0 35.4 - Pbrk 6.0 12.6 16.4 34.9 33.6 560.0 Beef and lamb 4.3 3.6 2.5 6.0 7.6 176.7 hilzIml oils 2.00 1.4 2.8 4.3 3.5 175.0 Poultry 0.27 1.4 1.5 2.3 3.2 1,182.2 Eggs 4.6 6.1 6.2 12.8 15.4 334.8 Fish 3.8 6.5 5.8 13.6 13.3 350.0 Fresh vegetables 279.0 246.0 259.6 329.3 360.1 129.1 Vegetable oils 10.0 9.2 6.3 11.0 13.6 136.0 Sugar 2.6 3.3 3.6 7.1 8.4 323.1 Cigarettes (packs) 16.2 23.9 22.6 34.3 29.8 184.0 Iaf tobacco, cut tobacco 0.3 0.2 0.1 0.36 0.26 86.7 Alcobol 0.8 2.1 1.3 9.0 10.2 1,275.0 Taa (liar%) 6.2 7.6 3.0 7.1 6.1 98.3 Pastry 3.3 3.3 4.6 10.6 11.0 333.3 Candy 0.9 1.8 0.9 4.4 3.2 355.6 Fresh fruits 11.4 7.8 6.6 56.9 70.8 621.1 Cotton cloth (chi) 25.9 18.4 10.9 12.4 10.0 38.6 aCi1cal-fiber cloth (meters) - 0.01 0.38 1.8 1.9 Wooen (meters) O.04 0.11 0.24 0.20 500.0 Silks and satins (meters) - 0.12 - 0.48 0.35 - (Dtton clothing (article) 0.60 1.08 0.9 0.6 0.65 108.3 aenical-fiber clothing (articles) - - - 0.96 1.11 IDlen clothing (articles) - 0.02 - 0.24 0.21 Sk-satin clothing (articles) 0.05 0.12 0.05 Underwar (articles) 0.72 Cotton and woolen jackets 0.02 and trousers (articles) &tshirts, pants (articles) 0.12 Scarves (articles) 0.10 0.39 0.48 0.51 510.0 (cnt'd) - 30 - Table 19 (contimued) Item and Unit (all in jin umless If 1955 is otberwise specified) 1955 1957 1965 1981 1982 100, 1982 is: Ylamel and woolen clothirg 0.11 0.02 0.36 0.41 372.7 Flannel socks (pairs) 1.44 2.27 Nylon stockings (pairs) 1.08 1.18 Rbber shoes (pairs) 0.12 0.24 0.20 0.12 0.20 166.7 Cloth shoes (pairs) 0.36 - 0.85 0.72 0.84 233.3 AU-plastic shoes (pairs) 0.12 0.36 0.35 Leather shoes (pairs) 0.24 0.12 0.48 0.47 Bath soap (bars) 12.5 12.9 6.2 8.2 7.4 59.2 Scented and medicinal soap (bars) 0.6 0.7 1.8 2.3 2.3 383.3 landry detergent 2.4 2.5 lbothpaste (tubes) 0.48 2.3 2.4 500.0 Miermos bottles (articles) 0.02 0.12 0.08 400.0 Coal, briquets 744.0 735.0 118.4 366.1 IPG (kg) 21.8 22.2 Table 20 demonstrates that purchases of durable consumer goods increase with increases in people's incomes. This trend shows up in virtually every category of durabler goods. In households with per capita monthly living-expense incomes of under 25V, the income is only enough to maintain day-to-day life, and they are unable to afford durable consumer goods. In households with per capita monthly living-expense incomes over 60V, purchases of durable consumer goods are highest, and they are moreover able to afford all categories of top-grade consumer goods. This means that if China is to popularize such durable consumer goods as color televisions, refrigerators, tape recorders, cameras and so on, in addition to other conditions we must make certain that people's incomes reach the over-60Y level. If there is no clear increase in people's incomes, it will be difficult to achieve the popularization of durable consumer goods. - 31 - Mable 20: Total Purchases of Durable C,nuuer Goods by 1,200 Staff-end-4rker blvolds, Beijing, 1982 hAerage per capita monthly living-egense income item Aole Below 20- 25- 30- 35- 40- 45- 50- Over Year 20 Y 25Y 3(X 35Y 40Y 45Y 50 60( 60Y Beycles 220 4 11 28 49 38 45 45 Sewirgmachines 23 1 3 5 3 6 5 IWtches 162 2 8 22 35 23 38 34 CLocks 93 4 11 12 22 11 13 20 Electric fans 162 2 7 17 29 35 40 32 Wsirgmachines 68 2 1 10 25 6 13 11 Refrigerators 17 1 2 2 6 6 lirniture (pieces) 994 8 11 51 160 154 131 255 224 Irliuding: wrdrobes 50 1 3 5 7 2 18 14 Sofas 41 1 5 6 7 12 10 Desks 64 2 6 8 7 10 16 15 Sprirg beds 13 1 1 4 5 2 Valve radio sets 8 1 0 1 2 4 Transistor radios 122 5 22 24 19 28 24 COlor televisions 2 2 B&W televisions 67 2 3 12 10 18 10 12 Tapeplayers 73 1 6 11 10 28 17 C(ras 45 3 12 7 10 13 Record players 12 4 1 4 3 - 32 - C. Some Suggestions 1. Low Wages and High Employment Are Unfavorable for Family Planning and Population Control Average per capita monthly income in staff-and-worker households increases as family size decreases, increases as the average number of jobholders increases, and increases as the number of dependents per jobholder decreases (Table 21). This means that the number of jobholders is an important factor influencing household income and living standards. In order to improve living standards and raise income, it is obviously necessary to increase employment and increase the population. Now, in a given family, increasing the population and increasing income would seem to be a contradiction, but ultimately they are not. Therefore, the policy of low wages and high employment is certainly contradictory to the policy of family planning and population control. Table 21: Relationship Between Living-Exense xme and Empoyment in Beijing Staff-eadWaorr Hueholds, 1982 Average Nuber of Average Per Capita Per Capita Employed Avrage Average Nuber of HMbthly Monthly Nlber of Population Persons Poplation Nibuer Dexlents Living- Living xiseholds of Surey in Survey Per of Employed/ Per Employed Exense Expen Surveyed busebldbis lkiuseholds Hmlseklvd weholiold Person Incame Below 20 3 18 3 6.0 1 6 18.2 24 25V 11 65 19 5.9 1.7 3.4 23.7 25-3CY 40 200 82 5.0 2.1 2.4 28.1 30-35V 85 375 170 4.4 2.0 2.2 33.0 35-4CM 187 809 411 4.3 2.2 1.9 37.6 40-45V 236 986 531 4.2 2.3 1.7 42.6 45-50Y 198 711 442 3.6 2.2 1.6 47.4 50-6M 249 904 632 3.6 2.5 1.4 54.5 Over 6CY 191 649 525 3.4 2.7 1.2 69.3 Total 1,200 4,667 2,815 3.9 2.5 1.6 46.6 - 33 - 2. Income from Mental and Physical Labor is Essentially the Same; Such Egalitarianism Is Detrimental to Achieving Modernization Looking at the survey data, we find that in areas where the intelligentsia is concentrated, such as research, education, health, etc., average wages and other forms of staff-and-worker income are not at all higher and are in fact lower than in sectors such as capital construction, communications, posts and telecommunications, where those who labor physically are more concentrated. This situation is detrimental to mobilizing the enthusiasm of the intelligentsia. In the socialist phase of development, there is still a major difference between mental and physical labor; this difference must be recognized, and the appropriate policy measures must be taken. In this regard, severe mistakes were made in the past; this situation has begun to change in recent years, but a good many problems still remain, and we must continue to take certain measures. 3. Whether the Supply Quota System for Cereals, Cotton and Oils Can Be Eliminated Should Be Discussed It can be seen from the survey that in 1982 average monthly cereal consumption per capita in Beijing staff-and-worker households was 25.8 jin, which is below the cereal quota standards; annual consumption of cotton cloth was 15 chi, which is also lower than the supply quota standards; although the amount of edible animal oils per person per month is half a jin, there is no restriction on negotiated-price oils, which can be bought any time. This situation shows that the limiting effect of the supply quota system for cereals, cotton and oils is now very minor, yet it is encumbered with much bothersome red tape, and perhaps consideration should be given to eliminating this system. However, because there is a low-price policy for supply-quota - 34 - commodities, after eliminating the system it will be necessary to correctly resolve the problem of how to handle pricing. 4. Low Rents Are Detrimental to Resolving the Tight Housing Situation As described above, the limiting effect on people of the supply quota system is already very reduced. However, even though there is no quota system for urban housing, the situation is getting ever tighter and is restricting people's activities; the limiting effect created by this is even greater than that of the supply quota system. The basic means of solving the housing problem is obviously to construct more housing. However, where will the funds for housing construction come from? One important source is an increase in housing rent standards. Rents in China are now too low. In Beijing in 1982 among 1,200 staff-and-worker households, rent averaged only 0.761 per person per month, which is 1.6 percent of the living-expense income of 46.75M. Needless to say, this proportion is far below the levels in advanced countries, and it is even lower than in many Third World countries. For this reason it is feasible to raise this proportion. Because housing is something that is basically not paid for, getting housing essentially amounts to getting income from the State. This situation encourages some people to grab up many housing units, thus making the housing squeeze more acute. Furthermore, the survey data show that the proportion of living expenditures accounted for by rent in high-income households is not higher than in low-income households but is rather even lower, which means that housing allocation is currently benefitting high-income people but not low-income people. In order to resolve the problem of irrational housing allocation, it is necessary to use a graded housing approach, so that people - 35 - who take up more housing space should pay higher rents. We have applied such methods as excessively high prices, negotiated prices and supplemental prices to other commodities; why can't we implement some high rents to solve the housing problem? 5. Some Characteristics Expressed by Engel's Coefficient in China Generally speaking, Engel's coefficient (the proportion of income spent on food) reflects rising and falling consumption levels according to changes in consumption structure. A drop in this coefficient reflects an improvement in the standard of living, while an increase in it represents a deterioration in the standard of living. The standard of living of the Chinese people will pass through a series of stages: poverty (pre-Liberation), adequate food and clothing (1949- 1979), modest well-being (1980-2000), and prosperity (after 2000). In the early days after Liberation, when the standard of living of the Chinese people was passing from poverty to adequate food and clothing, the Engel coefficient rose rather than fell. This rise was not due to a drop in the standard of living, but rather to an improvement in living standards. After Liberation, the laboring people's income increased, and their first concern was to solve the problem of getting enough to eat; clothing and articles for everyday use were second in their minds. This necessarily took the form of a rise in the Engel coefficient. After the third plenary session of the Central Committee of the Eleventh Congress of the Party, the rate at which the standard of living of the Chinese people was increasing was greatly accelerated; in par- ticular, peasant living standards rose sharply. This situation was reflected in the consumption structure, and it did not cause the Engel coefficient to drop very much; in a good many villages and cities, it rose in varying - 36 - degrees. Here of course there is the factor of a fairly sharp rise in food prices, but this also reflects the fact that after the problem of getting enough to eat had been resolved there still remained a problem of eating well. Once the matter of eating well is put on the agenda, the Engel coefficient will not drop but on the contrary will rise. This situation may still arise in the future. Therefore, we must not rush to a mistaken judgment on the lopsided basis of a rise in Engel's coefficient. 6. We Must Vigorously Develop the Service Sector People's expenditures on various categories of services are very low; this is the result of the slow development in China of the service sector. In this regard, in comparison with many other countries, there is a significant gap, and we must take measures to solve this problem. - 37 - III. SOME ANALYSES OF CONSUMPTION STRUCTURE IN HUBEI PEASANT HOUSEHOLDS A sampling survey of Hubei peasant household incomes was begun in 1954; with the exception of the two-year "Great Leap Forward" (1959, 1960) and the eight years of the "Great Cultural Revolution" (1966-1973), the year-by- year data are fairly complete, enabling us to undertake some preliminary analyses of the consumption structure among Hubei peasant households. A. Structure of Hubei Peasant Survey Households 1. Basic Conditions in Peasant Survey Households From 1954 to 1982, the basic conditions among the Hubei peasant households surveyed may be said to have been essentially stable, with only slight changes: average household population fluctuated between 4.6 and 6.3 persons. Beginning in 1976, an annual declining trend appeared, reflecting the first visible results of China's population policy. Since 1976 the average labor force per household has been basically stable at 2.7, while the number of dependents per worker has gradually declined since 1977. This corresponds to a gradual improvement in the peasants' standard of living (see Table 22). The figures in the Table 22 tell us that from 1954 to 1982 the changes in the basic conditions of surveyed peasant households in Hubei were as follows: (1) from 1954 to 1975, household size showed a slow, gradual increase, from an average of 4.6 persons per household up to an average of 6.3. After the smashing of the "Gang of Four," with the implementation of a family planning policy and the encouragement of eugenics and one-child-only, late marriage and late child-bearing, household size gradually and slowly declined, from 6.3 to 5.7 people. (2) From 1957 to 1974, the average number - 38 - Table 22: Basic Conditions: Hubei Peasant Survey Households Average number Number of Average number of full- and Average number survey of people part-time workers of dependents Year households per household per household (1) per worker 19 5 4 700 4.8 -- 19 5 5 579 4.6 -- 19 5 6 565 4.6 - - 1 9 5 7 900 4.6 2.2 2.1 19 5 8 330 4.7 2.3 2.1 19 6 1 120 5.3 2.4 2.2 19 6 2 400 5.0 2.4 2.1 1 9 6 3 400 5.1 2.5 2.0 1 9 6 4 400 5.3 2.6 2.0 1 9 6 5 400 5.4 2.6 2.1 19 7 4 750 6.3 2.7 2.3 1 9 7 5 750 6.3 2.7 2.3 1 9 7 6 750 6.2 2.7 2.3 1 9 7 7 747 6.1 2.7 2.3 1 9 7 8 1,072 6.0 2.5 2.4 19 7 9 956 5.9 2.6 2.3 19 8 0 954 5.9 2.6 2.3 1 9 8 1 954 5.8 2.7 2.2 19 8 2 948 5.7 2.7 2.1 (1) Full- and part-time workers refers to males aged 16-60 and females aged 16-55 who are able bodied. of workers per household showed a rising trend, from 2.2 people to 2.7; subsequently, from 1974 to 1982, the number basically stabilized between 2.5 and 2.7 people. (3) The number of dependents per worker from 1957 to 1978 went from 2 up. to 2.4, but after 1978 there was a clear drop, from 2.4 to 2.1 in 1982. - 39 - 2. Structure of Survey Households, Population, and Labor Force by Net Income Data from 1981 and 1982 are clearly different from previous years. By grouping the survey households according to average net income, we are able to examine the structure of the survey households in the light of net income level. First, let us look at the structure of survey households with different net incomes. Total income refers to the income obtained through all productive and nonproductive agricultural operations over a given period of time (such as one year) which can be directly controlled; net income refers to what remains after subtracting all costs from total income for that year. Table 23: Structure of Hubei Peasant Survey by Net Income Level 1 9 8 1 1 9 8 2 Average net income per person Number of % of total Number of % of total households households Total 954 100 948 100 Below 100Y 28 2.9 1 0.1 100-150 149 15.6 28 2.9 150-200M 255 26.7 124 13.0 200-300M 374 39.2 420 44.3 300-400Y 114 11.9 241 25.4 400-500M 24 2.5 98 10.3 Total over 500M 10 1.0 36 3.8 Including: over 10001 3 0.3 - 40 - The data in Table 23 show that from 1981 to 1982 the proportion of households with net income of less than 2001 declined sharply, from 45.2 percent to 16 percent, while the proportion of households in the 200-300M groups rose from 51.1 percent to 69.7 percent. This means that the number of households in the middle and upper-middle income range increased. Households with a net income per person of 400-5001 also increased markedly, from 3.5 percent to 14.1 percent. Now let us look at the population structure of the survey households at different income levels. Tables 23 and 24 both reflect the same sort of trend: when the average net income per capita in Hubei peasant households rose from 232.64Y in 1981 to 311.50M in 1982, the proportion of population groups with annual net income per capita of less than 2001 dropped from 45.9 percent of the total population to 17.07 percent, while the proportion of the 200-300M and 300-4001 groups rose from 51.5 percent to 71.4 percent; at the same time, the 400-500M and over-500M groups rose from 2.3 percent to 11.4 percent. This means that the low-income population declined, while the higher-income population greatly increased. Finally, let us look at the proportion of laborers in the peasant survey households at different income levels in the total labor force. The figures in Table 25 tell us that the proportion of lower-net- income laborers in the total labor force decreased, while more laborers entered the higher-income rankings. Laborers in the under-200M group went from 43 percent of the total to 14.63 percent, while the over-2001 proportion went from 57 percent up to 83.37 percent. - 41 - Table 24: Structure of Hubei Peasant Household Population by Net Income 1981 1982 Net income per capita Number of % of total Number of % of total people people Total 5,497 100.00 5,364 100.00 Under 1001 175 3.1 4 0.07 100-1501 899 16.3 70 3.2 150-200Y 1,461 26.5 739 13.8 200-300M 2,260 41.1 2,525 47.1 300-400Y 570 10.4 1,313 24.5 400-5001 105 1.9 473 8.8 Total over 5001 27 0.4 140 2.6 Including: over 1,0001 5 0.09 -- Table 25: Structure of Hubei Peasant Household Labor Force by Net Income 1981 1982 Net income per capita Number % of total Number % of total Total 2,543 100.0 2,584 100.0 Under 1001 58 2.3 1 0.03 100-1501 372 14.6 60 2.3 150-2001 690 27.1 317 12.3 200-3001 1,046 41.1 1,168 45.2 300-400M 295 11.6 683 26.4 400-5001 62 2.4 261 10.1 Total over 5001 20 0.8 94 3.6 Including: over 1,0001 5 0.2 -- -- - 42 - The above data give a general picture of the basic structure of the Hubei peasant survey households in 1981 and 1982. To summarize, the structural characteristics of peasant households in Hubei are: (1) Households are large, generally averaging four or five people; (2) the interpersonal relationships are complex, with many households consisting of two married couples of different generations; (3) each household averages more than two laborers with slightly more than two dependents each (including the laborer himself); (4) more than 70 percent of the total survey-household population has an annual net income of 200-400Y. And the trend showing up in these two years: the number of households or population with net incomes of less than 200M is declining, while the number of households and population with net incomes over 200Y is increasing. B. Income Structure of Hubei Peasant Survey Households There are three sources of peasant household income. First, the income distributed by the basic accounting unit of the collective economy and the income received directly from the commune or brigade enterprise. Second, income from household sidelines, such as agricultural or forestry products they produce themselves, income from raising livestock or poultry, and income from household handicrafts. Third, other non-loan income, such as money sent from abroad or brought back, subsidies for attending meetings received from State financial agencies, disaster relief income, wages of staff and workers living in the household, and so on. Obviously there is a clear distinction between peasant household income and staff-and-worker household income, which is that not all peasant household income is part of the net income; it includes repayment costs and tax payments on the means of production for their economic operations. - 43 - 1. Hubei Peasant Household Total Income and Its Sources Taking 1982 as our example, the total income structure of 948 Hubei peasant households is as follows: Table 26: Total Income Structure of Hubei Peasant Survey Households, 1982 Amount (M) % of total Total household income for the year 1,670,921 100.0 Income received from the collective 868,154 51.9 Total income from household sidelines 673,886 40.3 Other non-loan income 128,881 7.7 The figures in Table 26 show that slightly more than half the total peasant household income is distributed by the collective. But looking at the substance of what is so distributed we see that there is a major difference with the past, which is that there is a very large proportion of income from individual assignments and production contracts. Here we see that income from these two sources accounts for 63.3 percent of the income distributed by the collective (57.2 percent + 6.1 percent), and 70.8 percent of the income distributed by the basic accounting unit (64 percent + 6.8 percent). This reflects the huge success of the agricultural production responsibility system. - 44 - Table 27: Structure of Hubei Peasant Household Income Received from the Collective Amount (Y) % of total Total income received from the collective 868,154 100.0 Distributed by the basic accounting unit 777,277 89.5 (100.0) Including: Actual contract income from responsibility assignments 497,252 57.2 (64.0) Actual over-quota income from the production contract 53,255 6.1 (6.8) Distributed non-contract income 226,770 26.1 (29.2) Received from commune or brigade enterprise 33,865 3.9 2. Sources of Net Per Capita Income in Hubei Peasant Households The structure of the income from the collective is as follows: if the 161.81 in net per capita income from the collective is 100 percent, then the net per capita income of 102.61 from the assignment and production contracts accounts for 63.4 percent. The figures in Tables 28 and 29 tell us that of the 1982 net per capita income of 286.11, 202.81 (102.6 + 100.2), or 70.9 percent, comes from the two contracts and household sidelines. 