China - Livestock sector study

A WORLD BANK COUNTRY STUDY PUB-6589 CHINA The Livestock Sector FILE COPY  A WORLD BANK COUNTRY STUDY CHINA The Livestock Sector The World Bank Washington, D.C., U.S.A. The World Bank 1818 H Street, N.W Washington, D.C. 20433, U.S.A. AD rights reserved Manufactured in the United States of America First printing July 1987 World Bank Country Studies are reports originally prepared for internal use as part of the continuing analysis by the Bank of the economic and related conditions of its developing member countries and of its dialogues with the governments. Some of the reports are published informally with the least possible delay for the use of governments and the academic, business and financial, and development communities. Thus, 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. Any maps that accompany the text have been prepared solely for the convenience of readers. The designations and presentation of material in them do not imply the expression of any opinion whatsoever on the part of the World Bank, its affiliates, or its Board or member countries concerning the legal status of any country, territory, city, or area or of the authorities thereof or concerning the delimitation of its boundaries or its national affiliation. The most recent World Bank publications are described in the catalog New Publications, a new edition of which is issued in the spring and fall of each year. The complete backlist of publications is shown in the annual Index of Publications, which contains an alphabetical title list and indexes of subjects, authors, and countries and regions; it is of value principally to libraries and institutional purchasers. The latest edition of each of these is available free of charge from the Publications Sales Unit, Department F, The World Bank, 1818 H Street, N.W, Washington, D.C. 20433, U.SA, or from Publications, The World Bank, 66 avenue d'I&a, 75116 Paris, France. Library of Congress Cataloging-in-Publication Data China, the livestock sector. p. cm. -- (A World Bank country study) ISBN 0-8213-0940-4 1. Animal Industry-hina. 2. Livestock--China. I. International Bank for Recohstruction and Development. II. Series. HD9426 .C62C475 1987 338.1'76'00951--dcl9 87-16089 CIP PREFACE This report is largely based on the findings of a World Bank livestock sector study mission that visited China for three weeks in May-June 1986. The mission consisted of Mr. Thomas Wiens (mission chief and agricultural economist), Mr. Bjorn Almquist (agroprocessing specialist), Mr. John Blake (poultry specialist), Mr. Brian Brandenburg (veterinarian), Mr. Justus Damm (pig specialist), Mr. Paul Harrison (agricultural economist), and Ms. Zhong Tong (agricultural economist). Overall responsibility for oversight of the study was assigned by MAAF to the Planning Office of the Bureau of Animal Husbandry, and the task of executing the study and coordinating the Study Mission was assigned to staff of the General Corporation for Animal Husbandry, Industry and Trade (GCAHIT), a corporate subsidiary of MAAF with management responsibility for state-owned livestock breeding, production and processing enterprises. Agricultural Committees of the municipal or provincial governments of Tianjin, Fuzhou, Shandong and Jiangxi served as local hosts to the Mission, and information collection and site visits were arranged by the local Bureaus of Animal Husbandry, with the assistance and cooperation of the local Food Company, Feed Company, and State Statistical Bureaus. Staff of the Chinese Academy of Agricultural Engineering and Systems Analysis also assisted in information collection and processing in Shandong Province, and the Jiangxi Agricultural Development Corporation and the Dongxiang County Government hosted the Mission in Jiangxi Province. The tireless help, patience, and valuable insights of numerous individuals in these organizations were essential to the preparation of this report. The livestock study was structured in two phases: during the first phase, GCAHIT collected information on the markets for livestock products in two municipalities (Tianjin and Fuzhou), existing and potential sources of supply, and data from household expenditure surveys. This information was reviewed during a brief planning mission to Beijing during March 1986. The second phase, which culminated in the study mission, involved collection of detailed survey data on production and processing enterprises representing different scales, forms of organization, and locations of production in Jiangxi and Shandong Provinces as well as Tianjin and Fuzhou Municipalities. The mission visited these and other units for extended discussions, as well as receiving briefings from the relevant bureaus of the municipalities and provinces. A draft of the mission's report was submitted to the Chinese government in October 1986, and discussed with the government in November 1986. This version of the report has been revised in light of government comments. CURRENCY EQUIVALENTS The Chinese currency is called Renmenbi (RMB). It is denominated in Yuan (Y). Each Yuan is 1 Yuan - 10 jiao - 100 fen Calendar 1985 October 1986 US$1.00 - Y 2.94 US$1.00 - Y 3.69 Y 1.00 - US$0.34 Y 1.00 - US$0.27 WEIGHTS AND MEASURES Metric System FISCAL YEAR January 1 - December 31 TRANSLITERATION The Pinyin system is used in this report. ABBREVIATIONS ABC - Agricultural Bank of China AFS - Australian Friesian-Sahival AHB - Animal Husbandry Bureau AI - artificial insemination ATM - Agricultural Trade Market BSF - Bureau of State Farms CAHIT - (Provincial) Corporation for Animal Husbandry, Industry and Trade CMT - California Mastitis Test GCAHIT - General Corporation for Animal Husbandry, Industry and Trade (Beijing) GB - Grain Bureau MAA - Market Administration Agency MAAF - Ministry of Agriculture, Animal Husbandry, and Fisheries MOFERT - Ministry of Foreign Economic Relations and Trade NFIO - National Feed Industry Office NRC - U.S. National Research Council RCC - Rural Credit Cooperatives SEC - State Economic Commission SH - Specialized Household S&MC - Supply and Marketing Cooperatives SPC - State Planning Commission VT - veterinary technician CHINA LIVESTOCK SECTOR STUDY TABLE OF CONTENTS EXECUTIVE SUMMARY .............................................. Vii I. INTRODUCTION .............................................. 1 Study Scope .............................................. 1 Characteristics of Study Sites ............................. 2 Changing Policies and Performance .......................... 6 Organization of the Sector ................................. 10 II. THE MARKET FOR LIVESTOCK PRODUCTS .......................... 16 Organization of the Marketing System ....................... 16 Consumer Demand for Livestock Products ..................... 24 Issues and Implications ................................... 33 III. FEED SUPPLY AND PROCESSING ................................ 35 Aggregate Feed Supply Situation ............................ 35 Systems and Organization of Livestock Feeding .............. 41 The Feed Milling Industry ................................. 44 IV. LIVESTOCK AND POULTRY BREEDING ............................. 48 Dairy Cattle Breeding ...................................... 48 Pig Breeding ............................................... 52 Poultry Breeding .......................................... 61 V. ANIMAL HEALTH AND VETERINARY SERVICES ...................... 64 The Animal Health Delivery System .......................... 64 Disease Status in Dairy Cattle ............................. 66 Disease Status in Pigs .................................... 71 Disease Status in Poultry ................................. 73 Quarantine Stations and Abattoirs .......................... 76 VI. PRODUCTION ALTERNATIVES .................................. 80 Pigs ....................................................... 80 Dairy Cattle ............................................... 83 Poultry .................................................... 84 Locations for Investment ................................... 86 Viability of Specialized Households ........................ 89 Financial Viability of Livestock Investments ............... 90 VII. PRODUCT PROCESSING ......................................... 93 The Dairy Industry ........................................ 93 The Slaughtering Industry .................................. 96 Poultry and Eggs .......................................... 99 VIII. ISSUES AND IMPLICATIONS ................................. 101 Price Reform .............................................. 101 Organizational Reform ..................................... 104 Expanded Role of the Market ................................106 Enlarging the Role of Interregional and International Trade 107 Locational Strategy ........................................108 Form and Scale of Production Enterprise.....................108 Financing Policy ............................................109 List of Tables in Text 1.1 Production of Livestock Products, Study Sites ......... 5 1.2 Personal Consumption of Livestock Products, Study Sites, 1984 ... ................................. 5 1.3 Livestock Performance, China, 1978-85 ................. 8 2.1 Urban Income Elasticities of Demand, Fuzhou and Tianjin, 1984 ........................................ 26 2.2 Local and Budget Study Projections of Growth Rates of Consumer Demand, Fuzhou and Tianjin, 1984-90 ...... 26 2.3 Changes in Fuzhou and Tianjin Consumption Between 1984 and 1985 ........................................ 28 2.4 Estimated Direct Price Elasticities of Demand, Fuzhou and Tianjin, 1984 ............................. 30 3.1 Human Grain Consumption and Quality Feed Availability by Region, 1984 ...................................... 37 3.2 Composition of Available Feed, by Region (1984) ....... 39 4.1 Average Milk Production by Region and Individual Farms. 51 4.2 Services Per Conception and Dairy Herd Size ........... 53 4.3 Lean Meat Production Per Sow Per Year From Local, Local x Improved, and Improved Crosses ............... 56 4.4 Pork Meat Grades and Procurement Prices (Fuzhou) ...... 58 5.1 Livestock Per Veterinarian/Technician ................. 65 5.2 Disease Incidence Report of Pig Diseases in Tianjin Municipality, 1984-85 ................................ 73 6.1 Opportunity Cost of Land Occupied as Percentage of Livestock Sales Value ................................ 87 6.2 Summary of Financial Results of Various Livestock Investments .......................................... 91 8.1 Urban Expenditure on Dairy Products by Income Group, Tianjin and Fuzhou, 1984 .............................103 Figures in Text Graph 1 Pork Production and Feed Availability, 1952-85..... 7 Graph 2 Procurement Price Ratio: Pigs to Grain ........... 7 Chart 1 Organization of the Livestock Sector.............. 11 Graph 3 Seasonal Variations in Livestock Supply, Tianjin and Fuzhou, 1984 ................................... 22 ANNEXES Annex 1 - Livestock Supply and Demand, Tianjin and Fuzhou: Tables Annex 2 - Income and Price Elasticity Study, Tianjin and Fuzhou Annex 3 - Feed Balance Tables (Provincial, 1984) Annex 4 - Supplement on Dairy Cattle Production Annex 5 - Supplement on Pig Production Annex 6 - Supplement on Poultry Production Annex 7 - Financial Returns to Livestock Investments CHINA LIVESTOCK SECTOR STUDY Executive Summa 1. The study considers the most efficient means of satisfying the growing demand for pork, poultry, and dairy products in China's large municipalities. It is not concerned with the broad topic of livestock raising in the pastoral areas and only touches on questions of livestock product supply for rural consumers. The major issues hampering the development of the sector which are discussed include difficulties in making the transition from an administered to a market system; price distortions due largely to the scope of consumer and producer subsidies; insufficient reliance on interregional and inter- national trade to solve problems of feed and livestock product supply; lack of functional specialization within the industry; and various weaknesses of support services and the sectoral management structure. 2. The main report reviews the recent developments in China's livestock production, the organization of the sector, the emergence of Agricultural Trade Markets (ATMs) in large cities as a result of 1985 policy reforms, and influences determining future growth of urban demand for livestock products. Each component of the industry is examined in detail, including feed supply and processing, livestock and poultry breeding, animal health and veterinary services, alternative production systems, and product processing. The following summarizes major findings of relevance to Government policy. Enlarging the Scope of the Market 3. The livestock products distribution system in China was characterized in the past by compulsory procurements, administered prices, and rationed distribution. However, major reforms in 1985 were designed to facilitate an expanded role for market-oriented production in the sector, that is, for economic activities governed by consideration of profit-and-loss, which is in turn determined by flexible, market-clearing prices. At present market-oriented activi- ties are most important in poultry, beef and mutton supply, whereas in pork and dairy product supply, administered prices and administrative management are still dominant. The inflexibility of administered prices can easily lead to dynamic instability in markets where both supply and demand are strongly price-sensitive, and the pig market appears to be such a case, as the disequilibrium in China's pig market in 1986 indicates. Rather than dealing with the disequilibrium by imposing additional administrative control, the transition in the pig market to more flexible pricing methods should be speeded up. "Flexi- bility" here implies that each independent accounting unit within the distribution system should be encouraged to adjust prices within a wide range so as to maximize profits; which in turn requires that each unit be allowed to retain a significant proportion of profits. - vii - 4. At present, the free market, or ATM distribution system is supposed to help equilibriate the livestock market, by providing signals to producers and government alike of supply and demand condi- tions, and by providing an outlet for excess supply or pent-up demand due to deviations of official from equilibrium prices. It would better serve these functions if its scope were enlarged, which depends partly on reduction of subsidies in the state-controlled subsector and partly on decreasing barriers to trade (discussed below). One major shortcoming of the ATM distribution system is its reliance on expen- sive truck transport, and lack of access to cheaper long-distance transport modes. This could be alleviated if the Ministry of Railways were to set up an "outside-Plan" goods traffic program, perhaps con- sisting of two elements: (a) a set proportion of wagons suitable for conveying live animals or frozen/chilled product would be offered for short-notice use by any commercial unit (state or private) at unsubsi- dized freight rates; and (b) the railways or, failing this, commercial units would be encouraged to invest in refrigerated freight wagons or chilled milk tankers, which could be leased out and attached to regu- lar freight trains. Price Rationalization 5. During the process of transition to a market price system, there are some desirable reforms of administered prices which have major ramifications extending as far as the profitability of crop production and the level of rural-urban income differentials; and other adjustments with a relatively minor impact. The greatest difficulty is associated with elimination of subsidies in the sector. 6. The main price distortions in the livestock sector are those associated with the subsidized state procurement, processing and dis- tribution system and the supply of feedgrain at subsidized prices to livestock producers. These distortions are associated mainly with pig production (to a lesser extent with eggs and dairy products) and especially with supply to the larger municipalities. Towns at the country level or below appear to have been weaned from both forms of subsidies. The subsidies benefit urban consumers, employees of the state distribution system, and suburban farmers (especially Specia- lized Households, or SHs), at the expense of the state budget and commercial feedgrain growers. In the face of these subsidies, the private or cooperative distribution system, including the urban ATMs, is handicapped in its role in the municipal pork trade. 7. Removal of subsidies on egg and dairy products would be relatively painless. Egg and poultry meat production at present are profitable enough that producers could absorb the loss of feed and distribution subsidies. Urban consumption of dairy products is small enough that consumers could easily bear the costs of price increases sufficient to compensate dairies and dairy farms for loss of subsi- dies. In view of the recent price adjustments for milk, which proba- bly eliminate direct subsidies (but not rationing) in most locales, it appears the Government has reached the same conclusion. 8. However, elimination of subsidies to the suburban pig industry is far more problematic. Many marginal producers, particularily SHs depending mainly on purchased feed, would be bankrupted by higher feed costs, even if the market price stabilized at about the same level as the present procurement price. As supply contracted, the relative advantage would shift to the backyard (nonspecialized) producer of local breeds of pigs, whose production process relies mainly on non-purchased feeds and surplus labor. Urban consumers would be forced to pay higher prices at state markets, reflecting full costs of processing. Demand would shift further away from pork sold in state shops to the ATMs, and overall levels of consumption would contract. Improved financial performance for the state distribution system also would require finding other jobs for redundant workers, and shifting the pensioner burden to the state. 9. Considering the above, the subsidies to the pig industry should be removed gradually, following a strategy with multiple ele- ments, including: (a) reduce the transport constraint on inter- regional live pig or carcass trade; (b) improve the quality of chilled or frozen carcasses made available through that trade; (c) enlarge cold storage facilities to better meet peak seasonal requirements, while finding ways of utilizing these facilities in the off-peak season; (d) phase out the feed subsidy; (e) close down substandard abattoirs, and use other state-run abattoirs to serve the slaughter requirements of the ATM at fees close to marginal costs; (f) encourage expansion of ATM pork sales, and correspondingly reduce employment in the state distribution system (or encourage the latter to serve a wholesaling function for the ATMs); and (g) modulate the increases in the relative price of pork according to the rate of increase of consu- mer incomes. 10. This strategy is in direct contrast to the present municipal approach of increasing the extent of local self-sufficiency in pork supply. Elements of this strategy require actions by the central government, the provincial governments of supplier provinces, and most of the various bureaus involved in the livestock sector (including the Transport Bureaus). In the absence of the central and local coordina- tion this requires, the municipal governments would have little choice but to pursue self-sufficiency, at considerable cost to the national economy. 11. As an essential part of the "lean pig" policy, there should be an adequate differential in procurement prices according to the leanness of the product. If it is to provide an effective incentive, the differential must exceed the additional costs of lean pig pro- duction, now evaluated at about Y30 per pig. However, it should not exceed the value placed by the market on the leaner product. Invest- ments and policy measures supporting a "lean pig" policy would be premature in those areas where a sufficient consumer premium for lean meat over fat does not yet exist. Market price surveys are required to identify which urban markets offer an adequate premium, as this is a point of controversy in China. In such markets, it is important that (a) state retail outlets should offer cuts similar to those offered in ATMs, with equivalent price differentiation; (b) wholesale carcasses ex-abattoir should be graded by weight and leanness and priced accordingly; and (c) abattoir gradings of animals should be passed back in the form of higher prices to receiving stations and thence to producers. Wholesale carcass grading and pricing should also be introduced in inter-provincial trade, whether at list or negotiated prices, with allowance for storage costs and losses as well as market discounts for frozen pork. 12. The low retail price of fresh milk in urban areas has been central to pricing problems in the dairy sector. This has been justi- fied on the grounds that milk is supplied as a nutritional supplement for infants, the sick, and the aged, rather than a luxury food. In reality, a high proportion of dairy products have been consumed by local residents outside these target groups, and, in addition, foreign residents, tourists, and the bakery and confectionary trade all bene- fited unnecessarily from the same low prices. Also, the low profita- bility of fresh milk production due to price controls has created a strong incentive for dairies to enlarge the proportion of less nutri- tionally valuable products, such as milk powder diluted with sugar or milk chocolate. As the fresh milk rationing scheme lies at the heart of these distortions, consideration should be given to reducing its scope or phasing it out. 13. Judging from the present cost structure in the industry, producer prices for raw milk of about YO.56/kg, and a processing margin of about YO.25/kg for processors of pasteurized milk, should allow an adequate rate of return. Recent price adjustments come close to matching these figures. However, the same adjustments also in- creased the price of powdered milk, leaving the relatively high profi- tability of milk powder production unchanged. This pricing decision should be reconsidered: if the profitability of fresh milk production were increased relative to milk powder, suburban dairies might well be willing to shut down their powdered milk production lines, and rural dairies might find means of transporting liquid product to the urban market. It would also be an opportune time to introduce milk grading and pricing according to butterfat content, and for dairies to stan- dardize product at 2.5% butterfat and use excess butterfat to increase production of butter. Enlarging the Role of Interregional and International Trade 14. Feed Supply. China may be marginally short of quality feed for its present livestock population, in that estimated requirements of such feedstuffs (coarse grains, byproducts, and oilseed meals) slightly exceed potential supply, and not all of potential supply is in fact used for livestock. However, even considering use of rice and wheat as feed, and use of coarse grains for human consumption, there remain significant regional imbalances, with the Northeast having a surplus and the West, parts of the Southeast, and the three provin- cial-level municipalities in deficit. If nutritive balance is consi- dered, it appears that, compared to a desirable 15%, on average Chi- na's potential feed supply contains about 12% crude protein and, allowing for the underutilization of oilseed meals, actual feed use provides as little as 11% protein. Both the potential and actual protein shortages are most serious in those southern provinces which depend heavily on tubers as feedgrains and rapeseed and cottonseed meals as protein feeds. 15. In areas of the Southeast and West where paddy rice produc- tion is now predominant and feed deficits occur, there are three options to consider: (a) trade rice for feed grains imported from the north or abroad; (b) diversify the cropping system, introducing corn or other feed grains on fields or in areas which are marginal for paddy rice production; or (c) increase use of rice as a livestock feed, with appropriate balancers. The first option is advantageous at current domestic or international price ratios, but the price ratios would become considerably less favorable if exports of rice were attempted on any scale. The second option may prove feasible and should be pursued in the longer term, but will require a transfer of suitable corn varieties and cultivation techniques to a region with little experience in corn production. It will also require adjustments to the procurement price of corn in the south, which is now about one-half the negotiated price for imports from the Northeast. At least in the interim, increased use of rice as a feed grain appears to be economically justified. However, this does not alleviate the protein feed deficit in the rice producing areas. Relative to either North China production costs or international prices, the southern provinces lack any comparative advantage in the production of oilseed meals valuable as livestock feed. Therefore these areas in the short term should rely on imports of oilseed. meals or other protein sources rather than attempting self-sufficiency in protein feeds. 16. In according high priority to rapid development of livestock production, the Central Government should attach the corresponding priority to solving the national and regional problems of feed supply. It may in fact be economically advantageous to export Northeastern corn and soybean to the Soviet Union, South Korea, and Japan, rather than processing the soybean and transporting corn and soymeal to the South; but this advantage is fully offset if the Southeast must import the same amounts of feed from abroad, or if Southeastern feed mills and livestock production are crippled for want of raw materials. The advantage, from a national viewpoint, would derive from a strategy of using non-exportable surpluses of rice and limited amounts of local oilseed meals as feed stocks in the Southeast, and allocating the necessary foreign exchange to import sufficient amounts of appropriate balancer ingredients (such as soymeal or fishmeal) from abroad. As a corollary, the foreign exchange-saving potential of increased production of each type of feedgrain and oilseed meal producable in the Southeast should be evaluated, and its farmgate price should be adjusted accordingly. Research directed at adapting corn and soybean varieties to the Southeast also deserves priority. 17. Lack of high quality fodder, as well as heat stress, are serious limiting factors in dairy production in the Southeast. As longdistance fodder transport is uneconomic, this region should continue to rely more heavily on milk powder imported from other regions or abroad. More generally, the opportunity to exploit the subsidies foreign governments give to their milk producers should be seized to close the gap between domestic supply and demand during the period in which domestic dairy herds are being built up. Production of milk powder by municipal dairies should cease, and cheap imported milk powder should be used (a) for local bakery and confectionary requirements; (b) in recombination with local edible oils, to "stretch" fresh milk supplies and reduce seasonality of supply; and (c) as a substitute for fresh milk in calf feeding. 18. Production Alternatives and Domestic Trade. Losses and costs associated with fresh product delivery are most significant in the poultry and dairy industries. Consequently, production of eggs, poultry meat, and fresh milk is most rationally sited near the cities. However, the raising of calves or heifers in suburban areas is uneconomic and unnecessary; and that portion of suburban dairy farming should be relocated to areas where roughage and green feeds are relatively abundant. Over the long run, an increasing proportion of urban pork supply should come from rural areas with feed surpluses and low opportunity cost of land. This applies especially to the Southeastern cities, and particularily if these cities cannot rely on supplies of corn and soybean from either the Northeast or abroad. However, this strategy depends on a gradual solution to transport and storage problems, and upgrading the quality of pork supplied in interregional trade, through breed and feed improvement programs aimed at the rural (mainly backyard) producer, along with corresponding price incentives. As a short run measure, the efficiency of suburban pig production could be raised by space-saving pig shed designs, siting of farms in the distant suburbs, and improved feed conversion rates. 19. A major obstacle to further development of inter-regional trade is the present power of local governments to intervene in re- straint of such trade. This power was exercised in 1986, both to restrict flows of feedgrains from northern to southern provinces, and also to prohibit or tax exports of pigs from southern pig-surplus provinces such as Hunan to areas of excess demand, such as Guangdong (which exports its pigs to Hong Kong). Such a chain of derived demand should normally be favored as beneficial to profits of producers at each link in the chain, but local governments have proven more sensi- tive to the complaints of consumers or state trading organs about increased prices or commodity shortages. Consequently, China should consider a legal prohibition on such local government interventions in interprovincial trade, in the interest of the development of the national economy as a whole. Form and Scale of Production Enterprise 20. In pig production, there is a clear case for division of function: state- and collectively-owned larger farms should serve mainly as breeding farms for nucleus herds (abandoning fattening operations); SHs should serve as multiplier farms in the breeding system and fatteners for lean varieties; and backyard producers should maintain their present role as suppliers of most market pigs, using local breeds to the extent that they depend on low-quality feeds (or vice versa). Overall, Government's strategy for development of the industry should give increased attention to the backyard producer, in the expectation of achieving major efficiency improvements at relatively low cost through use of improved hybrids and feed balancers. 21. In dairying, there is no "backyard production", because the high costs of investment in animals place even the owner of one or two cows into the SH category; and a five-ten cow SH is either very wealthy or deeply in debt. There are both well- and poorly-managed state farms and SHs at present, and it appears that small SH herds, if grouped and managed cooperatively to facilitate silage use and AI, and given adequate support services, could function as well as state farms; but that is based on hypothesis more than observation, as SHs usually are geographically scattered (taking advantage of equally scattered natural fodder resources). A mixed system, with state farms taking the lead in promotion of SH dairy development, would be quite acceptable. However, if the land area managed by state dairy farms is not to be increased, inevitably a high proportion of the national dairy herd will be under SH management by the year 2000, in view of the planned six-fold herd growth. In the short run, improving the utilization of land and buildings on existing state farms, and the motivation or expertise of technical personnel, are the most pressing issues, but in the long-run development of support services for SH dairying must assume high priority. 22. In egg and poultry production, the superiority of imported breeds and intensive feeding systems argues for a scale of at least a thousand birds, except in semi-scavenging duck production systems, where units of a few hundred birds would be efficient. If sufficient flexibility is allowed for wages and incentive schemes on state or collective poultry farms, large farms would conserve on technical and managerial skills which are scarce in this industry. 23. Production and service enterprises in China have rarely been organized in a hierarchical system of functional specialization which locates functions, according to their criticality or complexity, in enterprises possessing appropriate levels of managerial and technical expertise as well as qualities and quantities of equipment. The breeding system for livestock is one area where a pyramidal form of specialization should be introduced, particularly in pig breeding. The organizational structure in feed milling should also be more hierarchical, to include nucleus mills providing pre-mixes and techni- cal services to satellite mills; and increased division of labor in producing a more varied set of products. Improving Support Services 24. Feed Milling. The feed milling industry needs to address the major feeding problems in the sector. In general, these are more critical in China for pigs than for poultry or dairy cattle, as concentrate feeds used by poultry farmers at present are of largely adequate types and qualities, and the limiting factor at present for China's small number of dairy cattle is the supply of bulk fodder rather than concentrate. However, even the better pig farms have purchased rations with deficiencies in energy, protein, microelements and vitamins (especially calcium:phosphorus ratios); nor is there sufficient differentiation of commercial rations either by growth stage of the animal or by variations in availability of home-grown produce. 25. Feed mills in China usually have technically-adequate milling and mixing systems, despite reliance on labor-intensive processes (which is probably cost-minimizing, but does make quality control difficult). However, equipment suitable for the introduction of trace elements and vitamins into the mix, or for pelleting of feed is lacking. The greatest shortcoming is the lack of quality control of the product, including the absence of labelling of composition of the finished product. 26. Even to meet present market requirements, the feed industry would need to produce something of the order of 10-20 million tons of formulated feeds and 15-25 million tons of balancers to upgrade the grain and bran which may be available at the farm level. In order to expand and at the same time improve technical quality of feeds, a number of measures are recommended, including: (a) upgrading of some existing mills, focussing on monitoring and control systems and pelleting lines; (b) establishment of a nucleus feed mill in each main consuming area, to manufacture pre-mixes for smaller feed mills; (c) establishment of effective feed quality-control laboratories independent of the feed mills; (d) strengthening the organization of the feed milling industry to achieve uniformity of feed standards and improved supply of ingredients and feed distribution; (e) streng- thening of research and development; and (d) enforcement of regula- tions to ensure that feeds produced by the milling industry comply with pre-determined standards in terms of nutritional value. The industry should also expand its product mix to include (a) compound feeds (complete rations), (b) concentrate balancers (high protein supplements), (c) pre-mixes for small-scale mills, (d) mineral-mixes for farm use as supplements, and (e) medicated feeds. 27. Breeding. At the present time, the main constraints to rapid genetic improvement in milk production are nutritional factors and, in southern China, climatic factors. In order to sustain a meaningful genetic improvement program, it is important to improve the quality and quantity of nutrient intakes, through standardization of dairy rations, improvements in grain and fodder procurement, the storage of fodders by ensilaging during the cold seasons, and the upgrading of fresh and stored feedstuffs by the addition of non- protein nitrogens and micro and macro mineral mixes. 28. A national dairy cattle breeding program, including a performance testing program, has been initiated, but further progress towards central coordination and standardization of the program is required, as well as increased attention to selection and retention for appraisal of locally bred bulls, in the interest of future savings in import costs of genetic materials. 29. Dairy cattle production in the south of China is subject to increased seasonal variability and reduced overall productivity due mainly to the heat. The problem can be dealt with either through environmental control (cooling) or through introduction of more heat resistant breeds. The economics of production using Bos taurus x Bos indicus crosses (such as the Australian Friesian-Sahiwal, or AFS) should be compared with that of production by purebred Holsteins under modified management systems; and economical systems for environmental control in the south need to be designed and promulgated. In both north and south China, AI services should be improved in order to maintain farmer confidence, wherever conceptions per pregnancy are now unacceptably low. 30. The lean pork policy should provide the impetus for a review of the entire process of genetic improvement in pig performance, especially in the areas of breed characteristics of local and exotic pigs, selection techniques and environmental constraints to genetic upgrading programs. Existing breeding farms should redefine their breeding programs to use a few breeds for a limited variety of crosses. Breeding activities should be focussed on select-ion, as well as mating and multiplication. Selection by visual, subjective evaluation must be replaced by measured, objective evaluation. This will require the procurement of scales and backfat measuring and other equipment, and liaisons with abattoirs to better evaluate carcass merits of improved pigs. Boar testing should be introduced as a separate, highly important breeding/selection pig station activity. Sow performance, for local, improved and exotic breeds, should be also be clearly evaluated by breeding farms. No commercial pig production activities should take place on farms where selection and breeding is carried out. Accurate record keeping systems are needed to evaluate the data collected. 31. The pig breeding system should be organized according to the pyramid design, with a system of multiplier farms, managed by well- trained specialized households, inserted between the existing breeding and commercial farm levels. Some existing breeding farms should be upgraded to a nucleus herd status, where extensive, strict selection would take place. Standardized, high quality feeds should be used on nucleus and multiplier farms. 32. In poultry production, efforts to experiment with crosses between local and imported strains, and between imported strains of different origins, should be abandoned as uneconomic. Breeding for chickens should concentrate on imported strains, whereas local breeds of duck seem quite adequate for the systems of production now in place. Efforts should be made to attract international breeding companies to establish Parent and Grand Parent stock farms in China so that a continuing flow of superior genetic stock will be ensured. Such ventures should incorporate substantial training packages extending down to the level of commercial production. 33. Animal Health. Clinical mastitis is the most commonly observed health problem on Chinese dairy farms, and the prevalence of subclinical forms is unknown but probably extremely high. A nation- wide investigation into the causes, incidence, treatment and control of mastitis is strongly advisable. To bring this problem under con- trol, the most necessary measures include use of udder disinfectants and retraining in milking technique. Milk quality control programs to screen for mastitic milk should be enforced by evaluating bulk samples taken at milk collection centers and dairy plants and by providing CMT test kits to producers. Dry-cow treatment should also be instituted. 34. Supplementing dairy rations with calcium and phosphorus should be considered to ensure that adequate nutrient intakes are met. Bone meal and shell meal should be readily available at feed mills or veterinary centres, and the use of dicalcium phosphate (DCP) as an economical source should be investigated. Further investigations into the nutritional, and possibly infectious, causes of infertility in dairy cattle are needed and should be coordinated with those related to AI failures. 35. In pig production, piglet mortality to weaning represents a serious production loss and investigations into the control of newborn pig diseases, especially into colibacillosis, should take priority. Improvements in the newborn pig's environment, such as bedding, the use of farrowing crates, and-night heating should be considered. Major efforts towards the improvement of pig performance should focus on pig nutrition. Calcium and phosphorus related skeletal weakness and lameness was observed, and requires correction. Similarly, multi- vitamin supplements in feed need to be introduced. Pig disease inci- dence, except for swine fever in areas where vaccination programs have yet to take hold, is not a constraint to better production. 36. Poultry production improvement at present is mainly dependent on feed quality and management factors; the latter are in need of considerable improvement on small poultry farms, where housing and lack of flock isolation are identified as constraints to better performance. Vaccination programs in breeder flocks, although standardized, need to be followed more faithfully. A compulsory Pullorum testing program should be instituted, and Pullorum test kits made widely available. An Avian Encephalitis vaccination program for breeding farms should be introduced, rather than relying solely on immunization in country of origin. 37. Veterinary Services. There is a strong need for additional and upgraded courses for university and technical school trained veterinarians, meat inspectors, livestock market inspectors, laboratory personnel, AI technicians and other service personnel. The emphasis of this training should be on the preventive, or cause-and- effect approach to recognition and control of production disease, as opposed to the prevailing symptomatic approach to infectious disease diagnosis and treatment (which should not, of course, be ignored). In addition, there is a particular shortage of service personnel in the animal nutrition and breeding sciences areas, which should be remedied through training programs. 38. Quarantine Stations. Health care and feeding of imported stock in quarantine should become the responsibility of the quarantine station, rather than the importing party. The station should charge the importing party for the costs of these services, and take respon- sibility for procurement of animal feeds and other supplies. 39. Meat Inspection. Meat inspection standards should be raised to where a meaningful screening and condemnation system exists. Accurate, objective diagnosis and recognition of carcass conditions without the pressure to pass questionable carcasses, should replace the present system. Inspectors should not be abattoir employees but should work independently of the abattoir, on behalf of a government agency. Abattoir hygiene measures need to be universally improved. The use of high pressure hot water supplies and suitable disinfectants should be introduced where necessary. Because typical abattoirs and dairy processing plants in China presently provide fertile breeding grounds for pathogens, there is a considerable need for better plant layouts, higher standards of materials, upgraded drainage, wall and floor surfaces, and lighting. In smaller abattoirs, meat contact surfaces should use stainless steel or tile rather than permeable wood. 40. Processing. The cost of fresh milk production in China is moderately high by world standards, but China also has access to cheap or subsidized powdered milk and butter oil products. The policy for domestically produced milk should be to satisfy demands for liquid milk as cheaply and as simply as possible. Because of excess demand for liquid milk, seasonality of supply should be ironed out in the simplest manner possible, including the use of recombined product to supplement winter shortfalls, production of ice-creams, yoghurts or flavoured milk to dispose of summer surpluses, and introduction of seasonal variations into the pricing system. There is no economic justification for the high proportion of dried milk being presently manufactured in China. Consequently, future policy should be to phase this out. 41. The marketing system for pork in eastern China is geared to butchers buying half-carcasses of meat, boning them out and selling them in small pieces to individuals. There is a strong domestic preference for freshly killed meat, and chilled meat is preferred to frozen. Neither these preferences nor the system of marketing are likely to change rapidly, therefore a processing system geared to providing fresh half-carcasses and without major investment in cutting facilities is appropriate. However, some freezing capacity and cold- storage will continue to be required, both to facilitate inter- regional transfers of pig meat and to iron out seasonal variations in supply. Carcasses prepared for freezing should not be skinned, and freezing and cold-storage technology should be upgraded to reduce shrinkage and retain quality. The growing consumer preference in at least some areas of the country for lean as opposed to fat pig meat, which is reflected in the prices paid by the consumer to the retailer but not yet in the prices paid to the wholesaler or producer, calls for grading of half carcasses by weight and leanness at the abattoir, as well as appropriate modifications to the pricing system. 42. Poultry are largely sold live and it is expected that this will continue. However, processed and chilled or frozen poultry are likely to be accepted by the catering trade and especially in association with the production of broilers or meat ducks. Frozen poultry would be helpful in ironing out peaks and troughs in production and facilitating inter-regional transport of poultry meat, which cannot now take place. Similar conclusions apply to eggs, where there is promise for development of small-scale service operations for sorting, grading, packaging, and distribution which could be extended to seasonal drying, freezing or egg preservation. 43. Financing Policy. Financing terms are important to the viability of the SH sector, where only a small proportion of investments in housing and stock can be owner-financed. The major source of finance has been lending from the Agricultural Bank of China (ABC) or Rural Credit Cooperatives, and there has been a strong tendency to use medium-term (3-5 year) loans to finance 80-90% of total investment costs. The use of such short maturities and low proportions of owner equity has allowed the SH sector to develop rapidly, with minimal interest burden, but timely repayment of principal will prove difficult, particularly as either increased sup- ply or reduced subsidization lead to declining profit margins in the sector. 44. Future financing should move to longer maturities and a higher proportion of owner-equity, to protect both the owners and financial system against price risks. For long-term lending, the present ABC interest rate structure provides for rates approaching 9- 10% (although few loans are now extended at these rates). The higher equity proportions and interest rates may discourage growth of SH production in some components of the industry, but they will also help to stabilize supply by discouraging entry of producers who lack the resources to withstand periodically unfavorable market conditions. Institutional Reforms 45. Higher-level governmental organs in China tend to be identified with a coordinating function or an administrative function, and occasionally both. The functional components of the livestock sector, including input supply, production, procurement, transport, processing, storage, and final distribution typically are administered by several different systems, and coordination is carried out at such a high level that it is inevitably weak. The coordinating bodies in the subsector are mainly commissions responsible for policy making and planning (such as the SEC and SPC at the central government level, or their counterparts at the provincial or municipal level). As coordinating bodies lack the direct administrative authority and the staff necessary for program or project execution, programs or projects which cut across either functional components of the sector or separate administrative systems are difficult to implement. 46. A deepening and extension of past organizational reforms in the sector would improve the prospect for coordinated execution of development programs in the livestock sector. The proposed program for development of the feed milling industry should logically be executed by the National Feed Industry Office, whose authority (espe- cially over mills under the Grain Bureau system) and staffing need to be further strengthened. In dairying, the growing responsibility of state farm veterinarians to service the needs of SH producers suggests that these vets should report to AHB rather than BSF, and that a fee- for-service approach or privatization should replace fixed salary scales. The dairy industry might alternatively be transferred out of the BSF system, with responsibility for production being moved into the system under the General Corporation for Animal Husbandry, Indus- try and Trade (GCAHIT) and, for services, into the AHB system. 47. In the processing and distribution subsectors, the most pressing needs are for better standards of control over animal health and sanitation, notably at abattoirs. The control system should be completely independent of the Food Company/Second Commercial Bureau system which administers the processing units. Probably it should reside with AHB, yet be separate from administrative offices concerned mainly with live animal health and receive greater emphasis. 48. In general, all production or service units in each administrative system should be given greatly increased freedom to engage in commercial transactions crossing functional and administrative lines (including those between state and private sectors). For some units, price reforms are required before this is feasible, because cross-subsidization within vertically integrated systems is currently the norm and its existence would discourage such transactions. For other units, encouragement and an appropriate fee or rate structure are all that is required. Vertically integrated activities (not necessarily involving integrated ownership) can be advantageous, e.g., in the poultry industry. The GCAHIT system should be encouraged to expand its organization and joint venture financing of such activities, in alliance with the private sector. GCAHIT should also engage in joint ventures with international breeding companies in the poultry industry to establish Parent and Grandparent stock farms in order to maintain a continuing flow of superior genetic stock. Issues Requiring Further Study 49. Future Supply Requirements. Urban consumer demands for livestock products, as with all other consumer goods, are influenced by changing income and prices, as well as population growth. Per capita incomes in major eastern cities like Tianjin and Fuzhou are expected to grow at annual rates of about six percent in 1984-90. Consumer budget survey data for Tianjin and Fuzhou in 1984 indicate that income elasticities (the percentage increases in demand which would result from a one percent increase in consumer incomes) for livestock products are moderately high. In general, consumer demand for pork and eggs can be expected to grow at rates perhaps 15-25% less than the growth rate of income, whereas growth of demand for poultry meat and dairy products will exceed that of consumer income. 50. However, this assumes unchanged relative prices, and both the effects of the 1985 price adjustments and theoretical considerations indicate that demands for various livestock products are also quite price sensitive. Present local projections of future supply requirements for livestock products do not seem consistent with estimates of relative income and price elasticities for these products. As the 1985 reforms have greatly changed the structure and reduced the level of consumer demand for livestock products, planning projections of supply requirements should be (annually) revised based on new estimates of income elasticities derived from post-1985 consumer budget survey data. 51. The Backyard Alternative in Pig Raising. At present, Government places considerable emphasis on development of specialized, commercial-scale pig production units. Consequently a growing gap is emerging between production performance levels in this (still tiny) segment of the industry and those of backyard producers, who mainly raise unimproved breeds with poorly-balanced feeds. If national sup- plies of protein sources and supplements, as well as the capacity of the feed milling industry, are regarded as short-run limiting factors in growth of pork production, is it more efficient to concentrate these supplies to achieve efficient conversion for SH producers, or to spread them more "thinly" as balancers to improve the diet of the backyard pig? The technical economic data required to answer this question seems lacking at present. Therefore an experimental study is recommended which would produce quantitative economic comparisons of the effects of supplying balancers to backyard producers with either local or improved pig breeds, versus supply of the equivalent amounts of feed to SHs in the form of complete commercial rations. 1. INTRODUCTION Study Scope. 1.01 This study considers the most efficient means of satisfying the growing demand for pork, poultry and dairy products in China's large municipalities. It examines such strategic issues as the choices among alternative sources of production, scale and management form of enterprises, and alternative sources of feedstuffs. It also explores major policy issues, including the roles of state and private sectors in the marketing system, pricing and subsidies, and the provision of services to the livestock sector. 