3. Relationship Between Hubei Peasant Household Size and Per Capita Total and Net Income As we have no comparable historical data, and with only one obser- vation in the 1982 data of a household at the lowest per capita net income level, we lack representativeness. For this reason, we can only do a cross- sectional analysis of the 1981 data. - 45 - 1mble 28: Net Per Capita Ixexai and Its Soirces in aibei Peasant S wvey Hxbjsholds, 1982 Total net incaie Average net incame Structure of for the year In per capita, populatim net per capita 948 h3useholds (Y) of 5,364 (1) incame (X) Tbtal 1,534,455 286.1 100 Includinqg: fran collective 868,154 161.8 56.6 Hbisehold sidelire 537,420 100.2 35.0 Nbn-loan incamt 128,881 24.1 8.4 TIal 29: Structure of Net Per Capita Income from the Cbllective, 1982 Net income Structure of net per capita (Y) income per capita (X: Armot of net per capita income from the collective 161.8 100.0 Tncluding: 1. Distributed by basic accounting unit 144.9 (42.3) 89.6 (26.1) Including: Actual income from assignmet contract 102.6 63.4 Over-quota inrcne fram production contract 2. Received directly fram oammune or brigade enterprise 6.3 3.9 Nbte: Figures in parentheses are the balance and proportion left after subtracting the contract inrames. - 46 - Table 30: Relationship Betwen Survey isdeod Size ard Net and Tlbtal Irhaoe Net income Net incane Net incane Net income Net imxu2 Net incme Net ircae per capita per capita per capita per capita per capita per capita per capita under 100 100-150 150-200 200-300 300-400 400-500 above 500 lHusebold size 6.25 6.03 5.73 6.04 5.00 4.37 2.7 (# of persons) Tbtal inxcm 99.44 140.52 201.08 249.41 364.51 468.61 766.44 per capita (Y) Net incxne 91.91 129.32 186.52 233.93 343.12 444.01 721.81 per capita (M) Table 31: labor Force Structure of Survey Households Net inccm Net income Net incaoe Net incane Net irrane Net incone Net incax e group: grop: group: group: group: group: group: Under 1OCY 100-150Q 150-200Y 200-300M 300-40CM 400-500M Over 500F Average labor 2.1 2.5 2.7 2.8 2.6 2.6 2.0 force per housebold Looking at the 1981 situation, we see that the smaller the household size the higher the per capita total and net income. Only in one group, with a net income per capita of 200-300, is the household size larger than in the 100-150Y and the 150-200Y groups, and its total and net income per capita are - 47 - higher than in the other two groups. Clearly this is due to the fact that in this type of household the labor force is larger. As for the household groups with small population and small labor force but a high income, it may be because they are engaged in some specialized production activity such as raising chickens, or they may be key households for grain and other crops, with a wealth of production experience, better production conditions, and a high degree of intensification, so that their incomes are higher. 4. Dynamic Analysis of Per Capita Net Income Levels in Hubei Peasant Survey Households In recent years the per capita net income levels of Hubei peasants have exceeded national levels; this was achieved after many ups and downs since the 1954 rural family livelihood survey. The two statistical tables included here very clearly show these changes. See Tables 31 and 32. During the First Five-Year Plan period, with the growth and completion of the farm cooperative organization movement, the proportion of income from the collective economy rose, total and net income per capita showed a growth trend. Only in 1956, guided by the "Left" ideology and with a decline in household sideline operations, did per capita income drop slightly. In 1958 began the Great Leap Forward and the organization of people's communes, which inappropriately changed the degree of socialization of the system of collective ownership and the administrative and management system; the emphasis was on making the communes "larger in size and more collective" and on "integrating government administration with commune management." The peasants' total and net income dropped sharply. With the absolute figures provided by Table 32, we can calculate the structural dynamics of the main sources of total and net income. - 48 - Able 32: Hubei Peasant Survey Huseholds by Per Capita lbtal anxd Net Irraie Unit: V 1. Net 2. Net 2. Incaue incae income 1. Ixncme fran 3. Other received fran 3. Net incmae I. Ibtal fran the household non-loan II. Net fran the h1 ehold fran other Year income collective side1ines incam incem collective sidelines mn-loan soarces 1 9 5 4 90.6 1.1 82.9 6.6 83.5 1.1 75.8 6.6 1 9 55 96.9 10.8 80.2 5.9 86.0 10.8 69.3 5.9 1 9 5 6 96.3 59.9 32.8 3.6 90.3 59.9 26.8 3.6 1 9 5 7 100.3 56.6 38.8 4.9 93.2 56.6 31.7 4.9 1 9 5 8 87.1 48.5 32.7 5.9 84.1 48.5 29.7 5.9 1 9 6 1 100.7 57.4 38.8 4.5 94.7 57.4 32.8 4.5 1 9 6 2 156.1 84.2 57.0 14.9 142.6 84.2 43.5 14.9 1 9 6 3 137.8 81.5 45.2 11.1 128.4 81.5 35.8 11.1 1 9 6 4 124.3 74.5 40.5 9.3 117.5 74.5 33.7 9.3 1 9 6 5 131.0 86.3 36.0 7.8 123.9 86.3 29.8 7.8 1 9 7 4 118.2 91.1 23.5 3.6 106.3 91.1 11.6 3.6 1 9 7 5 115.1 87.0 24.6 3.5 102.9 87.0 12.4 3.5 1 9 7 6 123.1 93.1 25.6 4.4 112.8 93.1 15.3 4.4 1 9 7 7 121.8 91.8 25.9 4.1 110.4 91.8 14.5 4.1 1 9 7 8 122.8 88.3 30.2 4.3 110.6 88.3 18.0 4.3 1 9 7 9 174.6 114.5 51.7 8.4 159.7 114.5 36.8 8.4 1 9 8 0 182.9 99.6 66.4 16.9 170.1 99.6 53.6 16.9 1 9 8 1 232.6 113.1 96.0 23.5 217.4 113.1 80.8 23.5 1 9 8 2 311.5 161.8 125.6 24.0 286.1 161.8 100.2 24.0 - 49 - Table 33: Structure of Sources of Per Capita Total and Net Incme in aibei Peasant Survey Hiseholds (%) 1. Net 2. Net 2. Inoe irncme inkcme 1. Received from 3. Other received fran 3. Net income I. Total fran the husehold nooan II. Net fran the hwsexila fran other Year inire collective sidelines incae incam collective sidelines nan-loan sources 1 9 5 4 100.0 1.2 91.5 7.3 100.0 1.3 90.8 7.9 1 9 5 5 100.0 11.1 82.8 6.1 100.0 12.5 80.6 6.9 1 9 5 6 100.0 62.2 34.1 3.7 100.0 66.3 29.7 4.0 1 9 5 7 100.0 56.4 38.7 4.9 100.0 60.7 34.0 5.3 1 9 5 8 100.0 55.7 37.5 6.8 100.0 57.7 35.3 7.0 1 9 6 1 100.0 57.0 38.5 4.5 100.0 60.6 34.6 4.8 1 9 6 2 100.0 53.9 36.5 9.5 100.0 59.0 30.5 10.5 1 9 6 3 100.0 59.1 32.8 8.1 100.0 63.5 27.9 8.6 1 9 6 4 100.0 59.9 32.6 7.5 100.0 63.4 28.7 7.9 1 9 6 5 100.0 65.9 27.5 6.0 100.0 69.7 24.1 6.2 1 9 7 4 100.0 77.1 19.9 3.0 100.0 85.7 10.9 3.4 1 9 7 5 100.0 75.6 21.4 3.0 100.0 84.5 12.1 3.4 1 9 7 6 100.0 75.6 20.8 3.6 100.0 82.5 13.6 3.9 1 9 7 7 100.0 75.4 21.3 3.3 100.0 83.2 13.1 3.7 1 9 7 8 100.0 71.9 24.6 3.5 100.0 79.8 16.3 3.9 1 9 7 9 100.0 65.6 29.6 4.8 100.0 71.7 23.0 5.3 1 9 8 0 100.0 54.5 36.3 9.2 100.0 58.6 31.5 9.9 1 9 8 1 100.0 48.6 41.3 10.1 100.0 52.0 37.2 10.8 1 9 8 2 100.0 51.9 40.3 7.7 100.0 56.6 35.0 8.4 From 1961 to 1965, an adjustment of the severe proportional imbalances in the national economy created by the "Great Leap Forward" was undertaken, which caused the peasants' per capita total and net income to undergo frequent swings. During the "Great Cultural Revolution," per capita income made no advances, with the overall levels being lower than during the "adjustment period." Two years before the smashing of the "Gang of Four," the situation was still fluctuating. - 50 - After the third plenary session of the Central Committee of the Eleventh Party Congress, there was a fundamental change in the situation; we have already analyzed this above. 5. Effect of Household Size on Per Capita Consumption and Consumption Structure Generally speaking, the larger the household size, the greater the per capita consumption level, and the higher the proportion of food in that structure. However, the situation reflected in the Hubei peasant survey households is not entirely like this. The figures in Table 34 clearly tell us that with a small number of people in the household per capita net income is also small. This is because there are fewer people working. In a household with a per capita net income under 100M, for example, even though there are four people in the household only one of them may be working, and he must support four people. Therefore, the average per capita annual consumption expenditure is only 92M in his family, and only 7.6+1. This amount also includes 2+V in State relief. Therefore the result may be that for peasants the more people, the more workers, the more income and expenditure, and the standard of living may be somewhat better; overall, this is true for households with a per capita net income of less than 200M. However, after the annual per capita net income exceeds 2001, the smaller the size of the household the greater the annual average consumption expenditure. In these relatively better-off households it is not that there are more laborers; generally one laborer supports slightly more than 2.6 people. In addition to the effect of family size on consumption expenditure, it also has an impact on the consumption structure. - 51 - Table 34: Relationship Between 9urvey Hxosehold Size and Cnainption Expenditure, 1982 Hlusehold Hbusehold Household Household Household Household Household with with with with with with with per capita per capita per capita per capita per capita per capita per capita net income net incame net income net incxome net income net income net income under 100X 100-150C 150-200Y 200-300Y 300X)-40( 400-500Y over 500Y I. Ho1usehold size 4.0 6.07 5.95 6.01 5.44 4.82 3.88 (persons/household) II. Thtal per capita 92.0 127.5 164.6 203.6 268.0 314.5 420.7 cnamsuption eaxpenditure (Y) Including: 1. Food (Q) 77.5 86.0 111.1 131.6 161.9 176.7 206.7 2. Clothing (V) 2.5 11.1 17.8 24.4 36.3 42.6 50.8 3. Fuel (1) 7.5 10.8 10.8 10.9 14.3 15.6 18.4 4. HDusirg (V) 0.5 8.8 8.4 14.1 21.4 33.1 89.2 5. Articles for 2.75 7.1 11.5 16.7 27.5 37.2 47.8 everyday use and other (Y) 6. Cultural and 1.25 3.6 4.9 5.9 6.7 9.3 7.8 life services (I) - 52 - Table 35: Relationship Between Household Size and Cwsuwption Structure, 1982 (%) Houehold HIbselld Household HLbsehold HIisehold H1uehold IHisehold with net with net with net with net with net with net with net per capita per capita per capita per capita per capita per capita per capita ha income ir coe incae income incame inc4re e under 10tF 100-150V L50-20(V 200-300V 300-40Cm 400-500{ ower 50(0 I. Household size 4.0 6.07 5.95 6.01 5.44 4.82 3.88 (persons/bsekold II. Thtal per capita 100.0 100.0 100.0 100.0 100.0 100.0 100.0 canstrctixn Ialuding: 1. Food 84.2 67.5 67.5 64.6 60.4 56.2 49.1 2. Clothirg 2.7 8.7 10.8 11.9 13.6 13.5 12.1 3. Ful 8.2 8.5 6.6 5.4 5.3 5.0 4.4 4. Hniusirg 0.5 6.9 5.1 6.9 8.0 10.5 21.2 5. Articles of 3.0 5.6 7.0 8.2 10.2 11.8 11.4 everyday use and other 6. Qultural and 1.4 2.9 3.0 2.9 2.5 2.9 1.9 life services - 53 - From the figures in Table 35 it can be seen that the impact of household size on consumption structure is directly related to per capita net income level. In the groups with per capita net income under 200Y, the smaller the household size, the greater the proportion of food, the smaller the proportion of clothing, the greater the proportion of fuel, the smaller the proportion of housing, and the smaller the proportion of cultural and life services. But in the groups with per capita net income over 200Y, the smaller the household size, the greater the proportion of food, with the proportion of clothing first rising slightly and then dropping slightly, and the lower the proportion of fuel; the proportion of housing rises sharply, the proportion of articles of everyday use remains steady, and cultural and life services drop somewhat. As for the causes producing this type of structure, if we look at the low-income households with annual per capita net income of less than 200, we find that these households generally have few laborers, with a high dependent coefficient. They must first ensure food needs, so that the proportions of food and fuel are higher. It is difficult for them to have any money left over to buy construction materials to build a new home, and in the areas of clothing and articles of everyday use all they can do is economize to get from one day to the next. In terms of households with annual per capita incomes over 200M, although these households have few people, their labor force is somewhat larger, relatively speaking, so that each laborer supports somewhat fewer people. They have no major problems in the area of food, so that as total income increases, the proportion going for food and fuel declines. They have money left over to build a home, and their situation in regard to clothing and articles of everyday use has improved. Therefore, in order to - 54 - improve their housing situation, they are willing to curb their spending on clothing and cultural and life services. This is representative of households with annual per capita net incomes of more than 500. 6. Preliminary Analysis of Consumption Structure at Different Net Income Levels in Hubei Peasant Survey Households Sections 3 and 4 above were a static examination of the consumption expenditure structures of Hubei peasant survey households from the viewpoint of a cross section of different net income levels and household sizes. The next section will provide some dynamic analyses of consumption structure according to systematic family livelihood data from the Hubei peasant survey households in chronological sequence. C. The Income-Expenditure Situation and Structure among Peasant Survey Households and Their Position Nationwide First, let us make some comparisons of the basic situation of the peasant survey households. Iable 36: Basic Situation of Peasant Hbuseholds in Hubei and Nationwide 1978 1979 1980 1981 1982 Unit Hibei Nationwide aHbei Nationwide Hubei Nationwide Hubei Nationwide Hubei Nationdde Number of survey houseohlds 1,072 6.095 956 10,282 954 15,914 954 18,529 948 22,775 Number of people regularly living in survey bousebold 6,453 34,961 5,647 15,153 5,586 88,090 5,497 101,998 5,364 124,286 Average number of people reguLr- ly living in survey household 6.01 5.74 5.9 5.66 5.85 5.54 5.76 5.50 5.66 5.46 Average anber of full- ard part- time laborers per household 2.5 2.27 2.6 2.38 2.6 2.45 2.67 2.53 2.72 2.58 Average number of persons supported by each laborer 2.4 2.53 2.3 2.38 2.3 2.26 2.16 2.17 2.07 2.12 Average number of rew rooms built per household during the year 0.2 0.11 0.2 0.22 0.2 0.23 0.25 0.27 0.20 0.29 Average number of roams used per lousehold at year's ed 3.2 3.64 3.3 3.84 3.5 4.06 3.6 4.28 3.68 4.56 Average area of roams in u_e per person at year's end (m2) 10.17 11.03 14.0 11.59 15.3 12.47 16.45 13.41 uble 37: Average Per Capita Net Ircome Level and Structure of Peasant HsehAolds, u.ibei arni Nationwide 1978 1979 1980 1981 1982 Hubei Nationwide aibei Nationwide Hiubei Natiaowide Hubei Nationwide Hubei Nationwide 1. Average per capita net inrime (C) 110.6 133.57 159.7 160.17 170.1 191.33 217.4 223.44 286.1 270.11 IiEludirg: (1) Incxse received fron the collective 88.3 88.53 114.5 101.97 99.6 108.37 113.1 116.20 161.8 140.12 (2) Net income frao hxisehold sideliries 18.0 35.79 36.8 44.0 53.6 62.55 80.8 84.52 100.2 102.80 (3) Other non-loan incxze 4.3 9.25 8.4 14.20 16.9 20.41 23.5 22.72 24.0 27.19 en 2. Proportions (Net ircone = 100a) (1) Net ixunai received fran tbe collective 79.8 66.28 71.7 63.66 58.6 56.64 52.0 52.00 56.6 51.87 (2) Net income fran housebold sidelires 16.3 26.79 23.0 27.47 31.5 32.69 37.2 37.83 35.0 38.06 (3) other non-loan inecme 3.9 6.93 5.3 8.87 9.9 10.67 10.8 10.17 8.4 10.07 - 57 - A comparison of the Hubei peasant survey households with those nationwide reveals the following characteristics: (1) The average number of people regularly living in each household is slightly higher than the average national level; although the number is declining both in Hubei and nationwide, the scale of the decline is somewhat greater in Hubei. (2) As for the average number of laborers in each household, the figure in Hubei is also somewhat higher than nationwide, although the figure in both cases is rising year by year. Moreover, the corresponding number of persons supported by each laborer is declining year by year; in 1982 in Hubei each laborer was supporting slightly more than two people, which is to say that one laborer was supporting one other person. (3) The average living space per person is increasing year by year; the area is greater in Hubei than in the nation as a whole, and the margin of increase is also higher in Hubei. In 1982 the average living area per person in Hubei was 16.45m2, slightly more than 3m2 greater than nationwide. Second, a comparison of average per capita net income in peasant survey households in Hubei and nationwide. The figures show that from 1978 to 1981 average per capita net income in Hubei was lower than nationwide; only in 1982 did Hubei surpass the national level, going to 106 percent of the national level. In 1982 the per capita net income of Hubei's peasants was more than 16V higher than national levels, primarily because their net income from the collective surpassed national levels by 21.68V. However, net income from household sidelines and from other non-loan sources was lower than the average national levels, by 2.6V and 3.19V, respectively. - 58 - In terms of structure, the proportion of per capita net income which Hubei peasants receive from the collective was consistently higher than the national level from 1978 to 1981, although in terms of overall trend Hubei is the same as the nation as a whole, with the proportion declining year by year. In 1982 the proportion received by peasants from the collective in Hubei exceeded the national level. Leaving out 1982, the proportion of net income from household sidelines rose consistently from 1978 to 1981, although it was always lower than nationwide. That means that the level of development of household sideline activities among Hubei peasants is lower than the average level for peasants nationwide. Third, let us again look at average per capita living-expense expenditure levels and structure among peasants in Hubei and nationwide. We shall first divide living-expense expenditures into two parts: consumer-goods expenditures and cultural and life services expenditures. Then we shall further divide consumer-goods expenditures into five categories: food, clothing, fuel, housing and articles of everyday use. In terms of consumer-goods expenditures, in 1979 and 1982 Hubei was slightly higher than the nation as a whole, while in 1980 and 1981 it was somewhat lower, although these expenditures rose consistently both in Hubei and nationwide. As for categories of expenditures, food expenditures in Hubei were consistently higher than national average levels, except for 1980 when Hubei was slightly lower than nationwide. In clothing expenditures, Hubei was higher than national levels in 1979 and 1982, and lower in 1980 and 1981. For fuel, Hubei was consistently lower than the national level except for 1979, when it was somewhat higher. For housing expenditures, although these increased annually both in Hubei and nationwide, Hubei was below the national - 59 - level. Expenditures and trends for articles of everyday use and other were the