1.02 In emphasizing problems of urban supply of the most important livestock products, the study deliberately ignores the broad topic of livestock raising in the pastoral areas and only touches on questions of livestock product supply for rural consumers. As the livestock sector in China is extremely large and diverse, this is partly due to an attempt to obtain greater depth subject to limited time and resources. A broader coverage of issues in the Chinese livestock sector was included in Annex 2 (China: Agriculture to the Year 2000) of the World Bank economic report, China: Long-Term Development Issues and Options (1985). In particular, the problems of (mainly ruminant) livestock raising in the pastoral areas differ from those of the "agricultural" areas (where crop cultivation is the primary activity). In broad terms, improvement of livestock raising in the pastoral areas has as crucial prerequisites the privatization of herd management, division of collective pasture, and creation of financial incentives for herd reduction and pasture improvement. Policy reforms which may fulfill these prerequisites have been initiated by Government, but time is required for implementation, following which a sector study focussed on the pastoral areas would be appropriate. 1.03 Within the non-pastoral (or "agricultural") areas, growth in urban incomes is a major factor enlarging the market for livestock products, particularily commodities such as Lean pork, broilers, and dairy products which are in short supply and have high income elasticities of demand. Within the rural areas, livestock products at present are mostly self-supplied as sideline activities by private farmers; whereas the volume and higher quality of supply requirements for urban areas demands production by more specialized and often larger scale units, with more elaborate support facilities. The greater scale in turn offers ample scope for efficiency improvements through introduction of foreign technology. Consequently, the problem of improving supply to urban markets is an appropriate starting point for World Bank involvement in the livestock sector. The supply problems are most acute for selected eastern cities with either exceptionally large populations or a limited agricultural hinterland. Tianjin and Fuzhou municipalities were chosen as study sites because their supply problems are particularily acute, and reflect differences between north and south China. - 2 - 1.04 Despite the focus on large municipalities, problems of production, distribution and service delivery in these municipalities are not dissimilar from those elsewhere in China's agricultural areas. After all, large municipalities in China include within their boundaries cultivated areas as large as those of small countries or entire provinces in large countries. Nevertheless, because livestock and feed production capabilities within these municipal boundaries cannot conceivably meet urban consumer demand, this study has also examined possibilities for interregional trade, in particular, the supply of livestock products from more distant agricultural areas to the municipal populations. Two study sites, Shandong and Jiangxi Provinces respectively, were included because they presently serve as suppliers to these two municipalities and the possible expansion of their roles deserved consideration. Characteristics of Study Sites 1.05 Location and Area. Fuzhou Municipality, the capital of Fujian province, is located along the southeast coast of China, facing the Straits of Taiwan to the east. Following recent expansion of its boundaries, the municipality currently includes five districts and eight counties. In 1984, Fuzhou's total municipal land area was 1.196 million ha, of which urban districts covered less than 4,500 ha while rural areas occupied the remaining 1.192 million ha. About 150,000 ha were cultivated, mainly with grains. Nearly 230,000 ha of grassland and wasteland exists, nearly half in large patches and much of it hilly with slopes under 25 degrees. Fuzhou hopes to convert 200,000 ha of this area to pasture for 120,000 head of cattle. 1.06 Tianjin Municipality is situated in the northeast portion of the north China plain, along the Bohai Sea, about 110 km southeast of Beijing. It is one of three centrally-administered municipalities, and is China's third largest city after Shanghai and Beijing (also centrally-administered). Tianjin's total municipal land area covers 1.13 million ha, of which 15,300 ha are in urban districts while the remaining areas are rural. About 625,000 ha were sown to crops, with nearly 80% in grain and 6-7% each in vegetables, oilseeds and cotton. Only some 35,000 ha of reclaimable wasteland is available, much of which is already committed to use in expansion of dairying and aquaculture. 0 1.07 Shandong Province is located between 35-38 N. latitude on the coast in the eastern portion of the north China plain. Its northern border approaches to within 100 km of that of Tianjin Municipality. The western part of the province includes large stretches of the alluvial plain of the Yellow River. About 7 million ha of land are cultivated, or 47% of the total land area of 15.3 million ha. Of a total cropping area of 10.5 million ha, 75% is sown to grain crops (winter wheat, spring and summer corn), 16% to cotton, 6% to peanuts, and the remainder to other cash crops. The province ranks third in the nation in terms of total sown area and grain - 3 - production (2nd in corn and tubers, 6th in soybean); it ranks first in production of cotton and oilseeds. 0 1.08 Jiangxi Province lies between 25-30 N. latitude about 300 km inland from the southeast China coast, to the west of Fujian Province. Most of its 16.7 million ha area is mountainous or hilly, leaving only about 2.5 million ha (15%) for crop cultivation. Because of a high extent of multiple cropping, sown area at present is about 5.5 million ha, of which 68% is sown to grain (mainly double-cropped rice), 4% to rapeseed, 2% to soybean, and 19% to green manure crops (mainly overwintering purple vetch). Jiangxi also claims some 3.3 million ha of underutilized upland regarded as suitable for pasture development. However, only 40% of this area has slopes under 20o and, even if fully utilized for, e.g., beef cattle raising, this would probably add only 3-6% to provincial red meat production./1 1.09 Transportation. Both Tianjin and Fuzhou have relatively convenient access to transportation. Fuzhou is linked to inland and coastal shipping routes through its location along the Min River and proximity to the coast. It has rail and road connections with the remainder of Fujian Province and with other provinces. Tianjin is the major port serving Beijing and the northeast hinterland. Recent installation of container facilities has increased Tianjin's already substantial international shipping trade. Its location at the convergence of five rivers permits significant inland water transport. 1.10 Shandong Province has two major ports linked by rail. to the main agricultural production areas in its interior. The area visited by the Mission, in the vicinity of the provincial capital of Jinan in western Shandong, lies on one of two major north-south trunk rail lines, giving direct access to, e.g., the markets in Tianjin or south China. Most surpluses of livestock products or feed grains would, however, move by coastal transport from the eastern production bases to other provinces or abroad. Northern Jiangxi is well-linked by rail with provinces to the north, east, and west; the main pig production areas, including Dongxiang County which was visited by the Mission, have access to these lines and thence, e.g., to the Fuzhou market. Only the mountainous southern part of Jiangxi, where much of the potential pasture land is located, is relatively isolated. In gene- ral, in all of these municipalities and provinces, a network cof paved secondary roads link county seats with each other and with major cities, but below this level, the rural road system is far from adequate. 1.11 Population. In 1985, Fuzhou's total municipal population stood at almost five million, of which 667,000 or only 14 percent were residents of the urban core. Tianjin's total population in 1.985 was about eight million, of which 4.3 million or 53 percent were within urban boundaries. Tianjin therefore has a ratio of urban to rural /l Based on a 1.2-2.4 t/ha dry matter yield, about 3.8 t/ha required per beef animal, with 50 kg meat offtake per year, 1.3 million ha of pasture would produce about 20-40,000 t of beef p.a. residents of 1.15:1 compared to Fuzhou's ratio of 1:6.3. Despite urban growth rates in recent years ranging from 2.2-5.4% p.a. and 2.1- 3.1% in Fuzhou and Tianjin respectively, urban planners optimistically expect future annual growth rates to decline to only 0.4-0.6 percent and 1.1 percent respectively. Shandong and Jiangxi have total populations (1984) of 76 million and 34 million, of which 49% and 22% respectively are "urban" (meaning residence under jurisdiction of city or town). 1.12 Income Levels. Local statistics suggest that per capita incomes of urban residents in Tianjin slightly exceed those of Fuzhou (Y677 and Y620 respectively). Substantial leaps in urban incomes in various years in 1982-85 were attributable mainly to public sector wage adjustments, and have been at least partially nullified by cost of living increases. Even higher growth rates in incomes of rural residents reflect recent improvements in the agricultural policy environment, which have allowed suburban farmers to specialize in high-value crops such as vegetables and receive increasingly favorable prices. In 1984-90, per capita urban incomes in both cities are expected to grow 50-60%, and rural incomes 75-100%. Although exact figures on urban incomes in Shandong and Jiangxi Provinces are unavai- lable, average urban wage levels in these provinces fell about 7-17% below those in, e.g., Tianjin. Per capita net incomes for the farm population in 1985 were Y408 in Shandong, Y377 in Jiangxi, and Y565 in suburban Tianjin. 1.13 Livestock Production. Shandong and Jiangxi are both provinces which have surpluses of grain and grain byproducts, and have successfully converted part of these into exportable surpluses of livestock products (Table 1.1). Jiangxi is strong only in pork production, whereas Shandong also produces significant quantities of other red meats, eggs and milk (mainly from goats). Because of the large expanse of rural area incorporated within municipal boundaries, Fuzhou and Tianjin have achieved a partial self-sufficiency in livestock production. Tianjin is particularily strong in egg and milk production, but correspondingly weak in pork; whereas Fuzhou's strength in pork and poultry meat is offset by deficiencies in other products. These strengths and weaknesses on the whole reflect productivity levels in the agricultural regions in which these municipalities are situated, rather than characteristics of local policy. For instance, per capita rural consumption levels of red meat and poultry in Tianjin seem very low, but in fact are slightly better than those in neighboring Hebei Province. 1.14 Livestock Consumption. Rural producers sell the majority of their production of livestock products to the urban market, which also receives supplies from more distant areas. Consequently urban per capita consumption greatly exceeds both total per capita production and the consumption level of rural residents (Table 1.2). The levels of consumption in Fuzhou and Tianjin slightly exceed the urban norms; also, these coastal cities are bases for fishing fleets and consequently aquatic products consumption is far better than - 5 - Table 1.1: Production of Livestock Products, Study Sites 1985 (kgs per capita) Municipality: Province: Fuzhou Tianiin/a Shandong Jiangxi Meat and poultry 14.3 N/A 16.8 18.6 Pork 11.5 6.8 14.5 17.1 Mutton 0.2 0.4 0.7 0.03/b Beef 0.1 0.1 0.6 0.1/b Poultry/rabbit 2.5 N/A 1.2 1.4/b Eggs 3.8 9.0 9.5 3.1 Milk 3.8 4.9 2.0 0.5 /a 1984 figures. /b Estimates derived partly from 1984 data. Table 1.2: Personal Consumption of Livestock Products, Study Sites, 1984 (kgs per capita) Urban Urban Urban Rural Rural China Fuzhou Tianjin China Tianjin Pork 17.1 15.0 18.5 Beef and mutton 2.8 1.2 3.7 ) 10.6 7.8 Poultry 2.9 5.1 2.5 0.9 0.3 Fresh eggs 7.6 7.6 12.8 1.8 3.2 Fish and shrimp 7.8 35.4 11.6 1.7 3.5 average./1 Both municipalities have attempted to meet the demands due to increasing urban populations and rising per capita incomes with increased supplies. During 1982-84, Fuzhou was unsuccessful in in- creasing per capita red meat supply, but did make major improvements in supply of poultry, eggs and milk. Supply improvements in Tianjin have been remarkable for all livestock products except mutton. /1 In fact, the personal consumption figures, derived from family budget surveys, understate actual urban consumption (and therefore the excess over rural consumption), as they reflect only direct purchase of livestock products and ignore quantities consumed in restaurants, worker's dining halls, etc.; also certain parts of animals which are nonetheless consumed (e.g., head, liver, feet), egg products, and canned meat are excluded from these figures. If, for example, the total amount of pork supplied to urban Fuzhou is divided by the urban population, then total direct and indirect consumption of pork (at carcass weight) would have been about 38 kg/capita in 1984, compared to 15 kg/capita direct pork meat consumption as implied by the consu- mer survey; and 11.7 kg/capita of eggs compared with 7.6 kg/capita. - 6 - 1.15 Further information on the two municipal study sites is given in the Annex tables. In the discussion which follows, other data collected from the study sites will be used mainly to illustrate analysis of general problems in the supply of pork, poultry and dairy products to the urban population. Changing Policies and Performance 1.16 China's livestock sector is dominated by household production units for whom pig and chicken raising is a sideline which takes full advantage of surplus resources (labor and crop byproducts). Its contribution to national income is only about 14% of value-added in farm activities, and 11% of net incomes of farmers. Even though China possesses about 5% of the world's cattle and 12% of its sheep, ruminants are mostly raised either on a subsistence basis or as draft animals, and contribute little to total meat production. Of the latter (measured at carcass weight), pigs contribute about 88%, poul- try 8%, and other livestock only about 4% /l. 1.17 Under this traditional system, over the last thirty years meat production has been closely determined by availability of feed in the rural areas, including mainly grain byproducts and such surpluses of grain above requirements for seed and human consumption as were retained by or resold to farmers (Graph 1). As long as the grain surplus was small, meat production could grow only as fast as bypro- duct availability, i.e., hardly faster than grain production as a whole. But because the income elasticity of direct grain consumption is low, a large proportion of grain surpluses can be used as feed; hence quantities of feed available are highly elastic with respect to increases in grain production beyond a minimum subsistence level. Consequently the several bumper grain harvests since 1978 have been reflected in significantly higher growth rates for meat production than experienced earlier (Table 1.3). 1.18 Although crop byproducts have little use except as animal feed, whether or not grain surpluses are converted into livestock production would normally be strongly influenced by relative procure- ment prices of animals and grain, and policies toward private (free market) marketing, as determined by Government. However, following communization in 1958 and continuing through 1984, pig procurements from farm households were compulsory and the Government appears to have made little direct use of the price instrument (Graph 2). Rather, feedgrains retained or distributed at subsidized prices in the rural areas (accounting in 1980 for about 60% of total livestock requirements of feed grains, not including byproduct) were closely tied to deliveries of finished pigs and eggs to the state, as well as /l For a broader description of livestock performance in 1979-82, see World Bank, China: Long Term Development Issues and Options, Annex 2 (China: Agriculture to the Year 2000), 1985-, pp. 72-87. -7- Graph 1. Pork Production and Feed Availability ia soo- (Index: 1952 = 100) Pork -l Grcin 500- 400- 300- 200- 100 ln r N LO 00 ) Ln I%nra to r- tCo w CO i 0 ' 0M COQ'4 Year /a Pork for current year, feed availability for previous year. Graph 2. Procurement Price Ratio: Pork to Grain (Index: 1952 = 100) 140- ---price 130 - 120- 110- 100- So l I I I I I I I I I I I I 1 I N I r- N in C 0 0 tO inO Wi r tY- e r Ca000$O '0 h0 Yeiar - 8 - Table 1.3: Livestock Performance, China, 1978-85 1978 1979 1980 1981 1982 1983 1984 1985 Yearend inventory (millions) Pigs 203.3 319.7 305.4 293.7 300.8 298.5 306.7 331.4 Cattle 70.7 71.4 71.7 73.3 76.1 78.1 82.1 86.8 Sheep and goats 169.9 183.1 187.3 187.7 181.8 167.0 158.4 155.9 Annual slaughter (millions) Pigs 161.1 187.7 198.6 195.0 200.4 206.6 220.5 238.8 Cattle N/A 3.0 3.3 3.0 3.1 3.5 3.9 N/A Sheep and goats N/A 35.4 42.4 44.8 48.7 49.2 51.8 N/A Slaughter rate (percent) Pigs 55.2 62.3 62.1 63.8 68.3 68.7 73.9 77.9 Cattle N/A 4.2 4.7 4.2 4.2 4.6 4.7 N/A Sheep and goats N/A 20.8 23.2 23.9 26.0 27.1 32.1 N/A Meat per animal (kgs) Pigs 49.0 53.4 59.8 61.0 63.5 63.7 65.6 73.7 Cattle N/A 77.4 80.9 82.4 85.8 90.7 96.4 N/A Sheep and goats N/A 10.7 10.5 10.6 10.7 11.1 11.6 N/A Total meat production ('000 t) Pork 7,890 10,014 11,314 11,884 12,718 13,161 14,448 16,547 Beef 310 230 269 249 266 315 373 467 Mutton 360 380 445 476 524 545 586 593 Subtotal 8,560 10,624 12,054 12,609 13,508 14,021 15,406 17,607 Milk production ('000 t) Total 971 1,307 1,367 1,549 1,959 2,295 2,596 2,894 Cow's milk 883 1,070 1,141 1,291 1,618 1,845 2,186 2,499 /a Updated from F. M. Surls, "China's Agriculture in the Eighties", in Joint Economic Committee, U.S. Congress, eds. China's Economy Looks Toward the Year 2000 (Vol. 1 The Four Modernizations). Washington, D.C.: U.S. G.P.O., May 21, 1986, p. 342. draft animal requirements. Relative prices of livestock and grain nevertheless may have influenced feeding techniques: for instance, marginal feed costs can be reduced by finishing pigs to heavier (and fatty) live weights on lower-quality feedstuffs. The trend toward heavier animals has been pronounced in recent years (Table 1.3). 1.19 It is possible that free market, rather than official procurement prices, have played a more significant role in determining - 9 - levels of livestock production, at least at the margin. However, the pig:grain price ratio on the free market, at least in 1984, may have been less favorable than the similar ratio of official procurement prices (rural free market grain prices were 54% and livestock prices 31% higher than average procurement prices). On the other hand, favorable free market prices have been largely responsible for the doubling of poultry production in 1980-85 (to about 1.9 billion birds and 1.5 million t of meat), as Government control over poultry raising has been limited (eggs are procured by state organs, but not under a compulsory delivery scheme). 1.20 Excellent grain harvests in 1982-84, combined with farmer disposal of accumulated stocks of grain (encouraged by high government procurement prices and declining market prices), put the government by 1984 in the position of procuring 30-40 million t more grain than urban, industrial, export and strategic reserve requirements justi- fied. Abandonment of the systems of quota grain procurement and tight rationing of feed grain use were matters of necessity as well as opportunity. Therefore, in 1985, the government cut back its grain procurements by 26 million tons (and reduced marginal procurement prices), forcing farmers to find alternative ways to dispose of grain surpluses; raised the procurement price of pigs and other livestock products to approximately the level of rural market prices; correspon- dingly increased retail prices in the cities and towns; and ceased to require compulsory deliveries of pigs to state procurement stations. 1.21 The full effects of the 1985 reforms are unclear, but the short-term effects were to (a) stimulate increased livestock produc- tion and slaughtering; (b) cause a constriction of urban demand for livestock products after May 1985; and (c) shift a significant propor- tion of the livestock trade from the state commercial organs to the urban free market and other, more direct links between producer and consumer. The reported effects on livestock production are impres- sive: the year-end stock of pigs rose 8.0% to 331 million (ending six years of stagnation); numbers slaughtered rose 8.4% to 239 million head; and total red meat production rose 14.5% to 16.5 million t. Production of cow's milk reached 2.5 million t (up 14.2%); and egg production rose about 23% (early report). End-of-1985 reporting also initially claimed tremendous increases in supplies to the urban areas. However, final statistics indicate that total sales of pigs rose by only 3%. Sales by state commerce in the second half of 1985 fell 40%; July-December pig procurements normally range from 99-107% of sales, but in 1985 were 114% of sales, implying overstocking by the state commercial organs and perhaps significant product losses. Direct marketing by farmers in urban free markets took up part of the slack, increasing in 1984-85 from 13% to 23% of total sales. The residual forms of distribution - presumably direct procurements by catering and other industries from producers or private wholesalers - increased from 9% of the market to 28%. - 10 - Organization of the Sector 1.22 The livestock sector, which includes supply of production inputs (feed, breeding stock, and services), livestock production, processing, storage, marketing and distribution, is administered in China by several different agencies. In general, the feed supply is controlled by Grain Bureaus ranging from the Central Government to the county level; Animal Husbandry Bureaus are responsible for production planning and services; Commercial Bureaus are responsible for processing, storage, transportation, and marketing. Responsibility for procurement for domestic consumption rests mainly with the Commercial Bureaus, and, for international trade, with agencies belonging to the Ministry of Foreign Trade and local governments. Interprovincial trade is coordinated by the Central Government (Ministry of Commerce) and negotiated bilaterally between provinces. Chart 1 describes the overall organization of the sector, which is described in greater detail below. 1.23 Ministry of Agriculture, Animal Husbandry, and Fisheries (MAAF). At the Central Government level, two bureaus under MAAF are involved in livestock production: the Animal Husbandry Bureau (AHB) is responsible for project planning, veterinary and quarantine work, technology transfer, personnel training, and varietal improvement for farm households under the collective system. The Bureau of State Farms (BSF) has the same responsibilities for livestock within the independent productive organization of the state farm system. Also under the guidance of MAAF, the General Corporation for Animal Husbandry, Industry and Trade (GCAHIT) is an independent accounting unit, a licensed, registered and capitalized corporation with investments in companies (usually joint ventures) distributed throughout the country. GCAHIT also has subordinate, specialized companies, such as the Beef Cattle Company or the Apiculture Company. The Corporation's authorized functions extend beyond MAAF's traditional role in production to include slaughter, processing, transportation, storage and marketing of livestock products. 1.24 Ministry of Commerce. Three bureaus are involved in the livestock sector: the Grain Bureau (GB) is responsible for the plan- ning and provision of feedgrains and oilseed meals and administration of much of the feed processing industry. The Animal Husbandry Bureau (AHB) supervises procurement, slaughter, and state marketing of live- stock products. The Price Bureau, operating under the supervision of the State Price Commission (which sets pricing principles and prices of main commodities), determines average levels and allowable ranges for specific commodity prices (list and negotiated procurement prices and wholesale and retail prices). The price ranges for livestock products are set by region, season and quality grade. In addition, the Supply and Marketing Cooperatives (S&MC) fall under MC authority. At the grassroots level, they manage supply of inputs other than feedgrains to the livestock sector and procurement of some livestock products, mainly from small-scale producer households. ！•必鑿 ••••．•一，•••••．••－一•••．••••．•．••．•一••．•••••••••二••．••．•．•••．••．•一•．•••二••．•．•••．••••．一•••••．•二••．••■••－．•．•••．•．•••．•••．•••二•■一••．••••••．•••，•■•．一•．•••．•二•二•••．，•．--．•••••．•••• 開正寫開叩籐閑方方口n頂乞魚開打 取取口驪U江細絮口取日刃阿 啊面他 - 12 - 1.25 Other Central Government Agencies. Under the Ministry of LLght Industry, the Food Processing Bureau is responsible for proces- sing of livestock products, including management of some major food processing enterprises. Because a number of different agencies engage in feed processing or control raw materials, the National Feed Indus- try Office (NFIO) was set up in 1985 under the State Economic Commis- sion (SEC) to help formulate policies for industry development; coor- dinate an investment program; determine standards for various feed- stuffs and regulations to enforce them; developing a testing program and appropriate organizations; distribute technological information; arrange for training and technical exchanges; and contribute to plan- ning exercises by the State Planning Commission (SPC). 1.26 Quasi-Governmental Organizations. Two professional or- ganizations play important roles in the sector: the National Feed Industry Association, under the authority of the SEC, was originally set up to serve most of the functions which have now been taken over by the NFIO (see para. 1.25). Its present role apparently is mainly to facilitate technical exchange and training, stimulate research, and serve as a professional forum for the industry. The National Animal Husbandry Association, under whose umbrella may be found the Dairy Association, Poultry Association, etc., serves a similar role for producers of livestock products. These secondary-level organizations are becoming a major vehicle for coordinating and controlling develop- ment of specialized household livestock production and, at the grass- roots level, providing extra services to specialized producers. 1.27 ProvincialZMunicipal Government. Although in China the Central Government sets basic policies and supervises or coordinates activities of agencies at lower levels of government, most direct administrative responsibility rests at lower levels, hence a surpri- sing degree of diversity can exist in practice among geographical administrative units. In terms of hierarchy, below the Central Government level, the governments of the provinces and three province- level municipalities (Beijing, Tianjin and Shanghai) are similarily organized, and organizational forms at this level are largely repli- cated at lower levels, accompanied by decreasing administrative inde- pendence. At the provincial (municipal) level, administration of the livestock sector is coordinated by two committees, usually chaired by a deputy governor or mayor: the Agricultural Committee Zl coordinates production-oriented agencies such as the Grain Bureau, Agricultural Department, and Bureau of State Farms; and the Finance Committee coordinates agencies responsible for procurement, processing and dis- tribution, such as the Second Commercial Bureau, Bureau of Industry and Commerce, International Trade Bureau, Price Bureau, etc. 1.28 Grain Bureau (GB). This bureau is responsible for (a) procurement of grain and oilseeds from the individual farmers and /I In some provinces, an Economic Committee instead of an Agricultural Committee coordinates production-oriented agencies in agriculture and other sectors. - 13 - state farms, through Grain Procurement Stations located in the rural areas; (b) processing and distribution of grain and oilseeds, inclu- ding allocation to lower-level geographic units and various indus- tries, including the feed industry. The Feed Company is organized as an independent accounting unit under the GB to supply feedgrains to farmers and mills, and manufacture and distribute compound feedstuffs. The Feed Company manages most feed mills at levels above the county, as well as transport teams which distribute simple and compound feeds to Feed Supply Stations in the rural areas or directly to large users. 1.29 Agricultural Department. Within this department, the Animal Husbandry Bureau (AHB) is responsible for livestock production plan- ning and oversight, formulation of and investment in development projects, organization and supervision of the veterinary and quaran- tine services, technical extension and training, varietal improvement, and other services, such as planning allocations of credit (which is extended to farmers by the Agricultural Bank of China only with go- vernmental approvals or recommendations). County (or district) AHBs, operating under the dual leadership of the provincial-level AH1B and county government, share the above responsibilities, with greater emphasis on service provision: veterinary stations, breeding stations for artificial insemination (AI) work, and breeding farms are directly managed by county AHBs. At the provincial level, the AHB also has its business arm, the Corporation for Animal Hubandry, Industry and Trade (CAHIT), which is allowed to venture into feed milling, product pro- curement, processing, and domestic or international trade (see para. 1.23). 1.30 Bureau of State Farms (BSF). Although at the Central Government level, the BSF has been merged into MAAF, at lower adminis- trative levels, the BSF remains a separate organization parallel to the AD which administers state farm production of milk, meat and eggs. Outside of the pastoral areas of China, BSF's role in the livestock sector is most significant in dairying. Under BSF, the Dairy Compa- ny/l administers state-owned dairy farms, milk collection stations serving specialized households, dairy processing plants, and the dis- tribution of dairy products down to the retail level. Guidance of the quasi-governmental Dairymen's Association also comes from the Dairy Company. In general, the responsibilities and services provided by the Agricultural Department for the collective sector are replicated within the BSF system for the state farms; and coordination between the former and BSF is weak at provincial and lower levels. l "Companies" such as the Dairy Company, Feed Company, Food Company, etc. differ in organizational form from "corporations" such as GCAHIT or CAHIT, in that the "companies" are administered directly by governmental bureaus, receive all investment funds from budgetary sources, receive subsidies to cover losses, and surrender profits to parent bureaus, whereas the "corporations" are semi-independently managed, accept non-governmental investment funds, and are responsible for their own profits and losses. - 14 - 1.31 Second Commercial Bureau. This bureau administers procure- ment, processing and marketing of many food products (not including grain, oilseeds and dairy products) through various companies, such as the Food Company, Restaurant Service Company, Non-staple Foods and Condiment Company, and the Sugar, Tobacco and Alcohol Company/1. It also has links to the S&MC at the grassroots level. The Food Company is an independent accounting unit owning and managing various enter- prises, including: (a) pig collection stations and slaughter houses; (b) processing plants for meat and eggs; (c) transport teams; (d) cold stores; (e) state wholesale and retail markets for fixed-price sale of livestock products. Processing plants under other Bureau companies (e.g., beer, wine, or candy production enterprises) are sources of byproducts which are valuable as livestock feeds (brewer's grain, bagasse, molasses, dregs from bean product production, etc.). 1.32 Price Bureau. At the provincial (municipal) level, the Bureau reports to both the Central Government's State Price Bureau and the provincial government. Within the ranges set by the Central Government, the Bureau determines local official prices for procure- ment and wholesale/retail sale of livestock products. In doing so, it considers the advice or requests of the agencies responsible for production and commerce, as well as any implications for the provin- cial (municipal) budget. 1.33 Bureau of Industry and Commerce. This agency has overall responsibility for administration of marketing, including supervision of prices, tax collection, and commodity quality standards. Under the Bureau, the Market Administration Agency (MAA owns and operates the agricultural trade markets (i.e., urban free markets) where livestock products are sold, and licenses private traders. 1.34 International Trade Bureau. Operating under the dual leadership of the Central Ministry of Foreign Economic Relations and Trade and local government, this bureau is responsible for procurement of livestock products for export and import of feedstuffs, pharmaceu- ticals, etc. It has considerable autonomy in determining contract prices for export procurement, which are usually higher than normal official procurement prices, in part reflecting greater attention to quality standards. 1.35 Production System. The production of livestock products in China is carried out on farms varying widely in scale and modernity of technology. Overall, individual households raising a few animals each as a sideline dominate the market (para. 1.16), but Specialized House- holds (SHs) raising larger numbers of animals are growing in impor- tance, especially in poultry and (in some areas) dairy sectors. In 1985, there were about 150,000 pig SHs accounting for about 3% of total pork production, and most prevalent in eastern, coastal areas where feed milling is better developed and improved, hybrid stock more readily available. State- or collectively-owned farms are significant mainly in dairying and poultry production (nationally their production accounts for about 3% of pigs, 4% of cattle, and 5% of sheep and goats), and many are not very large. The organizational structure of - 15 - the industry differs regionally. For example, in Tianjin, production by large state or collective farms accounted for about 80% of total egg consumption, all of milk consumption, and little or none of pork consumption, which depends mainly on procurement from small-scale farms. Fuzhou depends mainly on small private farms for its milk (90%), but depends to a somewhat greater extent on large farms for poultry and pork production. Sources of supply in 1984 for both cities, including amounts produced by state and collective enterprises, specialized households, and individuals, and the range of scale for the state and collective sector in Tianjin, are shown in Annex 1. 1.36 The feed and livestock product processing industries are almost entirely outside the private sector at present, although small, labor-intensive abattoirs serving the market economy are being set up as joint ventures among private middlemen. Feed processing capacity in Tianjin in 1984 totalled about 167,000 t/year; five state-owned mills with over 10,000 t/year rated capacity accounted for 48% of annual production (in 1985 it is projected that the share of these mills will fall to 37%). Mills owned by townships or villages with rated capacities in the range 2,000-6,000 t/year were generally produ- cing at only a fraction of their capacity in 1984-85, both because many were in startup phases and most had difficulty procuring adequate supplies of raw materials (most small mills depend on raw materials procured at negotiated prices). Feed mills in Fuzhou produced about 140,000 tons of feed; three large mills accounted for 46% of the total. Fuzhou also has abattoirs with a capacity of 4,500 pigs per day and a cold storage capacity of 11,000 tons; Tianjin has one large and relatively modern abattoir with 3,300 pig per day capacity, and a handful of smaller abattoirs. 1.37 Organizational Weaknesses. The most obvious problem is the division of authority over the livestock sector, in which each minis- try and/or bureau is responsible only for one or a few aspects. The only exceptions are GCAHIT and the Dairy Company, which have the authority to vertically integrate their operations, and are motivated to do so by difficulties in coordinating with other administrative lines as well as remaining price distortions in the sector. Coordina- tion is not assured even at the municipal level, where two different committees preside over production/feed supply and processing/distri- bution respectively. Coordination would be less of a problem if each farm or enterprise operated in a business-like way (e.g., products or services not tied to particular distribution channels) and if prices at each level were set by market forces. At present, however, none of the agencies directly involved in livestock has much control over prices, other than transfer prices within its own system; and adminis- trative control by agencies of subordinate operations remains tight. - 16 - 2. THE MARKET FOR LIVESTOCK PRODUCTS Organization of the Marketing System 2.01 In response to the ongoing economic reforms in China, livestock production has developed very rapidly, and the marketing system for livestock products accordingly has changed. The state monopoly of the distribution of livestock products has been abolished, and replaced by a market in which state and private trade compete. In both the urban and rural areas, the market has been undergoing gradual evolution from a simple to a more complex system. 2.02 Marketing 1956-85. The urban marketing of livestock pro- ducts was monopolized by the state from 1956 until the 1980s. The state Food Companies purchased live animals from the state and collec- tive farms and individual farm households. After slaughter, the red meat and mainly live poultry were sold in the state food stores in the urban areas. In the rural areas, on the other hand, most farmers consumed their own pigs, poultry and eggs at special times, e.g., during the agricultural busy season and certain holidays. 2.03 Under this system, farmers annually delivered set quotas of pigs to the Food Company's purchasing stations. These animals were raised largely on crop waste and byproducts, and rarely were more than two or three pigs raised per household because of the limited availa- bility of concentrate feed, the distribution of which was tightly controlled. In the cities, red meat and eggs were rationed and unra- tioned commodities such as poultry were infrequently available. In the rural areas, farmers were unable to eat meat very often. Surplus pork, mutton or poultry and eggs were sold in limited amounts by farmers illegally on the urban black market, and eggs legally in rural free markets. In addition to household supply, a small proportion of livestock products as well as most milk were produced by state and collective farms and distributed through the state procurement system. Urban supply of milk was confined to amounts made available under rationing for old, sick or undernourished people and children under three years old. In summary, this monopsonistic marketing system failed to satisfy the demand for the livestock products. 2.04 Following the rural economic reforms begun in 1979, the total supply of foodgrains increased rapidly, leaving a growing sur- plus in the hands of both the state and the farmers. The transforma- tion of this surplus into livestock products has emerged as a major policy issue. But because, prior to 1985, grain surpluses (above quotas) could be sold to the state at premium prices, the price ratios between grain and most livestock products provided little incentive for livestock production (see Graph 2) and farmers generally found it more profitable to sell their grain rather than use it as feed. 2.05 Since May 1985 the government has reduced the level of procurement of grain along with the marginal procurement price, and simultaneously has increased the official procurement prices of live- - 17 - stock products. This has stimulated the development of livestock raising and the emergence of specialized producers of livestock, and led to rapid increases in marketed volume. The old marketing approach was both unnecessary and unsuitable in this new situation, hence the government has opened up the market completely, allowing farmers to sell to the state commercial organs, directly to the catering trade, or to the rural or urban free market. In the process, the very concept of the "market" has changed, from a temporary, limited, and mainly local rural "free market" to the highly organized and increasingly comprehensive "agricultural trade market" (ATM). 2.06 Previously the "free market" was organized spontaneously by farmers and involved no fixed facilities: the marketeers put their commodities on the ground (e.g. baskets of poultry or eggs) or on crude tables or scaffolds; as most markets were outdoors, the market closed during inclement weather. The government was more concerned with controlling the scope of the market than with servicing its needs or encouraging its development. The volume of commodities sold was small relative to total retail sales, supply was highly seasonal, and prices unstable. This description is less applicable to the ATM. 2.07 The Agricultural Trade Market. Today, in municipalities such as Tianjin and Fuzhou, large multi-purpose markets have been established in either multi-story structures or under roofed shelters. This type of market is usually divided into several sections, inclu- ding meat (pork, beef, mutton and other red meats), poultry (chicken, ducks and geese), aquatic products, vegetables, fruit, dried fruit and vegetables, grains, edible oils, etc. The poultry section in turn may be divided into subareas for live poultry (usually located outside of the market building) and dressed poultry respectively. 2.08 A typical large urban ATM may have over 300 stalls, of which some are leased by marketeers on a long-term basis and others are rented temporarily or daily. The market administration agency provides the electricity, water and other facilities and the marke- teers pay rents based on the cost of these services. In Tianjin's Xingfu Road ATM, with an area of 12,000 sq m, there are 171. permanent and over 100 temporary stalls. Fuzhou's Xiliying Market, an enclosed market with 4,080 sq m area, has 267 permanent and 100 temporary stalls. 2.09 In these large urban markets, daily trading volume ranges from Y30,000-90,000, depending on market and season. Meat, aquatic products, and poultry (live and dressed) account for over 60% of the stalls. Although markets are open 24 hours per day, in the summer meat is sold mainly in the early morning. In order to guarantee the quality of livestock products and poultry, meat and poultry must be certified as vaccinated and in some markets, marketeers are required to have refrigerators for meat storage. 2.10 At present the large municipalities typically have opened many middle-size markets and small "street markets" in addition to a handful each of big markets. The smaller markets are often specia- - 18 - lized in a few types of commodities, open for a shorter period (e.g., early morning), and involve a smaller sales volume; but offer greater convenience to consumers because of the dispersed locations. 2.11 In addition to the ATM, state food markets continue to play an important role in the supply of nonstaple foods, especially in big cities like Tianjin, Beijing and Shanghai and some provincial capitals. For example, in Tianjin the sales volume of meat, poultry and eggs in the state food markets continues to be more than 80% of total sales volume of livestock products. Prices in the state food markets are currently 5-10% lower than those of the ATM for the same type of commodity. 2.12 Rural markets differ slightly from municipal markets in the extent of specialization, diversification and flexibility. The retail food markets, which sell mainly meat, poultry, vegetables, fruit, condiments and some snacks, are similar to urban markets, but rural areas also have live animal markets for sale to producers, and production input markets, where tools and implements as well as some kinds of raw materials are sold. In addition to permanent markets (usually located in rural towns), the rural areas also have periodic markets which operate on regular schedules (e.g., every three days) or only on special occasions such as during the Spring Festival (Lunar New Year). Previous to the recent reforms, farmers could buy meat only at specific times once or twice a week, whereas now they can buy at any time in county market towns. 2.13 The urban marketing of dairy products is carried out almost entirely through the state Dairy Products Company, because of rationed supply and public health regulations. Milk Purchasing Sta- tions under the Company are responsible for procurement from small producers, quality checking, and delivery to the processing plant (state farms deliver direct to the plant), where the milk is pasteu- rized and then sold to the consumers through milk distribution sta- tions and milk shops. Milk sale in the ATM is usually illegal, al- though 10% of milk volume in Fuzhou passes through the free market. 2.14 Functional Specialization. As the volume of livestock marketing and the significance of the ATMs increases, one would expect changes in organization to occur which would improve efficiency. Two such changes, each representing kinds of functional specialization, are indeed observable, the first establishing a division of labor between private and state marketing systems and the second a division of labor within the private system. 2.15 Generally, the suburban livestock production of major municipalities cannot meet the urban demand for livestock products. For example, both Tianjin and Fuzhou depend heavily on imports of livestock products from other provinces (see Annex 1). These imports, arranged by the state Food Companies, have had the disadvantages that costs, particularily due to cold storage, were higher than for fresh product and product quality suffered from the long time lag between slaughter and sale. However, the Food Companies can take advantage of - 19 - their cold storage facilities and access to the long-haul transport facilities to procure in large volume from the cheapest sources at times of surplus. This provides a cost advantage which enables state markets to compete with ATMs. 2.16 A second trend is a rapidly growing degree of market specialization. At present, urban markets have grown to include both wholesale and retail markets. For example, Fuzhou's Taijiang Market is mainly a wholesale market, in which about 70% of commodities are purchased by private retail merchants as well as other large-volume purchasers such as hotels and restaurants. Also, the number of mid- dlemen and specialized marketeers has been growing, as the increasing scale of household production precludes involvement by the farmers in the marketing of their products. Also middlemen are increasingly becoming specialized in such functions as procurement, transport and slaughter, wholesaling, or retailing. With the present freedom to procure and transport across administrative boundaries and to invest in transport equipment and refrigerators, middlemen in principle are able to procure from more distant areas and slaughter locally for immediate distribution through private wholesalers to private retail marketeers. For example, in Fuzhou's Taijiang wholesale market, the daily wholesale volume of a single stall during the Spring Festival can reach 30 tons for aquatic products or over 100 head of pigs. 2.17 A contrary trend, which nevertheless increases efficiency, is for state-owned food markets, hotels, and restaurants to exercise their new right to procure from any source to bypass the Food Company procurement system in order to purchase livestock products based on comparative price and quality from sources within or outside of a province or municipality. For example, Tianjin's Dadu Road state food market during the winter season purchases 40% of its pork and 30% of its eggs directly from outside of Tianjin, and 20-30% of its aquatic products from Shandong and Zhejiang Provinces. Also, other state-run organizations, such as enterprises belonging to CAHIT, have begun to procure, process and distribute livestock products in competition with the Food Company. These new paths of distribution apparently accounted for nearly 28% of pork sales in 1985 (para. 1.21). 2.18 Price Determination. Three broad systems of price determi- nation are now in effect in China, specifically (a) state fixed prices, (b) guidance prices, and (c) market prices. The central go- vernment's Price Bureau determines the level and range of regional price differentials for commodities under both the fixed and guidance price system. The main distinction between these two systems is that guidance pricing allows some discretion to local Price and Commercial Bureaus to adjust actual prices to local conditions, within a certain range around the regional guidance price (roughly plus or minus 10%). Some 17-18 major agricultural commodities remain under the fixed price system, including grain, oilseeds, cotton, sugar, and tobacco. The 1985 reforms shifted most livestock products from guidance pricing to market price determination, reportedly leaving only live pigs, wool (informally), and possibly eggs under guidance pricing. However, at - ZU - the province or municipal level and below, intervention in the pricing system may exceed the levels dictated by central government policy, and it appears that a high degree of government price control still exists in the municipal marketing system for livestock products. 2.19 Between 1981 and May 1985, two kinds of procurement price of livestock products were implemented at the local level - a quota procurement price under which farmers sold a set amount of product (entitling them to the use of collective land for feed produc- tion and/or to procurement of feed at a low list price, the feed amounts being in fixed proportion to the weight of livestock product sold); and negotiated prices for sales above quotas or not subject to quota procurement./1 Procurement prices were set according to grades of animal, and variations in grading practices provided some addi- tional local flexibility. Official wholesale prices were also deter- mined by the central government, and negotiated wholesale prices were agreed between buying and selling organs on sales between administra- tive units (e.g., in different provinces) outside planned (and subsi- dized) levels of transfers. On the retail side, the Price Bureau also set list prices under which livestock products were sold to consumers, at certain times and places with ration coupons. 2.20 During this period, because the margin between procurement and retail prices was insufficient to cover the costs of transportation, losses, processing and storage, and distribution, the government, through the Food Company, subsidized procurement and sale of livestock products to the extent of about 5-10% of retail prices. The subsidies on livestock products and vegetables, which were inten- ded to help guarantee urban supply, became an increasing burdensome part of Government's expenditure as production increased. The pricing and marketing system for livestock products (as well as vegetables) was changed in 1985 partly in order to reduce these subsidies. There is now a single procurement price for state procurement organs (to which the exchange for feed still applies) and a single retail price for state markets (or shops), slightly below that found in ATMs. For pork and eggs, the indicative official procurement prices and state retail prices and the allowable range of variation (for procurement prices, plus or minus 10%) are still determined by the Price Bureau at the central and municipal level. The minimum prices are less strictly enforced than the upper bounds, although state markets usually set prices above the indicative minima in order to preserve profit mar- gins. 2.21 The market prices which prevail in the ATMs are more flexible, being largely determined by equilibrium of supply and demand. However, there are two significant limits: a) to impose a measure of orderliness on the market, the ATM administration displays a maximum price each morning after consultation with marketeers, which /1 So-called "negotiated" prices are set on cost-plus principles, are subject to approval by Price Bureaus, and do not necessarily fluctuate fully in accordance with supply and demand conditions. - 21 - is enforced throughout the day (usually prices fall as the day progresses and the best produce disappears); b) as the state market still accounts for more than half of the sales of all livestock products except poultry, the difference between ATM prices and state market prices is limited by the willingness of consumers to pay a premium for quality, variety and choice of cut, provided that state markets have sufficient supply to meet existing consumer demand. 2.22 Within these limits, seasonality and special holiday demands are major factors causing short-run fluctuations in supply and demand and thus prices. Under Chinese dietary customs, in winter more meat is consumed than in summer, and consumption of food such as meat, poultry, eggs and seafood peaks during holiday periods. In urban areas, consumption of nonstaple foods also tends to be higher on weekends, when people have the leisure to prepare fancier meals. On the supply side, there are strong seasonal variations. Under tradi- tional small-scale animal husbandry, production tends to follow pro- nounced annual cycles. For example, because of differences in feed availability and related flock replacement patterns, egg supply is short in winter and surplus in spring. The seasonal variations in sales in Tianjin and Fuzhou are indicated in Graph 3. 