same as for housing expenditures. With the exception of 1979 when Hubei was somewhat lower than the national level, expenditures for cultural and life services were higher in Hubei than nationwide. In terms of the structure of these expenditures, consumer-goods accounted for more than 97 percent of all consumption expenditures, both nationwide and in Hubei, while expenditures for cultural and life services were less than 3 percent. Among consumer goods, food accounted for about 60 percent every year, but the trend was downward. In Hubei the food proportion went from 67.5 percent in 1979 to 61.9 percent in 1982. The corresponding food proportion nationwide dropped from 63.96 percent to 60.48 percent. With the exception of fuel, all the other categories -- clothing, housing, articles of everyday use and cultural and life services all show an upward trend. Iable 38: Average Per Capita Qmsmutimon EKqexnditure levels and Strurture in Peasant Survey fbiseholds, in aibei and Nati.mwide 1978 1979 1980 1981 1982 Hubei Nationwide Hubei Natiobdide Hubei Nationwide abbei Natiorwide Hudbei Nationwide I. Average per capita coasunptim exnditure (Y) 116.06 148.7 134.51 152.8 162.21 183.8 190.81 226.9 220.23 1. CtweerKgoods peqxiture 112.90 145.4 130.81 148.4 157.95 178.3 186.17 220.7 215.30 Food 78.59 100.4 86.03 98.3 100.19 114.7 113.83 140.6 133.20 Clothlng 14.74 19.2 17.64 17.7 19.99 22.6 23.57 28.3 24.77 Fuel 8.28 8.8 8.34 8.5 9.66 10.3 10.59 12.3 12.36 a, 1kusirg 3.67 6.8 7.66 11.6 12.80 15.6 18.67 18.5 22.58 Everyday articles and otber 7.62 10.2 11.14 12.2 15.31 15.0 19.51 20.9 22.39 2. OCltural and life services expenditure 3.16 3.3 3.70 4.4 4.26 5.5 4.64 6.2 4.93 II. Prtportion (ostlmr-er epxslditure = 100) 1. Cbxmer-goods experditure 97.28 97.8 97.25 97.1 97.37 97.0 97.57 97.3 97.76 Food 67.71 67.5 63.96 64.3 61.76 62.4 59.66 61.9 60.48 Clothdig 12.70 12.9 13.12 11.6 12.32 12.3 12.35 12.5 11.25 Fuel 7.14 5.9 6.20 5.6 5.96 5.6 5.55 5.4 5.61 H1uzing 3.16 4.6 5.69 7.6 7.89 8.5 9.79 8.2 10.25 Everyday articles and other 6.57 6.8 8.28 8.0 9.44 8.2 10.22 9.2 10.17 2. Culitural and life services es4xnditure 2.72 2.2 2.75 2.9 2.63 3.0 2.43 2.7 2.24 Mble 39: Proportion of (kxmodity lelitures in Peasant Casumr-Qiods Ependitures (acpenditure in each category = 100) 1978 1979 1980 1981 1982 Hbi Natioalde Hlbei Natiride Hlbei Natiamride Hue Natio id Hubei Nationie Conumer-goods epediture - 39.7 36.86 44.5 45.11 50.4 48.39 56.1 51.26 56.5 Food - 24.1 23.78 25.8 25.87 31.1 29.01 36.6 32.21 38.0 Clothing - 89.0 93.73 91.6 97.90 98.1 97.75 98.6 97.98 98.4 Fuel - 31.9 15.84 32.4 16.02 28.7 14.32 25.1 10.45 22.7 E,usirg - 95.1 - 95.6 95.87 88.8 95.83 94.2 97.09 96.1 &Eryday articles and other - 87.7 100.OD 88.9 95.75 96.3 96.68 98.2 99.34 98.7 - 62 - Fourth, comparing Hubei and the nation in terms of the proportion of commodities in the consumer goods purchased by peasant households, the overall expenditure situation for all categories of consumer goods is that the proportion of commodities in Hubei is consistently slightly lower than nationwide. This means that Hubei Province is somewhat less economically developed than China as a whole. In terms of the commodity expenditures in the various categories of consumer goods, the trend in Hubei is somewhat different from the country as a whole, which reflects Hubei's characteristics: First, the proportion of total consumer-goods expenditures in all household expenditures: although this has consistently increased both in Hubei and nationwide, the increase in Hubei has been faster. In Hubei it rose from 36.86 percent in 1979 to 51.26 percent in 1982, a 14.4 percent rise. During the same period, the proportion nationwide went from 44.5 percent to 56.5 percent, a 12 percent rise. This shows that although the starting point for the proportion of commodity expenditures in all categories of consumption expenditures by Hubei peasants was lower, the rise was faster. This means that the relationship between peasant consumption and the market in Hubei has tightened up, and the scale of the commodity economy has increased. Second, the commodity proportion of food: from 1979 to 1982, this rose in Hubei from 23.78 percent to 32.21 percent, a 12.2 percent hike. The increase in Hubei was somewhat slower than nationwide. This means that the commodity proportion of consumer-goods expenditures by Hubei peasants increased faster than nationwide; the main reason for this lies not with food, but with the other expenditure categories. - 63 - Third, clothing expenditures: from 1979 to 1982, the commodity proportion among Hubei peasants rose from 93.73 percent to 97.98 percent, while nationwide it went from 91.6 percent to 98.4 percent. That is, although the starting point for the commodity proportion of clothing expenditures in Hubei was higher than nationwide, the margin of increase was somewhat lower: in Hubei the margin of increase was 4.25 percent, while nationwide it was 6.8 percent. This makes it all the more clear that Hubei peasants rely on the market supply for virtually all their clothing. Fourth, the fuel situation is quite distinct. The commodity proportion of fuel costs both in Hubei and nationwide showed a downward trend, but the margin of decline for Hubei was -5.39 percent (10.45 - 15.84), while for the nation it was -9.7 percent (22.7 - 32.4). This means that in Hubei peasant reliance on the market for purchases of fuel did not change as much as in the nation as a whole. Fifth, housing costs: with the exception of minor water and electricity charges, all construction materials must be purchased on the market. This is true both nationwide and in Hubei; therefore the commodity proportion of housing-cost expenditures has been consistently very high since 1978. The proportion in Hubei is slightly higher than nationwide. The commodity proportion of articles of everyday use and other items has also been consistently very high, with Hubei near but slightly higher than the national level. In short, for the four years from 1979 to 1982, commodity expendi- tures for consumer goods by Hubei peasants have been higher than nationwide, primarily because the proportion of self-supplied food, clothing and fuel is greater than the average national levels, while for housing (mainly - 64 - construction materials) and articles of everyday use, Hubei peasants rely more heavily on market purchases. Overall, comparison of the various indicators shows that consumption income and expenditure levels and structure for Hubei's peasants are very close to the situation in the nationwide peasant survey households. The only difference is that Hubei's economic level is slightly below that of the rest of the country. For this reason it may be affirmed that the various data from the Hubei peasant survey households are highly representative of national levels. D. Some Summary Assessments and Conclusions From the various foregoing analyses, we can see overall that: 1. The present family livelihood situation among Hubei's peasants is fairly representative of the national agricultural and forestry situation. Average household size is five or six people, similar to the peasant family size in Mexico in 1968. The average household has two-plus laborers, with each laborer supporting an average of slightly more than two people. The living standards of most peasant families mean that they dress warmly and eat their fill. However, there are still families in difficult circumstances with annual per capita net incomes of less than 100Y, and this must be viewed seriously. 2. In terms of trends, up to 1975 peasant family size showed an upward trend, while after 1976 the trend was a slow downward one, the start of a tendency to go from large families to small ones. As China's family planning policy is more thoroughly implemented, this tendency will be strengthened. The impact of this change to smaller families on consumption demands cannot be underestimated. For example, people in larger families can share the use of durable consumer goods, while those in smaller families will have to purchase them on an individual basis. 3. Prosperous families among Hubei peasants are on the increase, while the number of families in difficult circumstances is declining. In the last two years, those with annual per capita net incomes of less than 200M accounted for 45.2 percent of the total, a drop of 16.0 percent; households with annual per capita net incomes in the 200-3001 and 300-400V groups accounted for 51.1 percent of the total, a rise of 69.7 percent; and households with annual per capita net incomes of 400-5001 and over accounted for 3.5 percent of the total, a rise of 14.1 percent. - 65 - 4. In the past we always believed that the more persons and the more laborers in China's peasant households the greater the income and the more prosperous the standard of living. The situation over the last few years shows that this is certainly not the case. The data from the Hubei peasant survey households tell us that in the groups with per capita net incomes from 100V to 300Y the household per capita net income in fact rises as the number of persons and laborers in the household increases; however, in the groups with per capita net incomes above 300M, income rises when the number of persons and laborers in the household decreases. This is a very noteworthy tendency: with the spread of the rural production responsibility system, as the rural commodity economy is enlivened and the peasants' scientific and cultural knowledge is increased, and with the development of scientific sowing and cultivation, small household size and a small number of laborers will be unimportant; high incomes can be achieved if the population's cultural, scientific and technical levels are high, or if they gain more experience in sowing and cultivation. 5. In terms of the composition of the sources of income of Hubei peasant households, since completion of the agricultural collectivation movement in 1956, most of the peasants' net income has come from the collectives. From 1956 to 1965, the portion coming from the collectives was about 60-70 percent, while household sidelines accounted for 20-30 percent. During the "Great Cultural Revolution," the "Gang of Four" was blowing a communist wind, "cutting off the tail of capitalism," and peasants were not allowed to engage in sideline activities, so the portion of net income from the collectives rose to over 80 percent. After the overthrow of the "Gang of Four," and in particular after the third plenary session of the Central Committee of the Eleventh Party Congress, with the dissemination of the agricultural production responsibility system and the implementation of the Party's various farm policies, peasant household sideline activities were renewed and developed. For this reason, the portion of the peasants' net income coming from the collective declined, while the proportion of income coming from household sidelines showed a clear upward trend. 6. Comparing the domestic consumption and production expenditures of Hubei peasant survey households, in the last two years (1981-1982), with the development of the production responsibility system and especially the double contract households, there has been a clear change. If total annual household expenditures are 100, then domestic consumption expenditures dropped from 86.4 percent to 82.1 percent, while production expenditures rose from 7.5 percent to 12.3 percent. Of production expenditures, the proportion of purchases of fixed assets rose from 0.4 percent to 3.1 percent. This means that economically the peasants are now able to acquire large-scale production implements and equipment like hand tractors, trucks, and so on. - 66 - 7. Of domestic consumption expenditures by Hubei peasants, consumer goods (commodity) expenditures account for more than 97.0 percent, while cultural and life services (labor services) account for less than 3 percent. This reflects the high proportion of self-supplied life services among China's peasant households, as well as the inadequate and life services facilities requiring expenditures. 8. Changes in the domestic consumption structure of Hubei's peasant households basically conform to Engel's law, although there are certain Chinese characteristics. First, looking at this in terms of a cross-section analysis: in the various household groups categorized by per capita net income level in 1982, as per capita net income rises, the proportion of food and fuel decline sharply; for example, the proportion of food went from 84.2 percent in households with less than 100Y to 49.1 percent in households with more than 500Y, the average being 64.2 percent; fuel went from 8.2 percent to 4.4 percent, averaging 6.2 percent. The proportions of clothing, housing and articles of everyday use rose sharply. Except for the under-1Q00 households, where cultural and life services accounted for 1.4 percent, in the various other groups it was basically stable at between 2 and 3 percent, averaging 2.5 percent. This basically conforms to Engel's law. Next, looking at this through a time-series analysis, we find: of domestic consumption expenditures, the food proportion went from 55.4 percent in 1954 up to 68.1 percent in 1965; data are lacking for 1966-1976; from 1979 to 1982, this proportion dropped from 67.5 percent to 61.9 percent. Obviously, in the first ten years after the founding of the People's Republic, as the economy developed and production increased, the proportion of food in domestic consumption did not fall at all but rather rose. This of course contradicts Engel's law. The reason for this lies in the fact that in pre- Liberation China the masses of peasants for a long time struggled along on the border of starvation; after the establishment of the people's political power, the first thing the people wanted was to eat their fill. This is a sort of compensatory food consumption. Engel's law does not apply here. This is perhaps a characteristic common to the early stages of development in - 67 - developing countries. Only after people have eaten their fill does the food proportion decline in accordance with Engel's law. 9. Material consumption among the Hubei peasant survey households also changed greatly. From 1978 to 1982, material consumption of basic consumer goods such as cereals, vegetables, oils, meat, poultry, eggs, fish, shrimp, sugar, tobacco, spirits and tea all increased; only per capita consumption of cotton cloth, woolen goods, silks and satins, and shoes decreased. Increasing the most were eggs, spirits and chemical-fiber cloth; next were edible oils, fish and shrimp, sugar and meat. What especially needs to be pointed out here is that in the last four years the use of durable consumer goods by peasant households has increased sharply. In more than 900 peasant survey households, there were only 66 bicycles in 1979, while in 1982 there were 234; the number of sewing machines rose from 91 to 214; the number of radios rose from 18 to 455; from 1978 to 1981 there were basically no televisions or tape recorders, but in 1982 there were 2. In addition, in 1981 and 1982 there were more than 1,000 pieces of furniture worth more than 50. This is the necessary result of peasants building more spacious houses. Although this tendency towards urbanization of peasant life is now only in its infancy, its impact on future changes in peasant consumption structure should be carefully examined right now. This market for durable consumer goods created by 800 million peasants is a tremendous force for stimulating development of China's industry. 10. Historically, the proportion of domestic consumer goods which China's peasants have supplied themselves has exceeded 50 percent; it has only been since 1979 that this has been declining. This directly reflects the process of growth and decline of the natural economy of China's agriculture and commodity production and exchange. The consumption situation among Hubei's peasants is like this. From 1954 to 1965, the proportion of domestic consumer goods which Hubei peasants supplied themselves fluctuated between 57 and 65 percent. Beginning in 1979, the proportion of self-supplied consumer goods gradually declined. For example: in 1979 63.14 percent of domestic consumer goods were self-supplied; in 1980, 54.89 percent; in 1981, 51.61 percent; and in 1982, 48.74 percent. Compared with the country as a whole, the self- supplied proportion is higher in Hubei. Correspondingly, the proportion of commodity consumption in Hubei is lower than in the country as a whole. The margin of difference is generally between 5 and 8 percent. This means that, overall, not only are the entire country's villages faced with the task of vigorously developing commodity production and exchange, but Hubei's peasants in particular must make a tremendous effort to develop a commodity economy. - 68 - IV. USING A STATISTICAL DATA APPROACH TO FORECAST CONSUMPTION STRUCTURE FOR A COMFORTABLE STANDARD OF LIVING In using a statistical data approach to forecast China's future trends in consumption structure, many comrades are using the consumption structure of other countries when the total national production value (or national income) is US $1,000 per capita to deduce China's consumption structure when the living standard of the Chinese people reaches a comfortable level. To forecast China's future consumption structure trends, there can be no doubt that we should have reference to the experience and data of other countries; however, we cannot treat items of reference as primary, or worse yet as the sole basis for our forecasts. In studying any question we must start with the situation in the country in question, and forecasting consumption structure is certainly no exception to this rule. We must make every effort to start from China's own situation, utilizing the statistical data we already have for China (primarily data from surveys of staff-and- worker households and peasant household family livelihood), along with suitable reference to some foreign data, to forecast consumption structure when the standard of living of the Chinese people reaches a comfortable level. Consumption structure is controlled by a great many factors, including production structure, production levels, people's consumption levels, consumption forms, consumption policy, retail prices, personal habits, consumption psychology, etc., all of which have a definite effect on changes in consumption structure. Therefore if we are to correctly forecast trends in consumption structure, we must do a concrete analysis of the various factors influencing this. We do not intend here to do a thorough going analysis of the impact of these various factors on consumption structure; we shall only - 69 - analyze the single most important factor, which is the effect on consumption structure of consumption levels or people's income. Hence we must begin with forecasting people's consumption levels. A. Forecasting People's Consumption Levels Forecasting people's consumption levels should begin with forecasting national income. China's 1980 national income was 366.7 billion yuan. According to an analysis of the various factors affecting the rate of increase in national income, we believe it is feasible that China's national income in the eighties will develop at an annual average rate of 6.5 percent, and in the nineties at 7 percent. Given this forecast, in 1990 China's national income will reach 688.3 billion yuan; in 2000, it will reach 1,354 billion yuan. In future distribution of accumulation and consumption, we must avoid returning to our past mistakes of high accumulation and low consumption, and we must ensure that consumption is well adapted and not increasing so rapidly that it would affect national construction. We must certainly adhere to a policy of continuing to improve the people's standard of living, on the basis of developing production. Given these various analyses, we believe that the accumulation rate for the next 20 years should be held steady at between 28-30 percent (with the corresponding consumption rate kept between 70-72 percent). To facilitate the various calculations in our forecast, we will take the middle value, 29 percent (which is the actual figure for 1982). According to the above two forecasts, in 1990 China's accumulation funds will reach 199.6 billion yuan, and consumption funds will reach 488.7 billion yuan. In 2000, accumulation funds will reach 392.7 billion yuan, and consumption