2.23 Government Interventions. Although the ATM is a very important supplement to the state marketing system, China is by no means a market economy, and Government continues to intervene in and attempt to manage the market. Much of its effort is directed at stabilization of both the average level and fluctuations in prices. The concern with average price levels stems from the desire to control urban inflation, which has a significant impact on wage levels and government budgetary subsidies. The concern with price fluctuation for nonstaple foods arises not only from the effects of "normal" influences of price and incomes on supply and demand, but also from the seasonality of supply and demand and poorly developed transporta- tion and procurement systems which are presently characteristic of China. 2.24 Price stabilization is feasible mainly because the state Food Company still controls the largest portion of the supply of livestock products (at least 60-70% of livestock products; the Dairy Products Company controls nearly 100% of milk supply). Although farmers can freely exercise their option to sell through the ATMs, most farmers continue to deal with the Food Company because a) the state procurement price is often higher than prices offered by private middlemen, due largely to Government subsidies on processing and distribution which remain substantial in the largest municipalities; b) high-volume private procurement and wholesaling is still undeve- loped, and livestock products are highly perishable; c) sales to the Food Company gain entitlement to purchase feed at list prices which are subsidized relative to negotiated or rural free market prices (see below); and d) official procurement prices and/or the volume of feed entitlements are adjusted in response to seasonal fluctuations in market prices, but probably not as swiftly as market prices. - 22 - Graph 3. Seasonal Variations in Livestock Supply 25000 . Tianjin 1984-85 Egga Poultry 20000 - M utto n 15000- Pork 10000. 5000 - ~~ , el LE.5r o w* ~ ~ U 00 k.u~ c^ s-2 c Oz-L9 m m c0 rS Month Fu2hou 1984-85 Egga Poultry 4000- Nutton Beef 3000-Pork 0 2000- 1000- o- 0 Monh - 5 Month - 23 - 2.25 Prices of livestock products can be stabilized mainly because the Food Company procures in large volume, at times and places where prices are low, utilizes its cold storage to hold product until required, and maintains prices in state markets (or stores) which are adjusted infrequently within a limited range. Procurement policy continues to attempt to guarantee a supply which meets peak seasonal and holiday demands. The Food Company in fact engages in head-to-head competition with private merchants, inasmuch as about 25% of stalls in every ATM are occupied by state-owned stores. 2.26 Government intervention is also used to improve the quality of products sold through the ATMs. Animals slaughtered pri- vately for ATM sale must be certified as vaccinated. In Tianjin, ATM meat marketeers are required to have refrigerators for meat storage. In Fuzhou, an attempt is underway to confine slaughtering to approved abattoirs, with health inspectors. 2.27 Taxes. Previously all live animals except poultry were procured by the Food Company, which was responsible for payment of the slaughter and commercial taxes. Farmers who slaughtered animals for self-consumption were supposed to pay the slaughter tax (3% for pigs, 10% for cattle), but not a commercial tax. However, as farmers now can legally sell produce in the ATMs, they do pay a 3% commercial tax on such sales, as well as a 2% ad-valorem market administration tax. The slaughter tax is paid at the point of slaughter and based on the live weight of the animal. 2.28 In some cities or provinces, the marketeers are required to pay an additional "market cleaning fee", e.g., in Tianjin, levied at 7% of the combined commercial and market administration taxes, or 0.35% of total sales value. In other places, the marketeers at large central markets are required to pay market depreciation charges, which are based on construction costs of the markets. For example, in Fuzhou, Xiyingli Market levies a Y12 per month depreciation charge for each stall. 2.29 The commercial tax goes into the Government's general revenues, just as investments in structures for ATMs come from general capital construction funds. The market administration tax is used directly to pay salaries of market administrators, other costs of market services and facilities, and cleaning (in the absence of a special fee). The services provided include arrangement of stalls, free weight checking for purchasers, assistance in commercial accoun- ting, advertising within and without the market, finding hotels for marketeers, emergency medical assistance for marketeers and consumers, mediating disputes between buyers and sellers, checking commodity quality, fire and transport safety, prevention of illegal business activity and so forth. 2.30 Consumer Benefits. The opening of the ATM has greatly improved supply to consumers. Previously in the state food store or market, the supply situation was not in equilibrium, of course, be- cause livestock production levels, were too low to meet consumer - 24 - demand. Usually, during the traditional holidays supply was brought up to relatively adequate levels, at a high cost to the Food Company, whereas at other times red meat, and especially poultry and eggs were in shorter supply. Beef and mutton were offered mainly for the mino- rity (Moslem) population. The variety and quality of livestock pro- ducts also were regarded by consumers as unsatisfactory. The meat was often frozen and very fatty, and in most cities retail markets offered neither choice of cuts nor separation of lean from fat meat. At present, due to the presence of the ATMs, both the variety and quality of livestock products have been much improved, and consumer prefe- rences among varieties and cuts of meat, poultry and eggs are better served. ATM meat is characteristically fresh: even in summer, ani- mals are slaughtered in the early morning hours and meat is sold throughout the morning, providing very fresh product. Similarily, ATM eggs are typically fresh rather than preserved with lime. In almost every market, pork is sold by different type of cut. 2.31 As the state food market has set hours of operation, it lacks convenience for consumers, especially two-earner families. The agricultural trade market has more flexible hours, permitting shopping from 5:30 a.m. until (in some markets) very late at night. Its conve- nience extends to hotels, restaurants and official dining halls as well as individual consumers. Previously these units relied entirely on the state food market for their supply of nonstaple foods; at present they may procure in the ATM or directly from producers, espe- cially for particular types or qualities of meat, poultry, eggs and seafood needed to meet their special demands. In general, the vigorous competition provided by ATMs should promote improved, or more business-like management of the state retail system. Consumer Demand for Livestock Products 2.32 Chinese Consumption Behavior. China is a large country with different regional dietary customs for both cereals and livestock products. Although pork is the main type of meat, accounting for over 90% of total meat consumption, there are definite preferences regar- ding other forms of livestock products. In general, northerners have a relative preference for mutton, beef, eggs and chicken, and aquatic products, in that order, whereas southerners would rank these in a different order - poultry (duck and goose), aquatic products, eggs (duck eggs in Fuzhou), beef and mutton (see Annex 2). In Fuzhou, more than half of the stalls in each ATM are for aquatic products sales, more than for pork, beef and mutton combined; in fact, few beef and mutton stalls are found. In Tianjin the situation is reversed, even though Tianjin, as a coastal city with fairly abundant seafood supply, is unusual by the standards of other northern cities. In addition to general regional differences in consumer preferences, various locali- ties may have traditional special preferences, such as that of Fuzhou residents for red-feathered chickens. 2.33 There are also some differences between rural and urban consumption patterns. Urban dwellers prefer high-quality and low- - 25 - calorie foods, such as lean pork. Rural residents accept fat meat, perhaps due to the greater energy requirements of physical labor and the customary use of fat as cooking oil. In urban markets fat is trimmed from pork to a standard of under 3 cm; whereas in Dongxiang County, Jiangxi, pork sells well in the free market even with. more than 5 cm of fat. 2.34 Income Elasticities. Aside from customary factors and. sea- sonal influences, variations in supply and demand are largely related to consumer incomes and relative price levels. The income elasticity of demand, defined as the percentage increase in consumption expendi- ture on an item due to a 1% increase in incomes, is the normal measure of the relationship between consumer demand and income. For this study, 1984 budget survey data for urban households in Fuzhou and Tianjin has been analysed to estimate income elasticities for all major livestock products, as well as other commodities in the consumer budget./l Estimates based on budget survey data are generally regar- ded as better estimates of long-run elasticities of demand than of short run elasticities. Some important elasticities are shown in Table 2.1, for consumers at average income levels in the two municipa- lities. 2.35 These elasticities generally are of magnitudes comparable to those found in similar studies for other developing countries, and are also consistent with similar estimates developed from 1982 Beijing and Tianjin expenditure survey data. The high elasticities for dairy products (despite rationing) and poultry confirm the importance placed by local governments on development of supply of these products. An elasticity for poultry exceeding that for eggs implies that there is a strong potential for development of broiler production. 2.36 The equations relating commodity expenditure to per capita income levels estimated from the urban budget survey data can also be used to project consumer demand for future years. The average elasticities at current income levels are not suitable for this purpose, since elasticities change as income grows. In most cases, the best approximation of the empirical relationship between commodity expenditures and income found in the survey data is strictly linear, suggesting a gradual increase in income elasticities between lower and higher income levels. 2.37 The projected average annual rates of increase of per capita consumer expenditure for livestock products between 1984 and 1990 are compared in Table 2.2 with local targets for per capita supply, assuming urban per capita incomes in Fuzhou and Tianjin grow at rates of about six percent per annum. Local targets for pork and eggs in Tianjin and eggs in Fuzhou are remarkably close to projections based on the budget studies, but in other cases seem out of line. However, /l From samples of 175 households in Fuzhou and 1,000 households in Tianjin, grouped into 10 and 11 income classes respectively. Elasticities estimated from linear or quadratic regressions of expenditures on incomes, weighted to adjust for unequal class size. - 26 - Table 2.1: Urban Income Elasticities of Demand, Fuzhou and Tiankiin, 1984 /a Fuzhou Tianjin Grain 0.34 0.45 Non-staple foods 0.75 0.76 Pork 0.75 0.87 Beef & mutton 1.03 0.45 Poultry 1.22 1.05 Fresh eggs 0.57 0.86 Other meat 0.31 0.87 Fish 0.66 0.85 Vegetables 0.66 0.69 Vegetable oil 0.67 0.73 Sugar 0.41 0.86 Condiments -0.05* 0.69 Other non-staple foods 1.16 0.59 Tobacco & wine 0.43 0.92 Other foods 1.22 0.78 Dairy products 1.57 0.66 Fruit, canned goods, drinks 1.11 0.79 Clothing 1.02 1.08 Daily use goods 1.62 1.46 * not significantly different from zero. /a From Annex 2. Table 2.2: Local and Budget Study Projections of Growth Rate of Consumer Demand, Fuzhou and Tianjin, 1984-90 /a Fuzhou Tianjin Local Budget Local Budget Targets Projection Targets Projection 1984 Per Capita Income 620 620 690 690 1990 Per Capita Income 1,000 1,000 1,050 1,050 Expected annual % increase 1984-90 in per capita demand for: Pork 0.0 6.5 6.5 6.5 Beef 30.8 ) 12.2 Mutton 19.1 ) 8.2 12.2 ) 3.5 Poultry 2.0 9.8 19.7 7.5 Eggs 5.5 5.1 6.3 6.4 Milk 8.1 12.1 14.1 5.3 Fish N/A 5.8 14.8 6.3 /a Budget projections based on linear regression estimates of relationships between per capita consumer expenditures and incomes (Annex 2). - 27 - the budget study projections ignore the impact on demand of the 1985 (or subsequent) price increases for these products; the projected growth rates at best should be regarded as upper bounds, as discussed further below. 2.38 Price Elasticity of Demand. In general, the demand for livestock products in China is highly price sensitive. Contributing factors include overall low income levels, a high proportion of total expenditure occupied by food (40-45% for urban residents in 1984), a high proportion for expenditure on livestock products in total food expenditure (39% and 47% for Tianjin and Fuzhou respectively), and the ease of substitution among livestock products and between meat and fish. 2.39 The May 1985 retail price increase for livestock products caused a significant reduction in urban consumer demand. It would be useful in planning further supply and price adjustments to quantitatively measure the sensitivity of consumer demand to changes in product prices (holding other influences on demand constant). The usual measure of this relationship is the price elasticity of demand, indicating the percentage change in quantity demanded when the price increases by 1%. Estimates of price elasticities for livestock products can be derived from (a) the observed reaction of urban consumer demand to the May 1985 price increases (Table 2.3); and (b) consumer budget survey data (Table 2.4). 2.40 The monthly consumption of livestock products over 1984-85 was influenced by five factors, which unfortunately cannot be comple- tely disentangled: (a) a rising trend in nominal family incomes over 1984-85 (estimated to amount to 20% in 1985 in Fuzhou); (b) variations in supply due to seasonal influences on production and consumption, changes in producer incentives and opportunities, and varying success of administrative planning; (c) the direct effect of higher prices on demand (own-price elasticities); (d) cross-price effects (cross-price elasticities) due to changes in relative prices within the category of livestock products; and (e) an unknown extent of price discounting which occured in both years whenever temporary excesses of supply over demand resulted from inflexible procurement procedures. 2.41 The effect of rising incomes can be controlled for, if it can be assumed that the 12-monthly percentage increase in consumer incomes during 1984-85 was about the same, regardless of the period over which it is measured (e.g., January-March or September-December). Comparison of consumer demands between the same periods in 1984 and 1985 also controls for seasonality if seasonal patterns were the same in both years. The most heroic assumption required is that levels of supply in 1984 and 1985 at all times satisfied consumer demand at current retail price levels. This assumption is clearly untenable for most livestock products (those which were formally or informally rationed in 1984 or early 1985; milk was still rationed after May 1985). - 28 - Table 2.3: Changes in Fuzhou and Tianjin Consumption Between 1984 and 1985: 1985 Consumption Estimated Retail Computed (1984 - 100) Price Price "Price Jan-Apr Sept-Dec Effect/a Increase Elasticity"/c Fuzhou Pork 126 77 -39% 62% -0.63 Beef 291 227 -22 113 -0.20 Mutton 230 266 16 133 +0.12 Poultry 116 85 -27 -1 +20.22 Eggs 151 71 -53 8 -6.63 Milk 107 110 3 0 N/A Tianjin Pork 105 52 -50 36 -1.38 Beef 151 85 -44 110 -0.40 Mutton 64 65 1 70 +0.02 Poultry 130 43 -67 44 -1.50 Eggs 157 57 -63 29 -2.18 Milk 97 104 8 0 N/A /a Assuming that the percent increase in consumption above the seasonal norm between Jan-Apr 1984 and Jan-Apr 1985 (when list retail prices were largely unchanged) reflects only increased incomes (caviat: it may also reflect changes in supply under rationing), whereas the percent increase (decrease) between Sept-Dec 1984 and Sept-Dec 1985 reflects both income and price effects, the price effect alone is derived by dividing the second column by the first and expressing the result as a percentage. /b Post-May 1985 list retail prices relative to 1984 list prices. /c The estimated percentage price effect divided by the percentage increase in price. Results meaningless if the quantity consumed in 1984 or 1985 is less than quantity desired by consumers at list prices. 2.42 Consequently, the increases in January-April consumption (Table 2.3) in between 1984 and 1985, which are supposed to reflect only the influence of increased consumer incomes on demand, for some products may reflect mainly changes in supply conditions under ration- ing (the relative percentage increases in consumption do not conform well with estimated income elasticities for different livestock pro- ducts). Also, many of the estimated "price elasticities" are implau- sible in sign and/or magnitude. Nevertheless, the estimation exercise demonstrates that the effects of the 1985 price increases on urban demand, when the countervailing effects of income growth are control- led for, were quite severe (reductions in demand probably in the 30- 70% range). This is not surprising, as studies of consumer price - 29 - elasticities in other developing countries would lead one to expect own-price elasticities for most individual livestock products ranging between -0.7 and -2.0. 2.43 The second method employed for estimating price elasticities relies on the fact that theoretically, price elasticities are determined by, first, the effects on commodity demand of changes in disposable income which arise from price changes, and, second, from the substitutability of various commodities. The consumer budget surveys, and estimates of income elasticities derived therefrom, contain most of the information needed about the income effect, and informed guesses can be made about substitutability relationships. The advantage of the estimation methodology (described in Annex 2) is that complete sets of estimates of own- and cross-price elasticities can be derived which are consistent internally and with the budget survey data, and are not distorted by the estimation problems de- scribed in paras. 2.40-2.42. The disadvantage is that the estimates involve an element of subjectivity, and their accuracy cannot be assessed by statistical methods. 2.44 The estimates of own-price elasticities for livestock products, derived by the second method (Table 2.4), suggest that demand typically is moderately price-elastic (-1.00 + .25) for those items with high income elasticities such as poultry in both Tianjin and Fuzhou as well as beef, mutton, and dairy products in Fuzhou only; and somewhat less elastic (between -0.50 and -0.75 for pork, - 0.4 and -0.5 for eggs) for other livestock products. High price elasticities of demand indicate that relatively small changes in administered prices can lead to significant changes in demand; but they also indicate potential for rapid market expansion if efficiency improvements permit reduced supply prices. The projections of 1990 demand for livestock products derived from income elasticities alone (Table 2.2) do not take into account either the effects of the major price adjustment of 1985 on consumer demand, or the probability that remaining subsidies on the procurement, processing and distribution of livestock products will be phased out. These projections should therefore be taken as upper bounds, and new projections should be made based on 1986 consumer budget data in order to take into account the reduction in demand following the 1985 price increases. 2.45 Subsidies. In compensation for the May 1985 price in- creases for livestock and aquatic products (but not the similar in- creases for fruits and vegetables), urban residents on the state wage system received a staged and qualified cost of living allowance set at Y4.50 per capita per month in Fuzhou or Y6.00 in Tianjin. (Price increases for pork, fish, and eggs alone would have cost the average consumer about Y6.10 per month in Fuzhou and Y3.70 per month in Tian- jin, had expenditure patterns remained unchanged). Administrative and professional workers received their increases from July 1985 (but with amounts in excess of Y30/worker/month deferred until July 1986). The allowance for workers in state enterprises was delayed until 1985 productivity gains were computed, at which time retroactive increases were extended subject to the limits of productivity increase. Over- - 30 - Table 2.4: Estimated Direct Price Elasticities of Demand, Fuzhou and Tian'iin, 1984 /a Fuzhou Tianjin Estimate Estimate Estimate Estimate A ab B / A /b B /b Grain -0.27 -0.19 -0.37 -0.27 Pork -0.70 -0.49 -0.69 -0.49 Beef & mutton -2.11 -1.44 -0.75 -0.52 Other meat -0.68 -0.46 -1.02 -0.70 Poultry -1.24 -0.86 -1.21 -0.83 Fresh eggs -0.53 -0.37 -0.53 -0.37 Fish -0.58 -0.41 -0.76 -0.53 Vegetables -0.37 -0.26 -0.43 -0.30 Vegetable oil -0.37 -0.25 -0.44 -0.28 Sugar -0.14 -0.10 -0.34 -0.24 Condiments +0.14 +0.10 -0.08 -0.06 Other non-staple foods -0.58 -0.42 -0.29 -0.21 Tobacco & wine -0.17 -0.12 -0.43 -0.31 Dairy products -1.08 -0.75 -0.58 -0.40 Fruit, canned food, drink -0.59 -0.42 -0.42 -0.31 Clothing -0.04 -0.04 -0.15 -0.15 Daily use goods -0.06 -0.06 -0.05 -0.05 Housing const./repair -0.01 -0.01 -0.02 -0.02 Cultural/educa. goods -0.00 -0.00 -0.04 -0.04 Publications -0.01 -0.01 -0.00 -0.00 Fuel -0.01 -0.01 -0.01 -0.01 Rent -0.02 -0.02 -0.02 -0.02 Water & electricity -0.01 -0.01 -0.01 -0.01 School fees -0.00 -0.00 +0.00 +0.00 Child care +0.00 +0.00 +0.01 +0.01 Transport -0.00 -0.00 -0.01 -0.01 Communications -0.00 -0.00 -0.00 -0.00 Cultural events -0.01 -0.01 -0.00 -0.00 Medicine -0.01 -0.01 N/A N/A Other goods & services -0.21 -0.21 -0.01 -0.01 /a From Annex 2. /b Estimates A and B differ in the assumed levels of the marginal utility of money (income) at different income levels. See Annex 2. all, real wages in the state sector declined in 1985 largely as a result of the price adjustments. Price increases for non-staple foods were far more severe (34-39%) in the province-level municipalities (Beijing, Tianjin, and Shanghai), where in the past this component of consumption has been more heavily subsidized than elsewhere. - 31 - 2.46 The May 1985 reforms substituted cash income supplements for (a small portion of) budgetary subsidies to the Food Company, leaving the "average" urban consumer's welfare relatively unchanged. However, it resulted in a significant transfer of income to livestock produ- cers, who received average procurement price increases (22.6% for red meat, 15.1% for poultry products, counting free market price in- creases) at rates equaling or exceeding wholesale/retail price in- creases to urban consumers. As a result, budgetary subsidies to the commercial system for procurement and distribution of meat and vegeta- bles (excluding the cash supplements to consumers) only fell by 15% in 1985, to Y1.5 billion. In the larger cities, it seems unlikely that the budgetary subsidies to the Food Company will be reduced in the long term by the reforms. Such reductions would have to come from several sources: (a) an increased margin between local producer and retail prices; (b) decreased urban consumption of subsidized (vs. unsubsidized) products; (c) an increased proportion of state procure- ment from low-cost sources, particularily outside the municipality; and/or (d) a decreased volume of sales through the state distribution system (relative to the ATMs). 2.47 Of these possibilities, it appears that the first has played little or no role in the larger muncipalities such as Tianjin and Fuzhou, where the desire to increase local self-sufficiency caused higher than average increases in procurement prices. For an extreme example, in Fuzhou the procurement price of pigs was increased 82% whereas the retail price was raised only 62%, decreasing the procure- ment-to-retail price differential from 18% to 5%. This differential is insufficient to cover costs of slaughtering, which in Fuzhou is subsidized to the extent of Y15 per pig (6-7% of retail prices). In Tianjin, losses on procurement, processing and distribution of pork in 1985 reportedly still amounted to YO.60/kg (20% of retail prices). As the main abattoir in Tianjin showed a profit, the losses apparently were those of the collection stations and the wholesale/retail net- work; and possibly were mainly the result of price discounting in reaction to excess supply resulting from a post-May decline in consu- mer demand. In both Tianjin and Fuzhou, eggs are presently wholesaled at YO.10/kg less than the procurement price, ignoring costs and losses to the Food Company in excess of YO.20 per kg. 2.48 Urban residents have adjusted their consumption patterns in response to price increases, to consume relatively less non-staple foods (para. 2.42). This has reduced the subsidy burden below what it would otherwise have been (ignoring losses in 1985 arising from unan- ticipated oversupply), but due to consumer income growth, 1985 total demand for meat in the municipalities did not decline in absolute terms. The growth of the ATMs and direct links between producers and large-scale consumers nationwide has allowed the Food Companies to drastically reduce their involvement in procurement and distribution of livestock products (para. 1.21), but in Tianjin and Fuzhou the reductions were largely confined to poultry (Tianjin) and beef and mutton (Fuzhou), which were not significant sources of losses in the first place. Also, the effect of higher municipal procurement prices has been to increase, rather than decrease, the extent of local self- - 32 - sufficiency (in Fuzhou, from 36% in 1984 to 71% in 1985), reducing the ability of the Food Companies to compensate for losses on local pro- curement with more profitable purchases of frozen meat from other provinces. 2.49 Aside from budgetary subsidies applied through the Food Company system, in large municipalities, even greater subsidies are extended to producers through the supply of feed at list prices in return for pigs and eggs sold to the Food Company or milk sold to the Dairy Company (although, nationwide, a significant number of provinces have eliminated these subsidies, or greatly reduced the quantity of feed so supplied ). Valuing the subsidy at the differential between list and negotiated feed prices of about YO.20/kg, it represents the following proportions of procurement prices in Tianjin and Fuzhou: Tianlin Fuzhou Pigs 27% 25% Eggs 9-12% N/A Milk 22% 25% Although the same feed:product ratios apply to all producers (2.2-3:1 for pigs, usually 1-1.5:1 for eggs, and 1:2 for milk, denominated in low-quality mixed feeds), specialized households and state or collec- tive farms are more likely to receive complete feeds or high-ener- gy/protein feedstuffs which are not made available to small-scale producers, who mainly receive grain byproducts. The list price provi- sion of feeds by the state through the Grain Bureau/Feed Company system in general serves (incompletely) only the needs of producers supplying the urban core of municipalities as well as some breeding farms in the peripheral counties. 2.50 A third, and relatively new form of subsidy to the producer, prevalent mainly in more prosperous coastal areas, involves cross-subsidization of livestock and other agricultural products from profits of rural industries, e.g., in Zhejiang and Jiangsu. This has the effect of promoting local consumption, but the township and vil- lage governments responsible have every incentive not to increase sales outside the local area. 2.51 The effect of subsidies to the production, processing and distribution system is to make available to urban consumers an increased total supply of municipally-produced livestock products at retail prices lower than would prevail without subsidies. If price elasticities of producer supply and consumer demand are both reasonab- ly high, it is also possible that present local producer prices are not much different from equilibrium levels without subsidies. Howe- ver, the social cost of the subsidies includes losses of efficiency due to (a) favoring of local production relative to cheaper long- distance supply, and (b) favoring of Food Company suppliers relative to other producers. - 33 - 2.52 In practice, the private trade in livestock products has some countervailing advantages which have allowed its growth. The MOC system has the salary burden of 1.1 million employees, of whom 17- 18% are retired. Its processing and distribution capacity is designed to smooth seasonal differences between supply and demand; therefore it must accomodate seasonal peak procurement volumes at least twice the normal monthly flows, leaving a large overhead burden in the off- season and the additional operating costs of cold storage. The pri- vate market, with quick turnover and short lines of supply, labor- intensive processing methods and leased equipment or facilities, and a higher profit margin due to better quality and consumer services, is able to offer competitive farmgate prices, albeit without the feed- grain entitlements obtainable from the Food Company. However, small- scale traditional producers, depending mainly on family-produced grain, byproducts and wastes, have little use for non-transferable entitlements, and larger producers use the private market to dispose of production surpluses above contractual levels. Hence, although the private market can compete even in pigs and eggs at its present low level of sophistication, remaining subsidies to the state system and limited access to long-distance transport or storage facilities place severe constraints on its ability to expand and "deepen", particularily in the large and medium-sized cities. Issues and Implications 2.53 The complete opening of the ATM is an important step in ecomomic reform and its functions, as discussed above, have been very useful. However, there are still some issues which need attention. First, if the ATM is regarded by Government as a permanent: institu- tion, larger numbers -of permanent markets should be constructed in both urban and rural areas. The simple market or temporary street market is not sufficient if the quality of the commodities is to be maintained and provides no protection for marketeers from inclement weather. Growth in numbers of permanent stalls, which provide a fixed place of business, is also important in facilitating development of specialization. Of course, the same result would obtain, with reduced investment cost, if state retail outlets were enabled to distribute products procured from the market. 2.54 Second, the seasonality of supply remains significant, despite the opening of the ATMs. It may be exascerbated by a tendency to maintain stability of official procurement and retail prices, even when the Food Companies have been given some leeway to vary them. Then, too, present urban consumption patterns demand some degree of seasonality - whether current seasonal variations in market supply are excessive or merely appropriate remains open to further study. 2.55 Third, there is the price determination problem. There is a residual fear that complete reliance on market supply could cause excessive price fluctuation or lead to deterioration in urban living standards. However, efforts to suppress and control prices have been costly in terms of government subsidies, and have contributed to - 34 - inefficiencies in the production and distribution system. The same subsidies could be applied as additional direct income supplements to protect urban living standards, allowing Government to concentrate on stabilization of the market through such means as investing in con- struction of cold stores and renting space to farmers for product storage. This issue will be re-examined below, after the full implications of the subsidies in the sector have been explored. 2.56 Finally, by its sudden and severe price adjustments, China has probably initiated its own "hog cycle", a phenomenon which was common during the development of livestock production in the U.S. and other countries. Because of the rapid growth of grain production and increase in demand for livestock products since 1979, Government has sought to speed development of livestock production, and used procure- ment price increases to stimulate increased supply of pigs, milk and poultry. Farmers have benefited from sale of their products to the government at prices which almost equal the retail price as well as provision of feed at low list prices considerably below market prices of grain. However, these policies have also driven up the market prices of feedstuffs and probably led to a temporary oversupply of pork, resulting in an estimated decline in profits per pig from Y36 in 1985 to about Y12 in 1986. Consequently, inventories of feeder pigs have fallen, farmers are selling their grain rather than retaining it for feed use, decreasing their purchases of piglets, and slaughtering their sows. 2.57 The result is likely to be a contraction of livestock production in 1987, until reduced production of livestock products again drives down feed prices, drives up ATM prices for livestock products and stimulates renewed production. This kind of cycle, in the West, typically was completed in four years as a result of lags built into the feed and livestock production process as well as in farmer decision making, and resulted in unstable supply of pork. Barring further intervention by Government, the period could be simi- lar in China and severity even greater. Similar cyclicality problems are expected (possibly with greater severity because of higher price elasticities) in the poultry meat and egg market. - 35 - 3. FEED SUPPLY AND PROCESSING Aggregate Feed Supply Situation 3.01 Overall Situation. Livestock product supplies to the eastern cities in China, including milk, meat and eggs, come largely from high quality dairy cows which provide the milk, pigs and poultry which together provide well over 90% of the meat and both hens and ducks which provide the eggs. All of these types of livestock obtain a very substantial part of their diet from high quality feed stuffs, i.e. grains and grain by-products and to a limited extent, protein meals such as fish meal, meat and bone meal, cotton seed meal, other oil seed meals, brewers wastes, etc. Near the eastern cities, only a small proportion of the dietary nutrition in aggregate comes from green fodder. 3.02 The main source of feed, particularly for pigs and poultry, is grains. In this respect these livestock are in direct competition with humans, in that in China direct grain consumption by humans is also high. In 1984, direct human consumption averaged about 250 kg per capita of unprocessed grain. Within the provinces and municipalities visited, rural consumption levels were as follows (urban consumption levels averaged about 180 kg): 221 kg/capita for Tianjin, 231 kg/capita for Shandong, 302 kg/capita for Fujian and 335 kg/capita for Jiangxi. The higher figures in Fujian and Jiangxi in part are due to the facts that (a) they are both major rice producing provinces and (b) the percentage of "unprocessed grain" which is eaten is about 70% with rice, whereas with wheat (the main cereal in Tianjin and Shandong) the conversion percentage from "unprocessed grain" to flour is about 85%. 3.03 Production of fine cereals, i.e. wheat and rice, which provide most direct human consumption, averaged 258 kg per capita for China as a whole in 1984 (see Annex 1, Tables 3.1 and 3.3), including seed required for replanting. The totals for the provinces visited amounted to 58 kg/capita for Tianjin, 175 kg/capita for Shandong, 280 kg/capita for Fujian and 439 kg/capita for Jiangxi. 3.04 High quality feed potentially available for pig, poultry and dairy cattle consumption includes (a) most of the coarse cereals and tubers produced (i.e. total production, less the amount retained for seed, and the relatively small proportion used for human consumption), (b) wheat bran (about 13.5% of wheat production), (c) rice bran (about 9.4% of production), (d) about 50% of oil seeds/cotton seed, (e) small quantities of meat and bone meal and fish meal, (probably amounting to not more than 3% of total meat production), and (f) cultivated green fodder, of which about half the production would be fed to cattle and the other 50% available for pigs, particularly lactating sows. In addition to this high quality food supply, ruminants obtain much of their total dry matter intake from natural pasture; while some additional low quality rice husks and vegetable waste are also fed to pigs and poultry. - 36 - 3.05 Aggregate production of coarse grains and tubers, rice and wheat bran, and oil seed meals is estimated to total about 175 million tons. Of this about 12 million tons needs to be deducted for seed, industrial use, and waste; about 40 million tons of coarse grains, largely tubers, millet, maize, and sweet potatoes, would also be used for human consumption. This leaves a net available supply estimated to be at most 130 million tons which could be used for livestock feed.lI 3.06 A rough estimate of the present livestock requirements and availability for feed based on good quality cereals or protein meals is summarized in Table 3.1 (from Annex 3, Tables 4-6). This calculation indicates a total demand for quality feed of around 134 million tons (excluding requirements for poultry meat which would amount to perhaps a further 5 million tons). Of this total, two- thirds is for pigs. Although the estimates are relatively inexact, the indications are that China may be slightly short of feed for its present livestock population, in that requirements are about 134 million tons, compared with a potential supply of under 130 million tons. However, the difference is small and the apparent feed imbalance can be readily made up from green fodders or roughages. 3.07 Analysis of this data by province, however, indicates that there are significant regional imbalances. On the basis of these estimates aggregate feed supply in the Northeast for livestock is plentiful; an estimated total requirement of 22 million tons being more than catered for by a supply of about 31 million tons of typical livestock feeds. In the Southeast, potential supply of 78 million tons also exceeds requirements of about 70 million tons (ignoring quality problems). In the West demand of about 42 million tons is only halfway met by an estimated supply of 21 million tons. When allowances are made for (a) human consumption of coarse grains, and (b) use of fine grains as feed, the Northeast and Southeast have surpluses, but the West has a substantial deficit. 3.08 There is however substantial intra-regional variation. In the Northeast, Inner Mongolia is 2.5 million tons in deficit and Beijing and Tianjin have total gross deficits of nearly a million tons each, while Jilin has a surplus of 5.6 million tons. In the Southeast, Jiangsu has a surplus of 7.2 million tons of total feed, but Guangdong and Guangxi both have deficits of about 4 million tons. Of the provinces visited, Shandong has a 5 million ton surplus; Jiangxi is more or less in balance, with a deficit of over 2 million /1 Of which, about 66% (87 million t) would be cereals, tubers or soybean, 11% oilseed meals, and 22% brans. The estimate is sensitive to the assumption that use of grain for "other uses", including seed, waste, and industrial uses (excluding production of noodles and bread, but including the catering trade) totals 15% of gross product of cereals, soybean and tubers at original weight, or about 61 million tons. If this estimate is too small, then the estimated residual availability of quality feeds is correspondingly too large. - 37 - Table 3.1: Human Grain Consumption and Quality Feed Availability by Region, 1984 Northeast Southeast West Total Fine GrainZa Ex-Seed & Waste 20.7 161.4 44.1 226.2 Less: Human consumption /b 46.2 154.9 56.4 257.4 Equals: Fine Grain Balance -25.5 6.5 -12.3 -31.2 Plus: Soybean Ex-S&W & Crushing 3.3 2.5 0.5 6.3 Fine Grain Bran 3.0 19.9 5.7 28.6 Course Cereals Ex-S&W 44.1 24.9 18.7 87.6 Tubers Ex-S&W 3.3 14.7 6.3 24.2 Oilseed Meals 3.3 9.5 2.0 14.8 Subtotal: Gross Feed /c 57.0 71.5 33.2 161.5 Less: Feed Requirements 22.0 69.6 42.5 134.1 Equals: Gross Feed Balance/d 35.0 1.9 -9.3 27.4 Net Feed Balance/e 9.5 8.4 -21.6 -3.8 a Rice and wheat /b Population times provincial per capita consumption /c Sum of feeds normally fed to livestock /d Balance between feed normally fed to livestock and livestock requirements. /e Assuming that surplus/deficit of fine grains for humans is available to or taken from livestock feed. tons of traditional livestock feed offset by its rice surplus; and Fujian has a substantial deficit of about 2.4 million tons. 3.09 Within these overall estimates, total feed requirements for pigs (179 million tons including 91 million tons of quality feed) have been calculated on the basis of pig numbers in the country. With more efficient feeding, however, the present amount of pork could be produced with far lower pig numbers. Current pork production of 14.5 million tons is the equivalent of about 20 million tons live weight. Given the genetic potential of pigs in China, feed conversion rates of 4 or 5:1 should be obtainable under good feeding regimes, and are being obtained on the better specialized households (see Annex 3, Table 10). Indeed, if the "quality feed" were well-balanced, 91 million tons of quality feed alone should be more or less sufficient to produce 14.5 million tons of pork. However, under such circumstances, the number of sows would be reduced by one-third and - 38 - the aggregate pig population halved: Possible Situation for Present Situation Same Pork Production a/ Total Sows 22 million 14 million No. of Other Pigs 285 million 141 million Total Pigs 307 million 155 million Total Slaughtered 220 million 199 million Age at Slaughter 15.5 months 8.5 months Carcass Weight 66 kg 73 kg Total Pork 14.5 million tons 14.5 million tons a/ Based on present performance of specialized households. 3.10 The efficiency losses inherent in the present situation can be viewed as a tradeoff for the maintaining the advantages of backyard production - minimum cash costs (for purchased feed), maximum use of low opportunity cost byproducts or waste products and labor, and a close link between pig production and crop cultivation, via the recy- cling of manure. Production levels under this system can also be expected to be less sensitive to economic parameters such as market prices of feed and pigs. However, the comparison also suggests that improved feeding practices in backyard production can raise efficien- cy, by reducing energy now lost in maintaining body weight over a longer feeding period. 3.11 For a balanced pig or poultry ration an average of about 20% of the feed normally needs to be "added protein", typically in the form of meat and bone meal, fish meal, or oil seed meals. Of this, fish meal/meat meal seldom total more than 5%, and therefore at least 15% normally needs to be high-protein oilseed meals. An analysis of the relationship between the availability of feed grains, bran and protein meals by province is given in Annex 3, Table 6. This indi- cates that overall, protein meals could amount to about 11% of the net total quality feeds potentially available to livestock (in fact, according to a 1982 MAAF briefing, some 70-80% of protein meals were used as fertilizers rather than feeds). 3.12 The regional composition of these quality feeds is summa- rized in Table 3.2. Overall, China's potential feed supply would provide about 12% crude protein rather than a desirable 15%. The potential is about the same in the Northeast and Southeast, and slightly worse in the West. However, the analysis ignores regional differences in the crush and utilization rates for oilseed meals, which are assumed to be used only for feed. In reality, probably no more than one-half of available oilseed meals are utilized as feed (the remainder being used as fertilizer), and provinces depending mainly on cottonseed and/or rapeseed meal (notably in south China) use only limited proportions as feed; this puts the south at a greater disadvantage than implied by these estimates. If the crush rate for - 39 - Table 3.2: Composition of Potential Available Feed y Region (1984) Grains/ Type of Feed Tubers Bran Oilseeds Total Average % Composition of Feed Northeast 79.9 9.4 10.7 100.0 Protein content 8.5 9.4 37.3 12.1 Southeast 62.2 25.5 12.2 100.0 Protein content 6.9 12.9 37.2 12.1 West 63.6 27.6 8.8 100.0 Protein content 7.5 13.2 36.8 11.7 National Average 66.7 22.0 11.4 100.0 Protein content 7.7 13.1 37.2 12.2 soybean and peanuts in the north were higher than at present, northern or northeastern provinces would also have a more considerable protein advantage. 3.13 The Southeast. Of the provinces visited, Jiangxi is worst off with respect to availability of pig feed, with gross livestock feed (available production of feedgrain, oilseed meals, and bran, ignoring human consumption balances) amounting to only 3.8 times the amount of pork produced and only 2.5 tons per sow. In Fujian the ratio of feed to pork is 4.4:1 (3.7 tons/sow) which is also low, at about half of total requirements. Fujian's situation is also compli- cated by a deficit in grain for human consumption. Conversely, in Tianjin the ratio is about 16:1 and in Shandong it is 20:1 indicating that both of those regions have adequate total feed supplies for their domestic pig production. 3.14 Jiangxi is a major producer of pork despite low feed grain supplies. However, it is the largest per capita producer of rice in China (437 kg per capita). Therefore, in addition to locally produced feed grains, pigs in Jiangxi are fed locally produced rice and rice products as well as feed grains and protein meals imported from other provinces. As an approximation, the total pig production in Jiangxi of 508,000 tons of pork would require about 3 million tons of total feed. This is very large compared with Jiangxi's production of 10,000 tons of corn, 280,000 tons of tubers, 35,000 tons of other cereals and 135,000 tons of soybean; but not in comparison with Jiangxi's production of 15 million tons of rice. Indeed, an estimated 34% of the feed for livestock production in Jiangxi comes from rice bran. The situation is similar although less pronounced in some other south- eastern growing provinces, particularly Hunan, Hubei, and Zhejiang. - 40- 3.15 The optimum future policy for grain supplies for feeding to livestock in feed grain deficit areas of Southeast China is not completely clear. The alternatives would appear to be as follows: (a) continue to produce rice on the existing rice areas and exchange this for imported feed grains (e.g., including corn from the north); (b) introduce corn on some of marginal (usually higher) paddy fields in place of rice; or (c) retain the present cropping pattern but increase use of rice as a livestock feed. 3.16 On the basis of the World Bank's projected long-run economic prices for rice, it would seem better to export rice and import maize provided that the internal transport and trade infrastructure could be adequately arranged for the enormous volumes involved. Rice, however, is not a homogenous commodity; the small portion entering international trade is of high grades and special varieties, and the vast majority of world-wide rice production is consumed in-country. There must, therefore, be doubts about the applicability of Bank price projections if a major policy of swapping rice for corn were adopted by China, the world's largest rice producer. 3.17 An analysis of provincial rice and corn yields indicates that the weighted average yield of corn overall is only just over half that of rice (Annex 3, Table 8). From the animal feedstuffs stand- point, unhulled rice, i.e. paddy, has an energy value 75-80% of that of corn and a protein value about 95% of that of corn. Overall its worth for livestock feeding within the Chinese system is probably about 85% of that of corn. If the possibilities of trade are disre- garded, therefore, and land is considered to be the limiting resource, it would appear to be more efficient for farmers in the Southeast of China, particularly in Hubei, Hunan, Guangdong, Jiangxi and Guangxi, to grow and feed rice rather than to change from rice areas at average yields to corn areas at average yields. 3.18 However, in the Southeast, the comparison drawn above is between rice yields in paddy fields and corn yields on poorly-managed uplands or as a vegetable on private plots. In recent years, corn procurement prices in the south have been too low to induce expanded area and good management (e.g., in Fujian about Y240/t, compared to Y300/t for paddy rice). Although experience with large-scale feed corn growing in the south is extremely limited, Fuzhou obtains an average of 5.3 t/ha on its 40 ha of experimental fields, higher than average local rice yields; and Jiangxi can get 3 t/ha on uplands where paddy rice cannot be economically grown. As the negotiated price of northeastern corn is now roughly double local procurement prices, an increase in local corn procurement price to over Y400/t is being considered in Fuzhou, and should be considered elsewhere. With higher prices, better seed varieties, and correspondingly better management; - 41 - and with additional upland areas (particularly the red soils) brought into cultivation, it is quite possible that locally-grown corn could become, not merely a supplement to paddy rice on non-competing ac- reage, but also a profitable substitute at least on upland rice fields. 3.19 The more serious problem in the south is the shortage of high-quality protein feeds. Although the southern provinces grow some soybean, cotton, and peanuts, the main source of edible oil is rape- seed; and with local crop varieties, use of rapeseed meal in livestock feed must be limited./1 Local yields of other oilseeds are as little as one-half of yields in the main producing provinces in North China. Local producers require a correspondingly higher price, at which processing into edible oil and meal is not commercially profitable (at either domestic or border prices). Therefore the south should not attempt self-sufficiency in oilseed meals, but rather import from elsewhere. At present the cheapest source is imports from abroad, although with greater northern or northeastern production of oilseeds, this situation could change. Systems and Organization of Livestock Feeding 3.20 Pigs. At the end of 1984 there were about 307 million pigs in 'China of which 22 million were breeding sows. Most of these are kept on individual peasant holdings, with about 11% kept by state farms and perhaps 5% by SHs. At the technical level the system of pig feeding, even on the SHs and state farms, was only poor to moderate. This can be considered a significant factor in the rather low level of pig performance obtained (Annex 3, Tables 9-10). In particular the very low national average level of daily weight gain which requires 15 months to produce a 66 kg d.c.w. pig can be attributed mainly to poor feed quality. 3.21 The first problem with pig feed on the farms visited (which were substantially better than the national average) was that in general the quality of the rations sold to farms was inadequate. The proportion of grain fed in the ration was too low and as a result insufficient energy was available. Protein levels were often too low, partly as a result of low absolute inclusions of protein and partly as a result of the low quality protein used. Levels of micro-elements and vitamins also were not always adequate and in particular calcium and phosphorus were usually not in the appropriate ratios. Quality control was generally poor, and precise analysis of ingredients seldom made. This has led to variable final quality. 3.22 Secondly, feeding at the farm level was not always appro- priate given the raw materials available. Insufficient attention was lI China has only begun to expand production of varieties of rape- seed which are low in glucosinalates, an anti-nutritive substance which limits the use of rapeseed meal in livestock production. These varieties are presently best adapted to the cooler climes north cof the Yangtse River. - 42 - paid to the fact that pigs at different stages of growth need diffe- rent types of feed. The most noticeable effect of this was that pigs about one month after weaning, which need high protein rations, were generally in poor condition as a result of being fed only average quality feed. 3.23 A major step required in the future will be finer matching of feed formulations to the types and ages of livestock for which they are intended. In general several different types of feed, each appro- priate to a different age group of animals, should be produced and sold to specialized households and larger producers. Balancers should be formulated to match the different kinds of home-grown produce available and sold to smaller producers (see Annex 5 for details). 3.24 Poultry. Poultry production is either geared to full use of sophisticated feeds or to scavenging and semi-scavenging systems. In general commercial poultry households make little use of green feed or by-products. Virtually all of their feeds are procured from feed mills at negotiated prices and the types and qualitites of concentrate feeds involved were largely adequate. As poultry production increases and more poultry feed is manufactured, it will be necessary to ensure that increasing quantities of fishmeal and soybean meal are available for poultry feeds. Care will need to be taken that other oilseed meals, particularly cottonseed meal and rapeseed meal, which can have certain detrimental effects as poultry feed ingredients, are kept to tolerable levels. 3.25 In addition to fully enclosed systems, some ducks were also kept on semi-scavenging systems. With these, producers directly fed about two-thirds of normal feed requirements in the form of whole rice, with the ducks obtaining the rest of their requirements, inclu- ding most of the protein, from worms, slugs, etc. in rice paddies. This system of production appears highly efficient from the standpoint of feed usage although it is labor intensive, requiring one herder per 150-300 ducks. It can be operated at stocking rates of 30-40 ducks/ha. 3.26 Dairy Cows. Total feed requirements for dairy production are miniscule compared with those of pig or poultry production at present. At the end of 1984, there were only 0.7 million mature dairy cows of good breed in the country, compared with 22 million breeding Sows. Limiting factors with dairy production are generally the supply of high quality bulk fodder rather than concentrate feeds, which can be readily bought from the feed milling sector. Shortages of bulk fodder are particularily acute in the suburbs of major cities. 3.27 The systems of bulk feed for dairy cattle vary widely be- tween region and from farm to farm depending upon the climate and the amount of arable land available for producing forage crops. Given the very high yields of grasses which can be obtained (up to 75 tons of green matter per ha) and the relatively low cost of labour, hand carried fresh grass is probably the cheapest source of high quality feed for dairy cattle. However, particularly in the north there are relatively long periods of the year in which grass hardly grows, - 43 - consequently, provision also has to be made for bulk fodder during the winter period. This is particularly important as milk supply in China is geared to the fresh market for which,demand is largely year round. Silage is often made in the north, i.e. Tianjin, and on some more sophisticated southern farms, but is not common in the south. The making of hay generally is not widely undertaken in southern China because of high humidity. 3.28 Because of the relatively small numberlf of dairy cattle, the proportion of their feed which can be obtained from various sorts of by-products in China is high. For example, in addition to the normal bulk feeds and concentrates, feeding in Tianjin included corn starch solubles, sweet potato, beet pulp, carrots, corn distillates, sugar beet tops, reed grass and vegetable wastes, while in Fuzhou, sugarcane tops were also fed. This wide use of by-products is currently feasi- ble because of the relatively low number of dairy cows. Should the dairy herd expand substantially, it will be less feasible to rely on by-products for the bulky fodder. 3.29 The need to ensile or treat low quality by-products in order to improve their digestibility, e.g. by using sodium hydroxide, and the associated technology required means that large dairy units are likely to be able to cope more effectively with providing bulk feed than individual small farmers. It has been found in many countries that it is extremely difficult to make high quality silage In small lots, so in those areas where silage is a winter feed necessity, SH dairying should only be promoted if it is feasible to group households to effectively make communal silage. 3.30 As far as concentrate feeds are concerned, the concen- trates available for dairy cows are adequate, or can be adequately formulated. The main potential problem is the calcium:phosporus ra- tio, which appeared to have varied significantly on some occasions from the ideal. This is likely to become more important as levels of technical performance rise, giving rise to greater milk yields and therefore a larger drain on the animal's own calcium resources. It is possible that there are also shortages of other trace minerals and vitamins, and the general criticism that feed quality is not adequate- ly monitored at the feed mill applies to dairy concentrates as well as to those for pigs and poultry. 3.30 Given the price structure of beef, it is not worthwhile keeping male calves on dairy farms and these are slaughtered shortly after birth. Heifer calves are generally brought up on whole milk rather than milk substitute but are weaned at relatively early ages. With the present financial price structure of fresh milk and dried milk in China, this is rational. If imports were allowed, however, it would be worthwhile feeding cheap imported powdered milk to calves rather than using fresh liquid milk, which could then be sold at relatively high prices on the local market. 1/ Dairy cows are estimated to consume less than 1% of total livestock fodder in China (Annex 3, Table 4). - 44 - The Feed Milling Industry 3.31 The feed milling industry in China is in its infancy. Most of the existing facilities were developed during the late 1970s and early 1980s. Aggregate processing in 1985 was about 15 million tons of feed, or 12% of the total of high-quality feeds potentially available for livestock use. Of the total, only 6 million tons (40%) were considered to be fully formulated feeds, i.e., roughly meeting NRC nutritional standards; the remainder were mixtures of grain and grain byproducts of widely varying nutritional composition. The in- dustry is heterogeneous with facilities varying from large mills with 30,000 tons per year capacity or more (defined on a one-shift basis) down to small 1,000 ton per year operations. Most of the small mills are county-owned enterprises, or else are milling facilities belonging to a large-scale production unit which utilize most of the milled feed on their own farms. 3.32 The feed mills operate under two pricing systems, specifically the list price system whereby feed mills process products acquired at subsidized prices and distribute them to farmers who have an entitlement to procure officially priced concentrates, and the negotiated price system whereby ingredients are bought at negotiated or market prices and sold at a markup to farmers (even feeds which must be sold at list prices contain increasing proportions of nego- tiated price ingredients, creating a cost-price squeeze). List price ingredients are available exclusively to mills under the Grain Bureau system (most large mills), which also are in the best position to guarantee supplies of most raw materials. In both situations the buying prices and selling prices for the feed mill are administered prices and the mills do not have the right to set their own price levels. 3.33 Technical Performance. The milling and mixing systems of feed mills visited were generally technically adequate and capable of being operated effectively. However, some mills were poorly equip- ped for introduction of trace elements and vitamins into the mix. The one aspect of design which was almost universally poor was concerned with pelleting. Although several of the mills visited had invested in pelleting equipment, none was operating effectively. In consequence virtually no feed is currently being sold in pelleted form, although this would be highly desirable both for poultry and for fish food. The operation of the feed mills was typically labour intensive, often with manual weighing and usually no advanced electrical monitoring and control systems in use. In view of the very low wage rates in China this lack of mechanization/electrification probably minimizes costs; however it does make it difficult to achieve high and consistent product quality. 3.34 The major shortcoming of the feed mills was the lack of technical quality control. Little sampling was done on the quality of ingredients coming into the feed mills and in consequence the degree of variability of final product is thought to be large. However, because the composition of finished products is rarely either included - 45 - on labels or tested for, the degree of variability is uncertain. As users of feed mill products, particulary poultry farms, become more sophisticated it will become increasingly necessary to ensure that nutritional composition of finished feed is known in detail and gua- ranteed. An associated weakness of the feed milling industry as a whole is the limited range of feeds and balancers produced and the weak linkages between the feed mill and the needs of livestock produ- cers. 3.35 Future Role and Development. Even at the present levels of livestock production close on 100 million tons of high quality feed is required for pigs and poultry. Part of this needs to be met by fully formulated feeds, and part by balancers to upgrade the grain and bran which may be available at the farm level. To fully satisfy the existing market, the feed industry would need to produce something of the order of 10-20 million tons of formulated feeds and 15-25 million tons of balancers. Clearly it will take a considerable time to reach this level. 3.36 In order to expand and at the same time improve techni- cal quality of feeds, a number of measures are recommended. These include: (a) upgrading some of the existing feed mills; (b) establishment of a nucleus feed mill in each main consuming area. The nucleus mill would inter alia manufacture pre- mixes to be supplied to smaller feed mills; (c) establishment of effedtive feed quality control laboratories independent of the feed mills; (d) strengthen the organization of the feed milling industry to achieve uniformity of feed standards and improved supply of ingredients and feed distribution; (e) strengthen research and development; (d) enforce regulations to ensure that feeds produced by the milling industry comply with pre-determined standards in terms of nutritional value. 3.37 Upgrading of existing feed mills would focus on monitor- ing and control systems and make provision for sampling of finished feed in the lines prior to bagging out. Secondly, existing pelleting lines should be redesigned to minimise the distance between the pellet mills and coolers and also to provide for the possibility of introdu- cing fat or molasses in certain feeds. 3.38 The plants destined to become nucleus plants would need to employ advanced processing techniques and should be designed with a separate pre-mix line. They should be combined with laboratories capable of analysing all nutritive components including energy, diges- - 46 - tible crude protein, amino acids, calcium, phosphorus and trace ele- ments, as well as undertaking physical feed analysis which,would allow them to investigate and remedy separation problems. 3.39 Besides its production role, it is recommended that the nucleus feed mill also have overall responsibility for technical assistance to other feed mills. It would undertake routine analysis of feed ingredients, monitoring of sampling procedures and analysis of finished feeds. It would'also be responsible for the;provision of feed formulae to satellite mills based on the ingredients and pre- mixes available. Consequently, as well as technical staff, personnel at the nucleus plant would need to have expertise in feeding and animal nutrition. 3.40 It is proposed that in the future the feed industry be geared to producing the following types of feeds: (a) compound feeds which are complete rations mainly for pigs and poultry, formulated to be fed as sole feeds without any additional feedstuffs; (b) different types of balancer feeds to be fed as supplements to roughage, succulent feeds, own-produced energy feeds, and/or industrial byproducts, with a digestible crude protein normally in the range 25-50% and adequate calcium, phosphorus, sodium, trace elements, and sufficient vitamin supplementation. The balancers would need to be tailored to specific needs in different areas; (c) micro-mixes (pre-mixes) defined as an intermediate product within the milling industry, prepared by the nucleus mills for use in small-scale mills. These include calcium, phos- phorus, sodium, etc. at levels to ensure that the final feed has the appropriate Ca:P ratio, trace elements, vitamins and certain amino acids, and growth promoters. They would need to be formulated for each class of livestock and would have to take into account the precise raw materials (including both energy and protein feeds) being used at the satellite mills; (d) mineral-mixes designed for inclusion at about 3-5% in the final mix, primarily at the farm level. Mineral mixes would contain no protein, but calcium, phosphorus, sodium, trace elements and vitamin supplements adapted to the requirements for the different types of animals. They would be mixed and used at the farm when all feedstuffs - grains, brans, oil cakes/meals, waste products, and others - are available; (e) medicated feeds, which are complete or compound feeds spe- cially enriched with preventive or treatment medicines, to be manufactured by mills according to prescriptions of a veterinarian. - 47 - 3.41 This proposed system of development for the feed indus- try using nucleus and satellite feed mills and pre-mixes with techni- cal assistance coming from the nucleus mills is designed to overcome the problem of relatively unadvanced technical facilities in the rural areas. Its effective operation would require close liaison between the nucleus and the satellite plants. If this can be achieved and routine sampling and analysis of feeds is carried out regularly, then the feed industry would be in a position to provide a confident deli- very analysis on each bag or load of feed produced, which would also have to be periodically checked and certified by independent laborato- ries under control of a regulatory agency. This in turn would allow more efficient feeding to be implemented at the farm level. - 48 - 4. LIVESTOCK AND POULTRY BREEDING Dairy Cattle Breeding 4.01 The Present Infrastructure. Some forty bull stud and artificial insemination (AI) stations are located in various provinces in China (the major ones in Harbin, Beijing, Shanghai, and Tianjin), and provide semen from improved bulls to state farms, collectives and SH farms. These are run by the AHB, and their activities are coordi- nated by the Dairy Association, which has set standards for a breed registry (now containing about 30,000 cows) and performance testing program, and also coordinates the importation, quarantine and distri- bution of breeding bulls, heifers and bull semen, through the Central Quarantine Station, the Veterinary Medicines & Instrument Company, and the China National Animal Breeding Stock Import and Export Corpora- tion. 4.02 Breed Improvement Policy. China's Seventh 5-Year Plan envi- sions a 111% increase of milk production by the year 1990, and an increase in per capita milk consumption from the present 2.1 kg per year to 21.0 kg by the year 2000, which translates into an increase of the present 700,000 head national herd of producing dairy cows, avera- ging 3 t/year of milk production, to the equivalent of 4 million head each producing on average 6 t per year. These goals, which imply a sustained 16% p.a. increase in milk production, will put considerable strain on the technical, extension, and production resources committed to breed improvement. To meet these goals, a "two-legged" policy of exploiting domestic resources while drawing heavily on imported gene- tic material has been formulated by the government. Aside from rapid expansion of the present herdby maintaining relatively low culling rates, emphasis will be placed on upgrading a portion of the 61 mil- lion local Yellow Cattle and 20 million water buffalo now suitable mainly for use as draft animals, through multiple crossings with breeds of higher milk yields, aiming at single-purpose milk animals in the suburban areas and dual or multiple purpose animals elsewhere. 4.03 To distribute improved genetic material, China relies al- most exclusively on AI. At present, most of the semen comes from foreign animals. However, only 10-15 per cent of the total semen volume is imported directly, and most of the gene pool imports repre- sent breeding bulls. The dairy breeds used are primarily the Hol- stein-Friesian and, for water buffalo, the Murrah (India) and Nili- Ravi (Pakistan). In addition to stud bulls, China has also imported increasing numbers of pregnant heifers both for breeding and produc- tion use (in 1985, 5,000 head were imported, an exceptionally large number). There have been problems with this program, notably high costs (transport and quarantine costs, including losses, nearly double the price of the animal) and difficulties due to stress and maladapta- tion to the Chinese environment. As there are now about 10,000 cows in the national herd with yields of 7 tons or more, a sufficient basis exists to begin a national selection program and phase out heifer imports, particularily of animals procured mainly for production pur- poses. - 49 - 4.04 Performance Testing. China has only just begun a progeny testing program for both imported and local bulls, and it appears to have penetrated only the topmost layer of stud farms to date. As practiced by the Beijing Dairy Stud Station, it is supposed to involve 200 services per untested bull distributed over a large number of different farms in a period of three months. Based on at least 50 daughters, a Total Production Index is computed for the bull based on first lactation data. A system to identify superior bulls has also been set up in the Beijing area, drawing from among the progeny of cows producing at least 8,000 kgs per lactation with 3.6-3.7% fat content. At present about 20 locally bred Holsteins (one-third of bulls on station) are under test at the Beijing station. 4.05 However, at the Tianjin Stud Station, the fourth largest in China, and at other smaller stations, the breed selection program so far consists of little more than a milk recording scheme to monitor half-sibling heifers as well as older, producing cows bred to stud station sires. No emphasis seemed to be placed on superior bull identification, as virtually all bulls on these farms were imported from abroad, where their superiority was already established. For example, the Tianjin Station reportedly requires that semen from young untested bulls be used on 30-40 heifers within a 3-month period. Heifers are expected to reach 5,000 kg of production in their first lactation (a low standard when compared to the local herd average of 5,805 kg). However, examination of progeny tests for four bulls suggested that in practice bull performance was only evaluated on the basis of between four and nine daughter records, collected over a 3 1/2 year period. 4.06 It is premature to evaluate such a new program, but it is apparent that further progress towards central coordination and stan- dardization of the performance testing program is required, as well as increased attention to selection and retention for appraisal of local- ly bred bulls, in the interest of future savings in import costs of genetic materials. One difficulty which is unavoidable in the early stages of the program is that both nutritional and record-keeping deficiencies, which are common on all but the best managed farms, will reduce the number of cows meeting production criteria set by the national selection program. 4.07 Selection of Dairy Cattle Breeds. The dairy breed. preferred in China and universally encountered from Heilongjiang in the north- east to Hainan Island in the tropical south, is the Holstein.-Friesian, locally referred to as the Black-and-White. While this selection of breed is broadly appropriate, questions may be raised about both the type of Holstein employed in breeding, and its suitability in south China. Most of the cattle on state farms and in stud stations derive from European Holstein-Friesian blood lines, exhibiting the somewhat muscular, stockier conformation of the dual purpose (milk and beef) Holstein breed in Europe in the 1950's and 60's. These cattle differ from the North American, larger framed Holsteins, which have been bred for the production of milk alone and are now producing some 500 - 50 - kg per lactation more milk than their dual-purpose European counter- parts under comparable conditions. Because of the higher genetic milk production potential of North American Holstein-Friesians (a maximum of 12,000 kg of milk per lactation along with 4 per cent butterfat) and the relative lack of local or international demand for beef, the Dairy Association might consider an early shift in emphasis in the breeding program toward the single purpose dairy type, following the example of the European breeders. However, at present nutritional and management factors are far more limiting than genetic potential, and it is unlikely that the genetic superiority of the specialized blood- lines will express itself unless or until these environmental con- straints are removed. 4.08 The differences in climate between temperate-zone northern and subtropical southern China appear to be reflected in an increase in seasonal variability and a decline in overall productivity on dairy farms, the farther south one looks. This is neatly reflected in cross-sectional comparisons of average annual milk yields per produ- cing cow for entire provinces: 2.9-6.3 t in the northern subtropical zone (Anhui, Hubei, Jiangsu, and Shanghai), 1.9-4.0 t in the central subtropics (Yunnan, Hunan, Zhejiang, and Jiangxi), and 1.3-3.0 t in the southern subtropics (Guizhou, Guangxi, Fujian, and Guangdong). Among sites visited, both Jiangxi and Fuzhou (Fujian), where hot, humid weather exists from April to November, showed considerable productivity drops during the hot seasons. For example, during the cooler months milk production at the Red Star State Farm in Jiangxi rose to 1.7 times its average level. Due to the small number and better management quality of farms visited, it is difficult to find a clear correlation between milk production and geographical region for the farms visited (Table 4.1). Nevertheless, heat stress was an obvious problem for Holsteins observed by the mission in the open in Fuzhou in early June. 4.09 The effects of heat stress on Bos taurus cattle imported into subtropical and tropical climates are well documented. The problem can be dealt with effectively either through environmental control or breeding. Some south China farms, such as Jiangxi's Red Star Dairy Farm, have installed water cooling systems relying on spray or mist; Wu Changmei's small SH operation installed one ceiling fan for every two cows (a fixed cost of Y65/cow). In both cases, under otherwise superior management, high productivity was obtained (see Table 4.1). Alternatively, a commercial Holstein/B. indicus, stable-cross replace- ment heifer could be imported, with the appropriate semen source to perpetuate this hybrid. Australian herd records indicate that the Australian Friesian-Sahiwal (AFS), a stable breed maintained at 50% Bos indicus, under tropical conditions has an overall performance (based on an index of milk production per day of intercalving inter- val) about 22% better than the Holstein-Friesian and 21% better than the Australian Milking Zebu (a Jersey-Zebu cross). The margin of superiority might be reduced or eliminated under subtropical condi- tions, however. In general, the economics of production using such crosses needs to be compared with that of production by purebred Holsteins under modified feeding and management systems; and economi- - 51 - Table 4.1 Average Milk Production by Region and Individual Farms Kgs Production Location Per Cow Per Day /a Tianjin Municipality (average) 18.6 Farms: Miao SH 16.6 Panzhuang State Farm 15.2 Shuanglin State Farm 22.3 Farm Average 18.0 Dongxiang County and Nanchang (average) 18.7 Farms: SH Dairy, 100-Cow Group 13.1 Red Star State Farm 16.7 Farm Average 14.9 Shandong Province (average) 9.0 Farm: Gaoguanzai Dairy Coop 13.4 Fuzhou Municipality (average) 8.0 Farm: Northern Suburb State Farm 9.3 Dongxiang State Farm 14.7 SH Wu Chang Mei 16.6 SH Xue Shan You 16.3 Farm Average 14.2 /a Based on 305 day lactation. cal systems for environment control in the south need to be designed and promulgated. 4.10 AI Service Delivery. AI is the chosen means for dairy cattle improvement and multiplication in China. Bull semen is distri- buted from stud stations to peripheral state and SH herds, and all bulls are maintained, tested and collected at these stations. This system is entirely acceptable if AI services meet the goal of obtain- ing adequate conception rates within 80-90 days after calving, in order to maintain average intercalving intervals of close to 12 months. This requires that cattle be inseminated at their second estrus after calving, with a repeat insemination, if necessary, 21 days later. 4.11 The effectiveness of AI is sensitive to three factors - detection of cows in heat, expression of estrus by cows, and insemina- tion techniques. Detection of estrus requires careful daily monitor- ing of post-partum cows; failures are usually attributable to over- - 52 - worked, inexperienced, or poorly motivated farm workers. Expression of estrus depends on physiological functions which can be interfered with by nutritional deficiencies such as shortages of vitamin A, phosphorus, trace minerals, protein and energy. Insemination failures usually relate to improper timing or technique, the first due to excessive travel distances, lack of sufficient technicians, or poor motivation, and the latter due to inexperience or carelessness. Defi- ciencies in all three factors may be influencing the efficacy of AI delivery on many of the dairy farms investigated. 4.12 The rate of services per conception is one indicator of the general efficacy of an AI program. The critical line in AI efficacy is typically set at 1.7 services per conception, above which serious problems exist related to one or all of the factors listed above. A positive correlation between this rate and the herd size was evident on farms visited (Table 4.2), similar to trends observed in other countries. Rates deteriorate mainly on state farms, where AI techni- cians are close at hand, semen is immediately available, and nutri- tional standards are relatively good. Hence the most likely cause of this deterioration is failure in estrus detection by farm workers due to inattentive cow observation. In smaller SH herds, observation is apparently better, as small groups can be watched more easily and owner-managers are better motivated. This indicates a need for im- proved motivation and/or training of the (mainly state farm) techni- cians involved in estrus detection. 4.13 Poorly executed delivery of AI services will greatly damage productivity. As China has selected Al as the basis for its breeding program, the natural breeding option has been largely elimi- nated. This option should be reconsidered and reinstituted in areas where conception failures are serious. Where more than 1.7 services per pregnancy are required, alternative means of settling cows should be available, if only on a temporary basis, until the AI problems have been resolved. The state farm herds appeared to be the more likely candidates for such a reconsideration, in view of the high reported services per pregnancy ratios. Pig Breeding 4.14 The Lean Pork Policy. MAAF has promulgated the "Three Principles" policy under which the local Chinese pig breeds are to be upgraded from unimproved, fat to improved, lean pigs (the latter presently account for about 8% of all pigs slaughtered). The policy's three strategies are to (a) utilize the existing national sow herd by (b) crossing local sows to improved or exotic (imported) lean boars, and (c) to raise the resulting improved crossbred Fl offspring on better quality feeds allowing for better efficiency of growth and lean meat ratio. 4.15 The available local pig breeds show considerable phenotypic variation but are more homogenous in their genotypic high rate of fat deposition with increasing weight and in their very low post-weaning - 53 - Table 4.2 Services Per Conception and Dairy Herd Size Farm and Location AI Services/Pregnancy Cow Numbers Gaoguanzai 67 dairy SHs, each Cooperative, Jinan 1.4 with 3-5 cows Nanchang SH 1.5 Wu Chang Mei SH 1.5 15 cows Miao SH, Tianjin 1.6 7 cows Nongxiang, Fuzhou 1.6 71 cows Northern Suburb, Fuzhou 1.6 160 cows Shuanglin, Tianjin 1.7 730 cows Red Star, Dongxiang 2.0 665 cows Panzhuang, Tianjin 2.5 180 cows. growth rates. Lean ratios of local breeds range from 35 to 45 per cent, compared to Government's goal of 51 - 54 per cent lean for commercial crosses. The three-five year target for accomplishing this goal nationwide is very ambitious in view of the size of the required breeding program. The introduction of lean, improved boars into the system can have a considerable influence on carcass improvement, as matings between local sows and improved boar breeds introduce 50 per cent of improved genetic merit into offspring. The increased use of AI would speed this process considerably. 4.16 Ten to twenty improved boars maintained by an AI station would be sufficient to inseminate 10,000 sows twice a year. Assuming a national AI service rate of 30 per cent and extrapolating to the national scale, China's sow herd of 22 million head would require a total of 780,000 improved boars./l At an average annual replacement rate of 30 per cent, 234,000 replacement boars would have to be pro- duced each year, by a national breeding sow herd of 95,000 head dis- persed on provincial breeding stations in proportion to pig population densities in the rural and suburban areas./2 An increase beyond 30 per cent of AI services would reduce the number of breeding boars and breeding sows, at the risk of a marked deterioration in service effi- ciency (as AI is most successful in areas where livestock populations are most concentrated). An AI program can disperse the genetic merit of a boar at a rate 33 times that of natural breeding. The genetic quality of the improved boars to be used must therefore be beyond question, and this calls for a standardized test station setting using a proven selection program. Such programs are widely used and docu- mented in the pig industries of developed countries and can be adapted readily for use in China. /1 AI: 6.6 million sows at 15 boars per 10,000; natural breeding: 15.4 million sows at 1 boar per 20. /2 Based on a low selection rate of 1 in 3 boars, and on 15 pigs raised to 90 kg per breeding sow per year. - 54 - 4.17 The Role of Local Breeds. Some 49 local breeds of Chinese pig exist, including lesser as well as better known varieties such as the Laiwu in Shandong Province, Taihu in Jiangsu, Jinhua from Zhejiang Province, the Beijing Black, etc. A few of these breeds (found mainly north of the Yangtse River) have high prolificacy, with litter sizes (from the second litter) averaging 13-16. Fl hybrid local x exotic sows preserve litter sizes of 10-11. Under backyard management, mortality before weaning is often extremely high, and even with the best local breeds, frequently only 9-10 pigs survive through weaning. This is attributable to the management environment, particularily low winter temperatures and sow crushing. The lack of farrowing crates on farms results in considerable crushing by the sow of her weaker pigs, especially with very large litters of pigs of a lower birth weight. 4.18 The mission did not observe many unusually large litters, perhaps because a high proportion of field visits were south of the Yangtse and to farms raising hybrid pigs from mongrelized local sows. Litter sizes observed on the hoof averaged less than 10, and, with a 15-20% mortality before weaning, numbers fattened per sow averaged about eight under SH production conditions, for those local, local x exotic, or exotic hybrids which were observed. It should be observed that, if the local sow is not one of the purebreds noted for prolifi- cacy, a major advantage of its use in commercial breeding disappears. 4.19 Local breeds are also credited with good adaptation to coarse feeding regimes (as well as preferred meat flavor). The correct interpretation of this statement is crucial in determining the future role of local breeds, as well as the scope for improving their manage- ment. Exotic pigs have been bred for maximum efficiency in conversion of quality feeds into relatively lean meat, and in this respect are clearly superior to local breeds. Local pigs, raised on a high- quality feed regime, would begin to convert feed into fat rather than meat at an earlier weight, thereby becoming less efficient converters (as fat deposition requires 2.5 times more feed energy than lean meat deposition). Such pigs could, of course, be marketed at lighter weights (before excessive fat deposition occurs); however, this would greatly reduce their average efficiency, which is boosted by the better marginal conversion rates during the final growth phase. The traditional feeding technique, which emphasizes the use of low quali- ty, high fibre feed and roughage, has an efficiency rationale in that it lowers energy and protein intakes, slows growth rates, and reduces the excessive fat deposition that would inevitably occur on high energy feeds. 4.20 National surveys of production costs/l verify that local pigs, raised to market weight using relatively coarse feeding techniques, obtain a better feed conversion rate (2.7:1) than is /l State Statistical Bureau, China Rural Statistics Yearbook 1985, p. 169 for 1980-83 and MAAF for 1985. The survey statistics, based on a sample size of several thousand pigs, reflect more intensive management than average (as deduced from national aggregate statistics on pig production). - 55 - likely for exotic or hybrid pigs fattened to the same weight on mainly quality feeds, if conversion rates are measured as the ratio of live weight gain to amount of quality feed consumed. Of course, this method of measurement ignores quantities of garbage, green feed and other forms of roughage (in these surveys, totalling about 700 kg/pig at fresh weight) fed to the local pig. These ingredients have little or no opportunity cost, but can have considerable food value (e.g., legumes or the tops of sweet potatoes). For the average pig in this sample, during 1980-83 quality feed increased by 36 kg, roughage decreased by 328 kg, and slaughter weight increased by 2.5 kg. This implies that the average food value of roughage, in "quality feed- equivalents", is about 12%; and that therefore, in 1983 roughage accounted for 25% of the average pig's dietary intake. The total feed conversion rate may thus be estimated as 3.5-3.6:1 or, on a carcass weight (without head) basis, 5.7:1. 4.21 A reduction in the percentage of roughage in the sample pig's diet from 37% in 1980 to 25% in 1983, together with probable growth in numbers of hybrid pigs raised, contributed to an improvement in the average conversion rate from about 5.9:1 to 5.7:1 during the same time period. As the latter figure already compares favorably with average commercial results of 5.3:1 in the U.S., it appears that local pigs, given better than average feeding and management, can approach the overall feed conversion ratios achieved in advanced countries, albeit via slower growth rates and much higher rates of fat deposition. Thus, the speed at which the "lean pig" should be intro- duced will depend largely on consumers' changing valuation of lean and fat, relative to the higher cost of production of lean, improved breeds, which is due mainly to reduced possibilities for use of low opportunity cost feeds. 4.22 Because of the speed with which the government hopes to upgrade the leanness of Chinese pigs, initial reliance on the local breeding sow is inescapable. However, Chinese breeding strategy seems to go a step further, in the interest of maintaining prolificacy and adaptability to coarse feeding, to the extreme of dictating preserva- tion of purebred local sows to serve as the maternal line in Fl crossings for commercial use. Whether or not this strategy makes sense, compared to the alternatives of gradually replacing local sows or encouraging further dilution of the maternal line with exotic genes, should also be determined on economic grounds, that is, the final value of pork produced per sow per year, relative to total feeding costs. Unfortunately, the reproductive advantages of the best local breeds are balanced by almost equal disadvantages in growth rate, killing-out percent, carcass lean and conformation./i 4.23 The drawbacks of the slow growth rate and low carcass lean percentages of local pigs are underlined by a rough comparison (Table 4.3) of total meat production per sow per year, as generated by local x local, local x exotic and exotic hybrids under intensive SH manage- /l M. Bichard and P.J. David, "Producing More Pigs Per Sow Per Year - Genetic Contributions," J. Anim. Sci. 1986. 63: 1275-1279. - 56 - Table 4.3 Lean Meat Production, Per Sow Per Year From Local Local x Exotic and Exotic Crosses/a Local x Exotic x Parameter Local Exotic Exotic Pigs raised/ sow/year 14.0 13.7 14.8 Market weight, 8 mos. live (kg) 93 97 105 Liveweight marketed/ sow/year (kg) 1,310 1,320 1,560 Lean carcass meat (%) 40 50 55 Lean carcass meat (kg) 524 660 858 Increased lean meat over local x local (%) -- +26% +64% Increased carcass retail value (%) /b -- +16% +46% /a Based on Annex 3, Table 10. /b Assuming the market price of lean is 260% greater than the price of fat, as observed in Fuzhou urban markets in 1986. ment conditions (in all cases, with maximum use of high-quality feed). The estimates imply that the two latter crosses would yield increases in lean meat output of 26 and 64 per cent, respectively, compared to local, unimproved breeds. In the southeastern urban markets, where lean meat now commands a sufficient premium over fat, the total retail values of the hybrid carcasses are likely to be about 16% and 46% higher respectively, with minimal differences in feeding costs. Under comparable feeding regimes, therefore, there seems to be no obvious economic advantage to local x exotic (improved) crosses, compared to exotic-type hybrids. 4.24 The comparison puts the local and even the local x exotic pigs at a disadvantage, however, because it ignores their relative adaptability to coarse feeding, which may be more economically significant in the Chinese context than prolificacy. With a cheaper diet composed of a higher proportion of green fodder, and despite the lower growth rate, the traditional producer can finish the local pig at a cheaper cost than either hybrid. For example, at present there is reportedly a 15-20% differential in production costs between local - 57 - and improved, hybrid pigs fattened to the same market weight. Without a retail price differential for lean pork over fat at least as large as found in a few urban markets in southeast China, production of the local x exotic pig may not be economically worthwhile. 4.25 Grading for Lean Pork. The speed with which the "lean pig" is introduced should vary with consumer tastes in each market, as reflected in relative prices of lean and fat. It appears that "condi- tions are ripe" for leaner pigs in the southeastern urban market and possibly the more prosperous cities in the north; but possibly not in other urban markets and certainly not in the rural areas. Overall, the procurement pricing system for pork which is currently in force does not provide sufficient incentives for the pork producer to pro- duce leaner pigs, regardless of retail market valuations. With minor regional variations, procurement prices are based on five grades of pork, based on weight and dressing percentage (Table 4.4). By Western grading standards, Grades II and III could include better quality pigs than Grade I, as heavy, local pigs could still dress out at 75 per cent. In Fuzhou, where hardly any lean pigs are raised, 50 per cent of pigs presented at the Freezing Factory slaughterhouse fell into Grade I, 30 per cent in Grade II, 20 per cent in Grades III and IV, and 1-2 per cent in Grade V. The procurement price differential between Grades I and V is only Y 0.24 per kg, or 12 per cent. At this small differential, most farmers do not have an incentive to invest the additional funds to switch to the use of improved breeds and feeds. 4.26 Government authorities are aware of this problem and Price Bureaus in some provinces and municipalities have given permission for a procurement price premium fdr "lean pigs", i.e., local x exotic crosses. However, this premium typically averages only about 5-7% of the total price, and appears insufficient to reflect either a signifi- cant improvement in lean yield or increased costs of production. The urban ATMs, at least in the southeast, obtain a large premium on retail prices for lean cuts, which sold for up to Y 5.80 per kg,, com- pared to average pork cuts at Y 3.20 and pork fat at only Y 1.60 per kg (implying roughly 16% and 46% increases in market retail value of the local x exotic and exotic carcasses, respectively, over local pigs). However, most pork production doesn't pass through the free market, and it is the state procurement system which tends to set the farmgate price of the pig; the lean meat and fat prices in state markets visited by the mission are around Y5.20/kg and Y2.20/kg, giving the local x exotic hybrid pig only an 11% retail price advan- tage over the local pig. 4.27 At present, pigs are usually graded by sight before slaugh- ter, and improved pigs given a higher grade, although the objectivity of grading comes under pressure during times of oversupply. As a prerequisite for widespread breed improvement, a uniformly applied pork grading system, independent of market forces, and which presents sufficient price incentive to produce leaner pork is urgently needed in areas where ATM retail prices justify such incentives. Grading is best performed by government-appointed personnel acting independently - 58 - Table 4.4 Pork Meat Grades and Their Procurement Prices (Quoted in Fuzhou) Grade Dressing % Live Weight Procurement Price (kgs) (Yuan/kg) I 75.1 85 - 115 2.24 II 71 - 75 85 - 115 2.18 III Not evaluated <85 and >115 2.12 IV Not evaluated <75 and >125 2.06 V Not evaluated >150 2.00 of the abattoir's management. Ideally, backfat measurements would be taken on split carcasses moving down the slaughter line, which togeth- er with weight data, could then be converted into a lean ratio related grade. This system would be easy to implement for large producers who deliver direct to the slaughter house. However, as reliable grading on the hoof is not so easily implemented at state-run procurement stations located some distance away from slaughter houses, some thought must be given to finding ways of passing premiums back to the small-scale producers who are responsible for the majority of pig production. 4.28 Pig Breeding Stations. China is beginning to formulate a national pig breeding system, based on central government-run farms for pure breed maintenance and improvement; provincial or locally-run farms responsible mainly for multiplication, along with some selection and culling; and county- or district-level multiplication farms which distribute to commercial farmers, who are responsible for the final hybridization for production purposes. The mission did not visit any of the twelve breeding farms at the top of this "official" pyramid. It did visit a number of (unofficial?) breeding farms or stations in Shandong, Tianjin, Dongxiang and Fuzhou which were under state farm, collective or municipal management. Some of these farms would qualify for nucleus herd status, based on the general management and care of their stock, including the Red Star farm, the Northern Suburb farm in Fuzhou, and the Laiwu farm in Shandong. However, despite different locations and administrative levels, most of the stations shared a number of limitations which are important to their roles in spearhead- ing the lean pig policy. 4.29 First, in order for these stations to function as selec- tion centers as well as the breeding and primary multiplication cen- ters that they now are, a comprehensive system of data collection and record keeping will have to be introduced. Recording activities are presently confined to information on farrowing dates, litter size at birth and weaning, age at weaning, conception rates and, occasionally, growth rates and feed consumption of porker pigs. None of the sta- tions had a weighing scale to monitor weight gains. Backfat measure- ment equipment was not present and such measurements on live pigs were - 59 - not taken. Only occasional carcass evaluations at the slaughterhouse were undertaken. No numerical scoring systems were in use to measure sow mothering ability, the indexing of porker pigs for performance or of boar superiority over other station mates. Without such criteria, selection was reduced to visual, subjective methods of evaluation which are not standardizable and are subject to error of interpreta- tion. 4.30 Second, a great variety of breed crossing trials with local x exotic and exotic x exotic breeds was being carried out to evaluate the effects of local breed upgrading by the use of imported breeds. There is no need to re-evaluate the performance of exotic x exotic crosses as such information has been widely documented for a great number of environmental conditions. The number of variables and the lack of accurate data collection which were encountered at such eva- luation programs would hide any performance changes that might: other- wise have come to light. Variations in feed quality, or the simulta- neous testing of 23 different types of feed within a 4-breed crossing trial, as was encountered on one station, would render any information gained meaningless. 4.31 Third, insufficient knowledge of the scientific procedures involved in pig breeding was revealed in the activities of many sta- tions. Lack of data collection, feed variability, lack of knowledge about exotic breed characteristics, and reliance on visual (phenoty- pic) versus performance (genotypic) selection were all indicative of this deficiency, which in most cases must be traced to the advice and training received by breeding station staff. The procedures involved are basic and well-tested, and need only be adapted for use in China. 4.32 Fourth, much of the opportunity for the stations to disperse improved stock quickly throughout the pig population is wasted by the failure to increase numbers of superior improved stock through a multiplier system, before their dispersal to farmers. While the immediate dispersal of small numbers of primary stock from stations to farmers may appear to be a quicker method to achieve on-farm improve- ment, it will in the longer term fall short of the multiplier system. There are presently too few improved herds on stations to quickly influence pig performance by direct dispersal, and all station-derived pigs should be utilized to the full by multiplying them elsewhere. 4.33 The Pyramid System. The breeding, selection and dispersal system recommended is the well-tested pyramid breeding concept, where the nucleus breeding stations maintain the nucleus herds at the pyra- mid's tip. Purebred exotic lines at this level are subjected to severe performance evaluation. Those pigs passing the evaluation protocol are then brought down to the larger multiplier herd system, which multiplies, for example, Landrace x Yorkshire crossbred replace- ment gilts and purebred Duroc, Yorkshire and/or Landrace boars for commercial use on a larger number of farms (these breeds are the main ones chosen for use in upgrading China's herds). At the bottom of the pyramid are the commercial producers who purchase suitable replacement gilts and boars from the multiplier farms to raise market pigs. The - 60 - system is adaptable to Chinese conditions by converting suitable breeding stations into nucleus stations. Joint venture or technology- transfer arrangements with international breeding companies could also be negotiated to produce replacement gilts and boars in those areas without suitable local breeding stations. 4.34 Decisions on the selection and importation of exotic breeds, and also on the selection of local purebreds for upgrading would be made at the nucleus level. Herds at nucleus breeding sta- tions must contain at least 50 breeding females per line or breed in order to maintain a meaningful selection pressure. Boars should be turned over quickly (maximum 1 year of service) in order to maintain the genetic diversity required for rapid genetic improvement. Selec- tion procedures should be strictly objective, scientifically grounded and uncompromising so that performance faults are largely eliminated at the nucleus level. 4.35 Multiplier herds should be established and stocked with superior pigs from nucleus breeding stations for multiplication. They may be managed by trained SHs with special expertise in pig raising, who would be responsible for carrying out specific breed crossings to achieve simple but specific goals are achieved. For example, SH multiplier herds can be used to multiply one or two exotic breeds as purebreds, such as Duroc, Landrace or Yorkshire/Large White boars and gilts. Other SHs can cross pure lines of Landrace and Yorkshire to produce crossbred L x Y replacement gilts, with the boars going to slaughter or for dispersal to local sow herds. Others could cross local breed sows to Yorkshire/Large White or Landrace, with the im- proved replacement gilts being dispersed and the males sent to slaugh- ter. Multiplier herds would be culled only for conformation faults (eg. weak legs and number of teats), based on advice from nucleus breeding station staff. They should be managed like nucleus herds, under a closed herd system with footbaths and access allowed only to that part of the farm where animals for sale can be seen by visiting farmers. Performance measurements at the multiplier level would be kept to a minimum, assuming that the stock originated from the nucleus breeding station and had been properly selected. 4.36 Pig AI Stations. Most of the AI stations visited operated in conjunction or cooperation with pig breeding farms supplying the AI boars. AI services to pig farmers are based on daily collection of fresh semen, which is transported in insulated flasks to the farms, and inseminated with an extender or in undiluted form. Insemination fees are about Y2.50 per successful pregnancy, after two insemina- tions. AI services in a county or municipality may be carried out by a main AI station, assisted by peripheral substations with access to more distant farmer clients. The proportion of AI services delivered varied by region and ranged from a high of 30-40 per cent in the Tai'an area in Shandong province to lower percentages in other areas visited. SH pig farmers also offered their own exotic or improved boars for service to other producers, at a charge of Y 5.00-10.00 per successful service. The delivery of AI services to local breeds of sow is generally quite satisfactory, with conception rates in the 80- - 61 - 90 per cent range (see also para. 5.31 below). Further utilization of AI should be encouraged to allow quicker genetic upgrading to take place. Additional AI stations, and especially substations which sup- ply outlying areas, should be considered, together with their infra- structure of transportation, trained AI personnel and farmer training programs. 4.37 The selection criteria for boars placed in the AI stations are a cause for concern as their genetic influence on the sow popula- tion may be 30 times that of natural service. Visual evaluation was again often the only criterion used for AI boar selection, although on several AI stations boars were also selected on the basis of sire and dam performance, even though such performance was not usually eva- luated with hard data. In the future, the AI stations will have to play greater roles than they do now. Close cooperation with nucleus breeding stations, and priority access to superior boars is required. Imports of breeding stock to farms should always include boars for dispersal to AI stations, not just to breeding farms alone, so that the boars in a station will present a wide variety of exotic genetic material and can be quickly turned over. The present turn over rate of AI station boars is equal to their useful lifetime of 3-5 years. This should be shortened to at most 2-3 years in order to prevent inbreeding and possible dissemination of genetic defects or inferior performance traits. Poultry Breeding 4.38 Local Versus Imported Breeding Systems. Many of the bree- ding farms visited utilized imported lines of broiler and layer birds, often renamed for local appeal, and many of the international breeding companies are represented. Local performance levels of these imports were quite acceptable, with good hatchability, egg production commen- cing between 150-160 days, egg production of 225-270 eggs per year, and egg weights of 50-60 g. Feed conversion ratios of broilers were somewhat weaker at 2.5-2.7 due to lower quality feeds, with market weights of 1.5-1.8 kg at 8-9 weeks. 4.39 The tendency to crossbreed imported hybrids with local strains of poultry to satify consumer demands for more "local flavour" or for local egg characteristics is self defeating, as imported lines are too highly specialized for outcrossing. Production then tends to deteriorate quickly within a few generations. Genetic improvement of local poultry breeds is not feasible in view of the extensive selec- tion work that would have to take place. Instead, the introduction of exotic lines, together with their integrated management systems, will turn out to be the more economical way to increase supplies of eggs and poultry meat. Duck breeding is a probable exception, where it is unlikely that small SH enterprises would benefit from imported strains in view of the limited feed resources available. Larger, commercial duck farms may want to investigate exotic duck breeds, although local strains perform quite acceptably under local conditions. - 62 - 4.40 International Poultry Breeders. There are approximately five commercial breeders of broiler type chickens and the same number of breeders of commercial egg type chickens who collectively supply 95 per cent of all breeding stocks in the world. These breeders have spent and continue to spend millions of dollars on research and deve- lopment. Poultry breeding requires highly specialized personnel, such as geneticists, nutritionists, physiologists, statisticians and vete- rinarians. The expense of developing a major domestic breeding pro- gram based on local breeds would be prohibitive and would probably fail to develop breeds as productive as those already internationally available. 4.41 A number of breeding farms visited were using crossed or backcrossed birds, mixing two or several strains. Experimentation with local x exotic crosses on a small scale could be helpful in upgrading village stock raised under foraging systems. However, these kinds of crosses will have no advantages over exotic lines on commer- cial farms, even as small as SH farms maintaining a few score of birds. 