funds will reach 961.3 billion yuan. - 70 - Consumption funds are used in two areas: people's individual consumption, and collective social consumption. In terms of the overall socialist historical period, collective social consumption has increased more quickly than people's individual consumption. China's experience over the last 30 years proves this point. This may also be the case for the next 20 years. In 1980 China's collective social consumption funds were 29.5 billion yuan; if we figure the average annual increase at 6.5 percent from now on, in 1990 the figure will reach 55.4 billion yuan, and in 2000 104.0 billion yuan. Subtracting collective social consumption funds from all consumption funds, the national income used for individual consumption, i.e., the people's individual consumption funds, will reach 433.3 billion yuan in 1990 and 857.3 billion yuan in 2000. As for forecasts of China's population, there are two cases. In the first case, assuming that each couple has 1.7 children, the 1990 population will be 1,085,620,000, and the 2000 population will be 1,164,050,000. In the second case, assuming that each couple has 2 children, the 1990 population will be 1,115,940,000, and the 2000 population will be 1,223,640,000. We are making every effort to achieve the first case, but the difficulties are considerable, and there is no absolute assurance that we can do this. Therefore we are taking the second case as the standard for our forecast. Given the above forecasts of individual consumption funds and population, our people's consumption level (total individual consumption funds divided by population) will reach 388M in 1990 and 701w in 2000. If this forecast is realized, it will raise the 1980 people's consumption level (227V) by 0.71 times and 2.1 times, respectively (average annual increase of - 71 - 5.8 percent over 20 years), which fully meets the strategic goal of doubling the income of people in urban and rural areas set by the 12th Party congress. Some comrades believe that collective social consumption funds are also used for the people's standard of living and that therefore forecasts of consumption levels should include them. If this is the case, the consumption level of the Chinese people in the year 2000 will be 7861. Other comrades believe that nonproductive accumulation is also used to improve the people's standard of living and that therefore in calculating people's consumption levels we should not only include collective social consumption funds and individual consumption funds, but also nonproductive accumulation. If we follow these broad requirements, the consumption level in the year 2000 will be 8931 (figuring that nonproductive accumulation accounts for one-third of total accumulation). The results of the above analyses and calculations are that in the year 2000 the consumption level of China's people will be 7011, plus 851 in average collective social accumulation funds and 1071 in average nonproductive accumulation funds, for a total of about 9001, which is a comfortable standard of living. There remain significant gaps in consumption levels between urban and rural areas. Henceforth, in order to gradually reduce the gaps between worker and peasant living standards, we should ensure that the rate of improvement in peasant domestic consumption levels is somewhat faster than that of urban consumption levels; in 1980 the peasant consumption level was 1731, while that for urban dwellers was 4681. In the next 20 years, if peasant consumption levels rise at an average annual rate of 5.9 percent, by 1990 the figure will be 3071, and by 2000 it will be 5451; if the consumption level of urban - 72 - dwellers rises at an averge annual rate of 5.7 percent, by 1990 it will be 8151, and by 2000 it will be 1,419M. Thus the ratio of peasant consumption level to urban consumption level would be 2.65 : 1 in 1990 and 2.60 : 1 in 2000. This would represent a lessening from the 1980 ratio of 2.71 : 1. The national consumption levels as well as urban and peasant consumption levels forecast above, whether for 1990 or 2000, are already being achieved in some areas of China by some people. Below we shall forecast national and urban-rural domestic consumption structure on the basis of consumption levels and structure already reached in some areas and among some groups. B. Forecasting Urban Consumption Structure 1. Consumption Structure in 1990 According to the above forecast of consumption levels, the consumption level of China's urban dwellers will reach 815M in 1990, a monthly average of 68M. This consumption level has already been reached or approached in 1982 by some urban dwellers nationwide and in Beijing. Let us first examine the consumption structure of these urban groups nationwide and in Beijing. In 1982, per capita purchases of major commodities by staff-and- worker households with average per capita monthly living-expense incomes over 60M nationwide were as follows: cereals, 353 jin; fresh vegetables, 425 jin; edible vegetable oils, 15 jin; pork, 52 jin; beef and lamb, 5 jin; poultry, 8 jin; eggs, 18 jin; fish and shrimp, 22 jin; sugar, 8 jin; cigarettes, 53 packs; spirits, 14 jin; cotton cloth, 17 chi; chemical-fiber cloth, 3 meters; woolens, 0.6 meters; silks and satins, 0.9 meters; cotton clothing, 0.5 pieces; chemical-fiber clothing, 1.3 pieces; leather shoes, 0.7 pairs; soap, 9.7 bars; coal, 584 jin. - 73 - Table 40: Consumption Structure of Staff-and-Worker Households with Average Per Capita Monthly Living-Expense Incomes Over 60Y, Nationwide, 1982 Absolute Relative Item figure (M) figure (%) Total living-expense expenditures 61.84 100.00 I. Expenditures for commodity purchases 56.76 91.79 Including: 1. Food 34.14 55.21 (1) Cereals 6.41 10.37 (2) Nonstaples 18.93 30.61 (3) Tobacco, alcohol, tea 3.47 5.61 (4) Other foods 5.33 8.62 2. Clothing 9.66 15.62 3. Articles of everyday use 7.00 11.32 4. Cultural and recreational items 2.89 4.67 5. Books, magazines, newspapers 0.56 0.91 6. Medicines and health products 0.30 0.49 7. Fuel 0.96 1.55 II. Noncommodity expenditures 5.08 8.21 Including: 1. Rent 0.95 1.54 2. Water and electricity charges 0.63 1.02 3. Miscellaneous education costs 0.10 0.16 4. Child care costs 0.10 0.16 5. Transport costs 0.92 1.49 6. Posts and telecommunications 0.12 0.19 7. Cultural and recreational costs 0.31 0.50 Per capita purchases of major commodities by staff-and-worker households with per capita monthly living-expense income over 60M in Beijing in 1982 were as follows: cereals, 360 jin; pork, 47 jin; beef and lamb, 10 jin; animal oils, 5 jin; poultry, 6 jin; eggs, 20 jin; fish and shrimp, 20 jin; fresh vegetables, 442 jin; vegetable oils, 19 jin; sugar, 10 jin; cigarettes, 45 packs; spirits, 16 jin; cotton cloth, 14 chi; chemical-fiber cloth, 3 meters; woolens, 0.48 meters; silks and satins, 0.72 meters; cotton - 74 - clothing, 0.6 pieces; chemical-fiber clothing, 1.3 pieces; leather shoes, 0.6 pairs; soap, 8.2 bars; coal, 360 jin; LPG, 28 kg. Table 41: Consumption Structure of Staff-and-Worker Households with Average Per Capita Monthly Living-Expense Incomes Over 601, Beijing, 1982 Absolute Relative Item figure (1) figure (%) Total living-expense expenditures 62.79 100.00 I. Expenditures for commodity purchases 57.52 91.61 1. Food 36.06 57.43 Including: Purchased from food-service 1.42 2.26 enterprises Purchased from collective 5.40 8.60 dining halls (1) Cereals 6.21 9.89 (2) Nonstaples 19.64 31.28 (3) Tobacco, spirits, tea 3.88 6.18 (4) Other foods 6.33 10.08 2. Clothing 9.68 15.42 3. Utilitarian articles 10.74 17.10 Including: Everyday consumer goods 6.38 10.16 (1) Articles of everyday use 6.72 10.70 (2) Cultural and recreational articles 2.63 4.19 (3) Books, magazines, newspapers 0.58 0.92 (4) Medicines and health-care products 0.24 0.38 (5) Housing and construction materials 0.17 0.27 (6) Other commodities 0.40 0.63 4. Fuel 1.04 1.66 II. Noncommodity expenditures 5.27 8.39 1. Rent 1.07 1.70 2. Water charges 0.25 0.40 3. Electricity charges 0.40 0.64 4. Educational costs 0.13 0.21 5. Child care 0.18 0.29 6. Transport 1.54 2.45 7. Posts and telecommunications 0.09 0.14 8. Cultural and recreational costs 0.29 0.46 9. Repair services 0.71 1.13 10. Medical and health care 0.21 0.33 11. Other noncommodity expenditures 0.38 0.61 12. Taxes and insurance 0.02 0.03 - 75 - Given the consumption structure for Beijing and nationwide staff-and- worker households whose consumption level reached 720Y in 1982, we forecast that the approximate consumption structure of urban dwellers in China in 1990 will be as follows: of monthly living-expense expenditures, commodity expenditures will account for 91 percent and noncommodity expenditures for 9 percent; in terms of commodity expenditures, food will account for 56 percent, clothing for 15, percent, articles of everyday use for 11 percent, cultural and recreational articles for 5 percent, books, magazines and newspapers for 1 percent, medicines and health-care products for 0.5 percent, and fuel for 2 percent; in terms of noncommodity expenditures, rent will account for 2 percent, water and electricity charges for 2 percent, transport for 2 percent, miscellaneous educational costs for 0.2 percent, and culture and recreation 1 percent. Purchases per capita of major commodities: cereals, 360 jin; pork, 50 jin; beef and lamb, 8 jin; poultry, 7 jin; eggs, 20 jin; fish and shrimp, 20 jin; sugar, 10 jin; fresh vegetables, 450 jin; cigarettes, 50 packs; spirits, 15 jin; cotton cloth, 15 chi; chemical-fiber cloth, 3 meters; woolens, 0.5 meters; silks and satins, 0.8 meters; cotton clothing, 0.6 pieces; chemical-fiber clothing, 1.5 pieces; leather shoes, 0.8 pairs; soap, 0.9 bars. 2. Consumption Structure in 2000 Given the above forecast of consumption levels, in the year 2000 the consumption level of China's urban dwellers will reach 1,419M, an average of 118w a month. This consumption level has already been reached or approached in a very small number of households, but concrete survey data are lacking, so it is difficult to make forecasts on this basis. Therefore we can only rely on historical data to make some general forecasts. - 76 - First, let us examine changes in the retail structure of social consumer goods: Table 42: Retail Sales Structure of Social Consuer Goods Books, Medicines and Articles of Cultural and magazines, health-care Year Food CLothing everyday use educational articles repapers products FUl 1952 56.4 19.3 15.0 2.6 0.8 2.6 3.3 1957 54.5 18.7 14.8 2.9 1.1 3.3 4.2 1962 51.6 14.7 19.6 3.3 0.8 4.8 5.2 1965 55.4 19.1 11.7 2.9 1.1 4.7 5.1 1970 53.4 23.4 10.5 2.5 0.4 5.0 4.8 1975 53.0 21.0 12.1 2.8 0.9 5.5 4.7 1980 51.2 23.0 12.9 4.1 1.3 3.7 3.8 1982 52.9 21.3 13.3 4.5 1.3 3.3 3.4 The data in Table 43 show that there have not been very major changes in the retail sales structure of social consumer goods over the last 30 years. These changes include: food, down 3.5 percent; clothing, up 2 percent; articles of everyday use, down 1.7 percent; cultural and educational articles, up 1.9 percent; books, magazines and newspapers, 0.5 percent; medicines and health-care products, up 0.7 percent; fuel, up 0.1 percent. Let us look again at changes in staff-and-worker household consumption structure: - 77 - Table 43: Oanges in Staff -sfldWrker HxJehold Onantion Structure Nationide (x) Year Food Clothirg Articles of everyday use Hbzsirg Fuel Cultural services 1957 61.2 9.4 13.4 2.1 3.1 10.8 1963 63.5 8.0 9.6 2.7 4.0 12.2 1964 59.8 9.4 13.7 2.4 3.7 11.0 1981 56.7 14.8 18.5 1.5 2.1 6.4 1982 58.2 15.0 16.2 2.0 2.0 6.6 Note: Accordirg to family livelihood survey data. The data in Table 43 show that there have not been very great changes in staff-and-worker household consumption structure over the last 25 years. These changes include: food, down 3 percent; clothing, up 5.6 percent; articles of everyday use, up 2.8 percent; housing, down 0.1 percent; fuel, down 1.1 percent; cultural services, down 4.2 percent. Given the above changes from 1952-1982 in the retail sales structure of social consumer goods, the changes in staff-and-worker household consumption structure from 1957-1982, and the forecasts for staff-and-worker household consumption structure from 1980 to 1990, we believe that the consumption structure for China's staff-and-worker households will be approximately as follows: of monthly living-expense expenditures, expenditures for purchases of commodities will account for 88 percent, while noncommodity expenditures will account for 12 percent. In terms of commodity expenditures, food will account for 50 percent; clothing, for 16 percent; articles of everyday use, for 20 percent; fuel, for 2 percent. This represents no major changes as compared with present consumption structure. - 78 - C. Forecasting Peasant Consumption Structure 1. Peasant Consumption Structure in 1980 According to the above consumption-level forecasts, the consumption level of China's peasants will reach 307Y in 1990. This consumption level has already been reached in 1982 by peasants in Shanghai, Beijing and Guangdong; 40 percent of Hubei's peasants also had reached this level in 1982. Below is a table showing 1982 peasant consumption structure in these four areas (see Table 44). TIable 44: 1982 Peasant Casumption Structure (%) in Shanghai, Beijing, OGangdong and Hubei Item Shanghai Beijing Guangdong Hubei lbtal livieg exenditures 100.00 100.00 100.00 100.00 I. Dnestic consumer good xpeenditures 98.68 97.51 96.26 97.45 1. Food 49.73 52.52 58.43 58.09 Staples 16.56 25.64 23.04 - Nonstaples 21.22 16.89 28.53 - 2. Clothing 8.69 12.08 6.57 13.39 3. Fuel 2.82 2.68 5.86 5.14 4. Husing 26.01 16.70 13.57 10.03 5. Articles of everyday use and other 11.43 13.52 11.83 10.79 U. Cultural ard life services 1.32 2.49 3.74 2.55 T'he data in Table 44 show that there are significant differences in peasant consumption structure from one place to another. The food proportion varies by 8.7 percent from the highest value to the lowest; the proportion for clothing varies by 6.82 percent, more than double; the proportion for fuel varies by 3.18 percent, also more than double; housing varies by 15.98 percent, also more than double; the proportion for articles of everyday use - 79 - and other varies by 2.73 percent; and the proportion for cultural and life services varies by 2.42 percent, also more than double from one place to another. With the differences between the highest and lowest values being so great, it is impossible for the peasant consumption structure in any of these provinces and cities to represent China's 1990 peasant consumption structure. In order to find an approximate outline of the 1990 peasant consumption structure, we have in some cases taken the middle value between the highest and lowest. In this way, we get the approximate peasant consumption structure for 1990 in China: of living-expense expenditures, domestic consumer goods expenditures account for 97.5 percent, while cultural and life services expenditures account for 2.5 percent; of domestic consumer goods expenditures, food expenditures account for 58 percent, clothing for 12 percent, fuel for 3.5 percent, housing for 14 percent, and articles of everyday use and other for 10.0 percent. 2. Peasant Consumption Structure in 2000 According to the above consumption level forecasts, China's peasant consumption level will reach 545Y in 2000, or an average of 45Y a month. This consumption level has already been reached in a good many rural 10,000Y households, but we are lacking concrete data on this and cannot proceed with any forecasts. This consumption level has already been met in most urban households. Therefore, we shall utilize the consumption structure of urban dwellers at this consumption level and other historical data to forecast peasant consumption structure in China in the year 2000. The living expenses of urban dwellers are of course different from those of peasants. At the same consumption level, peasant expenditures for housing are higher, while their expenditures for cultural and life services - 80 - are lower, and the proportion of food expenditures is higher. In making our forecasts we should not overlook these characteristics. Table 45: 1982 Consumption Structure of Urban Staff-and-Worker Households with Per Capita Monthly Living Expenses Over 45M Absolute Relative Item figure (M) figure (%) Total living-expense expenditures 49.91 100.00 I. Expenditures for purchases of commodities 46.03 92.23 Including: 1. Food 28.35 56.81 2. Clothing 7.75 15.53 3. Articles of everyday use 7.78 15.59 4. Housing 0.76 1.52 5. Fuel 0.83 1.66 II. Cultural and life services expenditures 3.88 7.77 Below let us look again at the history of changes in China's peasant consumption structure: - 81 - Table 46: Chnges in China's Peasant aGsnption Structure (%) Cultural and Articles of life services Year Food CLothing everyday use H3using Fuel experditures 1954 68.25 13.0 7.4 2.0 6.5 2.70 1956 68.14 13.27 6.42 2.02 7.51 2.62 1957 64.62 15.23 7.65 2.11 8.44 1.74 1962 60.87 8.39 9.41 4.95 13.85 2.53 1963 63.31 9.32 8.79 4.71 11.20 2.67 1964 67.11 9.17 7.68 3.39 9.93 2.72 1965 65.81 13.22 7.87 2.90 7.41 2.79 1978 67.71 12.70 6.57 3.16 7.14 2.72 1979 63.96 13.12 8.28 5.69 6.20 2.75 1980 61.76 12.32 9.44 7.89 5.96 2.63 1981 59.66 12.35 10.22 9.79 5.55 2.43 1982 60.48 11.25 10.17 10.25 5.61 2.24 The data in Table 46 show that over 28 years the proportion of food dropped 7.77 percent, that of clothing dropped 1.75 percent, that of articles of everyday use rose 2.77 percent, that of housing rose 8.25 percent, that of fuel dropped 0.89 percent, and that of cultural and life services dropped 0.46 percent. Overall, the changes are not very great. A similar phenomenon may present itself in the future. We forecast that the degree of change will be slightly greater. Given the consumption structure for staff-and-worker households with per capita monthly living expenses of over 45Y for China's urban dwellers in 1982 as described above, along with the changes in peasant consumption struc- ture from 1954 to 1982 and the forecasts of peasant consumption structure for 1990, we believe that in the year 2000 the consumption structure of China's - 82 - peasants may be as follows: of living-expense expenditures, expenditures for consumer goods will account for 95 percent, and expenditures for cultural and life services for 5 percent; in terms of consumer-goods expenditures, food will account for 52 percent, clothing for 14 percent, articles of everyday use for 15 percent, housing for 10 percent, and fuel for 4 percent. D. Forecasting National Domestic Consumption Structure Above we analyzed separately the consumption structures of urban dwellers and peasants; we now need to combine the two and make an overall forecast of the domestic consumption structure of China's people as a whole. There are several possible methods for combining the consumption structures of urban dwellers and peasants. The best method is of course to total up the various domestic expenditures for urban dwellers and peasants and directly find the proportions each accounts for. However, we lack data on total domestic expenditures in all these categories for urban and rural cases. Hence we can only combine the various urban and rural consumption structures as forecast above, using the weighted average approach. The difficulty in using this approach lies in selecting the weight factor. What weight factor is used has a major impact on the results. Here we are using the urban-rural population as the weight factor (i.e., the population ratio is the weight factor). When we forecast the 1990 consumption structure, the urban-rural population ratio is 2 : 8; when we forecast the 2000 consumption structure, the urban-rural population ratio is 3 : 7. Furthermore, in order to ensure that the urban and rural consumption structures can be combined, we must ensure that the categories in each are uniform. Hence we must merge and - 83 - make some adjustments in a number of categories of urban living expenses. After doing all this, the national consumption structure we arrive at is as follows: Table 47: National Domestic Consumption Structure (%) Item 1 9 9 0 2 0 0 0 Total living expense expenditures 100.00 100.00 I. Consumer-goods expenditures 96.6 93.5 1. Food expenditures 57.6 51.4 2. Clothing expenditures 12.6 14.6 3. Expenditures for articles of everyday use 11.6 16.5 4. Housing expenditures 11.6 7.6 5. Fuel expenditures 3.2 3.4 II. Cultural and life services expenditures 3.4 6.5 Finally, we must clarify one point: the basis for our forecasts is past statistical data. Future developments will of course not entirely repeat the past. Scientific and technical developments, changes in production structure, and other factors will cause the consumption structure of China's people at the end of this century to be significantly different from the consumption structure already achieved by some people. However, this type of change is slow and may not be very great, so that the consumption structure of some people today will gradually become the consumption structure of everyone in the year 2000; this conforms to the general laws of material development. - 84 - V. DEMAND SYSTEK ANALYSIS AND CONSUMPIION STRUCTURE FORECASTING This section will study the demand system of urban and rural dwellers and establish a measurement model in the light of data from Beijing and Hubei household income and expenditures surveys; these will then be applied to forecasting consumption structure in the future. A. Some Problems Requiring Attention in Studying China's Consumption Structure In studying China's consumption structure, there are several special circumstances that must be taken into account. These are primarily: 1. Collective Consumption This refers to material consumption in such sectors as education, health, science and culture. It does not include services provided by workers and personnel in these sectors. The figures for collective consumption are essentially equal to the operating expenses allocated by State financial or enterprise units to the scientific, educational, cultural and health sectors, minus the wages for these sectors. Most collective consumption is related to the people's material benefits. Because of collective consumption, people receive education, health care and so on, either free of charge or at low cost. We can get an overall impression of the position of collective consumption from the following 1982 national income application account: - 85 - Table 48: Application of National Income, 1982 Unit: 100 million V Application of national income 4254 Accumulation 1233 Of this, productive accumulation 607 Consumption 3021 Of this, people's consumption 2684 Social consumption 337 (approximately 1/3 of which is collective consumption*) * Collective consumptions is calculated this way: operating costs for science, education, culture and health expended by national financial agencies were 19.7 billion yuan; in the same year, wages of workers in these sectors were 9.6 billion yuan. The difference of 10.1 billion yuan may be taken as an approximation of collective consumption. 2. Free Health Care For staff and workers the State uses a free health care system and a labor insurance system, so that staff and workers do not pay for medical care or medicine. In national income accounts, free health care is entered as people's consumption; but in household income survey data, expenditures for people's consumption do not include free health care. In 1979 the average was about 40Y for each staff and worker. Precise figures for 1982 are not yet in, but it is said that the average is approximately 50M. - 86 - 3. Price Subsidies These are primarily used in the cities. The State-set retail prices for a number of key foodstuffs and domestic coal, rents for staff-and-worker housing, and urban transportation charges are lower than their costs. That portion of the price which is lower than cost is given in the form of subsidies by the State financial agencies to the enterprises. The beneficiaries are primarily urban dwellers. In 1981 State subsidies for grain, edible oils, meat, eggs, poultry and vegetables totalled 15.0 billion M; the average per city resident per month was about 6.25M. Current rent subsidies, as calculated roughly by the author, were approximately 1I per person per month. 4. Supply Controls Since the late fifties there have consistently been shortages in the supply of certain consumer goods. The situation in the early sixties was especially critical. Since the start of the eighties the situation has taken a turn for the better; before then, supplies of grain, cotton, edible vegetable oils, sugar, soap, fish and so on were controlled, and supplied by coupon or certificate. In many years supply coupons were also issued for pork, bicycles, sewing machines, watches, furniture and so on. In recent years, coupons and certificates have basically been eliminated, with the exception of coupons for grain, cotton and oils. Moreover, for grain and cotton the coupons in reality no longer have the effect of limiting consump- tion, inasmuch as the amounts allocated to the vast majority of households exceed the demand. For example, sample data for Beijing in 1982 show that the average monthly per capita purchase of grain (including prepared staples) was 25.8 jin, which is below the average monthly fixed amount for Beijing; average - 87 - per capita purchases of cotton cloth (including cotton clothing) for the year were about 14 chi, lower than the set amount of 18.1 chi. And in terms of edible vegetable oils, although the fixed amount (about 7 jin per person per year) was inadequate, the people are able to purchase higher-priced oil from State-run shops and markets, so that consumption actually is not limited (1982 Beijing sampling data show: average per capita annual purchases of vegetable oil were 13.6 jin). The most stringent supply control is on rents. Since the fifties, there has been a severe shortage of urban housing. Beijing sampling data show that in 1957 the average person had 4.42 m2, of living space; in 1982, the average was 5.42 m2. In a situation of acute housing shortage, the State is still following a low-rent policy, and this is worsening the supply problem. 5. Irrational Relative Pricing This is primarily manifested by prices for primary products that tend to be too low and prices for manufactured products that tend to be too high. In terms of consumer goods, this is primarily manifested by prices of basic foodstuffs and fuel as well as rents that are too low. In making inter- national comparisons, we must pay attention to this point. In forecasting, we must take into account the possibility of adjusting prices to make them more rational. Given the above circumstances, we believe that the following approaches should be taken in using the sampling data to study the laws of supply and demand in China's urban and rural consumption. (1) We should avoid using data from the 20-year period of 1959-1978 when there were more supply controls. Therefore we shall not do an analysis of the time-sequence data but rather only an analysis of cross-section data. - 88 - (2) Commodity categories should not be broken down too much. This is because, even though in the last few years all commodities have at a time been in short supply, if we list these commodities individually and calculate their supply-and-demand function, we may well get false results. If we consider broad categories of commodities, then supply limitations on detailed commodities will not be enough to have a clear impact on the correct computation of the demand function. (3) Urban staff-and-worker housing should be taken out of the demand system and treated separately. Since staff-and-worker housing has always been strictly controlled, its demand function cannot be found from sampling data. (4) Urban-rural comparisons of consumption structure must pay attention to differences in collective consumption, as well as to urban price subsidies for housing, transportation, food, etc. (5) Consumption-structure forecasts must take into account changes in relative prices. B. Extended Linear Expenditure System 1. Model We will use the "extended linear expenditure system" (ELES) as the frame of our analysis. It is an expansion of the famous "linear expenditure system" (LES). The extended linear expenditure system treats demand as a function of income and price. It is expressed by the following formula: Vi Pi *Yi + i (Y I . ; y) (2.1) in which: - 89 - Vi represents the expenditure of a consumer on Category i commodities or services; Pi represents the price of Category i coiamodities or services; yi represents the basic demand by each person for Category i commodities or services; Y represents the income of a consumer; a represents the marginal consumption trend for Category i commodities or services. From the formula (2.1) we can see that ELES includes three basic points: (1) Consumption expenditures are divided into basic demand expenditures and expenditures in excess of basic demand; (2) basic demand does not change with income; (3) marginal consumption trend is the same for all people. The linear expenditure system (LES) takes demand as a function of total expenditure and price. It is expressed by the following formula: Vi = Pi * Yi + 0 (V P i (2.2) Here V represents total consumption expenditure by a consumer; 0 represents the marginal budget portion (the proportion of expenditure in excess of Dasic demand used on Category i commodities or services); the other symbols mean the same as above. The basic theory behind ELES and LES are the same, but the latter takes the total consumption expenditure as a variable determined outside the model, while the former allows the total consumption expenditure to be determined within the model. For the ELES, the total consumption expenditure is sought from i in formula (2.1), obtaining: - 90 - V = (1 - -*) Pyj + (&i* )Y (2.3) The ELES may also be given the LES form. From (2.3) we get Y and replace it in (2.1) to get: Vi = PiYi + (i */Yi*)(V - IP.y.) (2.4) (2.4) corresponds to (2.2). From this we can find: i = Si (2.5) In the ELES, the parameters that must be calculated are Yi and gi*, and the data required are Vi, Y and Pi. In the LES, the parameters that must be calculated are Yi and ai' and the data required are Vi, V and Pi. A salient feature of the ELES is that if there is no price data Pi, it is also possible to calculate Pi Yi and i from cross-section data, and find the price elasticity. As for cross-section data, in (2.1) Pi Yi ano Bi * yJ are constants, so that they can be combined and expressed as a.: a, = Pi Yi + Si P.Y; (2.6) whereupon (2.1) becomes: Vi =i + B Y (2.7) We can use the ordinary least square method (OLS) to calculate the parameters in (2.7); after getting ai and oi , we can calculate Piyi according to (2.6). The formula for this is: iYi =ai + ai/(I- J i*) (2.8) 1/ Formula (2.8) is obtained thus: adding up demand for i in (2.6), we get - Pjy j )aiI(1 - ) ; this is entered in (2.6), and we get (2.8). - 91 - 2. Demand Elasticity We shall now use the linear expenditure system and the extended linear expenditure system to list the formulas for income and price elasticities. (A) Income elasticity. This refers to percentage of change in aemand with prices constant, due to a percentage increase in income. The defining formula is: ni= aQi *Y (2.9) in which Qi = Vi/Pi. n can only be calculated for the extended linear expenditure system. The formula is: ni ay *Qi- = Si 0 Vi (2.10) (B) Total expenditure elasticity. This refers to the percentage of change in demand, with prices constant, due to a percentage change in total expenditures. The defining formula is: i av * Q (2.11) The results of calculating ni for the linear expenditure system and thie extended linear expenditure system are the same. The formula is: n1 ,aQi v v= 6v (2.12) - 92 - (C) Price elasticity. This refers to the percentage of change in demand, due to a price change, with the prices of other commodities, income and total expenditure constant. The defining formula is: Tij = aPj * Q, * (2.13) When i = j, n ij is called own price elasticity; when i # j, n ij is called cross price elasticity. As for the extended linear expenditure system, calculating price elasticity assumes income is constant. The formula is: - aaQi Pi * pi,i nifi = - * =Qi ( i vi (2.14) "i Pi~~~~ij _ aQi Pi = -- *P /V (2.15) As for the linear expenditure system, calculating price elasticity assumes total expenditures are constant. The formula is: nii + aP * ai 'd1~ v i (2.16) aqi p~~~~~~ n Q P a * i py /v k2.17) Demand price elasticity also includes what is called "income compensation price elasticity." This means that when the price of a given commodity changes, if currency income remains unchanged, actual income will undergo a change. Changes in actual income may also affect demand. If we - 93 - take away the effect on demand of this type of change in actual income from the total effect in demand of price changes and then calculate elasticity, this is called income compensation price elasticity. Its defining formula is: 9 - i av*) P (2.18) in which aV*/aPj = -Q . When Pj goes up with APj, the consumer must expend APJQi on category j commodities, and this corresponds to the reduction in his actual income, i.e., AV* -AP Q . Making AP + 0 , we get aV*/aP; = -q . n i; is the same for the linear expenditure system and the extended linear expenditure system. The formula is: PiY i n*i 1- V (2.19) i n j * B(1 ii) *t2 .20 ) C. Analysis of Beijing Cross-section Data The data which we now have for Beijing are data from staff-and-worker household income sampling surveys of 1955, 1957, 1965, 1981 and 1982. The data for 1981 are divided into three income groups; those from 1982 are divided into nine income groups. The other years are not divided by income group. - 94 - Because there are only five sample points (data for five years) for the time-series data, we cannot fully meet the requirements of statistical analysis. Moreover, there were especially heavy supply controls in 1965, so we shall not analyze the time-series data. We shall only analyze the 1982 cross-section data. Again, in accordance with point (2) of the first section, we are only taking into account the consumer demands of the following broad categories of commodities and services: food, clothing, fuel, rent and other noncommodities. The 1982 Beijing data meeting the above analytical requirements are given the following table: Mble 49: Abstract of Staff-and-Worker Hkseohld Toe Srvey Data, Beijing, 1982 Per capita cmstumption (V/ennth) Per capita ultimate Nt:czmodity Number Number Number distributed Articles Includirg goods Ioane m of of Of inpe* Tbtal of every durable (not including group househlds jobholders people (V/mDnth) experxiture Food Clothirg Fel use goods Rent rent) Under 20V 3 018 3 18.17 18.17 12.02 2.59 0.76 1.36 0.00 0.44 1.00 2D-25V 11 065 19 23.69 23.78 15.78 2.70 0.54 2.38 0.35 0.56 1.82 25-3(W 40 2G0 82 28.05 28.08 17.03 3.49 0.83 3.68 1.56 0.70 2.35 3D-354i 85 375 170 33.02 32.42 20.21 4.17 0.78 4.07 1.60 0.71 2.48 35-4 C7 187 809 411 37.64 36.38 21.80 5.05 0.79 5.07 2.36 0.73 2.94 40-4_v 236 936 531 42.64 41.64 25.04 5.51 0.86 6.35 3.27 0.66 3.22 45-' > 198 711 442 47.44 45.55 27.15 6.32 0.88 7.0) 3.76 0.70 3.50 50-t 249 904 632 54.50 51.78 30.43 7.63 0.95 8.38 4.86 0.77 3.62 Over 3(w 191 649 525 69.31 62.79 36.06 9.68 1.04 10.74 6.38 1.07 4.20 Ibtal or ave.r.e value 1200 4667 2815 46.75 44.57 26.46 6.30 0.88 6.86 3.67 0.76 3.31 * Ultimate distributed income refers to income after shifted paymEnts. In the Beijing data this is called "living-expense income." We believe the latter is not as precise as the fomner. - 96 - According to point (3) in Section I, we will treat rent separately from the demand system analysis. The basis for the model of staff-and-worker demaand systea analysis is formula (2.1): Vi PiYi + i (3.1) in which Y* represents the ultimate distributed income minus rent. The i does not include rent. The other symbols have the same meaning as above. The parameters in (3.1) are calculated the same as above: first we calculate the a and Bi in Vi = a + Y ; then we calculate Pi yi according to (2.8) and ai according to (2.5). Because of differences in the number of people in each income group, we must calculate the parameters by weighting for number of people. The results are shown in Table 50. Table 50 makes no calculation of the laws of demand for housing. We believe with strict controls being placed on housing and with people being unable to rent freely, it is difficult to calculate accurately the marginal consumption trend for housing. As for current basic housing demand, we believe it is acceptable to use an average per capita rent of 0.76Y as an approximate value. The value of Piyi in Table 49 shows that basic staff-and-worker demand roughly corresponds to average per capita consumption in the 30-35Y income group. For a comparison of the two, see Table 51. - 97 - Table 50: Parameters of Demand Functions Calculated from 1982 Beijing Data -2 a B* °6* -R Piy1 i Food 5.3240 0.45966 O.O0059 0.9924 19.87 0.5422 Clothing -0.7469 0.15321 0.00016 0.9948 4.10 0.1807 Fuel 0.5387 0.00744 0.00004 0.8931 0.77 0.0088 Articles of everyday use -1.5067 0.18188 0.00022 0.9934 4.25 0.2145 Noncommodities (except rent) 1.2083 0.04554 0.00017 0.9380 2.65 0.0537 4.817 0.84773 31.64 1.0000 Table 51: Article of Non- everyday commodities Food Clothing Fuel use Rent (except rent) Total Calculated basic demand 19.87 4.10 0.77 4.25 0.76 2.65 32.40 per capita (M/month) Average per capita actual 20.21 4.17 0.78 4.07 0.71 2.48 32.42 expenditure in the 30-35M income group (M/month) - 98 - Such a calculation of basic demand has a certain reference value for the relevant agencies in setting welfare policy. Looking at the 1982 situation in Beijing, those whose income is below actual demand include the people in the three low-income groups ahead of the 30-35i group. They account for 10 percent of the sample total. Improvement in their situation is a more pressing matter than for other people. Below we are going to calculate demand total expenditure elasticity and price elasticity according to the last two parameters in Table 50. Calculating demand elasticity involves variables V and Vi. Here we shall only use the average levels of the values of V and Vi to calculate demand elasticity. In terms of studying consumption structure, it is enough to figure only total expenditure elasticity and price elasticity of demand. In Table 52, total expenditure elasticity of demand for food, fuel, and noncommodities is less than one; this means that the rate of increase in demand for these commodities by staff and workers is lower than the rate of increase in total expenditures. The total expenditure elasticity of demand for clothing and articles of everyday use is greater than one; this means that the rate of increase in demand for these two commodities by staff and workers is greater than the rate of increase in total expenditure. In Tables 53 and 54, the figures in a given row represent the effect of changes in the price of the various types of comaodities on the demand for a given commodity; the figures in a given column represent the effect of changes in the price of a given commodity on the demand for the various commodities. The figures in these two tables show: (1) the effect of identical prices (main diagonal element) is major; (2) of the impact of intersecting prices, only the effect of changes in food prices on demand for - 99 - the various coramodities is fairly great (first column of elements); (3) with no income compensation, a rise in prices of any major category of commodities may lower demand for all (major categories of) commodities, because the consumers' actual income is reduced; (4) if there is income compensation, a rise in prices of a given category of commodities only reduces demand for that category; demand for commodities in the other categories on the contrary goes up. This is because after income compensation actual income is unchanged, and only the relative prices (price ratios) are changed. Table 52: Total Expenditure Elasticity of Demand Among Beijing Staff and Workers (ni) Item Total Expenditure Elasticity Food 0.8977 Clothing 1.2566 Fuel 0.4381 Articles of 1.3699 everyday use Non commodities 0.7108 (not including rent) - 100 - Table 53: Price Elasticity of Demand Among Beijing Staff and Workers (n i) Articles Non- of commodities everyday (not including Food Clothing Fuel use rent) Food -0.6562 -0.0840 -0.0158 -0.0871 -0.0543 Clothing -0.5699 -0.4668 -0.0221 -0.1219 -0.0760 Fuel -0.1987 -0.0410 -0.1327 -0.0425 -0.0265 Articles of -0.6213 -0.1282 -0.0241 -0.5134 -0.0829 everyday use Noncommodities -0.3224 -0.0665 -0.0125 -0.0690 -0.2424 Note: Assuming total expenditure is constant. Table 54: Price Elasticity of Demand Among Beijing Staff and Workers (Income