4.42 There are a limited number of breeders who will supply pure lines of both layer and broiler type breeding stock. Such breeders are willing to contract with the government or private sector enter- prises to produce quality breeding stock from such pure lines. This is usually done as a joint venture, with the breeding company supply- ing the genetic "know-how", including a planned breeding program to improve the pure line stock. The main advantage of procuring pure line stock is a substantial saving in foreign exchange and an assu- rance of continued genetic development. Although the initial price of pure line breeders is high, yearly imports of breeding stock (P and GP) would not be necessary. There is very little difference in gene- tic makeup or in bird performance (within each breed, i.e. broiler or layer) amongst the leading poultry breeders. The difference is in the services rendered by the breeding company, which should be considered carefully when engaging a company in poultry development assistance. 4.43 It is frequently misunderstood that grandparent stock would be all that is required to perpetuate generations of pure line bree- ding stock. Grandparents are composed of at least four lines which could not be duplicated without going back to the original stock. In other words, it is not possible to carry on the breeding program by mating a hybrid male with a hybrid female. The quality of the bree- ding stock would deteriorate in a matter of a few years. 4.44 Specialized Production Lines. There is a clear tradeoff in chicken breeding between production of meat and production of eggs, which has resulted in development of separate varieties or lines of broilers, layers, and dual-purpose birds (the latter representing a compromise of objectives). In countries where the poultry industry is well-developed, market competition leading to low prices has favored growth of specialized broiler or layer production, and this trend can be expected in China as the industry develops. Consequently, the government must plan on setting up separate breeding systems for - 63 - broilers and layers, in the usual pyramidal form, i.e., pure Line farm - GP farms - P farms - commercial flocks. The size of a recommended breeding program at each level is described in Annex 6. - 64 - 5. ANIMAL HEALTH AND VETERINARY SERVICES The Animal Health Delivery System 5.01 Administrative Structure. Animal health programs at the central government level are administered by the Divisions of Preventative Medicine, Livestock Quarantine, and Pharmaceuticals under MAAF's Ani- mal Husbandry Bureau. The AHB carries out the national programs concerned with disease prevention, vaccination, planning of veterinary services, and pharmaceutical management. Its involvement in quaran- tine concerns only the domestic side; the General Quarantine Station, under MAAF but separate from AHB, is responsible for inspection and quarantine of imported and exported live animals. The Pharmaceuticals Division sets standards, issues licenses, and manages testing for locally produced or imported vaccines and pharmaceuticals. At levels below the central government, the organizational arrangements vary, but in general from the province-level down to the village, Veterinary Stations are responsible for service deliveries within the AHB admin- istrative system. 5.02 Veterinary Strategy. The overall numbers of livestock per veterinarian/technician (Table 5.1) in localities visited fall into ranges for each type of animal which appear favourable, e.g., relative to similar ratios in the U.S. Hence the major issue in China is not numbers, but rather quality of service. With the growth of SH and larger-scale livestock farming, and correspondingly higher animal densities, animal health care is moving increasingly into provision of services to producers who are undertaking considerable investment risk, and can be expected to demand greater accountability on the part of service providers. Commercial managers are attuned to economic performance, rather than the mere survival of animals. If commercial farms are to be maintained at high levels of herd performance and efficiency, the health delivery system must shift its emphasis from the diagnosis and cure of acute, clinical, predominantly infectious diseases of individual animals, towards the prevention of chronic, often subclinical, nutritional, infectious, or metabolic diseases which may affect the performance of entire groups of animals. 5.03 The emphasis of veterinary work in China is still largely confined to diagnosis from overt symptoms and treatment of individual- ly affected animals, whereas the emphasis in the future should shift towards herd health. This requires earlier and more accurate diagno- sis, disease reporting and disease prevention via appropriate vaccina- tion programs and improved management. Laboratory identification or confirmation of disease is needed both to identify subclinical di- seases and insure accurate diagnosis of clinical conditions early enough to control their incidence. Greater understanding by service personnel of the causes of diseases, as opposed to their symptoms, would facilitate preventive measures. Veterinary training in preven- tative measures, detection of subclinical cases and of nutritional or metabolic diseases should be particularily strengthened. At present, poultry health services in China are further along the path to a preventive approach, as this approach has been imported from abroad - 65 - Table 5.1: Livestock Per Veterinarian/Technician in Localities Visited Area Veterinarians/a Livestock per Veterinarian Cattle Pigs Poultry Zhuangchiu County /b 200 + 250 1,200 12,000 Tianjin Municipality 1,280 500 1,100 13,000 Dongxiang County /c 173 35 1,900 3,500 Fuzhou Municipality 888 7 400 5,000 U.S.A. 44,740 2,800 1,200 8.800 /a Includes technical staff. /b Shandong Province (near Jinan). /c Jiangxi Province. packaged with modern poultry production systems. 5.04 Veterinary Services at Farm Level. In China, veterinarians are considered valuable, respected members of their community, a position not always evident in other Asian cultures. From this posi- tion it is easier for them to pass on new information to farmers. The Veterinary Station, located at strategic points in municipalities, counties, townships, villages and state farms, is the basic field unit from which health services are dispensed. Most incoming requests from farmers for assistance are made in person, as telephone services are rarely available. Veterinarians usually make regular rounds of farms in their area so that livestock farmers see them frequently. Most farmers indicated that they requested veterinary assistance often, not only in cases of livestock sickness but also to gain advise on feeding and breeding. 5.05 Veterinaries with formal training from agricultural colleges or universities are more likely to become either teachers or adminis- trators at levels above the county, rather than field practitioners. Veterinary technicians (VTs) working at the field level are commonly trained at technical middle schools. Although their training permits them to identify diseases which fit the standard symptoms, they are not equipped for more subtle diagnostic work. Veterinary courses offer western and traditional (herbal) options, and most veterinarians visited used and dispensed western drugs as well as herbal medicines. VTs are often trained in AI, and are responsible for the delivery of AI services to farmers. Additionally, the VT has an important advisory role to play in animal nutrition and breeding, disciplines in which his present level of knowledge is weak, and for which supplementary - 66 - training programs are therefore required. 5.06 Either VTs or technically-trained inspectors monitor the health of animals from farm to market or slaughterhouse. Animals going to slaughter are first inspected at collection points near their farms, where they are passed for further travel. For pigs processed through state-run slaughterhouses, a second inspection awaits them at the slaughterhouse, with a third during the slaughter process. Ani- mals transported for live sale at livestock markets are inspected visually for evidence of disease at the farm and again at the market entry, where they are given a health approval mark which is in turn verified at the point of sale. In some locations, pigs are vaccinated at market entry if no vaccination certificate for a certain disease can be produced. 5.07 Disease Incidence Reporting. The field Veterinary Stations produce the basic disease incidence reports, but reporting is mandato- ry only for such diseases as swine fever, foot-and-mouth, Marek's, and any occurrence of new diseases. In the past, when livestock insurance schemes were widespread, statistical reporting of animal deaths was thorough; at present, with largely individual household management, reporting may not occur where insurance is not practiced and diseases do not fall within the mandatory reporting categories. Also, veterinary practice within the state farm system is semi-inde- pendent of AHB, and reports are not necessarily passed to AHB by state farms unless the situation has become serious. Reports originating from slaughter houses or export organs under the Ministries of Com- merce and Foreign Economic Relations and Trade also did not usually get to AHB. Under national Livestock Disease Prevention Regulations promulgated in 1985, all government agencies are to begin reporting disease statistics to the Veterinary Stations at each level. More detailed regulations on reportable diseases are being compiled. AHB is presently computerizing a slow, paper-oriented reporting system. Disease Status in Dairy Cattle 5.08 The control and monitoring of infectious disease in cattle is facilitated by the practice of stable confinement and their close contact with handlers. Skin testing for tuberculosis (TB) and blood testing for brucellosis were carried out on an annual basis for all dairy herds visited. TB skin testing was performed with both avian and mammalian tuberculin. No milkring field test for brucellosis was carried out, and heifer calves were not vaccinated for brucellosis, although these are regarded as effective methods of controlling the disease where outbreaks have occurred previously. Indeed, no vaccines of any type were either manufactured, imported or used on dairy cat- tle. Control of brucellosis relies exclusively on slaughtering (in a few provinces where the central government supports the work) or quarantine (elsewhere). Milk from tubercular cows reportedly is pro- cured at lower prices in, e.g., Heilongjiang, creating an incentive for culling. China is reported free from Foot-and-Mouth disease and rinderpest, but exceptional occurrences of contagious bovine pleuo - 67 - pneumonia and anthrax are listed./1 In Dongxiang, a disease named "3- day cow fever" was described which resembled anaplasmosis, one of the tick-born hemoprotozoal diseases, but no further evidence of ticks was found there or elsewhere. Cattle were sprayed "annually" for other ectoparasites, but such an occasional spray would not guard effective- ly against endemic tick-borne disease. 5.09 Disease conditions of newly imported cattle, as described, were not greatly different from adaptation problems found elsewhere. Conception failure, abortions, early delivery, climatic/nutritional adaptation failure and footrot are common conditions in imported cattle. In the most extreme case obvserved, in a Fuzhou herd, 5 of 38 imported pregnant dairy heifers died within 2 months of arrival from what was described as transportation stress, aggravated by failure to adapt to the hot, humid weather. 5.10 In spite of the often marginal nutrient intakes, metabolic diseases such as milk fever and ketosis were uncommon, possibly be- cause milk production is still too low to precipitate excessive de- mands on body calcium stores and on glucose metabolism. The Red Star herd in Dongxiang County, with a productivity level of 5,000 kg per lactation, had a 3-5 per cent incidence of milk fever in recently calved cows, higher than normal. This herd received daily intakes of calcium and phosphorus which were 33 and 11 per cent respectively below recommended levels, which was the likely cause. No evidence of milk fever in other herds on low mineral intakes was found, however. 5.11 Information on vitamin, trace mineral and macro mineral intakes and supplementation was scant. The Ca:P ratio is one impor- tant indicator, as it is the ratio, rather than the individual quanti- ties of calcium and phosphorus in the diet, that determines the ab- sorption of each element. Ca:P ratios in feeds given to lactating cows in nine dairy herds observed ranged from 0.9:1 to 1.8:1, with an average of 1.2:1, as compared to an ideal 1.4:1 ratio for lactating cattle. The problem is not critical at present average yields in these herds of 4,400 kg per lactation, but improvements in nutrition should include adequate supplementation of macro minerals. 5.12 The use of dicalcium phosphate (DCP) as a source of calcium and phosphorus supplementation is unknown in China. DCP, which has a Ca:P ratio of 1.3:1, versus a ratio of 36:1 in limestone and 2.1:1 in steamed bone meal, is closest to the ideal ratio and is the most commonly used supplement in Western dairy herds. In Chinese farms where bone meal and shell meal were used, supplementation was usually on a 1:1 weight basis, which would provide a Ca:P ratio of 4.6:1, which would lead to phosphorus deficiencies. Cows low in phosphorus are less able to utilize protein nutrients and are more susceptible to anestrus and repeat breeding syndromes. The bio- availability of calcium and phosphorus of plant origin is low and should not be solely counted on in the calculation of mineral intakes /l Animal Health Yearbook 1984, FAO/WHO/IOE. - 68 - from feedstuffs. Inorganic forms of macro and trace minerals are absorbed more efficiently and can be counted on to contribute to adequate intakes. 5.13 Foot and leg problems in dairy cattle were similar to those found elsewhere. Footrot was more common in the southern, warmer climate, particularily where continuous spray cooling kept the cat- tle's feet wet and soft, allowing for easier surface abrasions and the subsequent entry of footrot pathogens. Footrot was also prevalent in newly arrived cattle, due probably to their lack of resistance to new soil pathogens. Newly imported cattle should be walked between their exercise yards and stables through a shallow, wide footbath containing a copper sulphate and formalin solution. This would provide antisep- tic foot protection several times a day. Dryer, better drained exer- cise yards would also correct the basic causes leading to foot infec- tion. Hip and hock bruises, resulting from resting on non-bedded cement stable floors, were also common and relate to a lack of body condition and conformation. Foot and leg problems were encountered at an incidence of 5-10 per cent in the dairy herds visited. 5.14 Reproductive failure ranked high in the causes of low cow productivity, although not outside the normal range of losses. Exact data were not available but a culling rate of 5-10 percent was esti- mated for infertility in heifers, increasing to 15 per cent in older cows. The incidence of retained placenta, a failure of the afterbirth to be expelled from the uterus following calving and resulting in uterine infection, was quoted at 5-7 per cent, at the upper range of normal. This condition was especially prevalent in newly imported heifers which calved shortly after arrival. Large state farms ap- peared to have a higher incidence of infertility and often culled cows after two unsuccessful inseminations, provided pregnant replacement heifers were readily available. In private herds, three to even four inseminations were attempted before such cows were culled, as their cost of replacement to the small farmer was higher. 5.15 Digestive disturbances ranging from chronic indigestion and rumen stasis to bloat were rather common and were described on most farms. Such upsets related to the frequent changes in feed intakes resulting from the sudden changes in fresh or preserved feedstuffs offered. Without a gradual adaption period, the rumen flora cannot adjust to such changes and indigestion or bloat results. Corrective measures are difficult to institute where feedstuff shortages demand the use of diverse ingredients on short notice. Almost all digestive disturbances were treated by traditional herbal medicinal techniques (this is the normal practice for all livestock in China). 5.16 In young calves, enzootic pneumonia, colibacillary diarrhea and salmonellosis were the most common causes of disease and mortali- ty. Feeding of large volumes of full milk and resting on wet bedding or pen floors predisposes to these conditions, especially since stable floors are frequently cleaned by liberal hosings. Clinical calf pneumonia was observed in several calf units on larger farms, where treatment of acutely sick calves with antibiotics was the standard - 69 - practice. The grouping of large numbers of calves in poorly venti- lated quarters also predisposes to this condition; use of individual calf hutches should be considered as a remedy. At least 10 per cent of young heifer calves may be affected by one or more of these condi- tions, which can result in high mortality when not treated properly and promptly. 5.17 Mastitis, an acute or chronic infection of the mammary gland tissue, ranked after infertility as the most serious production di- sease. Knowledge about the incidence, etiology, treatment and preven- tion of mastitis was scanty at all levels of management, from district personnel down to the farm worker. Detailed questioning provided an estimate of 5-15 per cent incidence of clinical mastitis in the herds observed. The AHB estimates that subclinical (chronic) forms afflict 80% of cows. One state farm reported that, during each lactation, one half of the over 700 cows in lactation developed recognizable (clini- cal) mastitis. On this farm, up to 15 per cent of the milk from mastitic cows, or about 450 tons, was withheld per year. Incidence of mastitis tends to rise with increasing milk production and, in the nearly total absence of any measures to deal with the condition, it is the most serious production disease in Chinese dairy herds. 5.18 That the condition is as prevalent as it is, despite the absence of the often-blamed milking machine - known to be a major factor in its incidence - is not surprising. Udders and teats were washed with a cloth soaked in warm water only. Disinfectants were used in udder wash water on only one state farm visited. Without an udder wash disinfectant, dirt and mastitis pathogens are transferred by the wash cloth to the next cow to be washed, and so on down the milking line. The wash water and cloth then act as promoters rather than inhibitors of contamination. Also, hand milking techniques, as observed, consisted largely of strip milking, where the teat is strip- ped of milk between thumb and forefinger or even between fore and middle fingers. At a three-times-a-day milking rate, this irritating technique will result in internal tissue damage, predisposing to infection. 5.19 Rapid cow or herd screen tests for mastitis, such as the California Mastitis Test (CMT), were unknown to dairy managers. These tests can identify abnormal milk from clinically, and more important- ly, subclinically affected udder quarters. Bulk tank milk samples are taken at milk collection centers or at dairy plants but are only evaluated for coliform counts and milk solids, and apparently not for evidence of mastitic milk. Coliform counts only indicate milk conta- mination external to the cow, resulting from the unhygienic handling or transport.of milk after milking. Milk quality testing should also include the somatic cell count technique, used to indicate the pre- sence of leucocytes in milk which are evidence of udder irritation or infection. Actual identification, by bacteriological culture, of mastitis pathogens, followed by sensitivity testing to identify suita- ble antibiotics for treatment, should also be instituted by diagnostic laboratories so that this information can be passed to the farm. Such measures would deal with milk quality defects that originate before - 70 - milking. 5.20 There should be a national investigation into the incidence and types of mastitis in dairy herds of all types, carried out by the AHBs at provincial levels. Meanwhile, milkers should be quickly educated in the causes of mastitis, milking technique, udder disinfec- tion and CMT testing. Mastitis is a world-wide problem in dairy herds and methods of treatment and prevention have been well described and evaluated. Regional diagnostic laboratories, via local veterinary centres, should provide feedback on milk quality on a bulk (total herd) and individual cow basis, including treatment and drug sensiti- vity of herd pathogens. Local veterinarians should be re-trained in preventive methods and familiarized with the various field diagnostic methods available. Udder disinfectants and CMT test kits and reagents should be supplied to all dairy herds. Finally, a program of treatment for cows that have been dried off at the end of their lactation should be instituted, whereby a long-acting dry-cow mastitis control antibio- tic preparation is infused into each milk quarter. 5.21 While some economic penalty for sale of mastitic milk is necessary as an incentive to adopt control measures, rejection of milk from mastitic cows would be a devastating loss to the producers, and is probably not advisable. Indeed, it may be argued that the criteria defining acceptable milk standards in the developed world (such as bacterial count, somatic cell count, antibiotic residues) may be premature at this stage of development in China, especially since consumers are accustomed to boiling all milk before consumption. Nevertheless, there is evidence of, and justification for, consumer concern about fluid milk quality, which suggests a need for more stringent monitoring of the factors responsible for quality before, during and after processing. 5.22 Cattle Insurance. Private dairymen are at considerable risk because the size of their investment in cattle and sheds is extremely high, relative to the average income of farmers. Cattle insurance is one effective method for reducing risk, and insurance schemes to protect against death losses have been initiated under the auspices of the Dairymen's Association. For example, in one Fuzhou township where private dairying is prevalent, the scheme covered about 80% of the dairy households and 90% of the dairy cows. For an initial premium of Yl per cow, and an additional charge against every member of Yl for any cow which dies, death compensation of about 50% of the cow's market value is paid out, provided that the cow received timely vete- rinary care. The Association also ran a veterinary fee-for-service program for dairy cattle, and promoted, under government initiative, better sanitary regulations and quality standards for milk handling, with the authority to fine members. This type of Association activity will do much to improve the viability of private dairying and should be replicated elsewhere. - 71 - Disease Status in Pigs 5.23 Veterinary services to pig producers can take the form of a fee-for-service package, administered per sow and her offspring. For a reasonable fee, a veterinarian will provide diagnostic services, castrate the sow's male offspring and vaccinate for swine fever, erysipelas and paratyphoid. The spaying of female pigs, usually performed at 3-4 months of age, costs extra and is almost universally practiced in China, though uncommon elsewhere in the world. Vaccina- tion programs for pig diseases are well developed, especially for the control of swine fever (hog cholera), the most serious infectious disease of pigs in China. The government has recently increased efforts to control infectious diseases such as swine fever. The control programs are administered provincially, with counties and townships signing up for participation. Routine inspection and total, compulsory vaccination programs are instituted to control the disease. 5.24 The health implications of the growing volume of live pigs transported between provinces and counties is a cause for concern, as infections can be transferred in this manner, especially where swine fever virus or other pathogens can be present in asymptomatic carrier pigs. Newly arrived weaner pigs from other areas are apparently not initially quarantined on the farms of destination, although many of them were re-vaccinated on arrival. Weaner pigs arriving from Zhe- jiang province into Fuzhou were reported to have high incidences of salmonellosis, a typical stress-related transport disease of pigs. 5.25 Pigs in China are exposed to many of the diseases seen in other pig populations, although the one or two pig scale of most producing units tends to reduce the incidence of contagious diseases. The health requirements for importing pigs into China are very strin- gent, so that diseases such as Aujeszky's disease (pseudo rabies) and transmissible gastroenteritis (TGE) have apparently so far not been diagnosed. However, lack of accurate clinical and laboratory diagnos- tic procedures may cause such new diseases to go unidentified, so that a number of new conditions may in fact be present but unreported. Of other infectious pig diseases, erysipelas, enzootic pig pneumonia, acute Pasteurellosis and pig influenza were reported. Swine dysen- tery, a common disease in porker pigs kept under intensive conditions and high feed intakes, was rare. Enzootic pig pneumonia was endemic on large farms with high animal density, as in other parts of the world. This disease was more prevalent in Fuzhou, where up to 10 per cent of porker pigs were observed to have the disease. Mortality from colibacillosis in newborn pigs was also common in larger herds and causes considerable mortality. Electrolytes and oral antibiotics were used for treatment. Autogenous E. coli vaccines to control the di- sease were unknown. 5.26 Vaccines for pig diseases, all of Chinese origin and prepared in vaccine laboratories under AHB control, were widely available in multi-dose vials distributed through veterinary clinics. Vaccines supplied included modified-live swine fever vaccine, modified-live erysipelas vaccine, and a paratyphoid vaccine. Veterinarians usually - 72 - administered vaccines to pigs but trained farmers may purchase vaccine for use on their own farms. 5.27 Disease Incidence. The only detailed disease incidence report from any area visited was provided by the AHB veterinary staff in Tianjin (Table 5.2). Based on a pig population of Tianjin of 1.4 million, the reported incidence levels were very low, e.g. enzootic pneumonia incidence was reported at 0.046 percent, which compares with a normal endemic incidence of 10-25 per cent in pig populations. Reported incidence of swine fever was also very low, even though reporting of this disease is mandatory. This report thus may reflect the need for major improvement of the reporting system, rather than an exceptionally good disease situation (see para. 5.08). No similar di- sease incidence reports could be obtained in other locations. 5.28 Evidence of calcium and phosphorus deficiences were found on several breeding farms, particularly in Landrace and Landrace cross- bred pigs in the 50-90 kg weight range. Deficiences of zinc (parake- ratosis) and iron (anemia) were also noted. Many newborn pigs re- ceived iron given by the oral route, and injectable iron preparations at state breeding farms. Sows with litters that were herded outside their pens for foraging ingested sufficient iron from the soil to adequately supplement their needs, but close attention must be given where pigs are totally confined and have no access to soil. Orally- given iron is a low cost supplement but only 20 per cent is absorbed and utilized, so that it must be given repeatedly. Evidence of iron deficiency was never pronounced, indicating that at least borderline iron requirements are met in young pigs. 5.29 Reproductive failure in local breed sows did not appear to be a major production constraint, and most of the fertility problems, including sub-optimal reproductive performance, were confined to exo- tic purebreds and their crosses. Local sows exhibit very strong estrus cycles, are easily observed for signs of estrus, and conceive at rates of 85-95 per cent on artificial service using fresh (or fresh, extended) semen in two inseminations. Conception rates in Al- serviced exotic sows were reported to be 10-15 per cent below those of local sows. The less pronounced signs of estrus in exotics or hybrids required additional observation efforts which were seen by producers as a problem. Farrowing difficulties (dystocia) in sows were reported to be more prevalent in exotic breeds, especially where Duroc crosses were involved. Supervision during farrowing was practiced so that complications were usually noted and corrected early. 5.30 Injectable antibiotics, vaccines and other pharmaceuticals were widely available for treatment of pig diseases. However, most injectable antibiotics were only available in single-dose powder form, and were often being diverted from human to veterinary use. Large multidose vials of drugs, specifically prepared for veterinary and field use, were not available. Antibiotics for use in animal feeds or drinking water were not in evidence in most clinics visited. Mass medication is a useful method of treatment for groups of pigs, and the packaging of feed and water administered drugs should be encouraged to - 73 - Table 5.2: Disease Incidence Report of Pig Diseases in Tianjin Municipality, 1984-85. No. of No. of Disease Counties Affected Towns Affected Cases Mortality 1984 1985 1984 1985 1984 1985 1934 1985 Swine 3 2 5 4 135 94 60 50 Fever Erysipelas 4 3 6 14 117 86 62 48 Pasteurella 2 2 3 3 82 70 60 41 Pneumonia Enzootic 7 5 16 7 637 267 34 24 Pneumonia Paratyphoid 6 8 19 27 717 503 320 249 provide this treatment option to veterinarians and farmers. Drugs administered in the feed are presently given by topdressing, rather than being premixed in the feed by mills, which currently lack the required micromix technology. The proper mass dosing of drugs, an- thelmintics, trace minerals and vitamins is a concern that should be addressed jointly by VSB and the milling industry. 5.31 Certification of vaccination was commonly issued and was required to move pigs to different locations or to markets. Health examinations on the farm or at collection points were carried out to ensure that sick pigs are not being transferred. There are serious breaks in this system, for example, where pigs diagnosed with swine fever were nevertheless allowed to be slaughtered for home consump- tion. The chances of contamination and infection of other pigs by infected carcass parts is great. Similarly, vaccinating pigs at market entries will not prevent them from transmitting disease in view of the 7-10 day delay of onset of immunity. Disease Status in Poultry 5.32 The poultry health programs on large breeding and production farms and even on many of the smaller farms were better organized. than any of the health programs for other livestock species. This is mainly due to the ability to transfer complete poultry production systems from other countries where housing, feeding, stock management, and health have been combined into an integrated production system. Many of the health measures developed on the breeding farms spin off to benefit the smaller farms, where the day-old chicks are received along with instructions as to their proper care. The large farms - 74 - employ resident veterinarians who are in charge of vaccinations, hatchery and flock health, medication and other such activities. 5.33 Small farms are visited by clinic veterinary or technical staff and, in view of the low management standards, it is unclear to what extent they can achieve disease control. Many of the large breeding and production farms practiced complete isolation of the premises, where outside visitors were not allowed entry and where physical and disinfectant barriers were in place at entry points. However, this system breaks down almost completely where the smaller SH flocks are concerned, due mainly to the low standards of housing. 5.34 A Poultry Certification System has been imposed by the government which has set health standards for imported and locally produced breeding stock. Poultry production is expected to greatly increase around the large cities, where poultry units in the range of 200,000 birds each, are anticipated or have already been set up. Such systems will be integrated from breeding through marketing so that health services, meat inspection, meat processing quality control and hatchery health will be combined into an integrated system as well. 5.35 Preventive disease control in large flocks will become one of the main constraints facing the developing poultry industry in China. Diseases such as Newcastle, Marek's Disease and fowl cholera could have a devastating effect on commercial poultry production. These and other diseases can affect all segments of the poultry industry, through dissemination from hatcheries or breeding eggs and by horizon- tal disease transfer on farms. They can be controlled only by mandato- ry vaccination programs involving 100 percent of the bird population on a farm or village. Because of the newness of most large production and breeding units, severe disease problems are not yet common. Howe- ver, a gradual increase in mortality and morbidity on some large state-run layer farms is indicative of a disease buildup which is not being arrested by disinfection or vaccination. 5.36 The viscerotropic and neurotropic varieties of Newcastle Disease (VVND) virus are present in China. Fowl pox, Marek's disease, infectious bronchitis and lymphoid leucosis are also recognized, but no outbreaks of fowl pox or infectious bronchitis were reported on the farms visited. Marek's is becoming a serious problem; the vaccine previously employed in China is no longer considered effective. It was reported that, on one SH duck and chicken farm in Fuzhou, an outbreak of Marek's led to the ,death of 75 per cent of 4,000 layer pullets, which had not been vaccinated at the hatchery, at 70-80 days of age. 5.37 Salmonellosis is a problem in some areas of China. Although no data is available on the incidence, there is no reason not to believe that these infections are presently common in these flocks. Salmonellosis can cause morbidity and mortality up to 100% in young birds; in older birds there is very little mortality, but chronic infections with S. pullorum and S. gallinarum can lead to impairment of reproductive performance. There is an urgent need to institute a - 75 - Pullorum testing program on breeder farms so that S. pullorum infec- tions can be prevented from passing to day-old chicks by egg transfer. Periodic bloodtesting of all birds, the culling of positive reactors and, if necessary, of entire breeder flocks where infection rates are high will prevent long term production losses that are well recognized in countries with long standing Pullorum testing and control programs. The manufacture in China of Pullorum antigen and field test kits for on-farm use would assist in making the testing program quickly availa- ble to producers. 5.38 The incidence of coccidiosis in layer farms visited was found to be remarkably low as the raising of young pullets-to-lay in cages, rather than on deep litter, decreased exposure to this infection. Broilers did in fact have coccidiosis similar to those elsewhere and feeds medicated with coccidiostats were used as a preventive measure. Diseases of ducks as reported by duck producers included duck enteri- tis (plague) and duck cholera. Vaccines for duck plague and cholera were available and their use appeared to be widespread. Coccidiosis outbreaks in penned layer ducks raised on earthen floors has been reported. On smaller farms, this infection is still treated by tradi- tional herbal medicine protocols. 5.39 Most of the poultry vaccines now used in China are manufac- tured in regional laboratories, such as in Nanjing, Fuzhou, Beijing and Tianjin. Recommended vaccines and vaccination programs have been tabulated (Annex 6), although actual use of these protocols depends on disease incidence. Unfortunately, little research has been conducted to determine what types of poultry diseases are prevalent in China. Given the flow of day-old chicks from western countries, it is likely that most of the common poultry diseases recognized elsewhere do exist in China, but are currently not yet diagnosed or reported. As the poultry industry expands further, introduction of new diseases must be recognized quickly so that vaccines can be prepared. In particular, it would be prudent to begin production of an Avian Encephalitis vaccine for use only on breeding farms, rather than relying solely on immunization in country of origin, as imported chicks are now subjec- ted to neither quarantine nor A.E. testing. 5.40 Antibiotics in feed premix or water soluble form were availa- ble from veterinary centres and feedmills, but were in limited supply; many were produced in Shanghai. Several types of penicillin, tetra- cycline, chloramphemicol, potentiated sulfonamides, coccidiostats, furazolidones and electrolytes were also available. Label analysis of packaged drugs is often vague, causing difficulty in evaluating the types and contents to be administered. Label declaration of drug and vitamin formulations and quantities appeared to be seen as an exposure of trade secrets by the companies producing them, thereby creating a disservice to the user. 5.41 Hatchery Sanitation. Equipment and hatching procedures in hatcheries were generally acceptable and proper cleaning and fumiga- tion between batches of chicks was practiced. To prevent the build up of pathogens, the additional step of conducting spot check cultures - 76 - during a hatch would ensure that potential bacterial pathogens were monitored throughout the hatching period. The proximity of hatcheries to other poultry buildings on many farms was a cause for concern. Transfer of air borne pathogens from other poultry buildings into hatcheries or by other vector methods is a considerable risk where hatchery buildings are so located. Even with good sanitation prac- tices it would be difficult to avoid cross contamination. 5.42 Sanitation. Some form of footbath was used on most poultry farms visited. However, they were not always maintained properly. Neither disinfectant poured over a burlap sack nor the spreading of dry lime on the ground are effective sanitation methods. Footbaths should be used at the entrance of every poultry facility, with daily replacement of disinfectants, such as creosols and synthetic phenols. Many of the cage layer operations had improperly designed drinkers, where the water flow was unregulated and water dripped into the feed troughs placed directly below. During hot weather, wet feed becomes rancid and the resulting mold growth may lead to fungus infections. 5.43 Health Extension Training. Advisory services to small poul- try farms need to be improved by the training of additional personnel or by the re-training of existing ones. The preventive approach to disease control should be reinforced in curricula, and should include farm sanitation, vaccination programs and procedures, stress preven- tion, blood testing methods, simple diagnostic procedures, recognition of prevalent diseases, coccidiosis control, hatchery sanitation and poultry management and nutrition. Quarantine Stations and Abbatoirs 5.44 Quarantine Stations. China's quarantine stations are adminis- tered by the General Quarantine Station under MAAF (para. 5.01). Of the two quarantine stations visited, the Tanggu Animal and Plant Quarantine Station near Tianjin and the small inner city station at Fuzhou municipality are representative of the larger, regional sta- tions (Tanggu) and the local municipal stations (Fuzhou) that consti- tute China's animal quarantine service. Tanggu is one of four main stations located throughout China, each serving as a regional main entry and quarantine point for livestock and animal products. Three other such stations are located in Shanghai, Beijing and Guangzhou. Fuzhou belongs to one of 42 sub-stations across China. 5.45 The planned expansion of livestock populations has resulted in considerable increases in livestock and poultry imports, for which most quarantine stations now in operation are not prepared. Tanggu station is presently undergoing major upgrading, including new live- stock holding facilities to increase the capacity to over 1,400 cattle and over 2,900 pigs per year. Of the 45 staff members, 11 are involved with animal quarantine, of which five are graduate veterina- rians, several with additional overseas training. The laboratory personnel are undergoing training in advanced diagnostic procedures and new equipment is being procured. MAAF negotiates stringent health - 77 - requirements for imported stock with exporting countries and conside- rable retesting is undertaken during the 45-day quarantine period. The need to match the Chinese testing protocols with those of expor- ting countries has caused an upgrading of test protocols at Tanggu. These technical improvements are turning the Tanggu laboratory into the best equipped and staffed quarantine facility in China. 5.46 The construction underway will permit improved animal manage- ment during the quarantine period. Livestock passing through the old station were exposed to below standard feeding and management prac- tices resulting in deaths and health deterioration. Animal feeds have been the responsibility of the importing agency, which often resulted in the use of poor quality feeds. It is extremely important that livestock passing through the upgraded station not be subjected to adverse and stressful conditions in addition to those already under- gone during transportation. This will be determined by the station's management practices and attitudes. Qualified livestock attendants and prompt veterinary care should be available on a round-the-clock basis. Feed supplies should be procured by the station itself, rather than being left to parties independent of the station, and should consist of good quality fodders and grain concentrates in plentiful and timely supply. The station should charge a daily holding fee which includes animal care, feeds and other overheads, to be paid in advance, if necessary, by the importing party. The health and condi- tion of the quarantined stock should be assigned highest priority so that the large investments made are protected. 5.47 The Fuzhou quarantine station has six graduate veterinarians and three assistants on staff who deal with animal quarantine and the quality control of imported or exported animals and animal products. Its diagnostic laboratories are sparsely equipped and only basic bacteriological and serological testing can be undertaken. The sta- tion relies on the expertise and instrumentation of the larger region- al laboratories for any testing beyond its routine procedures and coordinates such activities through the Central Diagnostic Institute in Beijing. 5.48 The station's animal facility can hold 100 head of cattle, but none have passed through Fuzhou in recent years. The station's main activities now include the inspection of day-old chicks arriving by air, after which they are transported to designated poultry farms for quarantine and further testing. Half of the staff's time is involved with the export of live pigeons and aquarium fish. The station is not equipped, nor is it intended to be, for handling major import volumes or for advanced testing. Laboratory staff lacks the experience or equipment for the isolation and identification of pathogens from animal tissues. If Fuzhou or similar stations were selected to play a larger role in import and quarantine, a major upgrading program would be required. Even if this is not intended, the laboratory's diagnostic capabilities need improvement to meet the demands of local expansion in the livestock industry. 5.49 Meat Inspection. Abattoir inspections are the responsibility - 78 - of the Meat Inspection Division within the Food Companies which manage the abattoirs, although meat entering international trade falls under the separate authority of the National Commodity Inspection Bureau under MOFERT. Meat inspection procedures for pigs were observed at major abattoirs where qualified veterinarians and technicians were at work. At rural slaughterhouses, technicians with short periods of training were more often employed. An ante mortem inspection of live pigs usually took place to eliminate clinically sick and injured pigs. In larger abattoirs, the opened carcasses, liver, intestines, lungs and heart were tagged for identification. This procedure was not usually followed in rural abattoirs, where the smaller numbers of pigs slaughtered permitted matching by sight. Submandibular lymph nodes were isolated and examined for signs of infection but at no time were mesenteric lymph nodes seen to be cut for evidence of TB, a common condition in countries where pigs and poultry may share common space. Lungs and livers were examined for lesions by surface cutting to expose internal tissues. Examination of diaphragm muscle for evidence of trichinosis was carried out in large city abattoirs but not in rural areas. The adequate examination of organs which are discarded for rendering in western abattoirs, but habitually favoured by Chinese consumers, takes on a special importance in China. 5.50 In spite of the presence of an inspection staff, condemnation rates were generally too low to reflect a meaningful inspection sys- tem. For example, in the China Tianjin Meat Products Joint Processing Factory, where 600,000-800,000 pigs are examined each year and where processed sausages and other products are being manufactured, inciden- ces of trichinosis were quoted at 0.001 per cent, or equivalent to 6-9 pigs per year. Erysipelas, swine fever and pneumonia incidences were quoted at 0.01 per cent, or 60-80 pigs per year. Tuberculosis was never reported, and abattoir records were not available for further study. Pigs arrived with broken limbs at an average of 15-20 per day, an incidence rate of one in 113. 5.51 If these condemnation rates were genuine, it must be concluded that the inspection system was nonsensitive to the point of being nonfunctional. The effect of a nonfunctional inspection system on public health safety cannot be underestimated, as swine erysipelas, trichinosis, salmonellosis, certain streptococcal infections and cys- ticercosis can all cause human disease. If the inspection procedures now in place are to act as barriers to the transfer of disease from slaughtered meat, they are in need of upgrading through better inspec- tor training and vigilance by abbatoir administrators and government departments involved in abattoir compliance with set standards. 5.52 Smaller abattoirs in villages employ trained technicians and lay personnel to carry out meat inspection. At the slaughterhouse in Dongxiang City, 15,000 pigs are slaughtered per year, with another 20,000 pigs being slaughtered in smaller abattoirs belonging to the 6 surrounding townships. The remaining 56% of total pigs slaughtered are handled on the farm by SH families. The implications of on-farm slaughtering to the adequate protection of the consumer are consider- able. Adequate meat inspection services must be available to such SH - 79 - slaughter points in view of the large proportion of meat derived from these sources. If the quality of meat inspection in larger abattoirs is any guideline, there should be concern about the efficacy of such services in the rural areas, as well. 5.53 The hygienic standards of abattoirs visited ranged from the medium-sized Gushan abattoir in Fuzhou, which was recently upgraded, clean, well-organized and well-maintained, to one belonging to the Fuzhou Freezing Factory, in which no slaughtering had taken place for several months, and equipment was broken, dirty and neglected. In all abattoirs visited, hygiene practices left much to be desired. Largely because of a lack of high-pressure hot water supplies and of disinfec- tant stations for cleaning slaughter equipment, pig carcasses were not clean after scalding and dehairing. Meat contact surfaces, sometimes made of wood, were often deteriorated. Floor drainage, ventilation, and lighting were inadequate and there was no standard abattoir flow design. Meat inspector stations were not clearly defined and inspec- tors often walked between stations, inspecting up and down the slaugh- ter line, so that carcass components could be missed or not receive adequate attention. In none of the larger abattoirs were carcasses railed onto a section of inspection lines, where more time could be spent on examination. Such holding lines, in contrast to the straight, non-stop line normally present, should be installed in existing and new systems. 5.54 Considerable upgrading of abattoir installations and person- nel training will be required to raise the general standard of pig slaughtering. A more objective approach to condemnations should be adopted, which should not be influenced by the financial loss due to condemnation. Record keeping on disease conditions should be insti- tuted or improved. Veterinary services, veterinary facilities and other investigational units should develop better liaisons with abat- toirs and use them as a source of information on disease incidence by using their records, tissue samples and availability for blood samp- ling in disease incidence surveys. Abattoir personnel, in turn, should keep better, more detailed records on condemnation and other conditions, and should adopt a different attitude towards their res- ponsibilities in the maintenance of public health. Meat inspection and hygiene training courses for workers and inspectors should be instituted on an in-plant, compulsory basis with inputs by veterinary colleges and qualified meat inspection personnel. - 80 - 6. PRODUCTION ALTERNATIVES Pigs 6.01 Most of the pigs in China are produced at present under back yard systems in the areas where feed is available. In China as a whole the production of 180 million tons of rice, much of which is consumed locally, means that about 18 million tons of good quality rice bran is available for livestock feeding, much of it at the indi- vidual farm level. In addition wheat bran, much of which is collected at the urban level, amounts to about 12 million tons. Rice bran availability is equivalent to about 800 kg per sow and could make up 15-20% of the total diet for pigs on average. However in the major rice growing and pig producing provinces of the Southeast, rice bran amounts to at least one-third of the diet for pigs (Annex 3). 6.02 Because most of the pigs in China are presently fed on locally produced feeds, protein in the diet on individual farms will certainly be well below optimum. This is particularly the case in the south, where the majority of pigs are kept but protein supplies are low. One result of this is the relatively low weight gain in the Southeast where pigs achieve an average carcass weight of 66 kilos (live weight of about 90 kilos) at 15 months of age (Annex 3). This can be compared with the Northeast, where more protein is locally available and carcass weight averages 7 kilos heavier (live weight about 10 kilos heavier) for pigs of the same age. 6.03 There appear to be three main production options with pigs: (a) to produce pigs on the existing system with farmers fattening one or two pigs, relying to a significant extent on waste products and green feeds; (b) encourage individual farmers to become SHs producing perhaps 50-100 pigs per year relying largely on bought-in feeds; or (c) large scale production with modern units of several hundred sows producing several thousand pigs per year. 6.04 The average cost of production under the first system is almost certainly lowest, at least if the resources employed are eva- luated at opportunity costs. However, given the overall relationships between pig numbers, available brans (less than 1 ton per sow) and other available waste products, it is likely that the free resources for pig production are already nearly fully utilized in China. Thus any expansion of pig production from backyard systems will require the use of feeds which have a real economic cost (feed grains, oilseed meals, etc.). While it is possible to feed one or perhaps two fatten- ing pigs largely on byproducts, an individual household does not have sufficient waste and byproducts to feed one sow and its progeny to slaughter weight (this would require five to eight tons of dry feed equivalent per year - see Annex 3, Table 10). From an economic stand- - 81 - point, however, the opportunity cost of labour for the one to two pig backyard units is certainly close to zero. 6.05 With the SH maintaining about five sows, the system of production has to be geared to using feeds purchased or grown for livestock use (at least for 80-90% of the diet). In addition economic costs are almost certainly incurred in the provision of housing for pigs. For example, all SHs visited on the mission were using special- ized pig houses costing about Y800 ($250) per sow, which if depre- ciated over 15 years at a real interest rate of 5% would amount to a cost per pig produced of about Y6 (US$1.88). Furthermore, running an SH is a full-time job, typically for two people and the opportunity cost of their labour is almost certainly well above zero. On an SH producing 60 pigs/year, and employing the equivalent of one person full-time, the labour "cost" would be about Y14 (US$4.38) per pig sold, or 7-10% of the pig's value at sale. 6.06 The key economic question which emerges is whether the better levels of technical performance which probably can be achieved in SHs will outweigh the free resource advantages of the backyard producer. The answer to this on average is not clear, although undoubtedly some efficient SHs will be able to produce pork significantly more cheaply than the average backyard producer and, similarly, there will be many backyard producers who can produce more cheaply than the average SH. It is logical, therefore, that both production systems will and should continue. The present shortage of high quality feeds, and the feed- for-product system of rationing employed by the state, virtually guarantees this co-existence. 