Compensation) (n**ij) Articles Non- of commodities everyday (not including Food Clothing Fuel use rent) Food -0.1140 0.0450 0.0022 0.0534 0.0134 Clothing 0.1893 -0.2861 0.0031 0.0749 0.0188 Fuel 0.0678 0.0226 -0.1239 0.0268 0.0067 Articles of 0.2063 0.0688 0.0033 -0.2989 0.0204 everyday use lloncommodities 0.1081 0.0360 0.0018 0.0428 -0.1887 - 101 - Understanding price elasticity and income elasticity (or total income elasticity) is very helpful in trade and materials pricing work. D. Analysis of Hubei Cross-Section Data We have now compiled data from over a period of more than ten years on peasant household income surveys in Hubei Province. The 1981 and 1982 data have been divided into seven income groups, while the data from thle other years have not been. Considering the supply restrictions in the sixties and seventies, plus the fact that for most years there are no data on prices, we have not done any analysis of the time-sequence data but only a cross-section analysis of the 1981 and 1982 data. In order to improve the reliability of the analysis, we shall combine the 1981 and 1982 data and use them together (for a total of 14 sample points). We believe this approach is feasible wnen all price changes for all commodities are very small. In fact, during this period the change of commodity prices were not very significant (see Table 55). Table 55: Nationwide State-operated Industry and Commerce List Price Index (1981 = 100) I. Retail sales of consumer goods 101.3 Including: Food 103.1 Fuel 97.9 Articles of everyday use 99.0 Cultural and educational articles 100.2 Medicine 101.3 Fuel 100.8 II. Purchases of agricultural and sideline products 100.2 Including: Cereals 100.0 Industrial crops 100.1 Livestock products 100.3 Other agricultural and sideline products 100.8 - 102 - Among the above figures only the prices of food and clothing have changed significantly; changes in the prices of the other commodities are very slight. In terms of the peasants, the price indices for food consumption are considerably lower than the figure in the table. This is because peasant food consumption is largely self-supplied, and the self-supplied portion is calculated according to the State-operated industry and commerce list prices for agricultural and sideline products. The above price indices are national averages and not for Hubei. But according to personnel in the statistical agencies price indices for Hubei are only very slightly different from the nation as a whole. Given the above circumstances, combining 1981 and 1982 data into a cross-section sample will not create any major deviation. We use formula (2.1) to describe the laws of peasant demand, then calculate the parameters according to the relevant formulas in section 2, weighting the calculation for number of people. Here we shall first summarize the 1981 and 1982 data. For parameter calculation, see Table 58. w I-ble 56: samiary of Smmple Survey DEta: 1981 Hubei Peasant Hbisehold lIacom Per Capita Cbnsmpticn (Y/Year) Number Full- and Per capita Cultural Numzber of of part-time ret income Tbtal Article of and Income grap households people jobholders (v/year) expenditure Food Clothirg Fuel HDusirg* everyday use life services Under lOOY 2B 179 58 91.91 100.78 70.91 10.35 6.53 3.28 6.07 3.03 100-15UY 149 899 372 129.32 122.50 82.92 13.12 7.68 7.08 7.79 3.90 150-200Y 255 1461 690 186.53 166.85 109.88 18.85 10.03 10.56 11.98 5.55 200-300( 374 2260 1046 233.94 200.67 122.86 25.72 10.71 18.15 17.27 5.85 300-400V 114 570 295 343.12 237.30 137.71 30.06 11.99 30.79 20.72 6.02 400-500V 24 105 62 444.01 339.67 182.96 48.02 3D.83 35.54 40.59 11.78 Over 500V 10 27 23 721.81 537.52 293.44 98.56 33.30 31.89 71.26 9.07 Tbtal 954 5497 2543 Awerage 217.44 183.78 114.75 22.58 10.34 15.56 15.05 5.51 * ltis is the exriture for this year on house repair and construction, nDt the actual hiusi%g conmmption. Table 57: 9ummary of Sample Survey Data: 1982 Iibei Peasant Hbusebold Irxom- Per Capita Ccasumption (Y/Year) Number Full- and Per capita Oultural Number of of part-time net ihcnme Tbtal Article of and Income group xusehiolds people jbbolders (V/year) expenditure Food Clothing Fuel Hbusing* everyday use life services Urder lOOY 1 4 1 96.00 92.00 77.50 2.50 7.50 0.50 2.75 1.25 100-150Y 28 170 60 138.44 127.52 86.04 11.09 10.84 8.78 7.13 3.65 150-200W 124 739 317 185.88 164.64 111.13 17.85 10.81 9.20 11.52 4.95 200-300N 42D 2525 1168 246.79 203.61 131.59 24.38 10.93 14.07 16.75 5.89 300-400( 241 1313 683 349.29 268.05 161.86 36.33 14.34 21.35 27.46 6.70 400-50CY 98 473 261 438.78 314.45 176.69 42.56 15.59 32.98 37.23 9.27 Over 5OOY 36 140 94 598.97 420.74 2D6.74 50.81 18.35 89.22 47.81 7.81 Ibtal 948 5364 2584 Average 286.07 226.96 140.63 28.26 12.35 18.53 20.95 6.23 * This is the expenditure for that year on house repair and construction, mot the actual housing consumption. The housing expenditure for 1982 in the highest income group is extraordinarily high, perhaps due to an abnonnal sample point. In calculating the housing demand function below we get tWD sets of parameters: the first is calculated according to all sample points, and the second is calculated after dropping this abnonnal sample point. Table 58: Demand Function Parameters Calculated According to 1981 and 1982 Combined Data -2~~~~~~~~~~~~~~~~ a B aB* -R2 Piyi Food 54.8749 0.2891 0.00070 0.9394 92.63 (94.26) 0.4680 (0.4871) Clothing 0.9693 0.0971 0.00024 0.9360 13.65 (14.20) 0.1572 (0.1636) Fuel 4.8758 0.0257 0.00010 0.8546 8.23 (8.38) 0.0416 (0.0433) -8.8759 0.1033 0.00057 0.7515 Housing (-3.4238) (0.0791) (0.00034) (0.8350) 4.62 (7.35) 0.1672 (0.1333) Articles of everyday use -4.7786 0.0905 0.00018 0.9576 7.04 (7.55) 0.1465 (0.1525) Cultural and life services 2.8625 0.0120 0.00007 0.7440 4.43 (4.50) 0.0194 (0.0202) 49.9280 0.6177 (55.3801) (0.5935) 130.60 (136.24) 1.0000 (1.0000) Note: Figures in parentheses are those calculated after dropping the housing expenditure in the 1982 highest- income group. - 106 - For purposes of comparison, we use the combined 1981 and 1982 data to calculate Engel's double logarithmic function. The function formula is: lnVi = a + blnV. Results are given in Table 59. Table 59: Parameters of Engel's Double Logarithmic Function Calculated According to Hubei Cross-Section Data Food 0.6296 0.7936 0.00105 0.9815 Clothing -3.6840 1.2969 0.00197 0.9756 Fuel -1.3975 0.7193 0.00235 0.8962 Housing -6.5951 1.7586 0.00660 0.8674 Articles of everyday use -5.7044 1.6070 0.00182 0.9862 Cultural and life services -2.1583 0.7385 0.00278 0.8665 From the above statistics we see that the results in Table 59 are -2 better than those in Table 58, because its R value is higher. However, this type of Engel's function cannot reflect the effects of changes in relative price, so this article will not use it for demand analysis and forecasting. Below we shall use the second set of figures in Table 48 as the basis for further analysis of several questions. Basic Demand: The values of Pi iin Table 48 shows that basic demand of peasants must be essentially equal to the consumption level of peasants with per capita incomes of 140w; the people with incomes of less than 100Y and three-fourths of the people in the 100-150M income group have incomes lower than basic demand. In terms of the Hubei data, the number of people whose 1981 income is lower than basic demand accounts for approximately 15.4 percent of the total - 107 - number in the sample; the number of people whose 1982 incomes are lower thian basic demand account for approximately 2.5 percent of the total number in the sample. This reflects a major rise in peasant income froma 1981 to 1982. Marginal consumption trend: In Table 58 is about 0.6; this means that if peasant income increases by 14, then consumption increases by only 0.60, with the remaining 0.40 being used to increase savings, for cash on hand, or to purchase fixed assets. Compared with Beijing staff and workers, Hubei peasants marginal consumption trend is much lower. We believe there are three main reasons for this. (1) In these two years the system of fixing output quotas for each household began to be implemented, and peasants had to purchase some productive fixed assets, whereas staff and workers basically did not have this type of expenditure. (2) Peasants do not have free health care or retirement pay; with increasea income they can increase the funds they set aside for health care and retirement, whereas staff and workers basically have no need to do so. (3) In these two years peasant income rose very quickly, but there were significant differences in the rate of income increase from household to household. Under these circumstances, the marginal consumption trend as calculated according to cross-section data may be lower, because of consumption habits (adjustments in consumption cannot immediately keep up with changes in income). Total expenditure elasticity and price elasticity of demand: Considering that intersecting price elasticity is only important in food prices, we have here given only food-price elasticity for intersection price elasticity. In addition, the 1981 and 1982 average values were used to calculate V and Vi values required for elasticity. - 108 - The figures in Table 60 are similar to those in Table 52; the figures in Table 61 are also similar to the corresponding figures in Tables 53 and 54. This shows that worker and peasant incomes are subject to the same laws of distribution. Table 60: Total Expenditure Elasticity of Demand Among llubei Peasants (n i) Item Total expenditure elasticity Food 0.7834 Clothing 1.3219 Fuel 0.7838 Housing 1.6056 Articles of 1.7412 everyday use Cultural and 0.7064 life services - 109 - Thble 61: Price Elasticity of Dmand AmDng abiei Peasants Articles of Cultural and Iten Food Clothirg Fuel Housirg everyday use life services Identical price elasticity: AssuLing total ocenditure -0.6209 -0.5321 -0.2925 -0.6259 -0.6438 -0.2482 is constant Inoome compensation -0.1338 -0.3685 -0.2492 -0.4926 -0.4913 -0.2280 Fbod price intersecting elasticity: Asmning total expenditure - -0.6075 -0.3602 -0.7379 -0.8002 -0.3246 is constant Inoome aonpensation - 0.2146 0.1269 0.2769 0.2824 0.1134 - 110 - E. Forecasting Consumption Structure In this section we shall make use of the preceding demand model to forecast future consumption structure among urban and rural dwellers. In this forecasting we must take into account changes in income as well as ctianres in relative prices. The forecast of national income in China in the year 2000 in the article "A Forecasting Model for Production Starts with Funds as Limiting Condition" (He Juhuang, Quantitative Technical and Economic Research, 1983, No. 11) will serve as the basis for forecasting changes in income. In that article, China's national income in the year 2000 is forecast at approximately 1,200 billion yuan, at an average annual rate of increase of 6 percent. If the accumulation rate and the ratio of people's consumption to social consumption remain essentially constant, then the annual rate of increase in people's total consumption is also six percent; subtracting a one percent rate of increase in population, the annual rate of increase in per capita consump- tion is five percent. Assuming an appropriate reduction in gaps between workers and peasants, the annual rate of increase in per capita staff-and- worker consumption can be taken as 4.7 percent, with an annual rate of increase of 5.1 percent for per capita peasant consumption. Some preliminary calculations of rational price adjustments by price control agencies serve as the basis for changes in relative prices. In the view of some experts, with adjustments in prices according to average profits, fuel prices will rise at least 100 percent, and average prices of farm products will rise by at least 50 percent. We therefore assume that by the year 2000 food prices will have risen by 50 percent over what they are now, - ill - and fuel prices will have risen 100 percent, with prices of other commodities remaining constant. Although our forecast takes into account changes in relative prices, calculation of future consumption structure still involves using base-period prices. The demand forecast equation meeting this requirement is as follows: * * ~'-1 * Vt Ptr (vt 1' Ptr ) B * vt =Vo (1 + g) (5.1) 1'? ptv Pt I . t in which: Vt = per capita demand vector in year t calculated at base-period prices; * r = per capita basic demand vector calculated at base-period prices; vt = per capita total expenditure in year t calculated at base-period prices; Pt = diagonal alignment of price indices for all commodities in year t; Pt = total price indices for consumption in year t (consumption in year t as addend); 1 1 1- 1. I - 112 - B = marginal budget share vector; VO = base period per capita total expenditure; g = annual rate of increase in consumption. Below is a forecast of consumption structure in staff-and-worker and peasant households. Forecast of Staff-and-Worker Household Consumption Structure The demand function we obtained from the Beijing data serves as the basis for forecasting nationwide staff-and-worker household demand. This is because the income levels and consumption structure among Beijing staff and workers are representative of staff and workers nationwide. In the forecast housing demand will continue to be set aside, inasmuch as we have no empirical laws regarding staff-and-worker household housing demand. The data we already know are: r* B t 19.87 0.5422 1.5 4.10 0.1807 1.0 0.77 0.0088 2.0 4.25 0.2145 1.0 2.65 0.0537 1.0 VO = 38.66 (per capita monthly domestic consumption minus housing among staff and workers nationwide, 1982) g = 0.47; t = 2000 - 1982 = 18. These figures are entered in the equation (5.1), producing: 45.0 (food) 16.7 (clothing) 1.1 (fuel) V*2000 = (19.2) (everyday articles) 6.4 (noncommodities) 1'V*2000 = 88.4 - 113 - This includes the effect of changes in income and relative prices, with a forecast value of the per capita monthly consumption vector (not including housing) for staff-and-worker households in the year 2000 calculated at base-period prices. For a comparison of its composition with the survey data on nationwide urban staff-and-worker households in 1982 (see Table 62). Table 62 does not include housing. If housing is added in, the figures in the table would all be slightly lower. Forecast of Peasant Household Consumption Structure We shall use the demand function parameters (second set) obtained from the Hubei data as the basis for forecasting nationwide peasant demand, inasmuch as Hubei peasants' income and consumption are representative of the nation's peasants. The data we already know are as follows: r* B Pt 94.26 0.4871 1.5 14.20 0.1636 1.0 8.38 0.0433 2.0 7.35 0.1333 1.0 7.55 0.1525 1.0 4.50 0.0202 1.0 Vo 162.21 (annual per capita domestic consumption expenditure among peasant households nationwide, 1980) g 5 0.051; t = 2000 - 1980 = 20. Entering these figures in the equation (5.1), we get: 214.41 (food) 74.73 (clothing) 16.39 (fuel) V*2000 = (56.67) (housing) 63.98 (everyday articles) 11.97 (cultural and life services) - 114 - Table 62: Staff-and-Worker Household Consumption Structure Item 1982 2000 Food 0.595 0.509 Clothing 0.146 0.189 Fuel 0.019 0.012 Articles of everyday use 0.171 0.217 Noncommodities 0.069 0.072 (not including housing) Total 1.000 1.000 This includes the effect of changes in income and relative prices, with a forecast value of the per capita annual consumption vector for peasant households in the year 2000 calculated at base-period prices. For a com- parison of its composition with survey data on peasant households nationwide in 1982, see Table 63. Finally, we should point out the following: (1) the linear expen- diture system used above to forecast consumption structure is not very suitable for long-term forecasting with incomes undergoing major chlanges; if we do not take into account the effect of changes in relative prices, it may be somewhat better to use a double-logarithmic Engel's function for fore- casting. However, we believe that the impact of major changes in relative prices on demand is not to be overlooked. We have therefore used a linear - 115 - Table 63: Peasant Household Consumption Structure Item 1982 2000 Food 0.605 0.489 Clothing 0.113 0.171 Fuel 0.056 0.037 Articles of everyday use 0.102 0.146 Cultural and life services 0.022 0.027 Total 1.000 1.000 expenditure system capable of reflecting this impact as our forecasting tool. (2) Because the average income for the forecast period greatly exceeded the average income in the sample period, the accuracy of the above forecast cannot be very high. Some commodities (such as motorcycles and refrigerators) can be afforded by very few people at the income levels of tthe sample period, whereas there may be more people who can afford them at the income levels of the forecast period. On this point, the fuel consumption we have forecast for staff and workers and peasants may be somewhat low. - 116 - VI. SOME POLICY QUESTIONS REGARDING CHINA'S CONSUMPTION STRATEGY AND CONSUMPTION STRUCTURE China's original consumption strategy has had some merits and acIih'vements over its history, but some improvements must now gradually be In Old China the vast majority of the people suffered hunger and 0old, average life expectancy was only 35 years, and the degree of poverty and baickwardness was unique in the history of the world. After the founding of New China, the people's government under the leadership of the Communist Party took as its primary strategic objective the ensuring of the basic needs of the people's lives; it took as its basic measures the socialist transformation of city and village and the development of agricultural and industrial production. This brought about gradual improvements in the people's standard of living and rather quickly eliminated situations of dire poverty. People entered a phase in which they could dress warmly and eat their fill ,/ and average life expectance was pushed to 69 years. This was the great achieve- ment of China's original consumption strategy. However, in terms of the development of production and its continuity following the establishment of the People's Republic, improvements in the Chinese people's standard of living and in the quality of life have been less 1/ Some comrades in Chinese economic circles assert that improvement of the people's standard of living passes through four stages: poverty and hunger, meeting of the basic necessities, modest comfort, and abundance. China has now gone from the pre-Liberation poverty stage to meeting the basic necessities, and it is now moving towards modest comfort. This assertion, viewed in terms of the meaning of the Chinese terms in question, accords fairly well with reality, and it has the advantage of being succinct and general. Naturally it would be best to have a series of quantitative indicators. - 117 - than ideal. Moreover, in terms of future development targets and the realization of the future as forecast, the entire series of policies under our original consumption strategy should be conscientiously studied and the appropriate changes made. As everyone knows, people's consumption is an important component of the nation's economic life and a key stage in social reproduction, affecting all major programs and strategic policies in national economic development. It undoubtedly has an important impact on consumption levels and changes in structure. For example, national income production maintains a certain rate of development on the basis of continued improvement in labor productivity; by the year 2000 it must be maintained at an annual rate of increase of 6 to 7 percent, otherwise it will be difficult to realize our strategic consumption targets; the distribution proportions of national income applied to consump- tion and accumulation require that the consumption rate be regularly kept above 70 percent; if most of the nation's current human, material and financial resources are applied to the production of the means of production and the production of the means of subsistence, then how shall State invest- ments be distributed among agriculture, light industry and heavy industry; to what extent should the quantity and variety of society's natural material resource potential be used to ensure the demands of popular consumption and the demands of social production, and what is the rational proportion for distributing the two; population size and structure, including