6.07 This suggests a second question - should government policies and programs differentiate between backyard and SH producers, and what forms could such differentiation take? Several justifications for differential treatment have been noted above. First, backyard and SH producers to some extent serve different markets: many (but not all) SHs are involved mainly in breeding and, secondarily, in fattening improved breeds for urban or export markets which demand lean pork, whereas most (but not all) backyard production serves a rural market which either accepts or requires (for its value as a source of edible oil, or energy) a fattier product (para. 2.33). Second, at present, backyard production in most areas is based on the native black pig, whereas SH production is increasingly based on purebred or hybrid improved pigs. These breeds have specialized genetic abilities: the native black is an efficient converter of low quality feed into fatty meat, whereas the improved pig requires larger daily rations of quali- ty feed to obtain efficient conversion rates (paras. 4.19-21). 6.08 Aside from these justifications, there is an economic ra- tionale for differential price treatment of backyard and SH producers. The backyard producer has been willing to produce one or two pigs per year despite low pig-to-feedgrain price ratios because of his access to "free" resources (family byproducts and waste, labor, and shared space); an SH cannot survive at such price relationships. However, because the "free" resources are fully utilized, the marginal costs of - 82 - expanded backyard production are likely to be comparable to the ave- rage costs of SH production. If a single set of prices (for pigs and feed) were used to provide incentives for expanded social production, then those prices would have to be much more favorable to producers than in the past, and would bring about a large social transfer from consumers to backyard producers which reflected a rent on their "free" resources. Alternatively, the same rent could be retained for consu- mers by a pricing or subsidy policy which differentiated between average and marginal social costs. 6.09 Current government practice tends to differentiate along the lines indicated by these justifications, although not so much for the above reasons as to establish stable sources of commercial supply of higher quality meat. Producers for export and suburban producers, who are more likely to be SHs, typically receive higher than average prices; they also obtain privileged access to breeding farms offering improved stock or Al services, feed mills offering more balanced feeds, low-interest credit from the RCCs or ABC, and list-price feed entitlements in return for pig sales. Most of these advantages do not, however, apply to producers whose product is distributed through the urban Agricultural Trade Markets. Indeed, the differential treat- ment policy is most problematic because it is difficult to implement while simultaneously enlarging the role of the market in driving the sector. Also, uncalibrated favorable treatment of SHs has probably contributed to recent oversupply problems, and the SHs are less likely to survive the bottom half of the hog cycle. 6.10 By ignoring the backyard producer, a differential treat- ment policy may overlook his enormous potential - after all, more than 90% of the pigs in China are produced in this way; therefore small, but general improvements can have very large impacts on total sectoral efficiency. Strategies, therefore, need to be adopted to support and improve the efficiency of backyard production. These would involve two main elements - improving the genetic quality of pigs and impro- ving feed supply. 6.11 There is already a movement in China away from the local black pig and towards cross-breeds or hybrids involving leaner white pigs, e.g. Landrace and Yorkshire, mainly in areas oriented toward urban or export markets. As the appropriate price incentives become available, this breed change in China can be expected to occur more or less naturally, even if improved stock only "leaks" from the fa- vored producers to unsupported backyard producers. However, genetic change should not outpace improvements in feed supply. In particular, appropriate concentrate balancers need to be made available to back- yard producers. This will require that the county feed mills formu- late rations specifically designed to balance up the protein level of locally available raw materials, and also supply adequate minerals, vitamins, etc. 6.12 There is a need to come to grips with the economic tradeoffs among policies supporting backyard versus SH producers. As the same scarce quantities of protein sources and supplements could be used as - 83 - (a) a balancer for backyard producers raising either local or improved pigs, or (b) part of a compound feed for SH producers, which use is likely to add most to meat production (counting fat as of higher value in rural than in urban areas)? In other words, is it more efficient to concentrate supplies of protein sources and supplements to achieve efficient conversion for SH producers of improved pigs, ignoring existing imbalances in rations of backyard producers; or to use those supplies to improve the diet of the backyard pig? A controlled expe- riment, using existing backyard and SH producers, is recommended which would provide the data required to answer these questions. The design frame should allow for comparison of five production approaches: Backyard Producer SH Producer Local pig, balancer Improved pig, Improved pig, balancer compound feed/balancer Local pig, no balancer Improved pig, no balancer 6.13 Large scale state- or collective-run pig farms have the disadvantage that investment would be largely derived from the govern- ment budget, whereas SHs can mobilize private savings. At this stage of development in China there appears to be no real advantage to the former, as long as medium-sized private farms are politically accepta- ble. Large pig units are only justified when uniform management of large numbers of animals is required, as in the breeding of pure line livestock in nucleus herds and the production of boars for crossing. First-cross breeding sows could just as well be produced at the SH level under supervision. 6.14 The state- or collective-run units therefore should con- centrate on breeding, and drop sidelines such as fattening cull gilts and males from female producing lines, so that (a) maximum use can be made of the farms' facilities for their intended purpose, i.e. breed- ing, and (b) management effort is not diluted. At present, breeding farms often seem to use profits on fattening to subsidize losses or low profits on breeding, or are forced to fatten their own piglets for want of an adequate market. A solution may require either (a) in- creased prices of breeding animals relative to porkers, or (b) subsi- dies to breeding farms. This in turn requires that premium prices be offered for lean market pigs, encouraging increased demand for im- proved weaners; else the breeding farms would price themselves out of the market. Dairy Cattle 6.15 Dairying is relatively new to China and, outside of remote nomadic areas, there is no history of small farmer dairy production. In consequence, commercial milk production in China is based on exotic Holstein-Friesian cattle, maintained either by SHs or State farms. Backyard milk production is virtually non-existent. The systems of production of dairy cattle both on SHs and on large-scale farms are - 84 - relatively similar. Cows are stall-fed rather than being grazed during most of the year. They are fed cut grasses in the summer but a whole range of products including vegetable feed crops, silage, and concentrates in the winter and are hand milked, normally three times per day. AI is widely used in both cases. 6.16 State farms have advantages over SHs for a number of rea- sons. First, production of ensiled winter feed is far easier on large units than on small units. Second, AI is potentially easier to manage with numbers of cows large enough to justify on-the-spot employment of an AI technician (but see para. 4.12-13). A third point is that dairy cows are expensive animals in China, one cow being worth about Y3,000 (US$1,000), or about 1.5 times the annual net income of an average farm family. Therefore, for an individual farmer to own the 5-10 cows which are required to fully utilize family labour would mean either substantial concentration of wealth, or a precarious degree of indeb- tedness, neither of which may be desirable in the Chinese socio- economic context. 6.17 A number of potential management improvements can be made without radical changes in the existing context. First, many large- scale state dairy farms were located relatively close to urban centers on land which therefore has a high opportunity cost. In this situa- tion, it would be most efficient to use the land resource only for mature cows, which need to be near the market because of milk trans- port requirements. Replacement heifers, calves, etc. which now occupy about half the land and housing should be removed from the dairy herd soon after birth and reared in areas away from the towns where abun- dant grazing and other relatively cheap livestock feeds exist. This would allow scarce suburban feed and forage resources to be used for a larger number of milk-producing animals, with no increase in fixed investment within the urban areas. 6.18 Second, on some farms the investment in livestock does not match the investment in buildings. For example, at one farm near Tianjin, facilities had been constructed to accommodate 500 head of livestock, yet only 324 were actually on site. As building investment costs on dairy farms are normally about Y2,000 (US$625) per cow space, the additional overhead cost per year of only half filling the buil- dings with dairy cows would amount to about Y200 (US$62) per cow, or YO.04 (US$0.013) per litre of milk. Poultry 6.19 Egg production in China comes from both ducks and hens. Duck egg production (prevalent mainly in the south) is in part based on scavenging systems, while hen egg production is increasingly from large-scale battery units, using high quality genetic material and compound feeds. Overall, the potential for expansion of backyard production, however efficient, is limited, although in some areas it should be possible to increase the population of ducks on rice paddies and tidal flats. - 85 - 6.20 Laying hens managed mainly by SHs, which have recently become the big growth area, come from imported parent stock from leading poultry breeders worldwide. The breeding farms, which are normally state enterprises, sell day-old chicks to commercial egg producers. The management system followed on the larger SH units is to keep hens in batteries, normally four birds to a cage within spe- cially constructed sheds. Reported production on the farms visited was good with a feed conversion rate for hens during the laying phase of 2.8:1 over a 52 week period during which about 240 eggs per bird of 55-60 grammes each were produced. Despite the variation in feed quality and the probably below-optimal energy level in the rations, the SH performance appears to have been extremely good. However, this may partly be because intensive poultry is a new industry and as a result poultry farms are still relatively disease free. 6.21 The performance of modern hens in China is enormously supe- rior to that of backyard birds. Given this, any major expansion in chicken egg production should be through layers of high genetic poten- tial kept in batteries and fed top quality feed. The backyard option is much less valid with laying hens than with pigs and is probably not worth pursuing. The only major question for hens is whether to con- centrate this in the hands of SHs running from one to a few thousand hens, or to promote collective or state farms maintaining 50,000- 150,000 layers. 6.22 There doesn't seem to be an unambiguous answer to this ques- tion. For non-automated systems, there are no obvious economies or diseconomies of scale between, e.g., 1,000 and 150,000 hen units. Among the handful of layer farms visited, there were examples at either end of the range where investment costs per bird housed, unit production costs, and layer productivity were of comparable magni- tudes. At a scale of 1,000-5,000 hens plus reserves, owner equity is unlikely to exceed 10-20% of total investment, so that the ability of SHs to tap private savings is not compelling. The present scarcity of technical and management expertise in poultry raising, and the better enforcement of the Closed Flock system observed on the larger collec- tive and state farms (para. 5.33), favor a larger scale approach. Offsetting this, past practices for state farm labor and management compensation (low salaries but complete job security) are unsuitable to an industry where careless practices can lead to tremendous losses. Salaries on medium- to large-scale poultry farms are such a small element (presently under 5% of variable costs) that these farms should offer high wages and elaborate incentive schemes instead of job secu- rity, in order to attract the most responsible workers and managers. 6.23 For ducks, however, the situation is different. When enclosed and fully fed, laying ducks generally are less efficient in terms of feed conversion than hens, although their aggregate annual production can be greater than that of hens. Scavenging laying ducks are also highly productive and the same breed of duck can be used on a partially scavenging system as on a fully fed system. Consequently, the logical policy would appear to be to develop battery laying hens - 86 - in parallel with promotion of semi-scavenging duck units in those areas where feed resources are available. 6.24 Until recently poultry meat production was simply a by- product of egg production and relatively minor in China. Now, howe- ver, both broiler chickens and intensive meat ducks have been intro- duced, reflecting higher income elasticities of demand for poultry meat than for eggs. Both are very efficient in converting grain into meat compared with pigs. The level of technical performance with both broilers and ducks achieved in China is relatively good, perhaps again due in part to the newness of the industry and lack of disease pro- blems found elsewhere. Since both broiler chickens and ducks rely on high quality genetic material and need sophisticated feed, it is sensible to develop them in a specialized environment. Whether they should be produced on SHs or larger scale collective or state farms is subject to the same ambiguities as for layers (para. 6.22). Locations for Investment 6.25 From a social viewpoint, there are three major factors to evaluate in determining the locations for investments in expansion of livestock supply to the cities: (a) the comparative transport costs for livestock or livestock products versus the feed required to pro- duce them locally; (b) the relative opportunity costs of land in or near the cities compared to more distant areas; and (c) any other differences in the comparative marginal economic costs of production in these alternative areas. 6.26 Comparisons of transport costs must be location specific, so comparison will be made between production in the suburbs of Fuzhou Municipality, versus import from north China, e.g., Shandong Province. Transport costs of frozen pig carcasses by refrigerated ship from Shandong are about Y150 per ton, whereas sea or rail transport of grain or soymeal imported from the Northeast would cost Y30-40 per ton (at negotiated rather than list prices). As a ton of frozen carcass is equivalent to a liveweight of 1.4 tons, and a reasonable feed conversion factor would be between 4:1 and 5:1, the transport cost of the feed required to produce the same amount of pork locally would be Y200-250. However, these costs of transport for pork or feedgrain represent only 6% or 8-10%, respectively, of carcass price CIF Fuzhou. In short, long-haul relative transport costs slightly favor non-local production of pigs for the Fuzhou market; but this at present is more than offset by carcass shrinkage and the 7% lower market price of frozen pork. For poultry and eggs, because the feed conversion rate is under 3.0 and because of even greater losses due to shrinkage or brokens, a similar comparison would unambiguously favor local produc- tion. 6.27 The evaluation of the comparative impact of differing oppor- tunity costs of land in China is trickier, as there is no well- established land market, farms have either gratis use of land or have paid widely varying prices for land use rights, and the opportunity - 87 - Table 6.1: Opportunity Cost of Land Occupied as Percentage of Livestock Sales Value Land Percent of Total Sales Value/b Type of Occupancy Low-Value High-Value Farm Rate/a Rural Suburban Suburban Pig Rural 2.3% 11.3% 50.3% Suburban 1.8 8.9 39.9 Poultry Rural 0.9 4.3 19.1 Suburban 0.6 3.1 13.9 Dairy Suburban 0.8 3.9 17.2 /a Differed between rural (Jiangxi) and suburban (Tianjin and Fuzhou) farms, with rural farms using more land per animal for farm buildings, exercise yards, etc. than suburban farms. Ab Assumes total annual sales value (including byproducts) per animal housed (including reserves) of Y170 for pigs, Y23 for layers, and Y1,800 for dairy cattle; and annual land "rental" values per ha of Y900 for rural farmland, Y4,500 for low-value suburban, and Y20,000 for high-value suburban land. Land "rental" values may be converted to/from land "prices" by assuming a real discount rate of 5%. cost of land use is not reflected in enterprise accounting. Neverthe- less, it appears that subleasing of productive rural farmland, or voluntary return by farmers of land use rights to the collective, often involves compensation worth about Y900 per ha per year. At a real discount rate of about five percent, this would place rural land value at about Y13,600 per ha. Reported unit values of land occupied by suburban farms (ignoring some extreme cases) typically fall in the range Y70,000-300,000, or 5-22 times the rural land values. Taking these figures as reasonable representations of the opportunity costs of land, and considering only the average areas of land occupied by farm buildings and exercise yards on farms visited by the mission, the percentages of total sales value which the annualized sales value (or "rental" value) of land represent are shown in Table 6.1. The exer- cise suggests that livestock farms located in the distant rural areas can afford to ignore opportunity costs of land, but this should be a significant issue in considering suburban investments in livestock farms, more so than comparative transport costs. Because average pig farms occupy three times as much land per yuan of product as poultry or dairy farms, rural, as opposed to suburban production of pigs has a significant economic advantage, probably worth between 7-38% of total sales value. This could be cut to 3-4%, however, if the most land- saving designs and distant suburban land were used (farmland lying on the fringes of municipal areas may have low opportunity costs). - 88 - 6.28 The remaining differences in the costs of livestock pro- duction between rural and suburban areas are largely those of feed (70-90% of variable costs) and labor (2-15%). It may be supposed that, in the rural areas the cost of feed is cheaper due to possible use of waste materials. This is true on average, but not at the margin, as use is already made of virtually all waste materials avai- lable; and the relevant comparison is between marginal costs under alternative rural and suburban production systems. Marginal economic costs of feedgrain production are also lower in the distant rural areas (particularily if the opportunity cost of land is considered), but then the feed could be transported to the suburban production units instead of being used for local livestock production, and the conclusions of para. 6.26 can be applied. 6.29 In pig raising, however, a more subtle comparison is called for. Poor conversion rates are now being obtained in the rural areas where local feeds lack protein sources and micronutrients; and relatively small volumes of supplements could greatly improve these conversion rates, so that more pigs could be raised in less time from the same amount of energy feed. If this option is compared with efficient suburban production using entirely imported feeds, the po- tential improvements in feed efficiency in rural areas could greatly outweigh any comparisons of relative transport costs. They would require, however, investments in feed milling, rural breed improve- ment, and extension services. As the improved conversion rates would be reflected in a faster production cycle (as little as six months compared to the present fifteen), farm-level investment in pig housing would not need to increase. 6.30 Although the economics favor increased reliance on pig production in the distant rural areas, there are some practical prob- lems which currently strongly favor local production. Refrigerated rail cars are scarce and difficult to access, particularily at peak periods of production. In fact, access to rail transport of any sort requires advance planning and bureaucratic approvals by provincial Planning Committees and the Transport Bureau; and therefore private trade is largely confined to costly road transport, and official trade cannot readily take advantage of unanticipated trade opportunities. Long lead times in the planning of transport arrangements prevent expanded trade in live animals, contribute to a need for high-capacity cold stores, and result in losses of quality due to extended storage. For example, frozen pig carcasses from Shandong spend an average of three months in storage before shipment to Fuzhou; and then must be sold at a discount relative to fresh or chilled product. 6.31 The location of new investment in dairying needs to be geared to (a) access to fresh milk markets and (b) availability of high quality fodders both summer and winter. China's expanses of natural pasture, or land which can be economically developed as pas- ture, are distant from the major municipalities (with the notable exception of Heilongjiang Province, where milk production exceeds urban requirements), and at this stage of development long-distance - 89 - transport of fresh milk seems impractical. Therefore dairying located nearer the cities, based on intensive cut-and-carry feeding systems with substantial supplements of concentrate, is inescapable. Within this constraint, supply of high-quality summer fodder appears rela- tively ample in the north, where alfalfa and green-cut corn do well and ensilaging is commonly practiced; but winter fodder is scarce. At present dairy farms around Beijing and Tianjin import hay from the Northeast at a CIF price nearly equal to the local price of corn. The quality of fodder is a more serious problem in the subtropical south, where suitable crop varieties and experience with ensilaging are both lacking. 6.32 Rapid expansion of dairying will require massive new investment. Because of the very slow reproductive cycle of cattle compared with either pigs or poultry and the small existing base of dairy animals, some part of this will have to be in the import of breeding females. China does not share with, e.g., India a tradition of village subsistence dairying, and its existing stock of draft cattle and buffaloes produce little milk beyond calf requirements. However, if aggregate expansion proposals, which require a herd of four or more million milk cows by the year 2000, are to be approached, the present levels of imports, i.e., a few thousand per year, repre- sents a drop in the bucket. By far the major emphasis (and the most economic approach) will have to be placed on turning a fraction of the large base of native yellow cattle and buffalo into single or dual purpose dairy animals through a cross-breeding program. Assuming the improved animals remain in the hands of individual farm households, future milk production will increasingly depend on the development of a effective program of rural milk collection. Viability of Specialized Households 6.33 In the last few years, the use of Specialized Households for significant expansion of pig/poultry/dairy production has received considerable support from the government and financial institutions. Although SHs are privately-owned farms, they are locked into the state distribution system because their financial viability requires access to subsidized feed and credit which accrue only to suppliers selling to the state distribution system. The development of SHs is socially desirable in that the fruits of specialization often include better performance parameters and realization of some economies of scale. There are, however, several important issues concerning SH production of livestock in China which need to be recognized. 6.34 First, SHs have been artificially promoted rather than having emerged naturally over time. Therefore not all SHs have the technical expertise or management acumen to succeed outside the frame- work of subsidies. Also, because of the initial reliance on external technical assistance, technical designs, e.g., for buildings and other facilities, have often been transferred from the outside rather than evolving from within the subsector. In this process, there has been less than ideal attention to minimizing investment costs per animal, - 90 - or to suiting designs to the production structure of the enterprise. An example is in the design of pig buildings, where improved designs could greatly reduce investment per animal, as well as mortality losses. 6.35 The present financing system for SHs in livestock, as well as other types of rural production, is based on short- to medium-term (3 - 5 year) lending. This form of financing is only viable if the price relationships are extremely favorable to livestock production, which cannot be expected if supply continues to develop at a rapid rate or if existing subsidies are reduced. Under more normal circum- stances, in situations where major investments in buildings are required, ten-year loans would be required for dairy farms and a minimum of seven years for pigs and poultry. Moreover, the typical initial debt:equity ratio of SHs at present is very high, often 5 or 10:1, giving the enterprises and their sources of finance little protection against adverse trading conditions. Yet both pigs and poultry tend to be "boom and bust" industries, with highly cyclical trends in prices and profits. If loan maturities are not considerably lengthened and required'equity proportions increased, it is likely that a high rate of defaults and bankruptcies will be experienced in the future. Financial Viability of Livestock Investments 6.36 Annex 7 indicates the likely average financial performance for different types of (mainly SH) livestock investment in China, set out in the form of annual budgets, which are summarized in Table 6.2. The size of all operations is scaled to the level which would fully employ one worker for 365 days per year. The actual results achieved on any individual farm will be highly sensitive to the level of technical performance achieved, and therefore any financing policy should have considerable safety margins built in. 6.37 Poultry. With the exception of production of pullets for sale as layers, all of the SH poultry operations (including ducks) show reasonably attractive financial returns, after costing the ow- ner's labor. With poultry the main cost is feed, usually purchased at negotiated prices. The investment required varies widely by type of operation from a low of about Y4,800 for a semi-scavenging duck system (Y6,800 including investment in ducklings) to Y46,000 for a 1,000 bird laying hen unit (Y55,000 including the hens). The rate of return is higher for scavenging ducks than for other kinds of layer, but the latter have higher cash flow because of larger scale. Laying enter- prises with 30% owner-equity and 70% debt at 7% interest rates could reasonably repay the latter in 11 years for intensive systems and nine years for semi-intensive ducks. Meat producing poultry have rates of return of around 40% for broilers and intensive ducks. At present both broilers and meat ducks could be financed with 5 year loans, but as more producers get into the business, poultry meat prices will come down and margins will diminish. Table 0.R Summery of FinanciaL ResuLt. of Various Livestock Investments .............Reasonable Loan Period............ Minimum Loan (a) Interest 7%l Farmer (b) Interest 7%, Farmer Cash Cash FLow Period Contribution 80S; Contribution 30%; Investment FLow per YIOOO 71 Interest 501 of Cash FLow 75% of Cash FLow (Yuan) (Yuan) (Yuan] 100 Borrowed for debt service for debt service Growers 81,000 4,732 153 9 Years Not PossibLe 9 Years Laying Hens 54,520 10,544 198 7 Years II Years Laying Ducks 88,500 7,402 192 7 Years 11 Years - Semi Scavenging Ducks 8,000 1,460 215 6 Years 9 Years - BroiLers 18,000 5,288 406 8 Years 5 Years - Intensive Neat Ducks 15,400 8,775 440 8 Years 4 Years - Fattening Pigs (Nag.) 7,000 sme 180 12 Years Not PossibLe 11 Years Breading/Fattening (Nag.) 11,400 2,879 209 6 Years 10 Years Brading/Fattening (List) 11,400 4,214 870 4 Years 5 Years Dairy Cesl SpeciLaLised HousehoLds 28,000 3,584 187 Il Years Not PossibLe 10 Years State Farms 1,795,000 285,050 148 10 Years Not PossibLe 9 Years - 92 - 6.37 Pigs. The financial budgets compare different systems of pig production, including pig fattening using only purchased compound feed, and combined breeding/fattening using also home produced bran, coarse feed and green feed, with compound feed procured at either negotiated or list prices. With compound feed only and a conversion rate during the fattening cycle of 3.8:1, pig fattening has a rate of return of only 9%. This indicates that the present relationships between the negotiated price of concentrate feed and the free market price for pigs are unattractive for pig producers. Mixed breed- ing/fattening enterprises making use of home-grown bran, coarse feed and green feed typically achieve substantially better returns than fatteners using concentrate only; and those who have access to feed at list price (normally 1.4 kg concentrate feed per kg live weight sold) are able to make good profits from pig production in the present price situation. The return in the latter case is estimated at 35% and would allow the operation to be funded over 5 years. 6.39 Dairy Cattle. The financial budgets compare returns to dairying for SHs and state farms. Because of the high cost of dairy cows, the financial rates of return on these are relatively low at 10.9% and 11.9% respectively, and periods of 9-11 years would be required for repayment. Dairying is relatively unprofitable in China for three main reasons: (a) the capital costs of good quality cattle are relatively high (close to import parity prices); (b) the price of milk is relatively low; and (c) technical performance is only moderate compared with leading dairy producing countries. 6.40 For example, milk prices are 25% higher in the U.K. than in China, yet the price of a British dairy cow is substantially lower (about 30%). These factors more than offset the advantage of China's relatively low negotiated concentrate feed price, which is roughly half of the milk price (the two prices are about equal in the U.K.). However, the estimated budgets reflect relatively modest levels of performance and, should yields of 5,500 liters per cow be achieved, as happens on some farms (rather than 4,500 liters as assumed), the cash flow per Y1,000 invested would rise from its estimated level of Y148 to Y241, and the unit could financed over 7 years with only 50% of cash flow being used for debt service. - 93 - 7. PRODUCT PROCESSING The Dairy Industr 7.01 Overall Supply/Demand Situation. For China as a whole, dairy products are a relatively new, though rapidly growing, dietary component. The market is composed of two main segments: a nutrition- oriented segment geared to consumption of fresh or reconstituted milk mainly by young children and the sick or elderly; and a luxury market for milk-based drinks, ice cream, and yoghurt. There is also a limited market for butter and cheese, mainly for use by bakeries and the tourist trade. At present, the market demand for milk and milk products is considerably in excess of supply, and the nutrition- oriented supply is strictly rationed. 7.02 Most existing dairies are state-owned enterprises established within the last ten years. They typically have the capacity to handle 20,000 to 35,000 kg of raw milk per eight hour shift, and actual throughput rarely exceeds the nominal capacity. The focus of present dairy processing development policy is to increase scale of processing enterprises and utilization rates as well as total capacity in accordance with the rate of increase in the supply of raw milk. The new dairy plant now under construction in Fuzhou, for example, is designed for an annual processing capacity of 20,000 tons or about 50,000 kg per shift. Plans are to operate the plant two shifts per day--i.e., to process 30,000 to 40,000 tons of milk per year--once milk production in the area increases as expected. 7.03 Supplies of raw milk to the dairy plants vary seasonally with a peak period in the spring, tapering off towards summer and autumn and reaching a low point from September through winter. Though variations exist from one plant to another, deliveries during the slack season are approximately two-thirds of those during the spring months. One way to deal with this imbalance would be to base production levels on peak summer supplies and compensate for winter shortfalls by increasing the percentage of recombined ingredients in the product. An alternative strategy, in cases where recombination is not feasible, would be to gear liquid milk sales to winter production levels and to deal with summer surpluses through the production and distribution of summer seasonal products, such as ice cream, yoghurt, or flavored milk. Either approach is preferable to the present practice of using production of whole milk powder as the main balancing item. 7.04 Up to 75% of the raw milk supply to dairy plants comes at present from state farms though in most parts of the country measures are being taken--subsidized sale of dairy cows and provision of low- cost AI, veterinary and advisory services--to encourage increased production by dairy SHs operating within the collective sector. On the state farms, raw milk is usually chilled before direct delivery to the dairy plants. SHs are rarely equipped with facilities for cooling the - 94 - milk prior to shipment. Consequently, where SH numbers have reached significant levels, milk collection and cooling centers are currently being established as intermediaries between farm and dairy plant. In Fuzhou, for example, five centers are now in operation with two under construction and a total of ten planned by the end of 1986. At full development, each center is expected to receive milk from about 500 households. Handling capacities range from 6,000 to 10,000 kg per day, sufficient to accommodate twice present deliveries. In the case of the centers which are already operating, dairy farmers within a radius of 2-3 km bring in their milk by bicycle two or three times per day. Samples are analyzed on the premises for fat content and acidi- ty; an additional sample is sent to a nearby laboratory for bacterial analysis. 7.05 Dairy Plant Operations. Quality control is the major shortco- ming of the dairies visited. For example, while deliveries of raw milk are sometimes tested for density, acidity, dry matter, fat content and bacteria content, these analyses are not uniformly performed and very little milk is rejected. Cell count, as a measure of the quality of the milk, is reportedly never used, nor is the milk tested for protein. The price to the producer is not related to hygiene, quality or composition of the milk, including fat and protein content, so that there is little incentive for the producer to attend to these factors. 7.06 Maintenance of hygienic standards during processing is a problem for most existing dairy plants. Though buildings are normally spacious enough to allow access to equipment for cleaning and repair, the finishing of civil works is substandard for a food industry. Slopes of floors and gutters are inadequate to ensure complete removal from the main processing areas of milk residues which provide the perfect culture for spread of bacteria. Growth of fungi and molds on ceilings and walls is common due to lack of proper ventilation in the humid atmosphere characteristic of a dairy plant. Plant layout sometimes contributes to unhygienic conditions--e.g., the network of pipes for milk conveyance is often unnecessarily long which makes cleaning more difficult. The inferior quality of stainless steel used in manufacturing tubings, valves, and vessels, and inadequate corro- sion protection of non-stainless steel construction are also obstacles to maintaining the sterile environment associated with modern dairy plants. 7.07 The operational performance of the equipment in most dairy plants, the bulk of it manufactured domestically, is reported to be satisfactory. Dairies are typically provided with equipment for re- ception, filtering, cooling, intermediate storage, clarifying, pas- teurization and bottling. Class and plastic bottlers are used for the most part though more expensive plastic bags are now being tried in Tianjin, initially for the school milk system. New installations are increasingly relying on imported equipment to supplement equipment made in China. 7.08 Production Costs and Consumer Prices. The price structure for milk products was revised in mid-1986 to create incentives for expan- - 95 - sion of the dairy industry. Producer prices were increased nearly 20% in most areas to about YO.52-0.56/kg delivered to the collection station or dairy (compared to average costs of at least YO.42/kg including allowances for depreciation and interest). However, except in two provinces (Shanxi and Hubei) which have introduced a quality- related grading system, neither quality and/or fat content, nor season are factors in the price paid to the producer. This represents a lost opportunity from the standpoint of quality improvement, expansion of production and profitability of processing. For example, since there is scope for substituting domestically processed butter for imported butter, there is reason to promote higher fat content in the raw milk by providing adequate price incentives to farmers. A move in this direction has been taken by the Fuzhou Milk Company which is pegging its purchase price to fat content on a sliding scale ranging from Y 0.48/kg for less than 3.0% fat to YO.64/kg for more than 3.6% fat. The increase of the fat content from 3.0% to 3.6%, which is not unrealistic, would improve a farmer's annual gross income from a cow yielding 4,000 kg per year by Y 320. Standardization of milk for the consumer at an acceptable butterfat level--probably 2.5%-- and the production of butter from the excess fat would increase the profitability of the processing system and reduce the need to import butter. 7.09 A measure in force in some regions to reduce input costs of raw milk production is to supply dairy farmers with compound feeds, or other concentrate feeds, at subsidized prices. In Tianjin, for exam- ple, 0.5 kg of mixed feed can be purchased at Y 0.22/kg per kg of milk as compared with Y 0.50/kg which is the present negotiated price. This price advantage improves the profitability of dairy prodaction on the order of Y550 per cow annually on a farm with an average annual production of 4,000 kg of milk per cow. If it were to be abolished, a substantially higher milk price structure, from farmgate price through final product, would be necessary. 7.10 Total production costs of consumer milk vary substantially from one dairy to another largely as a result of differing levels of efficiency in dairy management. In one Tianjin dairy, for example, milk which had cost Y930/ton (US$0.29/litre) to purchase, pasteurize, pack and distribute was being sold at a controlled price of Y510/ton (US$0.16/litre). At another dairy visited, the total production cost was Y620/ton compared with a controlled market price of Y640/ton. The cost/price relationship for other milk products is usually much more favorable. For example, one dairy showed substantial gross profits for dried products, e.g., of Y640/ton for whole milk powder, Y 1,209 for honey milk powder, Y810 for baby milk powder, and Y782 for malted milk powder. 7.11 It is unlikely that even an efficient dairy plant could operate profitably in China producing only liquid pasteurized milk. The fact that production of the higher-cost dried milk products is financially more profitable for most dairy plants than production of fresh product is a function of (a) severe distortions in the current pricing system both for energy and milk products, and (b) the - 96 - existence of trade barriers preventing the import of such products at world prices which are far lower than the ex-factory prices of the domestically produced products. The recent price adjustments for milk products have not reduced this problem. There is no justification in economic terms for the rather high proportion of dried milk manufactured in China. Over the long term, if price distortions were rectified, by far the cheapest way for the population to have access to milk nutrients would be to increase the proportion of milk sold as liquid product. Production of milk powder may make sense only on a limited scale (a) to service outlying areas; and (b) where lack of a local market and/or transport constraints make distribution of liquid product infeasible. The Slaughtering Industry 7.12 Structure and Capacity. Pig farming is the major livestock activity in East and South China, accounting for over 90% of total meat production. Slaughtering facilities for pigs include large and medium-scale city or county-owned companies which handle from 500 to 3,000 pigs per day and small township abattoirs with a capacity of 500 to 1,500 pigs per year. However, a high proportion of all slaughter- ing activities are carried on at the township level or by the farmers themselves. 7.13 The abattoirs observed in Tianjin and Fuzhou were generally operated at well below capacity - slightly over one-half in Tianjin and much lower in Fuzhou. One reason for this is the high seasonality of both supply and demand. For example, about 85% of total pigs slaughtered at the largest plant in Fuzhou were processed during January-February. During peak periods of demand, local pig supply and possibly local slaughtering capacity are insufficient, and frozen carcasses are imported from outside. At other times, a larger propor- tion of pigs passes through the Agricultural Trade Markets, and slaughtering is done outside the state-run abattoirs. Overall, urban slaughtering capacity seems to have been designed to meet seasonal responsibilities of the Food Companies, which are even more peaked than total supply. The utilization could be improved if these abat- toirs were allowed and encouraged to service the private ATM trade on a custom basis, charging rates closer to marginal costs. As matters stand, they are a drain on the urban budget, in Fuzhou to the tune of Y15 per pig. 7.14 Operations. Slaughterhouses differ primarily in the level of mechanization employed. The larger facilities use mechanical dehiding equipment and cold storage; at the county level the facility is usual- ly entirely manual. This has a marked effect on labor productivity. In one of the large facilities visited, where annual capacity was about 1.2 million pigs, the labor input was 8 minutes per pig; in a 105,000 head per year establishment labor input was 44 minutes per pig and in an entirely manual county-level abattoir with a 15,000 head annual capacity, approximately 90 minutes per pig. The standard of equipment in nearly all facilities is poor due in part to inferior - 97 - material used in manufacture. Inadequate cleaning and maintenance routines add to poor performance. 7.15 The severe hygienic deficiencies of the slaughtering industry in China, discussed above (paras. 5.53-54), are worth reiterating. Many of the current shortcomings are related to the inadequacy of equipment and materials. For example, none of the slaughterhouses visited were provided with permanent or mobile high pressure cleaning facilities common in modern facilities, nor did they have a fixed steam pipe network which could be used for cleaning and sterilization of major equipment and installations. There is also a general lack of adequate cleaning agents, techniques and routines evidenced by the accumulation of carcass fat and other residues on cutting equipment, floors, corners and other contact surfaces. Inadequate building de- sign standards and substandard equipment are also contributing fac- tors. Many existing structural defects could be rectified by upgra- ding wall and floor surfaces, drainage, and lighting and installing hot water and high pressure water systems for cleaning. Replacement of existing wooden or other water permeable meat contact surfaces with stainless steel or tile would be low-cost improvements in many small slaughterhouses. 7.16 For the future development of the meat industry in China, it is essential that responsibilities for defining and monitoring stan- dards for building design, equipment, and abattoir operations are clearly assigned to an agency or agencies well-qualified for the task. Reportedly, the Commodity Inspection Bureau which is generally respon- sible for defining such standards for the food processing industry, has not been given this responsibility for abattoirs. Regulating the industry is especially important in view of increased interest in industrial food processing and the policy of creating export earnings from meat and meat products. 7.17 Marketing and Prices. The marketing system in China centers around butchers buying half carcasses from local abattoirs, deboning them, and sectioning the meat into small pieces for sale to individual consumers. There is a strong consumer preference for freshly killed meat, meat which is slaughtered between 2 and 5 a.m. and available in the market stalls from that time throughout the day. In the absence of fresh meat the Chinese consumer will accept chilled meat, but frozen meat is generally regarded as an inferior product. To a certain extent this perception is valid. Because of deficiencies in the system of freezing--lack of air-flow quick freezing equipment--there is loss.in meat quality, especially in the case of skinned carcasses which have no surface protection against dehydration. Quality loss continues when the frozen carcasses are kept in a cold store for several months and subjected to futher dehydration. 7.18 In addition to loss in quality, outmoded chilling and freezing techniques used in China at present result in shrinkage in bulk of the stored meat and therefore a loss in financial terms. For example, the estimated annual weight loss on 100,000 tons of pig carcasses stored for an average period of three months in the Tianjin - 98 - abattoir was 3.5%, worth Y 11.5 million at a retail price of Y3.30/kg. With the introduction of improved freezing and storage techniques, losses could be reduced to under one-third of this figure, and the quality of the frozen product would improve. While it may be necessa- ry to continue to cater to the demand for fresh meat, clearly the adoption of modern cold storage technology would help change consumer perceptions about frozen products. Indeed, as production and demand for pork expands, freezing will increasingly be required to facilitate inter-regional transfers of pig products generally and, specifically, to correct seasonal imbalances in supply. 7.19 Consumer preferences have been changing from the very fat pork typical of native pigs to the lean pork characteristic of cross- bred or exotic pigs. This preference is reflected in the higher prices paid by the consumer on the private, ATM market. However, state-owned retail shops do not necessarily offer the same selection of cuts and separation of fat from lean, and there is certainly no reflection of consumer preferences in the prices paid by the retailer to the wholesaler for carcasses, and, with rare exceptions, none in the prices paid by the processing plant to the farmer. As there is no obstacle to classification and pricing of pig carcasses leaving the slaughter house based on actual measurements, the introduction of this kind of grading system is an essential step in promotion of breed improvement. Pricing by carcass grade and weight range would also help eliminate the main objection to increased urban reliance on pork imported from other areas, specifically, the low quality of the fatty carcasses received at present. 7.20 Processed Meat Products. Meat processing plants, established in connection with abattoirs, are becoming increasingly important in China primarily to supply products for export and secondarily to respond to an expanding domestic market for processed meats. The Meat Products Processing Factory in Tianjin, for example, produces a wide variety of meat products such as frozen and cooked meat and about 50 types of Western style sausages. Edible fat is also produced, the pancreas glands of pigs are used in insulin production, and small quantities of scrap and offal are used in making animal feed. Of the total annual output of 83,000 tons, 80% is pork, 13% cooked meats including sausages, 6.5% edible fat and 0.5% animal feed. Of total output, about 1,500 tons are exported annually to Hong Kong. Techno- logy is imported and the venture appears successful. A similar venture is being planned for Fuzhou. 7.21 Existing meat processing plants are almost exclusively lo- cated in large cities. This is reasonable at present given the fact that marketing and distribution systems are better developed in the urban areas, and refrigerated transport capacity is very limited. As a long-run strategy, however, it may make sense both for hygienic and economic reasons to establish processing plants in the countryside, in regions with surplus pork production. By removing the fat on the spot for use in such industries as soap manufacturing, and leaving reduced, leaner carcasses or processed meat products for transport, the currently high costs of transporting live pigs to the cities would be - 99 - reduced and limited transport capacity could be used for higher valued products. Poultry and Eggs 7.22 At present, poultry production is largely a household opera- tion with the birds sold live. The opening of urban ATMs in 1985 has legalized urban free-marketing of live birds. In Tianjin, this effec- tively cut off the supply to, and probably demand for the products of one of the few poultry processing lines in China. The latter was constructed in the early 1980s with a 2,500 bird per hour capacity (about 7-8% utilized), but has now shut down. 7.23 Considering the thriving sales of live birds (or custom slaughter) in Hong Kong and Taiwan today, it is unlikely that frozen, processed poultry will find early acceptance among consumers. Factors favoring sale of live birds include: (a) at present, seasonality of production and consumption are even more pronounced than for pigs, making efficient utilization of processing facilities impossible; (b) poultry are consumed mainly during holidays or on weekends, when the consumer has sufficient leisure to process the bird; (c) at current high prices for poultry and low cost of labor, custom slaughtering is not expensive (one hour of unskilled labor costs about 10% of the price of one chicken); and (d) as birds sold for meat are now mainly culls, consumer "health inspection" of the live bird is the safest guarantee of product quality. However, the influence of each of these factors will diminish with development of the industry and growth of urban income levels. 7.24 It can be expected that poultry abattoirs will find a niche as part of the catering trade and in association with growth of production of broiler chickens and meat ducks or geese. Hotels and restaurants require a steadier supply of poultry, have the refrigeration facilities to store frozen or chilled carcasses, and would normally prefer to avoid self-processing. Broiler production is inherently less seasonal than sale of culled hens or roosters. Moreover, as one highly profitable joint venture between poultry SHs and the Jiangxi CAHIT has demonstrated, there is an urban market for fully cooked poultry or poultry specialty products sufficient to support one or more integrated or cooperative operation of modest scale in each urban area (consumers who lack the leisure to slaughter poultry also lack the even greater time required to stew it.) 7.25 Eggs are marketed either by the producer directly or, as observed in Tianjin, channeled through a poultry service center which acts as a procurement and marketing agent for a modest fee. At present the eggs are not cleaned or graded, although this could easily be done at the retail level. Where small-scale SH layer operations are or could be concentrated in individual villages, there is scope for further development of service operations acting as agents to sort, grade, package and distribute eggs. Such operations might expand to include processing lines for drying or freezing of seasonal surpluses, - 100 - primarily for a growing bakery and confectionary industry; or to exploit the market for specialty products such as the various forms of preserved eggs. In general, at this stage of development, processing of poultry and eggs should be done on a small- to medium-scale through labor-intensive private, joint venture or cooperative enterprises requiring mainly credit and modest amounts of technical assistance. - 101 - 8. ISSUES AND IMPLICATIONS 8.01 Although many of the improvements in policy or practice recommended above can be implemented ad hoc within individual adminis- trative systems or enterprises, some of the most important measures require coordination at the highest levels and the active involvement of agencies not strictly a part of the livestock sector. Implementa- tion of these measures would improve overall social efficiency, but would not be favored by all elements of the sector. This section will focus on these issues of systems reform, without reviewing in detail all of the more technical recommendations made in earlier sections. Price Reform 8.02 In the short run, the objectives of price reform should be to phase out subsidies and adjust administered prices to levels approaching market equilibrium prices. In the long run, the objective is to rely on market pricing, with government interventions confined at most to stabilization schemes. The discussion below stresses the short run objectives, but one should not lose sight of the long run objective. Also, there are certain reforms of pricing policy, the impacts of which are largely limited to the livestock sector; and other reforms which have ramifications extending as far as the profi- tability of crop production and the level of rural-urban income diffe- rentials. Among the former, reforms in the pricing systems for pigs and dairy products are most likely to be productive. 