age, sex, educational leave, profession, etc.; as well as foreign trade. These are all important problems in the macro-control of the national economy; this article cannot cover all of them in detail, and we shall focus our attention on a - 118 - number of policy questions directly affecting China's consumption patterns, such as welfare policy, consumption structure policy, etc. Regarding the characteristics of China's consumption patterns, the cities may be summarized as having a semi-rationing system, and the villages as well have some aspects of a rationing system. In fact, although most domestic consumption among China's urban dwellers relies on income from remuneration for labor, there is also a portion which relies on social welfare and financial subsidies. Before the production responsibility system was implemented among China's peasants, a considerable portion of individual income was distributed gratis by population or in the form of low prices. According to surveys, in some areas this type of free distributed income accounted for 15-30 percent of individual income. Now, although this problem basically no longer exists, urban and rural social welfare and financial subsidies, whether in terms of absolute figures or proportion of total consumption, are tending to expand year after year. People are now feeling more and more that this means of ensuring that "everyone is warm and eats his fill" and the consumption policy aimed at levelling out the standard of living are now clearly no longer entirely appropriate in a situation where "everyone being warm and eating his fill" has basically been achieved. A. The Problem of the Ratio of Social Consumption Funds to Individual Consumption Funds We know that the portion of national income which is used for consumption, i.e., consumption funds, is composed of collective social consumption funds and individual consumption funds. The proportion of social consumption funds in social consumption funds has risen from around 8 percent during the First Five-Year Plan period to about 11 percent now, while the - 119 - proportion of individual consumption funds has dropped from around 92 percent down to about 89 percent. This is undoubtedly the result of the rate of growth in social consumption funds exceeding that of individual consumption funds. Under a socialist system, This trend, generally speaking, conforms to economic laws. However, there must be a rational limit to it. This is because the use of social consumption funds for the people's consumption basically involved everyone eating out of a "big pot of rice" together, with free health care, housing subsidies, and so on. If the quantities in this "distribution according to need" are too big, the socialist historical production capacity levels and the people's ideological consciousness are exceeded, so that in areas where originally distribution according to labor should have been followed and individual income should have been relied on to solve the problems of life, this was replaced by collective social consumption, or "eating out of a big pot of rice," which may well lead to the masses being encouraged to think they should rely on the State for their livelihood, thus damping people's enthusiasm for being self-sufficient in solving the problems of life and creating severe material waste, which ultimately is not beneficial in raising the people's standard of living. We therefore believe that a full and conscientious effort should be made to ensure that the total individual income created by distribution according to labor serves as the quantitative basis for determining the individual consumption funds that can be used out of national income, thereby determining the proportions of the two types of consumption funds. Naturally, if this theoretical limit is to be carried out in actuality, a great deal of analytical and computational work must be done, and this requires the efforts of many people. However, by clearly defining the guiding ideology behind this - 120 - limit, we can at least wake people up; we must not start by abstractly "expanding communist factors," lopsidedly accelerating the rate of increase in social consumption funds and their proportion in national income. Clearly under China's socialist system, in terms of ensuring the necessary social consumption funds, such as workers' old-age pensions, stipends for disabled armymen, and so on, we absolutely cannot brush them aside; this is an unshakable principle. B. The Problem of Price Subsidy Policy We know that the financial subsidies and labor insurance benefits which are directly related to people's consumption are two important channels for getting social consumption funds into the people's actual consumption. The vast majority of China's current financial subsidies are price subsidies (including income subsidies to staff and workers when prices are raised). These are subsidies which the State gives to industrial and commercial enterprises or to consumers when the prices for certain commodities are irrational. When these commodities are essentials, the price subsidies directly affect the people's standard of living. China's price subsidy policy was originally intended in the First Five-Year Plan period to support medium and small enterprises and the development of industry in backward areas; subsidies were aimed primarily at enterprises suffering losses. Later, after the purchase prices for some agri- cultural products and the ex works prices of raw materials and fuels had risen, no corresponding hike in sales prices were made in order to stabilize the people's standard of living. Therefore it was necessary to apply subsidies to the losses suffered by the commercial agencies, or to give price and income subsidies to staff and workers after appropriate price adjustments - 121 - had been made, these staff-and-worker income subsidies including grain price subsidies, coal price subsidies, and nonstaple food price subsidies after price rises for these items. In 1982 the various categories of price subsidies in China accounted for about 74 percent of all financial subsidies. Of the nation's price subsidies, the vast majority, over 75 percent, are inverted purchase price subsidies for cereals, oils, meat, poultry, eggs, cotton, pigskins, etc.; in second place, approaching 18 percent, are loss subsidies for exported and imported cereals, cotton, sugar, chemical fertilizers, and agricultural chemicals; the remaining portion are for agricultural means of production supplied to agriculture and forestry at favorable prices. This illustrates that State price subsidies are directly and indirectly related to living standards. Before 1979, this price subsidy policy, with wages and prices essentially frozen, had an effect on ensuring the basic stability of the people's standard of living. However, it also covered up many problems with wages and prices that accumulated because of this freeze. For example, wages (including all types of labor remuneration) strongly violated the principle of distribution according to labor, and this greatly reduced the incentive effect of wages on labor; prices strongly violated the principle of exchange at equal value, and thus greatly weakened the adjusting effect of prices on economic activity. After 1979, when we raised the prices of agricultural products by a wide margin and reduced the price scissors differential between industrial and agricultural products and correspondingly ensured the stability of urban grain prices, the State financial organs allocated large sums of money to compensate for the deficits created by purchase price inversions. Added to this were the - 122 - subsidies for nonstaple foodstuff price hikes and other price subsidies, and the financial burden became heavier and heavier, reducing the funds the State could use for economic and cultural construction. Since 1980, the proportion of price subsidies in financial expenditure has reached one-third. Clearly we have already reached the critical point. Hence we agree with the views of many economic workers and economic theorists who are proposing that we should proceed gradually and in stages to cut back the proportion of subsidies to below ten percent. The method consists of raising those prices which should be raised (carrying out a reform of the pricing system), and transferring those subsidies which should be given to wages (carrying out a wage reform). It is possible to proceed from the easiest to the most difficult, from minor goals to major goals. For example, price-hike subsidies for nonstaple foodstuffs can first be converted to wages; then a further step can be taken, raising the sales prices of edible oils and cotton, eliminating the subsidies, and correspondingly increasing wages, and proceeding gradually in this way, until only those subsidies which cannot be eliminated are left. The key to determining whether a subsidy can or cannot be eliminated is whether, when both prices and wages have been correspondingly raised, the living standards of some people will decline or whether they will actually lose the means to support themselves. C. The Problem of Staff-and-Worker Labor Insurance Benefits Policy The cost of labor insurance benefits for staff and workers in enter- prises under collective ownership and under the system of ownership by the whole people, State organs and institutions is another major component part of the social consumption funds used out of national income. This includes wel- fare subsidies, labor insurance and labor protection. - 123 - Welfare benefits also consist of two parts. The first is collective welfare facilities (such as staff-and-worker housing, dining halls, nurseries, day-care centers, kindergartens, bathhouses, etc.) for which there is a low charge, the balance of which is made up by enterprise staff-and-worker welfare fund subsidies. The second is individual welfare subsidies, including staff- and-worker winter heating subsidies, home leave travel subsidies, transporta- tion cost subsidies, difficulty assistance, and so on. Labor insurance includes free health care, injury and disability relief, retirement pay, burial and bereavement pension, maternity leave wages, and collective labor insurance institutions: sanatoriums, rest homes, etc. Labor protection primarily involves labor protection gear and health care subsidies given to staff and workers according to the type of work they do. Experience over many years has shown that these labor welfare systems China has established have had a major impact on enriching the free-time cul- tural life of staff and workers, easing the burden of everyday life, resolving special difficulties encountered in birth, old age, illness, death, injury and disability, and protecting the safety and health of staff and workers in the labor process. However, in the last few years the labor welfare costs paid out by units under the system of ownership by the whole people have risen from 6.69 billion yuan in 1978 to 15.38 billion yuan in 1982, which corresponds to 21.7 - 124 - percent of total wages for that year.1 As reform of the economic system proceeds and the principle of distribution according to labor is more thoroughly implemented, a number of abusive "big rice pot" approaches have come to light in some labor protection welfare systems which have been extremely helpful in getting people to perceive that rapid increases in the amounts of labor welfare have led to waste in distribution. Here we shall concentrate on the issues of free health care and housing subsidies. First let us examine the free health care system. In 1982 China's free health care costs had reached 4.4 billion yuan. Free health care in the State budget amounted to about 30M per person per year, but in actual practice this was exceeded, and the figure reached more than 40V. This creates severe overspending on free health care costs. In addition to price hikes for some medicines, high costs of new medical treatments, the higher proportion of older staff and workers and other factors, the main reason for this was the complete detachment of the free health care system from individual benefits; with individuals being treated at hospitals and all costs paid, people were encouraged to come and "eat out of the big rice pot," creating an extreme waste of health care and medicines as well as a waste of staff-and-worker time. We might consider: (1) raising the registration fee from the original 0.1V to 0.5V or 1.OV; the original M.1y is too low and does not have much of a controlling effect. (2) all those who do not see a doctor for six months or a year could be given a bonus; for example, 10-30 percent of the free health care budget (30Q per person per year) could 1/ "China Statistical Yearbook, 1983", page 491. - 125 - be given, i.e., 3-9V. In consideration of aged staff and workers, this approach might be applied only to staff and workers now in service. Next, let us examine housing subsidies. Housing for Chinese peasants is what they themselves build and utilize without paying rent, so there is no housing subsidy problem for them. Except for a small amount of privately owned housing, urban housing is public housing for which a low-rent system is used. The average living space in China's urban housing is now 5.0 m2 per person, and in rural areas the figure does not exceed a little more than 10 m2. For many years there has been a severe urban housing shortage, and we have adhered to a policy of low rents and high housing subsidies. With the housing supply falling ever shorter of demand, these subsidies have had many negative results. The housing and housing subsidy situation varies in China from one place to another, so it is difficult to arrive at uniform figures for the whole country. But everyone recognizes that rents should at least meet the demands of depreciation, renovation and management costs. Overall, 4.51 in rent should be collected for every square meter of living space per year, which would meet this demand. But the annual rent collected at present is only 1.21/m2, which covers only one-half of the 2.61/m2 in annual maintenance and management costs; the remaining 1.41/m2 is fully paid by State financial subsidies. With individuals paying too little rent and the State offering housing subsidies, when there is a severe housing shortage some people may engage in the improper practice of grabbing up housing; moreover, this may artificially reduce the proportion of housing costs in household consumption expenditures, which hinders improvement of consumption structure and reduction - 126 - of buying-power pressure on food and clothing. If we further take into account that China's housing construction will now be developing at a faster rate, with a requirement of 8-10 m2 of living space per person for urban dwellers in the year 2000, the magnitude of the investment required is such that China's financial organs will have difficulty bearing it. Therefore the current low-rent high-subsidy housing policy has reached the point where it must be reformed. It is now time for the many proposals offered by those in China's economic circles to be examined and sorted out. The first step may be to raise rents to the point that they are able to make "housing support housing," i.e., the rent collected should be adequate to carry out maintenance of existing housing. At the same time wages should be increased accordingly. For the second step, we may consider doing away with the housing management tasks for enterprises and institutions, turning housing over to municipal housing authorities for management and further raising rents, so that the housing authorities are able to gradually go from a loss situation to an income and expenditure equilibrium; later a profit could be made. At the same time, wages should again be raised accordingly. This will convert housing subsidies into staff-and-worker wages. When rents have been raised to the point that "housing supports housing" and a slight profit is made, and staff and workers are able to pay rent by relying on their own wages, we shall begin to have a basic situation in which housing can be made a commodity. And ultimately we can consider making housing a commodity. - 127 - D. The Problem of Consumption Structure Policy China's present consumptions structure is the result of many years of economic, social, technical and cultural development; it may be conditioned by natural resources and ethnic tradition as well as international influences, and it may be even more conditioned by the influences of government consumption strategy and consumption structure policy. The latter -- government strategy and policy -- falls within the realm of people's wishes, wills, decisions and other subjective dynamic activity. It is not difficult to see what a great impact a correct consumption structure policy can have on the future quality of life of the people. China's current consumption structure - taking 1982 as our example -- provides consumers with a daily intake of nearly 2,800 calories, slightly more than 80 grams of protein, and slightly more than 44 grams of fat, which is enough to meet the requirements of human existence .1! Each person is provided with 30 chi of cloth to consume annually,2!/ which is enough to keep people warm. Thus the people are ensured of a standard of living which keeps them warm and fed. However, the consumption structure as it now exists has a number of points that clearly need to be improved. For example, the gaps between urban and rural situations are still quite significant; there is severe artificial pressure on housing; the nutrition structure could be improved; cultural and domestic consumption levels are low. 1/ "China Statistical Yearbook, 1983", page 509. 2/ "China Statistical Yearbook, 1983", page 507. - 128 - 1. Differences in urban-rural consumption structure reflect gaps between worker and peasant. First, peasants have a very high degree of self-sufficiency in their household consumption structure. For example, in a 1982 survey of national peasant family livelihood, the proportion of commodity expenditures in peasant domestic consumer-goods expenditures was such that, if the various categories of expenditures are taken as 100, the proportion of commodity expenditures for food was 38.0; the proportion of commodity expenditures for clothing was 98.4; for fuel, 22.7; for housing (construction materials), 96.1; for domestic articles and other, 98.7. Of total domestic consumer-goods expenditures, commodity expenditures accounted for 56.5 percent 11 But urban staff-and- worker household relied entirely on cash income for their household domestic expenditures. Second, in peasant consumption structure housing accounts for more than 10 percent, while for urban staff and workers it is only around 1.5 percent. 2 For their housing, peasants rely entirely on their own physical and financial resources to purchase construction materials, as well as for building, maintaining and managing their homes. But urban staff-and-worker household rely almost entirely on rented public housing, paying only a little more than O.1I per square meter per month. Urban housing distribution has become a welfare measure. Third, in terms of nutrition and food consumption levels, there are also considerable gaps between urban and rural conditions. First, let us look at the nutrition angle. 1/ "China Statistical Yearbook, 1983", page 501. 2/ "China Statistical Yearbook, 1983", page 495, 501. - 129 - Table 64 shows that in terms of calorie intake rural areas correspond to only 87.6 percent of the urban level, while protein in rural areas corresponds to only 92.6 percent of the urban level; for fat, the figure is 54.2 percent. Table 64: Nutrients Absorbed from Food Per Person Per Day, Urban and Rural Areas, 1982/a Nationwide Urban Rural Calories 2,779.0 3,087.9 2,707.2 Protein (grams) 80.5 85.8 79.5 Fat (grams) 44.4 70.6 38.3 /a "China Statistical Yearbook, 1983", page 509. Next, let us look at consumption levels for major foods. Table 65: Comparison of Average Per Capita Major Food Consumption among Peasant and Staff-and-Worker Households, 1982/a Peasants as percent Peasants Staff and workers of staff and workers Cereals (jin) 520* 289.11* Edible oils (jin) 6.86 11.55 59.4 Meat (jin) 18.10 37.32 48.5 Eggs (jin) 2.85 11.75 24.3 Sugar (jin) 2.37 5.59 42.4 * For peasants the figures are for raw grain, while for staff and workers they are for processed grain, so the two are not comparable. Note: Table 65 is from household income and expenditure sample survey data. /a "China Statistical Yearbook, 1983", pages 496, 502. - 130 - Although the figures given here are not entirely comparable, they clearly show the gap between urban and rural consumption levels for major foods. Our approach here should be to continue the various policies aimed at gradually reducing the urban-rural gap, such as developing rural diversification, expanding commodity production and exchange, and so on. Naturally, urban nutritional levels should also continue to be improved. 