8.03 As an essential part of the "lean pig" policy, there must be an adequate differential in live pig procurement prices according to the leanness of the final product. To promote production, the diffe- rential must exceed the additional costs of lean pig production, evaluated beginning with costs to the breeder, which should be passed on with reasonable profit margin to the fattener, who must also use higher quality feed and obtain a higher profit margin than fat pig producers to compensate for increased requirements of management skill, working capital and greater risks. On the other hand, the differential cannot exceed the value placed by the market on the leaner product. The latter may be evaluated by working back from ATM retail value of its product, deducting processing and transport costs and margins, to arrive at value at the receiving station. 8.04 If the market premium for leaner meat is insufficient to cover the additional costs of production of the lean pig, then the "lean pig" policy is premature. Surveys indicate that hybrid pigs with 50% lean meat cost about Y30 more to produce than local pigs with 35-40% lean; and therefore MAAF supports a procurement price differential of Y30-40 per pig. However, there appears to be some disagreement over the extent of the present market premium for lean meat between representatives of MAAF and MOC. MOC argues that an increase of leanness from 40% to 50% implies a replacement of fat by lean meat of about 7.5 kg per pig, and that the average urban price differential - 102 - between fat and lean is presently only Y 1.6-2.0 per kg, implying an inadequate producer price differential of Y12-15 per pig. MOC believes that only the Guangzhou and Beijing markets will now support a sufficient producer price premium. MAAF, however, considers that adequate market premiums are more widespread than claimed by MOC, although not in the rural areas and some urban markets. Before investments or policy measures are undertaken to further promote lean pig production, it is clearly important that the size and location of markets where consumer preference for leanness is reflected in adequate price premiums be identified through market price surveys. 8.05 In such areas, to support an adequate producer price diffe- rential, it is important that (a) state retail outlets should offer choice of cuts similar to those offered in ATMs, with equivalent price differentiation; (b) wholesale carcasses ex-abattoir should be graded by weight and leanness and priced accordingly; and (c) abattoir gra- dings of animals should be passed back in the form of higher prices to receiving stations or producers. Wholesale carcass grading and pri- cing should also be introduced in inter-provincial trade, whether at list or negotiated prices, with allowance for storage costs and losses as well as market discounts for frozen pork. 8.06 The low retail price of fresh milk in urban areas has been central to pricing problems in the dairy sector. This has been justified on the grounds that milk is supplied as a nutritional supplement for infants, the sick, and the aged, rather than a luxury food. In reality, a high proportion of dairy products is not consumed by these target groups, as urban budget surveys verify that disproportionate proportions of supply are consumed by wealthier segments of the population who also have the lowest ratios of dependents-to-total-persons (Table 8.1). In addition, tourists, fo- reign residents, and the bakery and confectionary trade all benefit unnecessarily from the same low prices. Also, the low profitability of fresh milk production due to price controls has created a strong incentive for dairies to enlarge the proportion of less nutritionally valuable products, such as milk powder diluted with sugar or milk chocolate. As the milk rationing scheme lies at the heart of these distortions, consideration should be given to reducing its scope or phasing it out. 8.07 It appears that unsubsidized dairy farmers can earn a rea- sonable rate of return at milk procurement prices which now approach YO.56/kg, and that a Chinese dairy producing and distributing pasteu- rized milk should be able to operate satisfactorily with a YO.25/kg margin between procurement price and retail price (roughly, YO.07/kg operating costs, YO.08/kg depreciation, and YO.10/kg profit; which compares to a YO.16-0.21 total margin at present). Unfortunately, the recent adjustments have also increased the price of milk powder, leaving the relatively high profitability of milk powder production unchanged. This pricing decision should be reconsidered: if the profitability of fresh milk production were increased relative to milk powder, suburban dairies might well be willing to shut down their powdered milk production lines, and rural dairies might suddenly find - 103 - Table 8.1: Urban Expenditure on Dairy Products by Income Group (Tianjin and Fuzhou, 1984) Tianjin Fuzhoi Income Group: Lowest Highest Lowest Highest Percent of population 11% 23 22 14 Pct. of value of total supply 8 31 12 30 Dependents/population ratio 0.48 0.22 0.59 0.35 that long distance transport of bulk milk in tankers is neither tech- nically nor financially infeasible; in short, the product mix could change rapidly in favor of liquid product. In addition, now would be an opportune time to introduce milk grading and pricing according to butterfat content. 8.08 The main price distortions in the livestock sector are those associated with the subsidized state procurement, processing and dis- tribution system and the supply of feedgrain at subsidized prices to livestock producers. These distortions are associated mainly with pig production (to a lesser extent with eggs and dairy products) and especially with supply to the larger municipalities. Towns at the county level or below appear to have been weaned from both forms of subsidies. The subsidies benefit consumers in large municipalities, employees of the state distribution system, and suburban farmers (especially SHs), at the expense of the state budget and commercial feedgrain growers. The private or cooperative distribution system, including the urban ATMs, which do not benefit from these subsidies, is handicapped in its role in the municipal pork trade. 8.09 Removal of subsidies on egg and dairy products would be relatively painless. Egg and poultry meat production at present are profitable enough that producers could absorb the loss of feed and distribution subsidies. Urban consumption of dairy products is small enough that consumers could easily bear the costs of price increases sufficient to compensate dairies and dairy farms for loss of subsi- dies. In view of the recent price adjustments for milk, which probably eliminate direct subsidies (but not rationing) in most locales, it appears the government has reached the same conclusion. 8.10 However, elimination of subsidies to the suburban pig indus- try is far more problematic. Eliminating both the feed subsidies and subsidies on processing and distribution would result in decreased quantities supplied and consumed, and a decline in unit revenues for producers.11 In this process, many marginal producers, particularily /l Offsetting shifts in supply and demand curves would occur, with an uncertain degree of change in equilibrium farmgate price. - 104 - SHs depending mainly on purchased feed, would be bankrupted by higher feed costs. As supply contracted, the relative advantage would shift to the backyard producer of local breeds of pigs. Urban consumers would be forced to pay higher prices at state markets, reflecting full costs of state-run processing and distribution. Demand would shift further away from pork sold in state shops to the ATMs, and overall levels of consumption would contract. Improved financial performance for the state distribution system also would require finding other jobs for redundant workers, and shifting the pensioner burden to the state. 8.11 Considering the above, the subsidies to the pig industry should be removed gradually, following a strategy with multiple ele- ments, including: (a) reduce the marketing constraints on inter- regional live pig or carcass trade; (b) improve the quality of chilled or frozen carcasses made available through that trade; (c) enlarge cold storage facilities to better meet peak seasonal requirements, while finding ways of utilizing these facilities in the off-peak season; (d) phase out the feed subsidy; (e) close down substandard abattoirs, and use other state-run abattoirs to serve the slaughter requirements of the ATM at fees close to marginal costs; (f) encourage expansion of ATM pork sales, and correspondingly reduce employment in the state distribution system (or encourage the latter to serve a wholesaling function for the ATMs); and (g) modulate the increases in the relative price of pork according to the rate of increase of consumer incomes. 8.12 This strategy is in direct contrast to the present municipal approach of increasing the extent of local self-sufficiency in pork supply. Elements of this strategy require actions by the central government, the provincial governments of supplier provinces, and most of the various bureaus involved in the livestock sector (including the Transport Bureaus). In the absence of the central and local coordination this requires, the municipal governments would have little choice but to pursue self-sufficiency, at considerable cost to the national economy. Organizational Reform 8.13 Higher-level governmental organs in China tend to be identi- fied with a coordinating function or an administrative function, and occasionally both. As Chart 1 made evident, the functional components of the livestock sector, including input supply, production, procure- ment, transport, processing, storage, and final distribution typically are administered by several different systems, and coordination is carried out at such a high level that it is inevitably weak. The coordinating bodies in the subsector are mainly commissions responsi- ble for policy making and planning (such as the SEC and SPC at the central government level, or their counterparts at the provincial or municipal level). As coordinating bodies lack the direct administra- tive authority and the staff necessary for program or project execu- tion, programs or projects which cut across either functional compo- nents of the sector or separate administrative systems are difficult - 105 - to implement. 8.14 There are at least four possible approaches to organizational reform within the existing system: (a) set up new offices under the coordinating committees and provide them with adequate staff and authority to execute programs; (b) shift administrative systems around to place related systems under the same administrative agency; (c) create new administrative systems with authority to compete with existing systems across functional or administrative lines; and (d) decentralize administration and set up coordinating committees or offices at lower levels, or else rely on market mechanisms to provide coordination. In fact, the present organization chart reflects recent application of each of these approaches: the creation of the National Feed Industry Office is an example of the first approach; the replacement of the Ministry of State Farms with the BSF under MAAF, and consolidation,of some offices within BSF and their MAAF counterparts is an example of the second; the creation of the GCAHIT/CAHIT corporate system is an example of the third; and the fourth is reflected in the increasingly direct dealings between state retail markets and producers of livestock products. 8.15 A deepening and extension of past organizational reforms in the sector would improve the prospect for coordinated execution of development programs. The proposed program for development of the feed milling industry should logically be executed by the NFIO, whose authority (especially over mills under the Grain Bureau) and staffing should be further strengthened. In dairying, the growing responsibi- lity of state farm veterinarians to service the needs of SH producers suggests that these vets should report to AHB rather than BSF, and a fee-for-service approach, or privatization, should replace fixed salary scales. The dairy industry might alternatively be transferred out of the BSF system, with responsibility for production being moved into the GCAHIT/CAHIT system and, for services, into the AHB system. 8.16 In the processing and distribution subsectors, the most pressing needs are for better standards of control over animal health and sanitation, notably at abattoirs. The control system should be completely independent of the Food Company/Second Commercial Bureau system which administers the processing units. Probably it should reside with AHB, yet be separate from administrative offices concerned mainly with live animal health and receive greater emphasis. 8.17 Production and service enterprises in China have rarely been organized in a hierarchical system of functional specialization which locates functions, according to their criticality or complexity, in enterprises possessing appropriate levels of managerial and technical expertise as well as qualities and quantities of equipment. The breeding system for livestock is one area where a pyramidal form of specialization should be introduced, particularily in pig breeding. The organizational structure in feed milling should also be more hierarchical, to include nucleus mills providing pre-mixes and technical services to satellite mills; and increased division of labor in producing a more varied set of products. - 106 - 8.18 In general, all production or service units in each administrative system should be given the freedom to engage in commercial transactions crossing functional and administrative lines (including those between state and private sectors). For some units, price reforms are required before this is feasible, because cross- subsidization within vertically integrated systems is currently the norm and would be disrupted by this freedom. For other units, encouragement and an appropriate fee or rate structure are all that is required. Vertical integrated activities (not necessarily involving integrated ownership) can be advantageous, e.g., in the poultry industry. The GCAHIT/CAHIT system should be encouraged to expand its organization and joint venture financing of such activities, in alliance with the private sector. GCAHIT should also engage in joint ventures with international breeding companies in the poultry industry to establish Parent and Grandparent stock farms in order to maintain a continuing flow of superior genetic stock. Expanded Role of the Market 8.19 The reforms in the livestock products distribution system in 1985 facilitate an expanded role for market-oriented production in the sector, that is, for economic activities governed by consideration of profit-and-loss and therefore determined by flexible, market-clearing prices. At present market-oriented activities are most important in poultry, beef and mutton supply, whereas in pork and dairy product supply, "indicative" (i.e., locally-administered) pricing and adminis- trative management are still dominant. Inflexible, administered prices can easily lead to dynamic instability in markets where both supply and demand are strongly price sensitive, and the pig market appears to be such a case, as the disequilibrium in China's pig market in 1986 indicates. Rather than dealing with the disequilibrium by imposing additional administrative control, the transition in the pig market to more flexible pricing methods should be speeded up. "Flexi- bility" means that each independent accounting unit within the distri- bution system should be encouraged to adjust prices within a wide price range so as to maximize profits; which in turn requires that each unit be allowed to retain a significant proportion of profits. 8.20 At present, the free market, or ATM distribution system is supposed to help stabilize the livestock market, by providing signals to producers and government alike of supply and demand conditions, and by providing an outlet for excess supply or pent-up demand due to official pricing errors. It would better serve these functions if its scope were enlarged, which depends partly on reduction of subsidies in the state-controlled subsector and partly on decreasing barriers to trade (see below). One major shortcoming of the ATM distribution system is its reliance on expensive truck transport, and lack of access to cheaper long-distance transport modes. This could be alle- viated if the Ministry of Railways were to set up an "outside-Plan" goods traffic program, perhaps consisting of two elements: (a) a set proportion of wagons suitable for conveying live animals or frozen/chilled product would be offered for short-notice use by any commercial unit (state or private) at special freight rates; and (b) - 107 - commercial units would be encouraged to invest in refrigerated freight wagons or chilled milk tankers, which could be leased out and attached to regular freight trains. Enlarging the Role of Interregional and International Trade 8.21 During 1983-85, China had begun to develop a national-level feedgrains market which, at times of bumper harvests, was moving large amounts of corn and soymeal from the Northeast to the Southeast. This has now come to a virtual standstill, due to a combination of poor harvests and a foreign exchange shortage which encourages the North- east to boost agricultural exports. At present, virtually every major corn or soybean producing province has banned or highly restricted feed grain exports to other provinces, and feed mills in the Southeast are facing acute shortages of raw materials. Simultaneously, the clamp-down on imports from abroad has constricted the dnly alternative source of supply. 8.22 In according high priority to rapid development of livestock production, the Central Government should attach the corresponding priority to solving the national and regional problems of feed supply. It may in fact be economically advantageous to export Northeastern corn and soybean to the Soviet Union, South Korea, and Japan, rather than processing the soybean and transporting corn and soymeal to the South; but this advantage is offset if the Southeast must import the same amounts of feed from abroad, or if Southeastern feed mills and livestock production are crippled for want of raw materials. The advantage, from a national viewpoint, would derive from a strategy of using non-exportable surpluses of rice and local oilseed meals as feed stocks in the Southeast, and allocating the necessary foreign exchange to import sufficient amounts of appropriate balancer ingredients (such as soymeal or fishmeal) from abroad. As a corollary, the fo- reign exchange-saving potential of increased production of each type of feedgrain and oilseed meal producable in the Southeast should be evaluated, and its farmgate price should be adjusted accordingly. Re- search directed at adapting corn and soybean varieties and production techniques to the Southeast also deserves priority. 8.23 A major obstacle to further development of inter-regional trade is the present power of local governments to intervene in re- straint of such trade. This power was exercised in 1986, not only to restrict flows of feedgrains from northern to southern provinces (para. 8.21), but also to prohibit or tax exports of pigs from south- ern pig-surplus provinces such as Hunan to areas of excess demand, such as Guangdong (which exports its pigs to Hong Kong). Such a chain of derived demand should normally be favored as beneficial to profits of producers at each link in the chain, but local governments have proven more sensitive to the complaints of consumers or state trading organs about increased prices or commodity shortages. Consequently, China should consider a legal prohibition on such local government interventions in interprovincial trade, in the interest of the deve- lopment of the national economy as a whole. - 108 - 8.24 To date China has taken advantage of international milk surpluses only to the extent of seeking World Food Program assistance in the form of milk powder and other dairy products. Considering the amounts of foreign exchange now expended on luxury consumer goods, and the general milk shortages in the cities, it is surprising that no foreign exchange is allocated for imports of milk powder (but is allocated for corn to feed the cows). It is arguable that, in view of the size of the domestic market, China's domestic production capacity for fresh milk should be increased rapidly. However, the opportunity to exploit the subsidies foreign governments give to their milk produ- cers should be seized in the interim to close the gap between domestic supply and demand. Production of milk powder by municipal dairies should cease, and cheap imported milk powder should be used (a) for local bakery and confectionary requirements; (b) in recombination with local edible oils, to "stretch" fresh milk supplies and reduce season- ality of supply; and (c) as a substitute for fresh milk in calf feeding. Lack of high quality fodder and heat stress are serious factors in dairy production in the Southeast, suggesting that greater use of imported milk powder is particularily advisable for this region. Locational Strategy 8.25 Losses and costs associated with fresh product delivery are most significant in the poultry and dairy industries. Consequently, production of eggs, poultry meat, and fresh milk is most rationally sited near the cities. However, the raising of calves or heifers in suburban areas is uneconomic and unnecessary; and that part of subur- ban dairy farming should be relocated to areas where roughage and green feeds are relatively abundant. Over the long run, an increasing proportion of urban pork supply should come from rural areas with feed surpluses and low opportunity cost of land. This applies especially to the Southeastern cities, and particularily if these cities cannot rely on supplies of corn and soybean from either the Northeast or abroad. However, this strategy depends on a gradual solution to transport and storage problems, and upgrading the quality of pork supplied through breed and feed improvement programs aimed at the rural (mainly back- yard) producer, along with corresponding price incentives. As a short run measure, the efficiency of suburban pig production could be raised by space-saving pig shed designs, siting of farms in the distant suburbs, and improved feed conversion rates. Form and Scale of Production Enterprise 8.26 In pig production, there is a clear case for division of function: state- and collectively-owned larger farms should serve mainly as breeding farms for nucleus herds (abandoning fattening operations); SHs should serve as multiplier farms in the breeding system and fatteners for lean varieties; and backyard producers should maintain their present role as suppliers of most market pigs, using local breeds to the extent that they depend on low-quality feeds (or vice versa). Overall, Government's strategy for development of the industry should give increased attention to the backyard producer, in the expectation of achieving major efficiency improvements at - 109 - relatively low cost through use of improved hybrids and feed balancers. 8.27 In dairying, there is no "backyard production", because the high costs of investment in animals place even the owner of one or two cows into the SH category; and a five-ten cow SH is either very wealthy or deeply in debt. There are both well- and poorly-managed state farms and SHs at present, and it appears that small SH herds, if grouped and managed cooperatively to facilitate silage use and ALI, and given adequate support services, could function as well as state farms; but that is based on hypothesis more than observation, as SHs usually are geographically scattered (taking advantage of equally scattered natural fodder resources). A mixed system, with state farms taking the lead in promotion of SH dairy development, would bet quite acceptable. However, if the land area managed by state dairy farms is not to be increased, inevitably a high proportion of the national dairy herd will be under SH management by the year 2000, given the planned six-fold herd growth. In the short run, improving the utili- zation of land and buildings on existing state farms, and the motiva- tion or expertise of technical personnel, are the most pressing is- sues, but in the long run development of support services for SH dairying must assume highest priority. 8.28 In egg and poultry production, the superiority of imported breeds and intensive feeding systems argues for a scale of at least several thousand birds, except in semi-scavenging duck production systems, where units of a few hundred birds would be efficient. If sufficient flexibility is allowed for wages and incentive schemes on state or collective poultry farms, large farms would conserve on technical and managerial skills which are scarce in this industry. Financing Policy 8.29 Financing terms are important to the viability of the SH sector, where only a small proportion of investments in housing and stock can be owner-financed. The major source of finance has been lending from the ABC or RCCs, and there has been a strong tendency to use medium-term (3-5 year) loans to finance 80-90% of total investment costs. The use of such short maturities and low proportions of owner equity has allowed the SH sector to develop rapidly, with minimal interest burden, but timely repayment of principal will prove difficult, particularily if either increased supply or reduced subsidization lead to declining profit margins in the sector. 8.30 Future financing should move to longer maturities and a higher proportion of owner-equity, to protect both the owners and financial system against price risks. For long-term lending, the present ABC interest rate structure provides for rates approaching 9- 10% (although few loans are now extended at these rates). The higher equity proportions and interest rates may discourage growth of SH production in some components of the industry, but they will also help to stabilize supply by discouraging entry of producers who lack the resources to withstand periodically unfavourable market conditions. ANNEX I Jable 1 LIVESTOCK SECTOR STUD Population Df Fuzhou An Tian.in, 1ft ('000) Fuzhou Tian.iin Growth Growth Growth Growth Year Total rate Urban rate Total rate Urban rate /a /a (1 /a Mn /a % 1981 3,511 594 7,603 1.5 3,928 2.2 1982 4,678 33.2 626 5.4 7,749 1.9 4,025 2.5 1983 4,749 1.5 640 2.2 7,853 1.3 4,111 2.1 1984 4,826 1.6 662 3.4 7,955 1.3 4,240 3.1 1985 4,866 0.8 666 0.6 8,058 1.3 4,305 1.5 1986 4,902 0.7 670 0.6 8,155 1.2 4,352 1.1 1987 4,941 0.8 673 0.5 8,253 1.2 4,400 1.1 1988 4,974 0.7 677 0.6 8,352 1.2 4,449 1.1 1989 5,004 0.6 680 0.4 8,452 1.2 4,498 1.1 1990 5,041 0.7 683 0.4 8,553 1.2 4,547 1.1 LI From 1985 on: local projections. - 111 - ANNFE Table LIVESTOCK SECTOR STUD 2=r Capita =.t Income, Fuzhou And Tianjiin, 1q81-S (Y RMB /a) Fuzhou Tianj in Growth Growth Growth Growth Year Urban rate Rural rate Urban rate Rural rate /b /b )(/b /b (%) /b ) 1981 418 175 521 298 1982 546 30.6 236 34.9 530 1.7 327 9.7 1983 609 11.5 287 21.6 561 5.9 412 26.0 1984 622 2.1 349 21.6 678 20.9 505 22.6 1985 748 20.3 385 10.3 776 14.5 600 18.8 1986 793 6.0 424 10.1 825 6.2 669 11.5 1987 840 6.0 466 9.9 877 6.3 746 11.5 1988 891 6.0 495 6.2 932 6.3 832 11.5 1989 944 6.0 545 10.1 991 6.3 927 11.5 1990 1,001 6.0 610 11.9 1,053 6.3 1,034 11.5 .A Y 3.2 US$ 1 in May 1986 L.h Fra 1985 onwards: local projections ANNEX i TabLe 3 CHINA LIVESTOCK SETO [ FM[|I IRiL SupLy of Livestock and Fish, Fuzhou and TianJin. 1981-90 Fuzhou Tianjin Year Pork Beef Mutton PouLtry Eggs MiLk Fish Pork Beef Mutton PouLtry Eggs MiLk Fish TotsL Suax Ltonal /a 1981 24,733 272 700 525 4,800 6,896 17,585 80,779 1,654 11,545 2,884 24,325 23,495 28,225 1982 24,382 219 1,193 650 8,915 8,878 18,705 95,497 1,578 13,751 5,095 29,165 28,275 35,031 1983 25,213 277 675 615 6,250 10,077 18,745 113,807 3,182 15,089 6,897 47,800 30,355 39,270 1984 25,382 225 473 4,290 7,779 11,449 24,000 122,431 5,286 10,019 7,138 64,801 32,745 35,385 1985 24,257 300 611 3,930 6,806 12,448 20,000 120,000 7,255 13,745 8,000 77,500 37,500 21,819 1966 28,000 375 795 4,800 15,600 14,000 N/A 131,648 8,051 15,015 9,792 81,800 43,303 24,807 1987 27,000 585 945 5,550 16,450 17,000 WA 144,540 8,800 16,500 12,100 8,020 48,940 28,600 198 28,000 785 1,145 6,150 18,050 18,500 N/A 158,807 9,788 18,019 14,904 90,537 57,837 32,923 1989 29,000 965 1,350 6,900 18,900 20,500 N/A 174,298 10,796 19,791 18,497 95,358 66,798 38,008 1990 30,000 1,170 1,560 7,950 19,500 21,500 N/A 191,202 11,822 21,599 22,735 100,489 77,299 40,151 Per Capita SanixL TkiLoraml /a 1981 41.7 0,5 1,2 0.9 8.9 11.6 29.8 20.8 0.5 3.0 0.8 6.2 6.0 8.7 1962 40.2 0.4 2.0 1.1 11.4 14.7 29.9 23,8 0.4 3.4 1.3 7.2 6.5 8.7 1983 39.4 04 1.1 9.6 9.8 15.7 29.3 27.7 0.8 3.7 1.8 11.7 7.5 9.6 1984 38.3 0.3 0.7 6.5 11.7 17.3 36.3 28.9 1.3 2.4 1.7 15.3 7.7 B.4 1985 38.4 0.5 0.9 5.9 10.2 18.7 30.1 27.9 1.7 3.2 1.9 18.0 8.7 5.0 1996 34.5 0.5 1.0 6.0 14.2 18.7 N/A 40.3 1.9 3.5 2.3 18.8 10.0 5.7 1987 35.5 0.8 1.3 8.3 14.5 22.5 N/A 32.9 2.0 3.8 2.8 19.8 11.4 8.5 1988 38.5 1.0 1.5 6.7 15.8 24.2 N/A 35.7 2.2 4.1 3.4 20.4 13.0 7.4 199 37.5 1.3 1.8 7.0 16.1 26.6 N/A 38. 2.4 4.4 4.1 21.2 14.9 8.5 1990 38.5 1.5 2.0 7.3 16.1 27.6 N/A 42.1 2.6 4.6 5.0 22.1 17.0 19,2 a/ From 1965 onwardso LocaL projections - 113 - ANMEX 1 LIVESTOCK SECTOR STDY Supoly .9.1 Livestock Products hy State Aad Market, Urban Fuzhou Alnd Tian.iin, 1984-8S (percent) Fuzhou Tianjin Product State Market State Market Pork 89 11 97 3 Beef 92 8 98 2 Mutton 91 9 96 4 Milk 90 10 100 0 Poultry 86 14 94 6 Duck Eggs 85 15 98 2 Chicken Eggs 83 17 98 2 19La Pork 87 13 Beef 18 82 Mutton 36 64 Milk 90 10 Poultry 92 8 Duck Eggs 76 24 Chicken Eggs 0 100 January-April 130a5 Pork 94.6 5.4 99.8 0.2 Beef 40.8 59.2 98.1 1.9 Mutton 59.7 40.3 95.4 4.6 Poultry N/A N/A 97.1 2.9 Eggs 89.2 10.8 99.4 0.6 May-December 1qLa Pork 81.2 18.8 99.5 0.5 Beef 13.5 86.5 89.2 10.8 Mutton 20.5 79.5 86.4 13.6 Poultry N/A N/A 66.9 33.1 Eggs 63.7 36.3 90.0 10.0 .g~il LiffESTOCKM ,OR UL MpnthLv i f .g Livestook .gE fifi (tane) 104 1BR Praet Jn Feb Har Apr My Jm Jul Aug Sep Ot N D ca Jan Feb Ner Apr Hay Jun Jul Aug Bap DOt No¥ S Pbrk 1,598 2,877 2,125 2,816 2,081 1,712 1,710 1,894 2,575 2,414 2,204 2,441 2,114 2,701 2,950 2,717 1,418 1,850 1,659 1,781 2,279 1,884 1,587 1,899 seet el a 82 21 10 a 18 19 18 mB 51 48 a f1 74 70 sa 78 73 75 84 100 109 189 m~tte./ 28 121 5 18 24 in 4 5 21 25 fl 28 114 184 71 47 57 en 47 42 58 68 sa 78 Peu try/A 374 491 298 805 820 418 814 208 817 872 878 426 462 6M1 487 165 124 log 20 m 800 820 80 m5 Ega 841 121 858 701 489 341 488 480 948 480 41 548 ED 879 770 8M1 449 588 408 087 388 408 410 505 NILk 1,00 1,00 1p211 1,182 1,72 894 72 788 m9 912 989 1,041 1,128 1,127 1,278 1,1B 1,117 D48 818 88 888 1, 1,084 1,187 Perk 12,499 8,550 9,0~ 9,987 8,5~8 7,848 8,251 9,508 11,105 10,85 11,81 15,051 11,10o 9,41 10,8 10,881 8,480 4,299 5,728 7,721 D,o1 6,181 5,864 6,78 *e~Lå 1,827 684 I 172 18 18 N a8 157 1,519 407 201 255 729 2,38 181 204 88 m0 88 127 427 28 825 9O Nuaton 2,817 1,01 M 111 - 820 272 885 882 891 419 518 8,188 1,844 80 480 420 21 218 271 814 258 406 752 - Paul try 2,780 a 18 408 2M 28 270 258 770 489 578 708 516 8,209 444 422 2E 154 me9 84 482 284 200 289 EggE 4,55 580 5,110 5,745 6,84 6,150 5,880 7,195 5,70 8@100 5,065 5,815 4,857 4,87 7,04 9,150 6,891 2,70 2,887 2.418 8,00 2,797 2,898 4,19 fitk 2,598 2,574 2,88 2,981 8,094 2,871 2,7891 2,5 2,64 2,688 2,587 2,845 2,299 1,48 8,894 8,884 1,914 1,898 1,95 1,814 8,1S 5 8,21 1.84 2,58 Fi@h 8,477 2,290 1,924 2,941 8,157 2,275 1,088 1,107 8,838 2,778 4,252 8,788 2,78 8,2 4 2,105 2,104 2,080 878 540 1,150 1,74 1,405 4,869 9,000 LA Ameulng Germes weight per heed of about 80 kg. & Aimng overage barae wess ight per head of 88.4 kg (baed on ~L Aripautterat Yma km . Lg A~ne* ore,age meat per heed of 12.0 kg (beed on ~ AoriouturaL Yarb~ j. Am Amsas fverage bird we1 gt of 1.5 kg, reported. - 115 - AIm 1 LIMES'TQC Sg[GWlr iaLy at Livestock Products jkl 3uld& Canasr fuzho andTlianJin. 1S85E (peroent) Fuzhou Tianl in Of which . TotaL IndividuaL Pro- Other Total IndividuaL Product Supply Consumers Other Catering sessing users Zg SuppLy Consumers Other Park 100 B2 8 12 4 22 100 82 18 Beef 100 81 89 21 1 47 100 75 25 Mutton 100 82 68 22 0 48 100 70 80 PouLtry 100 70 80 WA N/A WA 100 75 25 Eggs 100 70 so 25 5 0 100 s0 20 MiLk 100 56 44 WA N/A /A 100 100 0 Pork 100 57 43 'l 5 27 Beef 100 80 70 21 2 47 Mutton 100 81 69 22 0 47 Eggs 100 68 37 24 4 9 La Other users incLude dining haLLs, meatings, hotels, raiVLater transport, and military units. -116- le2his 2. LIVESTO C Sf.T0UDY0 Sources if SDjig if Livestock jL fi.h Products. iL&M Fuhoua MHd 1i1lia 1Mf [tons) Fuzhou Isgorty LocaL Other Fuitan State CoLLeoctive Specialized Provinces Province Enterprises Enterprisee Householda IndividuaLs TotaL Park 16,000 2,916 660 4,998 5,538 81,090 Beef 145 145 Mutton L 952 99 891 PouLtry /d 150 945 630 2,715 4,440 Eggs 8,470 8,470 HiLk 597 780 8,570 8,562 11,448 Tionlin Moorte LocaL Fram Other State ColLective Specialized Abroad Provinces Enterprises Enterprises HousehoLds Individuals Total Pork /a 74,200 210 700 2,100 51,590 128,900 Beef 6,310 100 748 7,156 Mutton 9,009 18,868 25,877 PouLtry 6,135 887 942 4,868 5,633 17,360 Eggs 25,805 4,543 1,532 15,098 298,973 76,350 MiLk 32,746 82,745 Fish 4,410 14,802 15,207 34,419 L Assumes average meet per head of 60 kg in Fuzhou and 70 kg in Tionjin. L Assumes everage meat per heed of 96.4 kg in Fuzhou. Lc Assumes average meat per heed of 12.85 kg in Fuzhou. L Assumes average bird weight of 1.5 kg in Fuzhou. - 117 - ANNEX 1 Table & CHINA LIVESTOCK SECTOR STUDI External Procurements Qtfig 1,yI Source, Fuzhou, 1_(M (no. of head) Live Frozen Pigs /a-- CarasesLb Imports frQm Fujian Ningzhu (110 km) 7,178 Putian (100 km) 8,232 Sanming (150 kam) 332 Subtotal 15,742 IMports Et= ther Provinces Zhejiang 2,410 5,601 Jiangxi 675 8,530 Jiangsu 1,735 13,292 Shandong 73,819 Henan 27,091 Hubei 10,305 Guizhou 459 Anhui 6,440 Sichuan 14,639 Subtotal i4820 360,176 19tal Imoorts 20,5362 16010176 La Weights unknown. Lh Averaging about 50 kg/head. inthL Prowuromant .g Live@to~k Pro :a j. Scuro ja Tionjin mad im (tone) PDrk Beef ~ton PbuL tr¥ E= MiLk Yar lath Locel/a iportedb Totak LoMal/a aportab Total LoceaL/ Iaportsd/b Total Locet/a SoLf/o ImporteWb Total LoaeL/a bport~d/ Total LooeL Maportadb 1184 Jan 2,345 7,80 10,805 2 84m o80 704 8 712 18 44 15 72 1,fl0 mo 2,870 1870 2,518 Feb 2,141 4,141 8,2012 8 sa 8 7 209 215 5 mg 5 48 2,270 1,250 8,5a0 2,84 2,574 Har 8,884 7.271 10,805 98 =04 882 72 382 04 17 82 15 84 88 4,000 4,08 2.804 2, ~ Apr 3,818 12,81 15,~ ta a 61 870 a 1 8 m8 19 95 18 80 7,85 a,o 11,15 8,W 2,980 MaY 4,84e 7,2 11 ,~2 et n0 m1 74 985 480 14 40 17 71 7,785 4,500 12,85 8,411 8,01 Ju 5,080 8,187 18,227 5 40 48 9 en m 15 28 17 m 4,80 8,000 7,800 2,04 2,870 Jul 3,740 8,258 9,98 a a8n a 450 450 8 i 1o ge 8,680 2,500 8,180 2,82 2,7k Aug 1,570 9,8=8 5,58 17 80 77 848 848 18 45 21 84 2,115 1,500 8,81 2,873 2.5b sep 2,129 5,88 7,792 21 418 508 5 581 5a 81 84 87 102 1,o80 750 2,90 2,88 2,848 00 Ot 2,425 3,275 5,700 5 552 847 181 2,528 2,899 50 80 59 1a 2,115 1,400 8,50 2,mi R,8 Now 2,588 4,2ac 7,194 584 2,150 2,759 am 2,01 2,488 88 15 9 200 2,520 1.205 8,75 8,844 2,587 Dec 4,837 20,340 24,677 89 888 878 188 1,08 1,222 18 e7 92 201 6,050 1,250 7,300 2,825 2,~ I 87-77 91.141 128.06 & H 1 .j .M 1.01 L.M 1000 m NZ «« 1A1 42.558 25.W 58^I 34.sL i ga.7 1em Jan 1,540 18,350 19,890 898 1,125 1,518 422 10s 528 48 47 98 194 8,805 1,800 5,105 2,a55 2,298 Feb 1,194 16,381 17,575 140 415 555 214 250 484 1 39 20 59 4,100 1,400 5,500 1,762 1,48 Mer 1,280 5,50 6,733 188 495 857 45 49 94 10 84 44 5,755 1,98 7,718 8,50 8,3a Apr 1,424 7,590 0,014 2,217 2,217 20 26 39 87 76 6,575 2,20 8,855 8,889 8,884 Key 2,478 4,08 6,541 1 8m 801 45 45 40 48 88 7,185 2,50 9,885 2,184 1,914 Jun 5,875 8,949 9,824 89 8 47 48 30 78 18 35 53 4,890 1,500 5,890 2,051 1,948 Jul 4,486 378 4,844 24 100 124 70 70 4 sa 87 2,0 82 8,455 2,017 1,950 Aug 1,894 1,148 3,042 70 79 148 2 83 85 11 48 21 80 1,29D 485 1,715 1,887 1,814 Sep 2,475 3,38 5,81 109 85 174 9 23 32 9 80 18 81 786 282 1,028 3,2 3,194 Oct 3,170 2,948 6,119 128 70 198 33 270 803 9 32 18 58 WA WA WA 3,352 3,291 Nov WA WA WA WA WA WA WA WA WA WA WA WA WA WA WA WA WA WA Oec WA WA WA WA WA WA WA WA WA WA WA WA WA WA WA WA WA WA Ttl 25.741 83.702 89.443 1IZ 4.87 5.4 ZZ4t 54 1.ZDZ lim. ini iZ. Z. 36.466 12.^ 48.031 26.ies 24.855 e LocaL reftre to Tienjin muncip&lity. . IMportad reftrs to domoetio importa. g Se~ men produced and sold by armnera themuLvo. Until May 19, park, beef and eutton mar* &upposd to be oLd ta the Stato. Ountitlis prodUced and sold by femera for produots other than poultry are not wveaLeble, a~though they bamcen Significent efter May 1, 190. - 119 - AMEEE 1 CHINA JJJ Procurement fPrSce fM Livestock ja Fish, Fuzhou. 1979-1I5 ZA (yumn/kg) 1979 1980 1991 1982 188 1984 1985 Hogs, Liveneight Zk 1.29 1.28 1.2 1.98 1.28 1.830 2.40 Sheep, Liveweight Le .95 1.04 1.04 1.04 1.04 1.04 2.00 Beef /d 1.40 1.54 1.54 1.54 1,54 1.54 3.20 Chickens, Liveweight /a 2.09 2.20 2.20 2.20 2.32 2.60 3.40 Ducks, Liveweight ZI 1.44 1,51 1.56 1.56 1.86 2.20 3.00 Nilk .40 .40 .40 .48 .48 .48 .49 Duck Eggs 1.75 1.81 1.91 1.94 1.94 1.94 2.90 grass Carp LA 1.10 1.10 1.10 1.10 1.10 1.10 2.80 Silver Carp Zh .94 .94 .94 .94 .94 .94 2.40 Crucian Corp Li .98 .98 .98 .98 .98 .98 2.00 /a Prices were State-administered through ApriL 30, 1985. Since Hay I 1985, they represent "target" prices, which may fluctuate within a prescribed range according to suppLy and demand. Zk Over 71% meat. Z2 Over 45% meet. Id 100% meat. ZA Over 1.25 kg (native chickens through 1984; over 901 foreign breeds since 1995). L Over 1.5 kg (crosebreeds of Nusoovy with meet ducks). ZA Over 1.0 kg. LZ Over 0.75 kg. Li Over 0.25 kg (African variety). - 120 - TabLe ii LIVESTOCK SECTi STUD Procurement Prices for Livestock, Tianjin. Pre- ad Post-May il La (yuan/kg) Pre-May 1985 Post-May 1995 Price Price AnimaL/Grade Standard (Y/kg) Grade Standard (Y/kg] Hoas SpeciaL Meat yieLd > 72% 1.71 let Grade 75-90 kg L.wt. 2.00 let Grade " " 89-72% 1,64 2nd Grade 90-100 kg L.wt. 1.88 2nd Grade " " 66-69% 1,56 3rd Grade 85-75 kg L.wt. 1.74 3rd Grade " " 63-66% 1.49 4rth Grade > 100 kg L.wt. 1.60 Fresh eggs 1.85 Fresh eggs 2.20-2.30 Live CattLe SpeciaL Meet yieLd > 150 kg 2.10 (ALL procurements at 1st Grade " " 120-150 kg 2.04 negotiated prices) 2nd Grade " " 90-120 kg 1.98 3rd Grade n n < 90 kg 1.92 Live! Shoes) let Grade Meat yieLd > 16 kg 2.10 (ALL procurements at 2nd Grade " " 12.5-18 kg 2.00 negotiated prices) 3rd Grade " " 9-12.5 kg 1.90 Live Goats let Grade Meat yieLd >12.5 kg 1.90 (ALL procurements at 2nd Grade " " 9-12.5 kg 1.80 negotiated prices] 3rd Grade " " 7.5-9 kg 1.72 f Prices were State-administered through ApriL 30, 1985. Since May 1, 1985, they represent "target" prices, which may fluctuate within a prescribed range according to suppLy and demand. ALL prices are for Liveweight. Meat yieLd refers to carcass weight, incLuding fat and bones but excLuding head, entraiLs, and hide. - 121 - Anals 1 IT.b Le ]1 LNFET!CK S1CIT SIT!DY Price . f lv.stock .nd Fish Products, Tinin, 1984 lAnnuall .nd 1B85 1.anthiy (yu@rVkg) Type of Avg. 18o5 Product Price 1984 Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nay Pork Z_ Whoteeale 1.96 1.98 1.96 1.96 1.986 2.68 2.68 2.68 2.88 2.68 2.68 2.68 Retail 2.20 2.20 2.20 2.20 2.20 3.00 3.00 8.00 8.00 3.00 3.00 8.00 Market 2.92 3.00 2.60 2.60 8.00 3.22 2.92 2.80 2.94 3.26 3.20 Pork Z Wholesale 2.00 2.00 2.00 9.00 2.00 2.78 2.76 2.78 2.76 2.78 2.76 2.76 Retail 2.28 2.28 2.28 2.286 2.26 3.12 3.12 8.12 3.12 3.12 3.12 3.12 Pork Ic Wholesale 2.04 2.04 2.04 2.04 2.04 2.80 2.80 2.90 2.80 2.80 2.80 2.80 Retail 2.40 2.40 2.40 2.40 2.40 3.80 3.30 8.30 3.30 3.30 3.30 3.30 Free Market 2.40 8.20 3.20 8.20 3.20 3.60 3.60 3.60 3.80 3.60 3.60 3.30 Beef Wholesale 1.80 1.80 1.80 1.80 1.80 3.78 3.78 3.78 3.78 3.78 3.78 3.78 Retail 2.00 2.00 2.00 2.00 2.00 4.20 4.20 4.20 4.20 4.20 4.20 4.20 Free Market - 2.00 2.00 2.12 2.12 4.40 4.00 4.00 4.00 4.00 4.40 4.40 Chicken ZA Wholeasle - 1.78 1.78 1.78 1.78 2.60 2.20 2.10 2.10 2.10 2.10 2.10 Retail - 2.00 2.00 2.00 2.00 3.10 2.50 2.40 2.40 2.40 2.40 2.40 Free Market - 3.00 8.00 8.00 8.00 3.80 3.40 8.40 3.40 3.40 3.40 3.40 Chicken ZS Wholesale 1.96 2.42 2.42 2.42 2.48 3.40 3.18 2.90 2.90 2.90 2.90 3.12 Retail 2.20 2.72 2.72 2.72 2.72 8.80 8.42 8.24 3.24 3.24 3.24 3.48 Free Market - 3.30 3.30 3.30 3.80 3.90 3.60 3.80 3.72 3.80 3.60 8.80 Chicken LI Wholesale 2.30 2.90 2.90 2.80 2.90 4.20 8.40 3.40 3.40 3.40 3.40 3.70 RetaiL 3.80 3.20 3.20 3.20 3.20 5.70 4.00 3.80 8.80 3.80 3.80 4.10 Chicken Lg Wholesale 1.90 1.98 1.96 1.8B 1.96 3.20 2.30 2.18 2.18 2.18 2.20 2.30 Retail 1.92 2.20 2.20 2.20 2.20 2.80 2.46 2.48 2.48 2.43 2.50 2.58 Free Market - - - - - 4.00 3.60 3.50 8.50 3.50 3.50 3.50 Chicken 1h Wholeale 2.20 2.20 2.20 2.20 2.20 3.58 3.18 3.00 3.18 3.18 3.18 3.30 Retail 2.56 2.46 2.46 2.46 2.46 4.00 3.42 3.42 3.42 3.42 3.48 3.70 Free Market 8.00 3.00 8.00 3.00 3.00 4.40 3.70 8.70 3.70 3.70 3.90 3.70 Chicken Li Wholeeale 2.40 2.40 2.40 2.40 2.40 4.12 3.80 3.80 3.80 3.80 3.90 3.80 Retal 2.60 2.80 2.60 2.80 2.92 4.70 4.20 4.20 4.20 4.20 4.30 4.40 Eggs L Wholesale 2.12 2.12 2.12 1.94 1.94 2.40 2.40 2.56 2.40 2.40 2.40 2.40 Retail 2.80 2.52 2.40 2.20 2.20 2.72 2.72 2.90 2.90 2.90 2.90 2.90 Market 8.00 2.74 2.40 2.26 2.58 2.62 2.70 2.90 2.74 3.00 3.04 Eggs Zk Wholesele - 2.10 2.10 2.00 2.00 250 2.50 2.64 2.84 2.64 2.84 2.84 Retail - 2.40 2.40 2.30 2.30 2.90 2.80 3.00 3.00 3.00 3.00 3.00 YeLL Croeker Fish Retail 1.34 1.84 1.84 1.34 1.34 5.08 5.08 5.08 5.08 5.08 5.08 6.40 HeirtaiL Fish Retail 1.04 104 1.04 1.04 1.04 3.20 3.20 3.20 3.20 3.20 8.20 3.20 Makou Fish Retail .86 .68 .66 .88 .66 1.50 1.50 1.50 1.50 1.50 1.50 1.60 Bighead Cerp RetaiL 1.10 1.10 1.10 1.10 1.10 2.20 2.20 2.20 2.20 2.20 2.20 - Zg With bone, skin and fat. 4b With bone without skin and fat. ZI Without bone, skin and fat. Z_ Cook, Liveight. Over 1.5 kg. ZA Cock, carceas. Over 1.25 kg. L2 Cock, not of entrails. Over 1 kg. Hen, Livmeight. 1.25-1.4 kg. /g Hen, carcas. 1.10-.25 kg. .1 Hen, net of entraiLe. 0.9-1.0 kg. ,i White-ohelLed. ZI gron-aheLLed. - 122 - AW 1 LIVESTOC SW[T (]@ JEtaL JaLe Prices fou Livestock AgL Fish Products. uhabLu 197"1985 (yusn/kg) 1979 1990 1981 1992 198 1984 1985 Park LA List 1.62 2.10 2.10 2.10 2.10 2.10 4.00 Negotiated 2.74 2.52 2.44 2.56 2.70 9.70 L Market 3.00 2.88 8.90 8.18 3.74 2.88 Beef Z3 List 1.48 1.98 1.88 1.88 1.98 1.68 4.00 Negotiated 1.25 1.25 1.25 2.76 3.28 3.18 Z Market 2.60 2.386 2.94 8.40 3.48 9.50 Hutton Z List 1.99 2.40 2.40 2.40 2.40 2.40 5.80 Negotiated 3.19 8.18 3.16 3.16 8.82 9.68 LA Market 3.00 3.16 8.44 4.18 4.1 4.386 Chicken Za List 2.24 2.82 2.62 2,82 2,82 3.04 3.80 Negotiated 8.90 3.70 4.20 4.12 8.90 zg Market 3.0 4.24 4.59 4.60 4.60 9.8 Duck /g List 1.72 1.80 1.88 1.98 2.48 2.84 3.20 Negotiated 2.80 2.44 2.50 2.86 2.90 La Market 2.80 2.90 3.28 3.89 3.48 8.20 Duck Eggs List 1.80 2.10 2.22 2.29 2.96 2.50 3.00 Market 2.80 2.42 2,68 2.84 2.94 2.94 Milk Negotiated .78 Market .48 .48 .70 .70 .70 .70 .70 SiLver Corp List 1.22 1.22 1.22 1.22 1.2 1.22 2.40 Market 2.40 2.14 2.34 2.40 2.90 2.80 Gress Corp List 1.42 1.42 1.42 1.42 1.42 1.42 8.00 Market 2.38 2.58 8.20 2.80 2.80 Lj Meat weight, headLess and no entrailg about 751 of Liveweight. Lb Met weight, bonaLess. ZJ Nest weight, with bones. Z# Livesight. L Sinos the Lifting of State-edtinistered prices on Hay 1, 1985, for aLL commodities except miLk, "List" and Pnegotiated" prices have been repLaced with a singLs officiaL retaiL prios market (ATM) prices have typicalLy been 10-20 higher. - 123 - ANNEX 1 L-IVESTOCK SECTOR STUDY Tianjin Fled BU Capacity Ad -Production (tons) Rated Actual Expected Enterprise Capacity 1984 1985 Egd Processing Sate Enterprise Ji County Feed Co. 25,000 20,135 28,000 Wuqing Feed Co. 15,000 22,900 22,500 Baodi Feed Co. 15,000 14,020 22,000 Jinghai Feed Co. 10,000 17,155 21,500 Ninghe Feed Co. 5,000 2,000 3,750 Tanggu Feed Station 5,000 3,490 4,250 Hangu Feed Station 3,000 1,465 4,000 Dagang Feed Station 3,000 2,940 8,000 East Suburb Feed Station 5,000 5,545 7,500 South Suburb Feed Station 7,000 8,375 11,500 West Suburb Feed Station 7,000 5,145 11,500 North Suburb Feed Station 7,000 4,515 8,500 Nankai Feed Station 5,000 12,800 15,000 Hedong Feed Station 3,000 3,635 4,500 Experimental Feed Factory 5,000 16,060 21,500 Huajin AHIT Co. 15,000 5,956 11,800 Subtotal 1150D0 1469146 205,800 Township And Village EL%d Mills Average, 29 Mills 2,780 395La 1,103 Subtotal (29 x Average) 80j600 11,460 32,000 Private Joint Enterprises East Suburb Dabizhuang-Nansunzhuang Mill 2,000 Lb Lb Total 2170600 1 S719596 240,0r.0 LA Nine mills did not start up production until 1985. Lb Not yet in production. - 124 - ANNEX 1 iilNA LIVESTOCK SECTOR SIUDY Tianiin Poultry and Dairy Earms Zi And Production Rated Actual Expected Enterprise Capacity 1984 1985 ('000 (eggs, (eggs, Poultry Farms layers) tons) tons) Hongguang Livestock Farm 200 1,500 1,916 Lizigu Poultry Farm 100 790 1,080 Wuqing Poultry Farm 100 554 950 Huangzhuang Poultry Farm 100 254 699 Shuanglin Poultry Farm 100 441 937 East Suburb Poultry Farm 100 57 683 Junliangeheng Poultry Farm 50 218 637 Beidagang Poultry Farm 50 264 492 Panzhuang Poultry Farm 50 250 534 Dazhongzhuang Poultry Farm 50 438 563 Worker-Peasant United Poultry Farm 50 199 380 Tanggu Poultry Farm 10 56 Huajin United A.H.I.T. Co. 220 515 2,376 Food Co.'s Beicang Poultry Breeding Farm 100 5 110 Total 158 ;482z 11412 (end-1985 (milk, (milk, Dairy Farms cows) tons) tons) Tanggu Livestock Farm 450 2,596 2,680 Junliangcheng Livestock Farm 675 3,223 3,814 East Suburb Livestock Farm 633 3,529 3,762 Shuanglin Livestock Farm 722 4,410 4,273 Beidagang Livestock Farm 700 2,968 3,557 Yangliuqing Livestock Farm 500 2,817 3,059 Worker-Peasant United Livestock Farm 961 5,636 5,799 Red Flag Livestock Farm 335 1,724 1,836 Panzhuang Livestock Farm 156 590 603 Hongguang Livestock Farm 740 3,750 4,767 Wuqing Livestock Farm 271 1,045 1,113 Apricot Tree Livestock Farm 127 412 607 No. 1 United Farm 128 449 Total 6,98 320697 36316 - 125 - CHINA LIVESTOCK SECTOR STUDY ANNEX 2: INCOME AND PRICE ELASTICITIES OF DEMAND 1. Introduction. Quantitative estimates of price and income elasticities of demand are useful in planning (a) orderly growth of production capacity, i.e., increases in supply which can be absorbed by the market without serious effect on the profitability of existing producers; and (b) an orderly transition from administered pricing to a market price system. Econometric estimation of such elasticities for China from time-series data is currently impossible, as unrationed supply of important commodities is a very recent phenomenon and the necessarily lengthy data series reflecting consumer response to income or price changes is therefore unavailable. Consequently, use of a wholesale estimation method employing cross-sectional data is justified. An appropriate technique has been suggested by Frisch for computing all direct and cross-price elasticities from estimates of income elasticities based on family expenditure studies.L 2. Frisch's technique has the disadvantage that it depends on the assumption that consumer utility functions reflect want- independence for the commodity categories which are its arguments, i.e., that the marginal utility of additional consumption of each commodity is independent of the quantities of other commodities consumed. This assumption is reasonable if the "commodities" involved are highly aggregated (e.g., "food", "clothing", etc.), but unrealistic if any degree of substitutability obtains among commodities (e.g., grain, pork or poultry). An expanded version of the Frischian scheme, which works around this objection, has been applied by Bussink to Pakistani data./2 Bussink's approach is to replace the assumption of want-independence with assumptions made about the degree of substitutability (or complementarity) among particular commodity classes in the consumer's utility function. The methodology is pragmatic, inasmuch as it provides a means to translate subjective judgments about relative substitutability into quantitative estimates of price elasticities. However, even if the subjective judgments are good ones, the absolute magnitudes of the resulting elasticity estimates are somewhat arbitrary, although the relative magnitudes or rankings may be taken seriously. /l R. Frisch, "A Complete Scheme for Computing All Direct and Cross Demand Elasticities in a Model with Many Sectors," Econometrica 27: 117-96. /2 W. Bussink, "A Complete Set of Consumption Coefficients for West Pakistan," Pakistan Development Review 10 (2), 1968: 193-221. The study was replicated for Egypt in T. B. Wiens, "Estimation of Income and Price Elasticities for Egyptian Consumers from Family Budget Study Data" (MATHTECH, Inc., Washington, D.C., August 17, 1979). - 126 - 3. The Data. This study applies Bussink's methodology to 1984 family expenditure survey data for urban residents of Fuzhou and Tianjin Municipalities (Tables 1-2). The 1984 data was the most recent available. However, rationing was at least formally present in 1984, and can be expected to bias the results; a radical price adjustment occured in 1985. Consequently, the study merits replication with 1986 or later data, although the results have at least heuristic value. The data as processed was grouped into 10 or 11 per capita income classes, and contained expenditure information on 29-30 commodity groups as well as savings. The commodities within the normal published categories gf Non-staple Foods and Other Foods were disaggregated to show expenditures on various kinds of meat, eggs, and dairy products, as well as other food commodities for which relations of substitutability or complementary exist. An attempt was made to segregate expenditures on consumer durables, which fall within several categories in the Chinese survey format, and combine them with cash savings to estimate a more meaningful relationship between savings and income. Because the subcategories thus segregated differed as between Fuzhou and Tianjin, the results for both municipalities are not wholly consistent (this does not affect elasticity estimates for foodstuffs). 4. Income Elasticity Estimates. The relationships between per capita commodity expeiditures and income were estimated by linear regressions, weighted by the square-root of the number of persons per income class in the sample to reduce heteroskedasticity. Double-log, semi-log, and quadratic equation forms were also tried, but with few exceptions failed to improve fit; consequently only simple linear forms were used in final estimates (Tables 3-4). Although the estimates of individual commodity regression equations were unconstrained, in fact for both Fuzhou and Tianjin, the sum of the intercept terms for all commodity expenditures and savings is very close to zero, and the sum of the marginal expenditures and savings coefficients close to unity, so that the normal constraints on linear expenditure systems were essentially satisfied. For further use, the 10-11 per capita income classes were grouped into three ("low", "middle", and "high") income groups, as shown in Tables 3-4. 5. Substitutability Matrix. Bussink's methodology, as described above, introduces a matrix of subsitutability factors designed to translate subjective judgments about the extent to which commodities were substitutes or complements into a quantitative measure of the effects on the marginal utility of goods i of increased expenditure on goods j (see Mathematical Annex for details). This factor was defined so as to have a range of -1 to +1, where +1 represented perfect substitutability, 0 represented "want- independence", and -1 represented perfect complementarity. Presumably (although this was not demonstrated) two commodities which were more "substitutable" (or less complementary) than any two other commodities would have to have a substitutability factor lying higher on this scale than those of other pairs of commodities. The choice of numerical values is otherwise arbitrary, and will affect the magnitude of estimated price elasticities. 6. However, the derivation of these factors is not without theoretical problems. For example, although the value of the substitutability factor would be +1 for perfect substitutes, it would - 127 - be the same for commodities which are less than perfect substitutes in an intuitive sense. E.g., if consumers always regard two kgs of chicken as equivalent to 1 kg of pork, then pork and chickens would have a substitutability factor of +1 on the Bussink definition (assuming relative prices reflect these consumer preferences). On the other hand, under perfect complementarity, e.g., consumers always require a certain amount of soy sauce to consume with a certain amount of pork, not only do the mathematical characteristics of the utility functions fail to satisfy those assumed by Frisch and Bussink (they are not everywhere differentiable), but also (a) the symmetricity breaks down (e.g., soy sauce might be "perfectly complementary" to pork, but not vice versa, since soy sauce is also used for purposes other than flavoring pork); and (b) there seems to be no clear reason why perfect complementarity should require a substitutability factor of -1 (as opposed to -5 or -0.3). 