2. The proportion of housing in China's consumption structure has been artificially depressed. This is primarily because of the State's financial subsidies for urban housing which have not been entered as people's housing consumption. Moreover, even though the subsidy is figured at 1.4w per m2, raising it from 1.2V to 2.6V still only deals with the maintenance and management costs while ignoring depreciation, which does not raise taxes or profits. Therefore, our housing authorities should be enterprise agencies, but to date they are only subordinate organs of welfare agencies. Looking at the situation now, we can see that in the early days after Liberation when the people's government was just being established, such a national contractual welfare-type housing strategy may have been useful in solving the urgent needs of homeless urban workers and the destitute. But to continue in this way after the people's urgent housing needs have basically been met does more harm than good, and perhaps only harm and no good. We have done some analyses of this above and will not repeat them here; we only wish to point out that if we consider our target of 10 m2 per person for urban living space in the year 2000, our present urban housing policy absolutely must be changed. If it is not, State investments will not be forthcoming (it is not that there will be absolutely no investments, but with limited total investments the proportion of housing construction investment cannot be very great or they will encroach - 131 - on the need for other types of construction); if it is not, the financial organs will not be able to bear the housing subsidies; if it is not, there will be other negative results that will be difficult to avoid. Clearly, adopting the path of increased rents, higher wages, and making housing a commodity is the correct approach. 3. In our consumption structure, nutritional structure is also a problem that merits our earliest possible attention. This not only affects the people in all of China's present age groups, it also has to do with the eugenic health and growth of future generations. In terms of daily intake of calories and protein, China has now reached the level of a moderately developed country; but comparatively speaking we are still far below the developed countries. See the per capita daily nutrition intake figures from the National Bureau of Statistics. Table 66: Per Capita Daily Nutrient Intake in Several Countries Calories (calories/ Protein (grams/ Fats (grams/ /day/person) /day/person) /day/person) World (1977) 2,571 68.8 62.2 China (1980) 2,592 77.3 39.0 India (1978-1980 average) 1,998 48.5 29.5 Soviet Union (1977) 3,460 103.4 101.5 USA (1978-1980 average) 3,652 106.7 169.2 It should be pointed out that the figures in Table 66 are out of date. Since 1977 it may be said that the world nutritional situation has - 132 - changed, and China's has changed even more markedly. According to 1982 statistics, the nutrients which the Chinese people take in from the food they consume has increased sharply over 1980: calories have gone up to 2,779, protein up to 80.5 grams, fats up to 44.4 grams. That is, leaving aside fats, calories and protein levels have surpassed 1977 world levels. However, compared with the developed countries, we are still far behind. Nonetheless, we must not blindly pursue them. What we should pursue is nutritional science to benefit physical health, a nutritional level and structure in keeping with China's own natural, economic and cultural characteristics. In this regard, we have a whole series of shortcomings: average per capita daily intake of calories is still somewhat low, especially in rural areas. Protein and fat intake is even more inadequate. Moreover, in terms of the structure of the sources of the aforementioned three nutrient products, the proportion of animal products is rather low, while the proportion of vegetable products is too high. There are two sets of data that prove this. First, in China's food consumption structure as calculated by comrades in the Ministry of Agriculture, Animal Husbandry and Fishery according to 1980 data, vegetable products account for 93.9 percent and animal products for 6.1 percent. The world levels are 83.1 percent for vegetable products and 16.9 percent for animal products. Second, there are statistics for 1982 from the National Bureau of Statistics. These statistics divide the three nutrient categories taken in from per capita daily food consumption into animal and vegetable product sources. The figures in Table 67 show the following proportions for the three nutrient categories taken in from animal products in China in 1982: calories, - 133 - 7.7 percent; protein, 7.1 percent; fats, 47.5 percent. According to the data from the comrades at the Ministry of Agriculture, average 1977 world levels for intake from animal products were: calories, 16.9 percent; protein, 34.7 percent; fats, 53.8 percent. This means that the average worldwide proportion of animal products in 1977 was higher in all categories than our 1982 levels. Table 67: Proportions of Nutrients Supplied by Animal and Vegetable Products Absolute Figure Percentage Calories Total 2,779.0 100.0 Animal 214.7 7.7 Vegetable 2,564.3 92.3 Protein (grams) Total 80.5 100.0 Animal 5.7 7.1 Vegetable 74.8 92.9 Fats (grams) Total 44.4 100.0 Animal 21.1 47.5 Vegetable 23.3 52.5 Furthermore, some nutritional experts have found that in addition to an inadequate supply of animal protein there is a slight insufficiency of calcium, phosphorus and iron. This is especially harmful to child development. It should be pointed out that the low proportion of animal products and the low fat intake may be of value in preventing cardiac and circulatory - 134 - disease, but it may not be beneficial for human physical and mental development. Naturally we cannot imitate the developed countries in pursing excessive calorie and animal product intake; it is better to pursue goals established by nutritional science. Therefore our nutrition structure policy should be aimed at raising the quantities of all nutrients, while at the same time increasing the proportion of animal products; particular attention should be paid to promoting those foods rich in protein, calcium, phosphorus and iron. 4. There is another clear inadequacy in China's consumption structure, and that is the low proportion of cultural consumption. This reflects the slow development over many years in China of education, science and cultural activities; the material conditions demanded by the people's cultural life are wanting. This also reflects the fact that China's popular cultural and scientific quality is still low. In terms of policy, we fear that it is related to the people's inadequate comprehension of the immense economic and social benefits of improving the quality of popular culture. A vivid example of this: according to a survey of income and livelihood differences among 60 peasant households in Xinjin County done by the Sichuan Bureau of Statistics, peasants with high cultural levels as compared with those with low cultural levels are characterized by the following: Income levels are higher. Per capita 1982 net income among peasant households with senior middle school educational levels was 409.89V; among those with a junior middle school educational level, it was 327.35V; among those with an elementary school educational level, it was 285.561. Peasant households with a senior middle school educational level had incomes 25.21 - 135 - percent higher than those with a junior middle school educational level, and 43.19 percent higher than those with an elementary school level. Income increases faster. Per capita net income over the last four years increased 184 percent for peasant families with a senior middle school educational level, 110 percent for those with a junior middle school educational level, and 94 percent for those with an elementary school educational level. Farm management levels are higher (more so-called get-rich skills). Peasant families with elementary school educational levels engage solely in traditional production projects, whereas those with junior and senior middle school educational levels are beginning to engage in large-scale breeding activities and planting industrial crops. Economic benefits of production are higher. In 1982 the net income generated by each laborer was 788.26Y among peasant families with senior middle school educational levels, which was 24.49 percent higher than for junior middle school educational levels and 24.78 percent higher than for elementary school educational levels. According to another survey of 40 households, the income generated by every one yuan invested was 4.531 for senior and junior middle school educational levels and only 3.86Y for those with elementary school educational levels. In addition, in terms of consumption structure, the proportion of costs ;expended on eating by peasant households with a higher cultural level is lower than that among peasant households with a lower cultural level, while - 136 - the proportions for clothing, housing and articles of everyday use are somewhat higher.±/ It is clear that universal education and raising the people's cultural and scientific levels is a key means of improving the construction structure of the Chinese people. It is also an important policy for the entire national economic development strategy. E. Policy Quiestions Regarding the Industrial Structure Corresponding to Consumption Structure People's social experience tells us that consumption is the starting point and end point of production; consumption promotes and guides the development of production, and changes in people's consumption structure necessarily lead to a corresponding change in industrial structure. This law is valid in both socialist and capitalist societies. For example, in the history of Japan's economic development there was what was called the "three- fold consumption revolution" which brought about the "three-fold great development of industry." China is now going through its own form of that process. In recent years the Chinese people have been scurrying to buy bicycles, watches, and sewing machines, and now they are competing to get radios and tape recorders, electric fans, televisions, and washing machines. It is impossible to say for certain that in future there will not be a rush to buy refrigerators and other top-grade durable goods. It is easy to imagine that the impetus coming from such consumption will produce a major effect on China's electronics industry, machine manufacturing and a whole series of related industries (including progress in science and technology). With 1/ See the Sichuan Daily, July 1, 1983. - 137 - China's socialist planned economy, allowing the proportion of national income used as consumption funds to become inflated is harmful to the people's long- term interests. But neither is it proper to ignore the stimulating and guiding effects of consumption on production. In view of the prospects for changes in China's urban and rural consumption levels and structure, we believe that in determining an industrial structure policy the following questions merit discussion: 1. Social reproduction should be able to follow along with the continuously expanding and alternating circuit of production-distribution- exchange-consumption; theoretically speaking, the key conditions for this are: (1) whether production level and structure are adapted to consumption level and structure; (2) whether production and structure of domestic consumer goods are adapted to the people's consumer-goods demand and structure; (3) whether production and structure of the means of production are adapted to the demand and structure of production consumption; (4) whether production of the means of production and production of the means of subsistence are suitably proportioned. In terms of objective economic management and national economic planning, only if the various relationships above are handled properly can the goal of socialist production - continuously meeting the ever growing popular demands for material and cultural life -- be achieved. We believe that the planning process is very important in meeting this demand. Our historical experience shows us that the "food, clothing, housing, useful-articles, agriculture-light industry-heavy industry" sequence conforms better to our situation than does the "iron and steel, machines, heavy-light-agriculture" sequence; therefore, in terms of developing production and improving living standards, it is also more substantial. Hence we believe that the first point - 138 - to consider in industrial structure policy is using domestic consumption structure and future changes therein as the basis for adjusting industrial structure. The sequence is: consumption structure - consumer goods structure - consumer goods production structure - structure of the production of the means of production of consumer goods - structure of the production of the means of production of the means of production of consumer goods - and so on, creating a comprehensive production structure for the national economy. Here an input-output analysis will to a certain extent aid us in concretely solving this problem. 2. In China's consumer goods production structure, agriculture has always been the basic, key sector, both at the present time and over a long period of history. In terms of China's present consumption structure and its future development, the agricultural development policy, structurally speaking, must push for vigorous development of plant cultivation and stock raising, but in order to improve China's food nutrient structure there must be an especially vigorous increase in the production of animal products. For plant cultivation, grain production and production of other crops must be vigorously developed; with a population of a billion people, grain production is always going to be critically important, but we can certainly not look only to grain. We must also vigorously develop diversified food production, especially the production of woody plants, as they not only provide high- quality nutrients but also have the effect of improving and protecting the environment. The proportion of animal raising in China's agricultural structure tends to be low; this is one manifestation of the backwardness of China's agriculture. Of meat foods, the proportion of grain-fed hogs is much higher than that of grass-fed food animals. In terms of feed sources and the - 139 - nutrient value of meat, and increasing the production of other animal products, grass-fed animals are better than hogs. Obviously, henceforth the vigorous development of grass-fed animals should be an important aspect of improving China's agricultural structure. Furthermore, in addition to supplying primarily food (about 80 percent of agricultural products), agriculture also provides the raw materials for clothing and other useful articles. This should not be overlooked. 3. In China's consumer goods industrial structure, the second major sector after agriculture is the consumer goods industry. China's consumer goods industry is facing several problems of the same type, such as the com- prehensive utilization, intensive processing, and improved utilization rate for raw materials (including agricultural raw materials, industrial raw materials, and mining raw materials); scientific and technical progress and renewal and transformation of production equipment; overstocking and understocking of a product structure nature created by ineffective market information, and so on. These must all be solved conscientiously. But in terms of sector structure, within the consumer goods industry the food industry, the durable goods industry, and the housing construction materials industry are particularly worthy of mention. The food industry: Comprehensive utilization of resources and intensive processing hold great economic potential. According to calculations by the agencies concerned, the utilization rate for China's food resources as compared with that of developing countries is much too low. Only one-fifth of China's food resources are converted to foodstuffs; the remaining four-fifths are unused. For example, the starch utilization rate in China averages less than 60 percent nationwide, while it is 98 percent in the United States. - 140 - Vigorous development of the convenience food industry and promotion of the socialization of household food preparation and scientific food nutrition are important means of improving consumption structure and the corresponding industrial structure. Developing the food industry among China's rural masses is highly significant for fully utilizing agricultural raw materials, improving the eating structure of the peasants, improving nutritional levels, and pushing agriculture's semi-natural economy towards development of a commodity economy. In particular,-this will have a long-term impact on rationalization of China's rural industrial structure. Durable consumer goods: As China's economy develops and the people's scientific and cultural levels improve, expenditures for durable goods will further increase their proportion in the consumption structure. Therefore in the production of durable goods, at the same time as we increase output, we must improve quality and lower cost, adapting to market demand and opening up markets for new products of fine quality and low price. Market volume can also be expanded by adopting such approaches as installment payments and layaways. Housing construction materials industry: housing may be termed as an expensive, very large form of durable goods. Housing construction requires appropriate quantities of construction materials such as steel, cement, glass, bricks and tiles. According to estimates by some comrades, if we are to reach the standard of a housing unit for every family by the year 2000, approx- imately 200 million housing units must be built with the next 20 years; if the area of each unit averages 60 m2, that will mean 12.0 billion m2 for the nation. If each housing unit requires about 80 tons of construction materials and products, the construction materials industry will have to provide 16.0 - 141 - billion tons of construction materials and products over the next 20 years. This is a tremendous figure! And then there are the types, specifications, and capacities for all these materials to be taken into account. It can easily be imagined that an impetus this great development of the construction materials industry will be for all sectors of China's heavy industry to increase their production. Clearly, this is an important strategic measure for economic development. It should be pointed out that in studying and setting China's consumption and consumption structure strategies it is not enough to deal solely with the several problems raised above. We must analyze in depth China's population, resources, ecology, pricing system, wage system and other problems and their stimulating or restricting effects on consumption levels and structure. Only after many analyses and comprehensive and detailed research can we offer opinions and proposals. Hence the views we have offered here (some of which match the suggestions of others) are only exploratory views; we invite readers' criticisms and corrections to benefit further research. 4HG 3881.5 .WS7 W67 NO.755 ILI, XUEZENG. THE STRUCTURE OF CHINAI SI ~-- * DO ESI QONuMPLT _- a The World Bank Headquarters European Office Tokyo Office 1818 H Street, N.W. 66, avenue d'lena Kokusai Building Washington, D.C. 20433, U.S.A. 75116 Paris, France 1-1 Marunouchi 3-chome Telephone: (202) 477-1234 Telephone: (1) 723-54.21 Chiyoda-ku, Tokyo 100, Japan Telex: WUI 64145 WORLDBANK Telex: 842-620628 Telephone: (03) 214-5001 RCA 248423 WORLDBK Telex: 781-26838 Cable Address: INTBAFRAD WASHINGTONDC ISSN 0253-2115/ISBN 0-8213-0640-5