7. Fortunately, in the current disaggregation of commodities, complementarity is only expected between condiments and other foods, and between dairy products and sugar (as even fresh milk is invariably sweetened). In choosing absolute magnitudes for substitutability factors, a maximum value of +0.3 has been chosen (between poultry and pork), compared to +0.4 in Bussink's Pakistan study. All other factors have been sets at +0.2, +0.1, 0, or -0.1. Substitutability or complementarity has been evaluated only for commodities in the food group, although some of the other commodities (e.g., books and school expenditures) cannot be characterized by want-independence. Although values (Table 5) were assigned by an individual familiar with Chinese food consumption patterns, the subjective nature of these values and their impact on estimate direct- and cross-price elasticities should be clearly recognized. 8. Income Elasticity of the Marginal Utility of Money. A choice of values for this variable (Z) at each income level is required to complete the calculations. As Bussink points out, the empirical data (mostly for developed countries) suggest an average value in the range -1.5 to -3.0; within this group of countries, there appears to be some negative relationship between the average value of Z and the average per capita income of the country. Moreover, common sense suggests that Z will be rather high for individuals at the bottom of the income scale within a country -- if one is near subsistence, the first addition to income will be much more valuable than the next, etc. On the other hand, the value of Z for wealth individuals cannot be too near zero, lest extremely high price elasticities be implied -- it is hard to believe that the rich are violently price-sensitive. 9. As Bussink concluded, the above evidence suggests an approximately hyperbolic function with a value at mean levels of income prevailing in China within the range found for other countries. The functions chosen for this study were of the form: Assumption A: Z - - a/Y Assumption B: Z - -1.0 - a/Y The coefficient "a" was chosen separately for Fuzhou and Tianjin (1,395 and 1,525 respectively) to yield values of 2 at the mean levels of income in each municipality of -2.25 and -3.25 for Assumptions A - 128 - and B respectively. The second function puts the average Z at a level slightly higher than has been found for developed countries, and also flattens out the Z-curve to prevent Z from falling to low for upper income groups. Overall results are plausible under either choice of Z-function, and both sets of results are reported in this annex and in the text. 10. Price Elasticities of Demand. The alternative estimates of direct- and cross-price elasticities derived from the data and assumptions described above are shown in Tables 6-8. If want- independence is assumed (Table 6), price elasticities depend only on income elasticities and budget proportions, as well as assumed Z- values. In this case the relative magnitudes of own-price elasticities tend to parallel those of the income elasticities, and the absolute values tend to be small. Under the assumption of want- dependence (Table 7-8), measures of substitutability (and complementarity) have been introduced into the calculations which have an impact on the price elasticity estimates. The impact is small where little substitutability or complementarity occurs, but tends to increase the absolute value of own-price elasticities where substitutability occurs and decrease them where complementarity occurs. That is, where it is easy to find a substitute for a commodity, the effect of an increase in its price on the demand for this commodity will be greater than otherwise; conversely, complementarity tends to stabilize demand when the price of one of a pair of goods is increased. Also, it should be recognized that the greater degree of disaggregation within the food commodity group allows for considerable substitution possibilities and results in higher absolute values of estimated price elasticities. A further disaggregation of other commodiites would have similar effects on their elasticity estimates (however, the price increase referred to would then be for a smaller subset of commodities, rather than the entire group). 11. Cross-Price Elasticities. Under want-independence assumptions, cross-price elasticities are due entirely to the income effect, that is, the effect of a price increase for one commodity on the amount of money left over for the consumer to spend on other commodities. In this case, cross-price elasticities are small except where (as with foodgrains and housing) the relationship is between a commodity on which a large proportion of income is spent and another commodity with a high income elasticity of demand. Under want- dependence assumptions, cross-price elasticities are a combination of income and substitution effects, and can have any sign. Given the substitutability matrix assumed, positive and large cross-price elasticities are found among the meat group (the elasticities are not symmetrical). For example, a one percent increase in the price of pork may increase the demand for poultry by something in the range 0.18-0.38 percent, other things (including the price of poultry) being equal. The reverse relationship, and all other cross-price elasticities estimated, are not so large in magnitude. - 129 - ANNEX 2. MATHEMATICAL APPENDIX (From W. Bussink, ok cit., Appendix A) Variables are verbally defined as they are introduced. A complete set of definitions is given (Subsection B.5) at the end of this Appendix. All incomes and expenditures are defined in per capita terms. Except for Subsection B.4, all relations refer to one particular income group. B.1 Income Elasticities per Income Group For each level of income Y, the savings relation derived in Section III of the main text specifies the absolute savings level S, the average savings rate s and the marginal savings rate i. Similarly, the Engel functions specify for each consumption item the absolute expenditure level Ai, the average budget proportion a and the marginal budget proportion, a. These proportions relate to total expenditures A but, given the savings rate, they can be converted to proportions of Y: A A a i , (1-s)a ..........................(1) dY Ai (1-s) at A dAi = (1-&)& ....................... (2) dY dA (1-s) The income elasticity is defined as dAi A a i (1-s).a.(3) Ei di i ( i-s)a, Since all items on the right-hand side of (3) are given, the income elasticity can be calculated. B.2 Price and Cross Elasticities per Income Group B.2.1 The decomposition of the price and cross elasticites: The price and cross elasticities eik are divided into a substitution part S and and income part V : ik ik ei ik ik......................4 - 130 - Using p and q to denote prices and quantities, respectively, the substitution component is defined as dq k __i ki S . .. .. .. .. .. .. (5) ik dp qk under the condition that the marginal utility of money is kept constant. Because of this last condition, the decomposition of the total price elasticity into income and price effects, as adopted in this paper, is different from the one introduced by Slutsky. Under the Slutsky definition, the indifference level is held constant, and this does not imply a constant marginal utility of money (income). However, changes in income influence the consumption pattern exclusively through a change in the marginal utility of money. Therefore, it is more efficient to include all such effects in the income part of the price elasticity, and to exclude them from the substitution elasticity. B.2.2 Derivation of the substitution effects: If U is an indicator of the total utility of a "representative consumer," the marginal utility of income can be defined as dU and the marginal utility of good i as ui = dU ........................ ........ (7) dq i In equilibrium, the marginal utility of one piaster spent on any good should be equal to the marginal utility of income (or money): - w i pw .. .. .. (8) The elasticity of marginal utility of good k with respect to consumption of good i can be defined as Zi duk . qj d2U .i dqi uk dqi k uk d2U Ai = ut A . ... (9) dAi dAk * i - 131 - Consider the effect of an increase in p by 1 percent, leaving all other p and w unchanged. Using (8), this implies that in the new equilibrium, u will be 1 percent higher, and that all other uk (k#i) remain unchangid. One can then write: ESij jk = 0 for kWi (10) ESij zji 1 If there are m sectors, the m relations (10) express the mS , S i2-S m as functions of the Z . Assuming price increases in oger sectors, the same relations can e developed for all other S . Thus, all S can be expressed in the Z . In matrix terms, the S m rix is the i4erse of the Z matrix. The Zik can be determined as follows: Analogous with (9), define the elasticity of the marginal utility of money with respect to income as: dw Y d2U Y .................. (11) dy w dY2 v Consider the effect of an increase in Y by 1 percent, leaving all p unchanged. According to (11) and (8), w and all ut will then increase by Z percent. One can then write: ZEi ik - Z fork=1toa .. .. .. .. .. .. (12) ik These m relations determine the system if all Z can be expressed in the m diagonal elements Z . This can be done in kreasonably realistic manner by assuming relationsimong the second derivatives of the utility function: U,ik = d2U ............................ (13) dA, dA This derivative indicates the change in marginal utility of expenditure on k, caused by an increase of one piaster in expenditures on i. If two goods are perfect substitutes, u/ik will be equal to u'ii; if they are not related, u'ik will be 0. - 132 - Therefore, a substitutability factor Xik (-1< x <1) is hereby introduced such that u'ik - 0 if Xik 0, and u'ik - u if Xi - 1. Because ulik = utki the link between u'ik and ui has to be symmetric in i and k.* Thus, u v (u' + u) (14) ik Xik ii kk '-- ik ii kk Substitution of (14) in (9) yields: A iai+Z(5 ik Xik (uii+u kk kxik (ii a k kk)(1 Thus, the relations (15) and (12) express the unknown Z in X* Z and E , which are all known. Once the Z have been derive k the reations (10) yield the-Sik* ik B.2.3 Derivation of the income effects: Consider the effect of an increase in p by 1 percent leaving all other p unchanged. This will tend to increase tAe total amount spent by ( S a + a percent. If, however, total income is also unchanged, these icNaes ii expenditure decrease the amounts available for consumption in the same way as a decrease in income would. Thus, the income elasticities determine the division of this income effect, such that: Vik - Si a + a ) Ek.. .....(16) Given this, the total price and cross elasticities are given by (4). B.3 The Special Case of Want-independence Under special conditions, the relations derived in this Appendix can be greatly simplified. These conditions are those of want-independence. A commodity is defined to be want independent if any increase in its consumption does not influence the utility of any other goods. In other words, good i is want-independent if Xik = 0 for all k#i. This implies that all Zik and S = 0 for k#i. Since u ik and u ki are second derivatives of a continuous function, symnetric in i and k. - 133 - In this case, relation (12) becomes Ez Z or z ... .. ... .. (17) and the relation (10) is simplified to: 1 E S z - 1 or S 1 i.(18) ii ii ii z Z ii Furthermore, (16) becomes Vi= - aiEk 1 + .. (19) which implies for (4) e Vk - a E 1 ( E) for k*i (20) ik ik k Z and e S +V - -aE +- ii ii ii z ii z Given these simplications, it is clear that the assumption of want independence is an attractive one. It can be proved that this particular simplified formula has geneal validity: it also applies when the goods are not want independent. In other words, the income part of the price elasticities is the same, whether goods are want-independent or not. - 134 - B.4 Weighting of Elasticities over Income Groups Composite elasticities can be derived by weighting over income groups. In the case of price and cross elasticities, such weighting will generally be indicated, since price changes for a certain commodity will normally be identical for all income groups. However, in the case of income elasticities, weighting over income groups only makes sense if all incomes go up in the same proportion. This will generally be an unwarranted assumption, and weighting will then lead to loss of information. Nevertheless, to limit the amount of coefficients to be presented, income elasticities have in this paper been weighted to arrive at coefficients for three main income categories. In weighting elasticities over income groups, the weights Wn have to be proportional to the amounts spent on good i by each income group n . Thus: n n ~n YnAin.. **....... .. .. .. .. .. .. (21) n y Ai in which yn gives the income distributions, such that ZY = 1 n The weighted elasticities then are: - Z n E ' (22 i n i .. ...... .. ....... ....(22) and e F. Wn~en(3 ik n k k* .. .. .. .. .. (23) B.5 Symbols and Definitions All variables are defined per income group, except as otherwise noted. All incomes and expenditures are in per capita terms. A = total expenditure in Pt./month. A, expenditure on good i in Pt./month. A1 at = - proportion of expenditure spent on goods i. A a = = proportion of income spent on good i. - 135 - dA a - - marginal proportion of expenditure spent on good i. dA~ -A d - marginal proportion of income spent on good i. A dq ._ dAi A i Y i Y a T---- . - income elasticity of good i. i dY qi dy E Wn En - weighted income elasticity of good i. dqk i eik i . - elasticity of good k with respect to price of i. eik iVne q nk k ek weighted elasticity of good k with respect to price of i. Pi price of good i. q- quantity of good i. S - total savings in Ra./month. S S = E proportion of income saved. .dS - marginal proportion of income saved. dqk S k -d -- * substitution component of elasticity of good k Ak i qk with respect to price of i. U = utility indicator for "representative consumer." dU ui - = marginal utility of good i. dq1 - 136 - d2U ud U = second derivative of utility with respect to amount spent ik AdAk on goods i and k. V = eik -S = income component of elasticity of good k with to price of i. Wn E i = proportion of good i consumed by income group n. i ynA dUJ W = - marginal utility of money (income). Ud ik U ik = substitutability factor betwen Xik u l (uii+u. kk) goods i and k. ii k Y = A+S = total income. yn =proportion of income earned by income group n. dw Y Z = dy . w = elasticity of marginal utility of money with respect to income. dk iosmto o odi U . -- elasticity or marginal utility of good k with respect 1 jqik uk to consumption of good i. - 137 - TabLe I LIVESTOC SURlB Vljf.l HousehoLd Income and Expendituresa Fughou, 194 ISMaLe Surveyl Income cLass (Y/month)s 20-25 25-30 30-35 35-40 40-45 45-50 50-55 55-60 80-70 Over 70 Average # of HousehoLds I 9.00 5.00 18.00 14.00 17.00 48.00 19.00 18.00 19.00 9.00 175.00 # of Mambers per Family I 5.80 5.67 4.95 4.38 4.20 4.33 3.91 3.91 4.02 3.00 4.30 # of BEpLoyed Members 2.58 2.33 1.93 1.12 2.00 1.54 2.43 2.11 2.50 2.20 2.42 Ter Caitam TotsL Income I320.78 355.92 489.08 527.03 554.39 817.43 705.90 780.44 857.84 1,131.22 820.24 TotaL Expenditure /a 307.10 379.85 428.84 487.98 479.12 543.23 531,80 858.00 732.18 927.28 547.88 Grain 45.58 52.57 81.81 57.84 59.10 80.44 59.55 70.52 85.52 73.12 80.60 NonatspLg Foods 14.8 180 1!7 212.2g 1g S 21 S 29 37 301.39. 334.38 2 Park 18.20 27.1 28.40 ",79 38.06 42.21 38.23 38.39 54.59 55.44 38.77 Beef & Nutton 2.83 7.20 1.97 2.52 1.29 4.83 4.35 4.52 4.03 10.12 3.93 Other Meet 9.45 15.27 5.54 9.08 800 10.23 8.84 8.04 11.51 13.13 9.42 PouLtry 8.23 10.93 7.88 12.14 15.82 19.54 14.57 15.18 27.19 38.80 18.85 Eggs I18.89 10.53 18.07 18,07 17.18 17.11 19.28 18.48 27.89 24.88 18.41 Fish I27.87 42.03 35.18 48.30 48.15 40.79 52.79 49.34 58.05 58.95 45.83 VegetabLes 14.88 25.27 24.20 24.13 25.42 25.94 32.55 30.32 34.44 37.27 22.97 VegetabLe OIL l.84 8.04 8.70 10.71 8.38 9.84 9.28 8.78 12.49 13.84 8.89 Sugar 4.99 5.04 4.48 5.92 5.48 5.02 7.30 6.72 4.55 9.21 5.54 Spices 13.10 18.00 9.18 9.13 28.88 11.88 12.00 10.73 13.99 13.43 13.24 Other NonstepLes 21.19 15.52 27.88 29.50 4.41 32.78 53.32 50.88 44.55 63.48 37.48 Other Foods I 28 16 Il.1 27.551 358 32.55 395 438 58.51 64.04 71.59 39.98 Dairy Products 1.94 3.07 5.98 7.95 4.31 7.34 4.95 8.78 13.17 28.29 7.55 Tobacco, Wine & Tea 22.03 18.21 11.55 13.28 21.02 18.58 28.99 23.89 10.83 33.88 18.72 CLothing 10.85 7.29 18.02 15.01 17.78 28.02 17.78 28.49 24.24 48.82 21.78 DaiLy Use Goods 7.99 8.17 15.04 18.75 15.98 18.58 23.28 22.85 33.38 73.20 20.43 Housing Cont/Repair .90 1.02 .57 1.03 1.68 2.59 6.10 10.55 5.33 5.22 3.34 CuLturaL & Edu. Goods .84 00 1.2 2,77 .82 2.32 2.24 3.88 2.35 2,58 2.08 Books, Newspaperse,tc. 2.13 3.48 5.58 5.40 5.38 5.07 5.08 7.95 7.81 7,45 5.53 FueL I8.28 8.22 9.72 10.28 8.19 9.50 11.81 24.01 11.87 13.25 11.02 Rent 2.11 2.03 5.12 10.48 8.25 9.53 9.84 8.37 13,92 17.38 8.91 Water & ELectricity 3.30 1.51 8.12 9.00 8.40 7.43 9.27 85 9.42 17.01 7.80 SchooL Fees 3.58 8.14 4.91 5.10 3.19 8.42 7,08 4.02 Se5 7.07 5.30 ChiLd Care AS .11 5.33 2.31 1.98 .88 1.09 2.77 .17 .00 1.81 Transport .40 3e20 2,82 4.71 5.27 4.78 1,25 4,50 7.05 3,15 4.05 Communications .14 00 .23 2.37 Be 83 .31 1.4 1.15 .53 .81 CuLturaL Events 1.23 1.33 1.22 .88 4.73 2.93 4.91 5.80 3.88 2.83 3.05 Medicine 3.27 1.78 3.83 4.1 4.88 4.81 2.97 8.15 5.87 12.73 4.71 Other Expenditure Zb 1 37.88 71.89 80.54 77.74 80.50 104.25 148.87 131.42 160.81 205.31 107.29 Savings Ic 13.88 -23.83 42.22 39.07 78.09 74.20 74.00 104.4 125.88 203.98 72.40 DurabLo Goods 1e43 3,03 20.66 13,30 41.84 44.85 2838 53.99 45.71 98.73 35.57 ZA Not inaLuding purchases of consuer durabLes. Z-b Other expenditure 1 ttaL expenditure - the 8. of Listed expenditures - durabLes /C Savings - cash savings + durabLe goods purchased. - 138 - ASM L JA" L LIVERFOK Iggl]) kUgg HousehoL Incomes jol Expenditures. Tiandin. 18W 1ljAAv AburueyL Income CaLsW (Y/month)l I 20-U 5-80 30-35 35-40 40-45 45-6D 50-55 55-60 60-85 65-70 Dver 70 Aarege # of HousehoLds 2*00 7.00 33.00 83.00 12.00 131.00 160,00 1pe,00 115.0 68.0 173.00 1,000.OG 0 of Nombere per FaniLy 865 4.77 3.74 3.60 3.80 3.73 3.73 3.66 3.67 3.60 3.48 3.70 # of EmpLoyed Members 2,21 1.77 1.0 2.13 2.11 2.20 2.35 2.33 2.61 2.63 2.75 2.37 E" cgiga Total Income 299,65 333.23 30,17 453.43 516.39 573.15 629.21 699,92 747.00909.13 997.36 679 11A Expenditure ZA 29. 3ZLM 352.04 4IL.7 428.11 4841 512.2 51241 57I.0 843.50 72.38 HL.7 Grai n I67,.48 56.55 63,58 69.26 Be,79 72.04 77,79 77.21 79,15 64,56 92.91 77.23 MonetasLe fggl El 1l8 11. 152.21 1.01.2 1221 008 W. 11.4 54 184.20 Fork I8,90 28,48 21.94 26.99 29.93 33,95 35.58 41,04 44.56 43.92 52,75 36.33 Beat & Mutton 10.65 4.05 5.0 6.46 6.01 6.74 6.19 7.91 9.73 6.B0 9.74 7.29 Other Moet 2.97 1.96 3,97 3,49 4.59 5,42 5.66 5,94 6.53 6,25 7.71 5.75 PouLtry .59 1.7 4,70 4.48 5.19 5,71 9.51 7.00 9.82 9.78 9.91 7,01 Eno. 12.47 11.73 18.65 16.60 20.39 22.99 23.57" 25,35 27,61 29.75 35.47 25.50 FIu sh 7.46 9.43 11.30 12.07 11.5B 15.86 19.39 19.51 241"4 20.00 22,48 17.68 VegetabLes 93.36 19.91 19.6 15.05 24.47 25,32 25.16 29.05 29.13 30.40 34.57 2.5 Vegetable OIL I9. 10.04 9.31 10.70 9.32 10.15 11.22 12.70 12.98 14.37 16.28 12.12 suger 8 .0 1,06 1,57 2.54 2.50 3.29 2.55 2.90 2.09 3102 4.16 2.97 Bpi we IN2 4.34 6.04 6.48 6,05 6.51 7.14 7.58 7.72 9.01 9.73 7.41 Other NonstapLes e.55 19.42 16.4a 4s.es 26.61 29.w 30.23 31.42 33.90 40.62 3.41 33,45 Ot~her fe2ft I. JLMR aZf. 11.& &La 54-41 ff . HIZ ZLK 7L-07 ILK M&ZI ZRjj Dairy Products 1.13 38 4.72 8.30 5.83 7.20 7.84 5.70 6.39 10.99 9,15 7.16 Tobacco, Wine & Tea 71 22.62 20,69 27.9 23.05 29.46 30.4 35.6 374 42.57 46.94 33.72 CLothing 36.99 45.12 51.17 0.75 64.15 75.82 07.22 99.06 9.55 11S.40 133.35 in.32 Daily Use goode 3910 9,48 12,21 14.41 17.00 15.94 .59 24.29 20.33 52.11 40.29 24.75 Housi a Const/Rapair .00 .00 .30 62 .34 1.19 .73 .89 .78 20.04 7.10 3.03 CuLturmL & Edu. Boods 4.70 2.78 9.7 11.05 16.05 18.62 14.29 17.97 20.91 20.29 33.81 19.33 looks, Newspapersato. 3.78 1.14 3." 2.66 5.25 4.62 4.51 4.04 3.59 4.59 5.94 4.51 Fu*L I5.69 9.11 9.77 9.79 9SIM 11.48 11.55 11.61 11.97 11.64 14.17 11.60 Ront I2.99 4.20 4.66 4.93 9.35 6.41 7.85 9.50 10.55 14.96 14.04 8.23 Water & Electricity 6.09 3.21 4.48 4.27 5.12 4.1 5.19 5.9a 5.se 6.51 7.99 5.65 SahaoL Fees 2.94 1.75 1.46 1,45 2.29 2.48 1.66 1.50 1.16 1.48 1.52 1.75 ChiLd Care 8677 O0l 6.4 3.3 9.51 7.80 5.03 4.47 5.13 4,35 2.35 5.12 Transport 3.94 2.17 2.61 6.46 4.37 5.8 5.58.45 5.94 6.86 6.27 6.04 Communi cetions .13 A1 .10 .09 .34 .45 In 03 .52 .30 .99 AB CuLturaL Events .80 A4 .6 J7 .89 IN 1.09 1.48 1.33 1,23 1.61 1.14 Other Expenditure iS.24 10.7 13.99 6.81 12.90 18.74 17.36 15.70 15.15 5.02 17.43 14.29 SOVInga Zi2I 9.07 25.57 40.13 39.00 88.29 14.97 117.01 137.77 16.08 1685.54 24.00 107.60 DurabLe GoodsI 0.00 41.44 9.12 12.17 29.75 29.80 31.2 27.50 34.07 92.47 9.25 ZI Not inaLuding purchases af consumer durabLea. Zk Other expendituresm toteL expenditure - the esum af Listed expenditures - durabLas Z9 Savings 0 ash eswingsa+ durebts goods purchaed. LIve8tom illil NE Incam Etonticitie .g[ De~ad j.igøft auremelan E*tiM&t**. Fuzh1.10M g - At Survoy Moen - -At M~.n of Categories 1-4 - - At Man of Categaries 5-7 - At Man of Categories 8-10 [ftan incae Y 820.241 (Men Income Y 48.18) (Mgan inc=me y 6EM.21 (Mon income Y 985.91) maermssn Cefficient /a B audent msara /b nome andat s8ar/ Incone Budoet mara /b Incone udBt Smre /b Income EtGsticity ELasti ci ty Etaticity Eteaticity 0pendent Varibe Intercapt Inome R* Average Marginal of Omand Aerage Nrginet of Ommand Average Marginat Of Demand Averaga Marginal of Demand grain 42.781* .0287* .90 .104 .035 .337 .133 .035 .284 .104 .035 .338 .084 .035 .415 menetapLa foode 84.200* .2527* .997 .412 .308 .749 .455 .309 .878 .412 .309 .750 .9 .800 .8m Park 11.110 .0448* .990 .072 .054 .750 .090 .054 .860 .072 .054 .751 .087 .054 .808 Best .M8 .0m .979 .008 .008 .ag .008 .00 .8m .008 .009 .99 08 .008 994 Other met 6.976 .0089* .55 .018 .005 .298 .021 .005 .20 .018 .005 .290 .018 .005 .372 POut try -4.929 .0947* .114 .0M4 .042 1.2M1 .01 .042 1.361 .034 .042 1.229 .087 .042 1.155 Egge 8.939* .0159* .978 .038 .019 .585 .0s .019 .478 .03 .019 .58 .029 .019 .844 Fi sh 17.7849 .0451* .ø87 .084 .055 .858 .09 .055 .576 .084 .055 .86 078 .05 .728 Vegtable 10.4~5* .026e* .994 .0m .038 .881 .07 .03 .578 .049 .083 .882 .045 . .731 UegetabLe OIL 8.80* .00* .9m .016 .011 .674 .019 .011 .5M .018 .011 ..75 .015 .01l .748 ~uger .85* .005 .072 .010 .004 .441 .012 .004 .358 .010 .004 .44m .008 .004 .524 SpiDe 18.85 -.0007 .e88 .OM -.001 -.04D .00 -.001 -.020 .021 -.001 -A41 .015 -.001 -.057 other nentapLes -8.487 .044* .02 .08 .078 1.158 .054 .079 1.82 .088 .079 1.152 .071 .0 1.105. Tobac c U&n 11.841 .0111 .92 .M .014 .415 .041 .014 .384 .m3 .014 .418 .027 .014 .498 Other foæd -11.107* .015* .994 .0 .101 1.818 .075 .101 1.340 .0m .101 1.217 .089 .101 1.147 Da ry produoto -.8m .0ffl .027 .917 .M7 1.0 .018 .027 2.11 .017 .027 1.595 .00 .027 1.387 Ctohing -.5w .0000* .0 .04 .044 1.010 .048 .044 1.029 .048 .044 1. .048 .044 1.015 DaiLy Us oDds -18.18* .0629 .947 .047 .07 1.820 .0 078 2.199 .04 .0 . 1.O .M .M 1.381 Kdsmig as~t./rpsl r -0.871 .0114* .7a .006 .014 1.84 .m .014 2.241 .008 .014 1.888 .010 .014 1.380 CuLtturet a adm. goed* .004 .04 .917 .00 .0 .W .004 .004 .918 .004 .004 .m ooko, n popere, eto. 1.800 .0~2* .978 .010 .008 .745 .At1 .08 .673 .lg .m .749 .010 .008 m.80 Fusl 3.464 .0122 .m .021 .015 .728 .0m .015 .854 .02 .015 .729 .019 .015 .768 Rent -.528 .0104* .721 .01 .023 1.221 .017 .023 1.344 .018 .023 1.220 .020 .023 1.149 Vater a eseotrilty -.40 .010 .7 .015 .018 1.048 .015 .016 1.M9 .015 .018 1.048 .015 .018 1.064 9hsot fes 4.907 .008 .928 .00 .001 .09 .012 .001 .081 .009 .001 .095 .o .001 .115 Chitd mar* 3.376 -.0020 .559 .0 -.0 -1.881 .004 -.00 -.7M .0M -.003 -1.713 .000 0 -7.153 Tramsport 1.30 .004 .66 .007 .005 .702 08 .005 .624 .007 .005 .708 .007 .005 .787 Coun c ati san .2N .0M .88 .0M .001 .72 .5 .001 .851 0m 2m 7E .01 .0 .79e CiLturet oens .W54 ml048 .M .0M .006 .991 .0m .006 .987 .006 .006 .991 om .00 .0o Nedilin -.8m .0088* .849 .010 .011 1.105 .00 .011 1.159 .010 .011 1.105 .010 .011 1.078 Other goods a service -24.044 .2119* .99m .178 .212 1.224 .157 .212 1.340 .173 .212 1.223 .184 .212 1.151 Svinga -72.584* .288* .90 .188 .295 1.894 .120 .2M 2.391 .18 .285 1.81 .202 .205 1.415 * Y-astaetil signifiant *t the = Lovet. /a Eetimated from wøightad regresaione on grouped dets, the weights being the equere roots of numbers of poræne In omch atagory. _ Camputed fr= regrossion-stimated Linsear equations et welghted meane of per capita iname in esch category. CHIs LIVE8 E C T E Inom ELaticitia Af Dowand. imer Mresin i Entimatea. Tianlin. - At Survøy Haen - -At Naen of Categaries 1-4 - - At Naen of Categories 5-9 - -At Man of Categaries 10-11 (Nme anome Y 877.89) (Men inome Y 415.42) (Maen income Y 832.83) (Nan incm* Y 21.15) Budnet Shere /b Inmæ n udamt Shar /b Income Iudot ner /b Icm e Gudent Sare /b Inc=me Raressian Coefficienta / ELetialty ELeoticity Elasticlty ELemticity Dependant Varimbe Intercapt Incme R*2 Aroregn Norginmt of Danend Averege MorginmL ar Omand Averege Morgimi of Oeand Averege MarginmL of Demand Brin 42.907* .0511* .999 .114 .051 .447 .154 .051 .331 .119 .051 .480 .098 .051 .523 Nonetapte fæad 44.575* .2080* .99 .274 .209 .790 .315 .200 .m .278 .208 .747 .25 .209 .11 Park 3.829 .031* .998 .044 .039 .877 .047 .039 .814 .044 .036 .889 .04 .0a .906 0e«f 4.299* .D49* .96 .011 .05 .45 .015 .0 .321 .012 .005 .419 .010 .005 .512 Other meet .9* .0077- .997 .009 .009 .841 .010 .008 .784 .0g .0g .92 .009 .008 .g8 PauL try -.287 .0118* .9m .011 .012 1.087 .011 .012 1.~ .011 .012 1.040 .012 .012 1.027 Egge 3.749~ .852* .999 .041 .085 .84 .044 .035 .798 .041 .035 .o8m .039 .035 .8 Fiah 8.0= .028* .992 .029 .O4 .B4 .081 .024 .78 .028 .024 .891 .027 .024 .877 VgatabLeo 8.569* .0295* .998 .04 .029 .700 .050 .029 .588 .048 .029 .985 .089 .029 .760 Vegetable at .177* .0145* .997 .019 .014 .755 .022 .014 .854 .019 .014 .742 .019 .014. .807 Suger .738 .0098* .987 .005 .004 .787 .005 .004 .889 .005 .004 .755 .004 .004 .818 pi ae 2.648* .0077* .99 .012 .008 .884 .014 .008 .548 .012 .00 .848 .011 .009 .729 1 Other non.t.pL. 14.038 .0914* .99. .052 .031 .89 .095 .031 .4.2 .054 .031 .56 .047 .01 .874 Tobaon & Ine 2.591 .0495* .997 .050 .04 .9M .0f .04 .84 .050 .04e .m .04 .04 .944 Other food 15.4* .015* .997 .104 .091 .791 .119 .081 .8m .100 .091 .789 .09 .091 .M9 Deiry producto 2.21 .0074* .7S .011 .007 .898 .013 .007 .58 .011 .07 .878 .010 .007 .754 CLothing -7.414 .1471* .999 .188 .147 1.090 .129 .147 1.138 .185 .147 1.097 .189 .147 1.05 Dal Ly Use ganda -11.013 .0545* .94 .087 .054 1.47 .0M .054 2.090 .9 .054 1.52 .042 .054 1.808 Hauming oanet./repei r -8.785 .0178* .505 .005 .018 3.799 -.004 .018 -4.90 .004 .018 4.728 .008 .018 2.181 Cutturl A oduationL -8.995 .0891* .98 .O9 .039 1.Jæ .022 .089 1.739 .028 .089 1.887 . .M9 1.27 ooke, nmu pare, et. 2.332' .0092* .972 .007 .003 .45 .009 .008 .380 .007 .003 .489 .008 .003 .51 Fuel 8.3859* .0078* .99 .017 .009 .454 .029 .009 .38e .098 .008 .487 .015 .009 .O Rent -2.340 .0178* .88 .014 .017 1.250 .012 .017 1.45 .014 .017 1.273 .015 .017 1.173 Noter & eteatricity 1.247* .0m* .9 .008 .007 .781 .010 .007 .8m .009 .007 .789 .00 .007 .m9 choot fae 2.584* -.0012* .958 .003 -.001 -.470 .005 -.001 -.244 .003 -.001 -.42N .002 -.001 -.789 Child care 9.518* -.0MB8 .98 .007 -.007 -.877 .018 -.007 -.401 .008 -.007 -.774 .004 -.007 -1.740 Trenspørt 1.30 .009* .987 .m .007 .773 .010 .007 .878 .009 .007 .761 .009 .007 .822 C~unlcatione -.443* .0014* .987 .001 .001 1.913 .000 .001 4.518 .001 .001 2.045 .001 .001 1.541 Cultural evente -.058 .0019* .9m9 .002 .002 1.050 .002 .002 1.094 .002 .00 1.054 .0M . 1.038 Other goodea &ervi om* 9.83* .0M .MR .021 .07 .11 .00 .007 .21 .022 .007 .29B .017 .07 .380 govinga -101.090* .310* .988 .182 .311 1.922 .098 .311 4.59 .151 .311 2.00 .201 .311 1.54 t-*atiøtic eignifimnt et the 5% Levet. Jo Eatimated fr=m weighted regresione on grouped date, the weighto being the aquere rat. of naubere of pernn In ech ætegry. I Computed fr=m regroeon-etioated Limer aquetione at wolghted mann of per capita Incomes in escach ategery. - 141 - LIVESTOC SCUQ jjRV SubstitutobiLity Natrlx 11 G P B 0 P E F V V 9 C 0 T D F r 0 e t a g i i u o t o a r a r o h u g s g g. g n h b i u i kf i L ha d e a ri n r t t 0 i r . y t r a I a + N y b L i N P i L n o r o O L t n i 0 t t 's 8 n d a Grein 1.0 .1 .1 .1 .1 .1 .1 .1 0 0 0 0 0 .1 0 Pork .11.0 .2 .2 .8 .1 .2 0 .1 0 -.1 0 0 0 0 Bseef a mutton .1. 1.0 .1 .2 .1 .2 0 0 0 -.1 0 0 0 0 Other met .1 .2 .1 1.0 .1 .1.2 0 0 0 -.1 0 0 0 0 PouLtry .1 .8 .2 .1 1.0 .1 .2 0 0 0 -.1 0 0 0 0 Egge .1 .1 .1 .1 .1 1.0 .2 0 0 0 -.1 0 0 .1 0 Flih .1 .2 .2 .2 .2 .2 1.0 0 0 0 -.1 0 0 .1 0 VegetabLes .1 0 0 0 0 0 0 1.0 0 0 0 0 0 0 .1 Veg. OIL 0 .1 0 0 0 0 0 0 1.0 0 0 0 0 0 0 Sugar 0 0 0 0 0 0 0 0 0 1.0 0 0 0 -.1 0 Condiments 0 -.1 -.1 -.1 -.1 -.1 -.1 0 0 0 1.0 0 0 0 0 Other nonstapLe .1 0 0 0 0 0 0 0 0 0 0 1.0 0 0 0 Tobce a& win 0 0 0 0 0 0 0 0 0 0 0 0 1.0 0 0 Del ry products .1 0 0 0 0 .1.1 0 0 -.1 0 0 0 1.0 0 Fruit, etc. 0 0 0 0 0 0 0 .1 0 0 0 0 0 0 1.0 la "+" vaLues indicate substi tutes, "-" campLments. - 142 - ANNEX CINA LIVESTOCK SECTO SURVEY irect Price Elasticitie Undgr Want- Independence (Mea Income) LA Fuzhou Tianjin Assump- Assump- Assump- Assump- tion A Lh tion B Lh tion A Lh tion B Lh 1- Grain -.152 -.087 -.235 -.177 2- Pork -.340 -.211 -.410 -.293 3- Beef/mutton -.423 -.287 -.197 -.137 4- Other meat -.116 -.077 -.391 -.270 5- Poultry -.607 -.405 -.475 -.329 6- Eggs -.230 -.149 -.402 -.286 7- Fish -.298 -.181 -.393 -.277 8- Vegetables -.289 -.184 -.323 -.230 9- Veg. oil -.280 -.187 -.332 -.232 10- Sugar -.196 -.132 -.373 -.256 11- Condiments .021 .013 -.308 -.213 12- Other nonstaple -.564 -.365 -.283 -.204 13- Tobacco/wine -.166 -.107 -.434 -.312 14- Dairy products -.887 -.604 -.294 -.204 15- Fruit, etc. -.545 -.352 -.397 -.295 16- Clothing -.477 -.312 -.559 -.429 17- Daily use goods -.891 -.604 -.682 -.484 18- Housing const/repair -.875 -.597 -1.970 -1.355 19- Cult./educ. goods -.397 -.270 -.627 -.441 20- Publications -.303 -.204 -.198 -.136 21- Fuel -.303 -.201 -.209 -.145 22- Rent -.575 -.388 -.563 -.391 23- Water/electric. -.480 -.323 -.373 -.257 24- School fees -.051 -.034 .163 .112 25- Child care .454 .307 .393 .271 26- Transport -.270 -.183 -.349 -.241 27- Communications -.333 -.227 -.650 -.445 28- Cultural events -.461 -.313 -.525 -.360 29- Medicines -.536 -.363 N/A N/A 30- Other goods/services -.647 -.378 -.149 -.104 LA Ignoring possible relationships of substitutability or cample- mentarity between pairs of different commodities. Price elas- ticities are due to income effects alone. Lb Assumptions regarding the income elasticity of the marginal utility of.money (Z) were as follows: Fujian: Assumption A - z = -1,395/Y Assumption B - Z = -1 -1,395/Y Tianjin: Assumption A - Z = -1,525/Y Assumption B - Z = -1 -1,525/Y I ii ncu§äiiiiiiiiiiiiiiiiii -ev"5er:gbgb imgn-gåwi-fi e tettvtertseiit Räiiiiiiiiis i-9Hi inneiiiiiiiiiiiiiiiiiiisiiiebic.iäiiiiiiiiHi qiii a 323-§§r gb!.9r99%w9.=fäfiäiiis ka~ 0.1 ¶ m * O @ .. . . . . . . ...... ......c s=444E lg 52t1?t¶? 2f4 ffiffENONERf~fff fiegis=5@ngiif%s§19ian»asuiutie Eff naaeRäk§esånssEf a se-aie5623senasskieitgi= ifefäiiisiffiiiiiii iiiiiiiiiiiiiiiii iiiiiiiiiiiiiii lulit MITi ml i HålIi i i i.iiki i iii ii s .ii.141111i6 sasaussesesassus :asgnseses*aa uss uaseeassusas.s H iiiiiiHH Hi; iiiiiiiiiiiiiim i iii jIii iiiiiiiiii ii ess sasssssssaasass ssusasasssaussesas esansas1!enesfuse - 144 - Djri fri ELiit iftms% le& ~l~ Aea maions. Faib m g DatstaryS ¶ . I 0 4 I 7 3 I l6 l- oein -.194 -0.01 0.03 0.029 -.006 -4.018 -9.049 -0.024 4.00 -0.033 Ø- Park 4.006 -0.493 0.325 0.10 0.1 9 4.000 0.000 -0.032 0.003 4.023 - 9ase/mMton .006 0.034 -1.444 -6.002 6.064 0.00 0.031 -0.004 -0.012 -0.004 oser m~ 8.004 0.033 4.015 -.44 -4.017 0.003 0.027 -4.010 4.020 -4.008 Peut try 0.005 0.007 .273 0.00 -4.06 0.013 0.029 4.011 -0.034 -.009 o- Es 0.001 4.006 6.025 0.013 -0.010 -0.365 0.023 4.017 -0.019 -0.007 7- F:Oh -0.013 4.007 0.345 0.153 0.020 0.065 -0.411 -.030 4.047 -0.017 - V satog 0.004 -.02 -.045 4.014 4. 2 -6.020 -.022 -0.29 -.022 -0.016 %- VW. al L 4.004 0.01 4.033 4.013 -4.029 -4.007 4.009 4.007 -0.253 -.005 lo- ~ r -.003 4.006 -.006 4.001 -.011 -.006 - -. .005 4.005 -.099 li- CandImeta 4.005 4. 4.024 4.004 4.04 4.013 -0.024 4.014 -.011 4.010 i Other nonstepte 0.009 4. 4.046 4.014 4050 4.020 -.021 -0.024 -0.021 4.026 1l- Ta~a*/nen -0.006 -0.019 4.026 -0.007 4.037 4.014 -0.017 -4.017 -.017 -4.012 14- Del ry producto 0.010 -.100 -0.045 -4.010 4.019 0.021 0.018 -.005 -0.002 -.053 10- Frult, etc. -0,012 -4.023 -9.030 4.000 4.044 4.016 -.020 0.014 -4.021 -0.014 »- Ctothino -.010 -0.024 -.033 -.009 -0.047 4.017 -4.021 4.021 -0.021 -4.015 1?- Deily ,aa Oawda 4.009 -4.022 -.029 4.000 4.041 -0.015 4.019 4.029 -0.019 -0.013 1 - mousing weneJrepsir 4.002 4.004 4.005 -0.001 4.007 4.003 4.003 -0.003 -0.003 -0.002 I- Out 4./aec. eods -0.001 -0.002 -0.003 -.001 4.004 4.002 4.002 4.002 -0.002 -0.001 W- PuhtiGstigne 4.003 4.006 4.000 -.002 -.012 4.004 -4.005 0.005 -4.005 -0.004 01- FuL -.005 -4.013 -4.017 -0.005 -0.024 4.009 4.011 -0.011 -0.011 -.00 rh t 4.004 -0.010 4.013 -0.004 -0.019 -4.007 4.000 -.000 4.00? 4.006 u- hter/etatric. 4.003 4.000 -0.011 -4.003 4.06 4.006 -0.007 40.007 -0.007 -0.005 Sa esaL fafst 4.003 -4.006 -4.000 -0.002 -4.012 -.004 -0.005 -0.005 -4,005 -.004 W - hiLd ar* -0.001 4.002 4.003 -.001 -4.004 -4.001 -0.002 -0.002 -0.002 -.001 W Treaep9rt 4.002 -.005 -0.006 -4.002 4.009 -.003 -.004 -0.004 -.004 -.0<3 - emIetiana 4.000 4.00 -4.001 -.000 -.002 -0.001 -0.001 -0.001 -0.001 -4.001 UL IOPu L earente 4.001 4.003 -.005 -4.001 -0.006 -4.002 4.003 -0.003 -0.003 -.0 MSdi lclnwe 4.002 -4.005 -4.007 -4.002 -4.010 -4.004 4.004 -0.005 -.005 -0.003 W Mer god/arviwa -9.040 -0.118 -0.160 -4.043 -4.227 4.03 -0.101 -0.103 -0.104 4.073 Dotsagrys 11 12 1 14 1 is 17 is le 20 - Din 0.012 -0.110 -0.032 -.002 -.110 -.094 -#.1ø2 -ø.00 -.002 0.ø e- Fork -030 -,.060 -,01g 4.23 4.0 4.051 409 4.9 404 4.034 - mf/uentfi -.000 -0.006 -0.002 -0.03 -0.005 4.005 -.009 4.009 -.004 .003 ~- er a eet 0.002 -0.010 4.005 -.052 -0.017 4.015 -.029 -0029 -0.013 -0.010 - ouLtry -.020 -0.019 4.006 -.059 -.020 -.01 4.031 -0.031 -0.014 -0.011 - -.000 4.031 -0.009 0.010 -.030 -0.027 -.052 -.051 -0.023 -0.017 7- Fl 4h -0.032 -0.074 4. 0.02 -.0n -03 -0.123 -0.2 055 -0 0- VagtabLs 0.00 -0.044 -0.013 4.001 -0.013 -.037 4.072 -.072 -.032 4.024 o v-g. i .002 -0.014 4.004 .-0022 -0.014 -0.012 -4.024 4.023 -0.012 -.000 1- ar0:000 4.-010 40.003 0.023 4.0 4.09 401 .017 -o.00c0 4.00 Sedlgte 0. -0. -0 01 -0.0 -4.024 .05 .05 -0 .:025 18 ld..m a 8:lj 18j# tm 1i 4:W 4E 405 4.15 4.05 4.2 lo- er t&. M.002 4.040 -4.02 0 .065 4 .410 -.034 -.067 -0.066 -0.030 -03 -- ab mIn e 4 4.022 4.069 4.040 -03. -4.070 0.9 03 -0 02 -- l t eaod. ,i0.@ 4. 0 2 4.0 0.2 06 - .0 61 . 4 0 -0 0 a ". 'i :2 - 400 .012 .06 -0.46 0 .1- 4.012 4.02 4.05 .00 is- CL0l "gså. Øa. .000 4.0042 -4.02 4.067 4.0 -4.03 -§.on7 4.007 4.031 4.02 17- De-lL -~Im de 0.001 -0.011 -.00. 402 4020 4.0 4.0.2 4.1 4.000 -0.1 ID- nea G0.002 4.022 4.2 0 4.03 402 .019 -.036 -0.036 4.06 -002 l6- -at tec. 0:02e 46.07 4.005 4.028 46 41 4,0 - 4. 0 4.023 4,:02 t SemL ts 0,0 -0.02 4.003 -4.028 4.2 4.0 4.018 4.010 4.0 4006 06 l -ta 0. -004 -.. -. 0 -03 4.0 -0.0 .0.06 .0.0, 4.002 06. aspt 0.000 4.00 4.002 4.013 4.00 4.007 4.013 -.023 -0.006 4.004 ... erIa, at- - - 01. 4.0 -0.02 -4.= -0.01 -4.M@ -0.4 .Øl -0. ,,. ~ fø:*e-- 600 -401 -403 -0.010 -0.60- 4, -0. 91 -0.010 4 -.0041 -.0d, 0- ein -d,.. 0.00 -0.02 4.000 -.00 3 4.0 4. 60 4.00 4.03 4.00 -0.001 W renpor tLnM 0003 4.0 4.002 4.00 .00 4.003 -4.00 4.00 4.00 4.00 0- 0herwmet 0.00 -.09 4.0 4-.02 4.00 4.007 4.01 -0.015 -0.00 4.05 w- curt O ee -0.00 4.20 4.00 -4.032 4.01 4.00 -.32 4.03 -4.01 -.1 - Igs -0.017 .7 - -.002 -.00 0.0 2 -.055 4.0 1 4.027 4.031 -.1032 ~er .g,im 4.0 7. -4.0M 0 .3 4.7 -. .030 4037 4.015 4.007 -.062 0- S tt" 4.024 4.6 4.0 4.002 0.007 4.003 4.03 4.-00 -.05 -0.1 l- amst n« 4.00 -.01 4.013 -.001 0.012 -0.609 -4.01 -4.015 -.011 -.01 0- Me -.00 -.01 -.07 -ø.00 0.029 4.009 -.012 4.026 4.010 -0.02 - o~mato -4.017 -4.029 4.005 4.003 0.023 4.016 -0.09 4.027 -.01 -.02 S Os*r ~nt -4.02 04 -4.047 4.:#9 4.004 0.05 4.2 -4.2 -0.05 4.041 4.043 a Ta / 4.017 4.033 4.0230 -0.003 0.02 4.0017 4.09 -.012 -4.03 -.5 7s- Fl a -4,1 4.011 -4.666 4.002 0.002 4.007 40-64 40094 -004 4011 S .egal,n . 4.0 4 -4.. 4.04 -03 . 0. .4 4.027 4.030 4.027 4.06 l>- .t. im 84 2 .044 4:E 11:7 #, 4 07 : -4.:22 4 4 4.04 4.013 is .e«,....e.. 4. .0 4.. ....3 .03 -..27 -.01 -.029 4-03 4.02 ~ - -O.003 4.017 4.1 4.002 0.005 4.003 4.004 -4.009 4.017 4.001 S alt. ea s 4.01 4.029 4.824 4.03 0.U 4-01 -4.3 4.032 40 4.0 ~er i ms ta '<.pL 4OM00 4 M.0 4.07 44M0 tø1 4.021 404 L -4.m -4.00 4.007 i eh avie 4:2 48 4.0 4.8 0 : 4.-: 4.03 4.034 4.00 OM r" m 4.009 4.01d 4.815 -4.002 0.04 -.m -.011 .015 -4.017 -4.00 Fm- -te/eIwar. 4.-m -0 4 .023 .002 0.12 4.007 -4.009 4.013 -.015 4.015 34- IsaL in ,.1k 4011 4,010 4,002 0,00 -4,g 4.007 .9 4.012 -4.022 06 Sid es" -4.002 -4.004 -4.003 4.000 0.003 -4.002 4.002 4.003 -4.004 4.004 V T apS 4.004 4.009 4.007 4.001 0.007 -000Ø 4.00 4.007 4.000 -0.006 l- Kommitn i -g001 4.002 -4.002 -.0 0.001 -.001 -4.00 -.002 -0.002 4.002 0- Sitet "e"lm -4.003 -4.1k 4.03 -4.002 0.00 -4.003 4.004 4.005 4.006 4.006 3- Uadia os -4,005 4.010 4.000 -000 0.06 4.005 4.06 4-.00 4-.009 -0.009 - 1ar gs-/earvima 4.113 4.210 41 4.m :.173 -0.01 .120 -.1764-.205 -0.213 - 145 - LMOETOCK ,OETOR giM Di1ret Pd EleatjGitia " ",t-"nnce Ase=tions. Tien.inl Dataery e I 3 4 5 0 7 e i l 4 1- grin -0.365 -0,048 0.106 0.123 0.04 -0.030 -0,048 0.003 -0.073 -0.051 0- Park 0.001 -0.693 0.169 0.24? 0.33 -0.031 0.023 -0.021 0.080 -0.024 3- Bef/utton 0.010 0.039 -0.753 -0.006 0.006 0.003 0.056 -0.001 -0.014 -0.011 4- Other ment 0.014 0.053 -0.002 -1.020 -0.021 0.006 0.06 -0.007 -0.013 -.008 5- POUL try 0.012 0.102 0.092 -0.023 -1.207 0.003 0.057 -0.007 -0.020 -0.008 g- Egg 0.006 -0.030 0.029 0.027 0.001 -.530 0.057 -0.021 -0.017 -0.007 7- Fioh -0.000 0.014 0.14? 0.201 0.135 0.03? -0.763 -0.014 -0.017 0.001 0- Vegeseh es 0.011 -0.02 -0.024 -0.041 -0.042 -.02 -0.028 -0.430 -0.021 -0.027 D- Veg, oi L -0.007 0.033 -0.018 -0.029 -0.041 -0.011 -0.014 -0.00 -0.396 -0.011 1 ~ ar -0.003 -4.003 -0.000 -0.001 -0.002 -0.005 -0.006 -0.002 -0.002 -0.344 11- ODndiete 0.001 -0.029 -0.020 -0.035 -0.037 -0.02? -0.02? -0.007 -0.003 -0.006 øoher nonstapLe 0.005 -0.037 -0.028 -0.04? -0.053 -0.038 -0.037 -0.032 -0.029 -0.035 13- T~ ~ecn/wine -0.013 -0,026 -0.013 -0.026 -.031 -0.026 -.02 -0.020 -0.022 -0.025 14- Død ry producte 0.014 -0.010 -0.025 -0.036 -0.02 .020 0.036 -0.009 -0.005 -0.008 15- FruI t, etc. -0.030 -0.053 -0.025 -0.051 -8.063 -0.052 -0.052 0.021 -0.045 4.051 10- Clothing -0.059 -0.118 -0.059 -0.117 -0.143 -0.116 -0.115 -.092 -0.090 -0.112 17- Daily us* goods -0.016 -0.031 -0.016 -0.031 -0.030 -0.031 -0.031 -0.025 -0.026 -0.030 10- Housing oonst/rapeir -0.002 4.004 -0.002 -0.003 -0.004 -0.003 -0.003 -0.003 -0.003 -0.003 1- Cut./educ. goods -0, 12 -.025 -0.012 -0.025 -0.030 -0.024 -0.024 -0.019 -0.021 -0.024 60- PubLications -0.W3 -0.006 -0.003 -0.006 -0.007 -.006 -0.006 -0.005 -0.005 -0.006 R1- Fuel -0.00 -0.015 -0.008 -4.015 -0.01? -0.015 -0.015 -0.012 -0.013 -0.015 2k- fant -0.006 -0.012 -0.006 -0.012 -0.014 -0.012 -0.012 -4.009 -0.010 -0.011 £3- wter/LectriC. -0.004 -0.007 -0.004 -0.007 -0.00? -0.007 -0.007 -0.006 -0.006 -0.007 R*» OchooL fesa -0.001 -0.002 -4.001 -0.002 -0.003 -0.002 -0.002 -0.002 -0.002 -0.002 25- odhiLd osr -0.004 -0.007 -0.004 -0.007 -0.009 -0.007 -0.007 -0.006 -0.006 -0.007 20- Transport -0.004 -0.008 -0.004 -0.00 -0.010 -0.00 -0.00 -0.006 -0.007 -0.006 27- coamunIGtlons -0.000 -0.001 -0.000 -0.001 -0.001 -0.001 -0.001 -0.000 -.000 -0.001 0- cuLture.t @vente -0.001 -0.001 -0.001 -0.001 4.002 -0.001 -0.001 -0.001 -0.001 -0.001 20- Other goode/ervices -0.00? -0.019 -0.009 -0.019 -0.023 -0.01? -0.018 -0.015 -0.016 -0.018 Category 11 12 13 14 15 1e 17 18 19 20 1- grmin -0.020 -0.055 -0.096 0.125 -0.078 -0.102 -0.140 -0.419 -0.12? -0.041 3- Pork 0.100 -0.016 -0.025 -0.031 -0.021 -.02 -.040 -0.120 -0.037 4.01- 3- Oeef/iutton -0.02 -0.005 -0.00? -0.029 -4,007 -0.010 -0.014 -0.042 -0.013 -0.004 4- oer mest -0.026 -0.003 -0.005 -0.029 -0.004 -0.006 -0.00 -0.025 4.008 -0.003 5- PouLtry -0.032 -0.004 -0.006 -0.026 -0.005 -0.007 -0.009 -0.027 -0.008 -.o003 e- Egga -0.095 -0.015 -0.023 0.005 -0.020 -4.028 -0.038 -0.114 -0.035 -0.011 7- F ah -0.064 -0.010 -0.016 0.09 -0,014 -0.019 -0.026 -0.079 -0.024 -0,00 0- V:gtabLoa -0.023 -0.017 -0.027 -0.033 0.005 -0.032 -0.044 -0.133 -0.041 -0.013 0- Veg. oi L -0.004 -0.00 -0.012 -0.00 -0.010 -4.014 -0.020 -0.059 -0.018 -ø.06 ID- ~ugr -0.003 -0.002 -0.003 0.034 -0.003 -0004 -0.005 -0.015 -0.005 -0.001 11- cDndionts -0.082 -0.005 -0.009 0.001 -0.006 -0.009 -0.012 -0.037 -0.011 -0.004 12- Other nonetapte -0.030 -0.292 -0.037 -0039 -0.031 -0.044 -0.060 -0.181 -.0M -0.01 1- Tobeco/wine -0.020 -0.017 -0.434 -0,01? -0.023 -0.032 -0.045 -0.133 -0.041 -0.013 14- Del ry producte 0.001 -0.004 -0.007 -4.575 -0.006 -0.006 -0.011 -0.034 -0.010 4.003 10- Frut, sto. 4.041 -0.036 -0.056 4.038 -0.422 -0.067 -0.092 -0.275 -0.005 4.027 10- CLothing -0.092 -0.07? -0.124 -00 -0.106 -0.147 -0.202 -0.606 -0.196 -0.060 17- Deiy use good@ -0.025 -0.021 -0.033 -0.023 4.026 -4.039 -0.054 -0.162 -0.050 -0.016 10- Housing oon*t/rpeir -0.003 -0.002 -0.004 -0.003 -4.003 -0.004 -0.006 -0.010 4·.006 -002 1l- Cutt./eo. goOds -0.01? -.017 -4.026 -0.018 -0.022 -0031 -.042 -0.127 -0.039 .-0012 28- PubLiotions -0.005 -0.004 -0.006 -0.004 -0.005 -.007 -0.010 -4.031 -0.009 0.003 El- FueL -0.012 -0,010 4.016 -0.011 -0.014 -0,019 -4.026 -0.07 -0.024 -0.00 22- nent -0.009 -0.006 -0.012 -0.00 -0.011 -0.015 -0.020 -0.061 .019 -0.006 W3- Water/eLOrr. -0.006 -0.005 -0.00 4005 0.007 -0.009 -0.013 -0.038 -0.012 -0.004 £4- ~Ghoot fee -0.002 -0.002 -0.003 -0.002 -0.002 -0.003 -0.004 -0.012 -0.004 -.001 25- hild care -0.005 -0.005 -0.007 -0.005 -0.006 -0.00? -0.012 -4.036 -0.011 -0.004 20- Transport -0.006 -0.005 -0.000 -0.006 -0.007 -0.010 -0.014 -0.041 -4.012 -0.004 27- Comunictions -0.000 -.00 -0.001 -0.000 -4.001 -0.001 0.001 -0.003 -0,001 -0.000 £0- CuLturaL @vants -0.001 -0.001 -0.002 -0.001 -0.001 -0.002 -0,003 -0.008 -0.002 -.001 £1- oMer goode/serviffs -0.015 -0,013 -0.020 -0.014 -0.0?7 -0.024 -0.033 -0.098 -0.030 -0.010 Categoryi m1 22 68 24 25 8 27 29 29 30 1- grain -0.043 -0,117 -0.078 0.034 0.01 -0.073 -0.137 -0.110 -0.030 0.000 2- Park -0.012 -0.033 -4.022 0.010 0.023 -0.021 -0.039 -0.032 -0.009 0.000 3- Beef/autton -0.004 -0.012 -0.00 0.003 0.006 -0.007 -0.014 -0.011 -0.003 0.000 4- waner est -0.003 -0.007 -0.005 0.002 0.005 -0.004 -0.008 -0.007 -0.002 0.000 5- P*ultry -0.003 -0.006 -0.005 0.002 0.005 -0.005 -0.009 -0.007 -0.002 0.000 o- Egg -0.012 -0.032 -0.021 0.009 0.022 -0.020 -0.037 -0.030 -0.000 0.000 7- F1h -0.006 -0.022 -0.015 0.006 0.015 -0.014 -0.026 -0.021 -0.006 0.000 0- VegetabLes -0.014 -0.037 -0.025 0.011 0.026 -0.023 -0.043 -.035 40.010 0.000 g- Veg. al L -0.006 -0.016 -0.011 0.005 0.011 -0.010 -0.01? -0.015 -0.004 0.000 10- auger 4.002 -0.004 -0.003 0.001 0.003 -0.003 -0.005 -0.004 -0.001 0.000 11- cOndIMenta -0.004 -0.010 -0.007 0.003 0.007 -0.006 -0.012 -0.010 -0.003 0.000 1 - ther nonet*pLe -0,010 -0.050 -0.033 0.015 0.035 -0.031 -0.05? -0.04 -0.013 0.000 18- Tobmoeo/wlne -0.014 -0.037 -0.025 0.011 0.026 -0.023 -0.044 -0.035 -0.010 0.000 14- Deiry produ"s -0.003 -0.009 -0.006 0.003 0.007 -0.006 -4.011 -0.00? -0.002 0.000 10- Frult, eto. -0.028 -0.077 -0.051 0.022 0.053 -0.048 -0.090 4.072 -0.020 0.000 10- CLothing -0.062 -0.16? -0.112 0,04? 0.117 -0.105 -0.198 -0.160 -0.044 0.900 17- DelLy ~e goods -0.016 -0.045 -0.030 0.013 0.031 -O.02 -0.053 -0.043 -0.012 0.000 10- Heuaing aonst/repe -0.002 -0.005 -4.003 0.001 0.003 -0.003 -0.006 -0.00 4.001 0.000 19- DuLt./edo. goods -0.013 -4.035 -0.024 0.010 0.024 -0.022 -0.041 -.033 -0.009 0.000 W- Pubtlmtione -0,003 -4.009 -0.006 0.002 0.006 -0.005 -0.010 -4.00 -0.002 0.000 e9- FusL -0.00 0 -0.2 -0.015 0.006 0.015 -0.014 -.026 -0.021 -4.006 0.000 e- Rat -0.006 -0.017 -0.011 0.005 0.012 -0.011 -0.020 -0.016 -0.004 0.000 RO- Wtr/eLeatrlo. -0.004 -0.011 -0.007 0.003 0.007 -0.007 -0.012 -0.010 -0.003 0.000 24- SchooL fese -0.001 -0.003 -4.002 0.001 0.002 -0.002 -0.004 -0.003 -0.001 0.000 M5- ChlLd ere -0.004 -0.010 -0.007 0.003 0.007 -0.006 -0.012 -0.010 -0.003 0.000 M- Transport -0.004 -.011 -4.00 0.003 0.000 -0.007 -0.013 -0.011 -0.003 0.000 =7- wmunictimns -0.000 -0.001 -0.001 0.000 0.001 -0.001 -0.001 -0.001 -0.000 0.000 eS culturet avente -0.001 -0.002 -0.001 0.001 0.001 -0,001 -4.002 -0.002 -0.001 0.000 RO- Mher gooM/gervi00s -0.010 -0.027 -0.016 0.006 0.019 -0.017 -0.032 -0.026 -0.007 0.000 - 146 - La~I Di ret Pfie Etaoticitine j -ans. Tianjin cht.prys 1 2 à 4 5 0 7 8 g lð 1- grain -0.26 -0.065 0.057 0.052 -0.006 -0.052 -0.064 -0.023 -0.077 -0.066 2- Park -0.60 -0.487 0.110 0.159 0.248 -0.033 0.004 -0.024 0.05 -0.029 8- e=It/Mutton 0.006 0.024 -0.517 -0.00 0.0 -4.001 0.035 -4.009 -0.012 -0.011 4- Other at 0.0g 0.033 -0.003 -0.701 -0.017 0.002 0.043 -.007 -0.011 4.00 5- PDuLtry 0.006 0.067 0.062 -0.01? -0.031 -4.001 0.03 -0.007 -0.017 -0.009 Ø- Egs -0.001 -0.03? 0.013 0.007 -0.013 -0.374 0.02 -0.023 -0.021 -0.015 7- Flsh -0.004 0.002 0.09 0.130 0.003 0.01? -0.530 4.015 -0.010 -0.007 0- VgtabLes 0.002 -0.031 -4.022 -0.040 -0.043 -0.031 -0.030 -.303 4.024 -0.029 a- Veg. alL 4.007 0.017 -0.015 -0.025 -0.034 -0.012 -0.015 -4.010 -0.276 -0.012 10- Rugar -0.002 -0.003 -0.001 -0.002 -0.003 -0.004 -0.005 -4.002 -0.003 -0.237 11 ondie" te -0.001 -0.023 -0.015 -0.027 -.030 -0.023 -4.023 -4.007 -0.005 -0.007 12- ~ther nonstapte -0.040 -.026 -0-0. -0.04 -LM 4.0 43 -0.032 -0.03B 1- Tobaco/ins . 16 -0.032 -0.016 -0.031 -0.03 -4.031 -0.031 -.025 -0.026 -0.030 14- Dail ry product* 0~ -0.010 -4.018 -0.027 -0.023 0.011 .M -4.M -0.M -0.063 15- Frult, etc. --.m --40- -,30- -0,g6i --g0 . - -*,0l1 --.,gt-. --4,6 ---4..4 -- 0 10- Ctothing 4059 -4.118 -.059 -0.117 -0.143 -0.116 -4.115 -.092 -4.09 -0.112 17- Dei Ly u. goodi -0,016 -0.031 -4.016 -0.031 -0.03 -0.031 -0.031 -0.025 -0.026 -0.030 10- HOMBni annZ/rpair -0.002 -0.004 4.002 -0.003 -0.004 4.003 4.003 4.003 -0.003 -0.003 1- Oult./adc. goodt -0.012 -0.025 -4.012 -0.025 -0.030 -0.024 -0.024 4.019 -.02 -0.024 20- Publieations -0.003 -.006 -4.003 -0.006 -0.007 -4.006 -0.006 -0.005 -0.005 -0.006 21- FuaL -0,0. -0.015 -0.008 -0.015 -0.019 -0.015 4.015 -0.012 -0.013 -0.015 22- pent -0.006 -0.012 -0.006 -0.012 -0.014 -0.012 -0.012 -.009 -0.010 -0.011 23- Watsr/*LctPic. -004 40.007 -0.004 -0,g07 -0.00 -0.007 -4.007 -0.0 - 4g.06 -,07 +- ~ooL fase -0.001 -0.002 -0.001 -0.002 -0.003 -0.002 -.002 -0.002 4.002 -.002 25- CilLd cara -0.004 -.007 -0.004 -0.007 -4.00 -0.007 -0.007 -0.006 -4.06 -0007 20- Transport -0.004 -.00M -0.004 -0.000 -0.010 -.00 -0.0M -.006 4.007 -0,.M 27- Omeuntiøn, -,000 -0.001 -.000 -0.001 -0.001 -0.001 -4.001 -.000 4.000 -4.001 29- CutturaL «ant. -0,01 -0.001 -0.001 4.001 -0,002 4.001 .091 -.001 -.001 4.M0 29- Other g d8/sirvions -0,009 -0.019 -0.009 -0.019 -0.023 -O.019 4.019 -0.015 -0.016 -.010 Categarya 11 12 IS 14 15 10 17 10 10 20 1- sin -0.039 -0.059 -.093 0.061 -0.02 -.110 -0.151 -0.452 -0.139 4.044 2- Park -0.078 -0.019 -0.029 -.030 -0.025 -.035 -.0M -0.143 -4.044 -4.014 a- sffi/utton -0.018 -0.006 -.009 -0.022 -4.000 -0.011 -.015 -.045 -0.014 -.004 4- Other mat -0.020 -0.004 -.006 -.022 -0.005 4.007 4.010 -0.030 -.009 -.003 5- PuL try -0.024 -0.005 -0.007 -0.020 -0.006 4.00 -0.012 -.035 -0.011 -0.003 e- Egs -0.074 -0.018 -4.028 0.049 -0.024 -0.033 4.045 -0.135 -0.042 -4.013 7- FIh -0.050 -0.012 -0.01? 0.062 -0.016 -0.023 4.031 -0.094 4.029 -0.009 - VsgtabLoe -.025 -0.020 -.031 -0.032 -4.007 -0.037 4.050 -4.151 -0.046 -0.015 0- Veg. niL -0.007 -.009 -0.014 -0.009 -4.012 -4.016 4.022 -0.067 -0.021 -0.007 i ~ ugar -0.003 -0.002 -0.003 0.022 -.003 -0.004 4.006 -0.017 -.005 4.002 11- Candinents -0- - 0.009 4.002 -0.007 -0.010 -0.014 -0.042 -4.013 -0.004 12- Oher nonstapL -O.032 -0.210 -0.041 -0.03B -0.035 -0.04 -.066 -0.19 -.061 -.020 13- Tab.o/1n. -0.025 0.021 -0.312 -0.024 -4.028 -0.039 4.054 -.162 -0.050 4.016 14- Dal ry product* -0.002 -0.005 -0.00 -0.396 4.007 -4.00 -0.013 -.038 -0.012 -.004 15- Frult. etc. -.048 -0.042 -0.066 0.045 -0.312 -0.07 4.107 -0.321 -0.099 -4.032 10 Clothing -.092 -0.079 -0.124 -0.0M -0.106 -0.147 -.202 .606 4.16 -.060 17- DeiLy u~. g~ods -0.025 -0.021 -0.033 -.023 4.028 -.03? 4.054 4.162 -0 4.016 1- Housilng mnet/repair -0.003 -0.002 -0.004 -.003 -4.003 -0.004 -.006 -0.018 -0.006 -.002 10- CuLt./ode. gode 4.019 -.017 -0.02 -0.018 -0.022 -0.031 -0.02 -0.127 4.039 -0.012 Ø- Pubiation -0-5 404 4006 -0.004 -.005 -0.007 -0.010 4.031 4.009 -.003 21- Fumt -0.012 -0.010 -0.016 -0.011 -0.014 -0.01? 4.026 -0.079 -0.024 -.000 22- Fant -0.009 -.00 4.012 -.009 -4.011 -0.015 -.020 -4.061 4.019 4.006 23- Water/laectric. -0.006 .005 -4.00 -.005 -4.007 -.009 4.013 -0.038 -0.012 -0.004 24- School fm& -0.002 -0.002 -0.003 -0.002 -0.002 4.003 4.004 -0.012 -.004 4.001 25- ChiLd care 4.005 -.005 -0.007 -.00 4.006 -.00 -0.012 -.036 -0.011 4M4 W- Transport -0.006 4.005 -0.00 4.006 -0.007 -0.010 -.014 -4.041 4.012 -.004 27- cmwniceaione -0000 4.000 -0.001 -0.000 -4.001 -0.001 -0.001 -0.003 -0.001 4.000 20- CutturaL 'vent. -0.001 -0.001 -4.002 -0.001 -0.001 -4.002 4.003 -000 4.002 -0.001 20- Other god/GarviO -0.015 -0.013 4.020 -0.014 -4.017 -0.024 -.033 -0.098 4.030 4.010 Categarys 21 22 a 84 25 m 27 U te 30 1-- rn -004 -.126 -.004 0.037 0.067 -.07t 4.148 4.11? -.032 0.000 2- ark -0.015 -0.00 -.027 0.012 0.028 4.025 4.047 -.03 -0.010 0. 9- B=f/mutton -.005 -0.012 4.00 0 0.04 0009 4.008 -.015 .-0012 -0.003 0.000 4- other mat -.003 -4.009 -.006 0.002 0.006 4.005 -.010 -0.008 4.002 0.000 5- tuL try -.004 -0.010 .006 0.003 0.007 4.006 -0.011 4.00 -.002 0.000 0- Egge -0.014 -.038 -0.025 0.011 0.02 -4.023 -0.044 4.036 -.010 0.000 7- Fiah -.010 -0.026 -9.017 0.0M 0.018 -.016 -0.031 -0.025 407 0.00 - VgtabLee -0.015 4.042 -.020 0.012 0.029 4.026 4.049 4.000 -0.011 0.000 1- Veg. al L -.007 -.019 -.012 0.005 0.013 4.012 -.022 -.010 -.005 0.000 1a- ugar -.002 -.00 -.003 0.001 0.003 4.003 -0.006 -4.004 -4.001 0.00 11- Condlmnt -.M4 -0.012 4.0M 0.03 0.M 407 4.014 -0.011 -4.«3 0.00 1 ~- Oher nontpLe -O.M0 -.M -.037 0.016 0.03 -4.034 4.065 4.0m 4.014 0.M 18- Taouo/lne -.016 4.0 -0.030 0.013 0.031 -.028 4.M3 -0.043 -0.012 0.000 14- Dal ry producte -0.004 -0.011 4.007 0.003 0.007 -0.007 -.013 4.010 4.003 0.M 15- Fru t, st. -.033 4.090 -.05 0.026 0.0 2 4.M6 -4.105 -*.M -0.023 0.000 le- CLt~ilng -4.062 -0.19 4.112 0.049 0.117 -0.105 4.19 -0.160 -0.044 0.000 17- Daily u pode -.016 -0.0M -0.030 0.013 0.031 -.028 4. 3 4.043 -0.012 0.000 19- Houeing acnot/rapeir 4.002 -4.m -4.003 0.001 0.003 -0.003 -4.006 -4.05 -0.N1 0.000 10- CuLt./eamo. goade -0.013 -.035 -.024 0.010 0.024 -.022 -.041 4.033 4.009 0.000 20- Publiati ons -0.003 -0m9 -.006 0.002 0.Nå 4.0M 4.010 -0m 4.002 0M 21- FusL 4.00 -1.022 4.015 0.006 0.015 -0.014 4.026 -.021 -O. m.- ant -0•006 -0.017 -0.011 0.005 0.012 4.011 -.020 -0.016 -0.004 0.000 13- Mator/Otatric. -0.004 -0.011 -0.007 0.003 0.007 -.007 -.012 -.010 .003 0.000 24- Saboat fe 4.001 -0.003 4.002 0.001 0.002 -.002 -.004 -0.003 -.001 0.000 25- OIiLd ar. 4.04 ¯ ie7 0.003 0.007 4.006 ~4.012 -O.00 400 0.000 M- Transport -.004 4.011 -L000 0.003 0.000 4.007 -0.013 -0.011 4.003 0.000 27- Demuniatione -O.0 -.001 -0.001 0.000 0.001 -0.001 -0.001 -0.001 -4.000 .000 29- CuLtur*L oventa -0.001 -.002 4.001 0.001 0.001 -0.001 -.002 -0.002 4.001 0.000 æ- OMher goode/servime -0.010 -.027 -0.018 0.00 0.019 -0.017 -0.032 4.026 4.007 0.000 - 147 - a Ia±a i lLa ERTDM IL| !ml, 1 T~lat OUMer Total Totat flow PopuLe un = Mo. Uheet Mos + Crn Ceruat Coumu Totalt Tubera y n "gran hemt Provinue tion Arm Prod Pred Wheet Prod Prd Carmte Care&L Prod Prod Pred Drarjt MiL1 [MM hel [mm0 tonal NationL 1,080.5 144,121 179895 97,,15 908p70 73p418 28~ 18,99 =9,i5 2,475 9,@6 407,30 U,f1l elijing 9.5 m m 799 m m in 1,145 8,1 5 15 2,175 in Tiohjin 8.0 - ia m 405 m 210 775 1,249 40 0 1,315 57 Hebui 54.9 8,70 750 7,190 7,10 6,400 29,75 9,155 17,065 1,380 275 18,700 1,037 Shanzi 8.0 4,155 8 2,90D 9816 2,850 2,40 4,5 7,~ 740 10 8,720 38 Inner IngoLla 19.8 4,8=0 0 t,44 1,500 1,465 ,900 9,715 5,915 500 280 5,45 200 Linoning 98.5 3,737 8,310 1 3,85 7,150 ,905 10,110 18,44 im 8M 14,080 315 JlIn E.8 4,00 1,910 150 2,05 11,040 2,070 13,110 11,175 300 8113 18,845 200 HeLongjiang N.9 8,8 1,94 å, 5,08 6,4 2,590 8,~5 14,04 8 2,908 17,575 8w *totL E la A ~ 96 Z19 E z ia lsj= E RZ 9 I6 11 Z4I6E 7aiN 95.095 m @hanghei 12.0 740 1,790 140 1,820 m 5 80 s5 2,515 0 10 2,525 196 Jiangau 61.7 8,584 17,4~5 ,040 11985 8,090 8,095 ,175 31,710 1,311 430 3,585 2,^85 ZhwijIang 9.o 4,W 1519 1,1 1,g0 145 1,02 1,170 17,90 780 120 18,170 1,557 Anhui 51.0 7v117 11,380 8,485 17,= 410 m55 1,045 18,870 8,470 8f fi ,0E 1,41 FuJian 0.8 2,887 7,310 1N 7,595 0 70 70 7,575 m a 8,505 713 Jiangxi 84.8 5,457 14,80 100 15,00 10 a 45 15,075 20 135 15,490 1,417 Shandong 78.4 10,772 575 19,71 1,8M 9,00 1,80 11,1M 4,4 5,m 60 30,40 1,790 He~nn 78.5 11,425 ,890 18,A30 18,^0 5,80 1,226 6,48 2,75 2,590 970 28,035 2,458 Hubel 48.8 7,396 15,870 3,790 19,40 1,91 58 1,90 21,330 1,0M 22) 22,830 1,983 Hunan 55.8 7,839 24,185 - 84,500 240 m 40 24,0 9,15 225 28,180 2,817 guangdong 81.7 6,258 17,990 la 1,0 70 a t 18,10 1,45 120 19,725 1,702 mnxi 88.1 4,511 10,51 15 10,87 1,0 4 1,09 11,750 910 180 12,190 1,004 fubtoteL 77.gZ ZZ 5 j g ij9.155 1Z.20 W-fZi 240.20 19.21 sichuen 101.1 11,7s4 80,84 7,010 87,870 8,50 1,880 8,110 95,480 4,995 30 40,795 2,a1 Ouizhou 29.3 2,91 4,070 980 4,40 2,175 Ril 2,80 6,990 m15 IN 7,580 481 Tunnen 8.o 8,=98 5,005 en s,~0 9,740 778 m,ig g,e 805 100 10,050 511 Tibet 2.0 21 8 140 145 5 s 84 40 5 1 411 1 Bhenxl 29.7 4,70 m 4,48 5,490 8,130 ag 9,~80 8,80 e 108 10,985 8s5 ganmu 0.2 3,41 in 8,415 8,40 871 - 1,ve m 4,=5 - m 5,e5 481 inghet 4.0 507 0 675 875 0 v7 170 45 55 m 1,010 91 inoxie 4.1 w 410 730 1,170 115 190 m 1,45 40 21 1,54 141 Xuijlan 18.4 2,7e 370 a,030 9,4o 1,30 in 1,=95 4,80 20 20 4,70 445 Bubtotet gI3 5ai 6ILN i E 31, 00 11E 8 6 LM 21,276 E ZL= E E Re hiZE £. Asmuming 9.4 of rim and 18.5% of h*e are high qu&Lity bran. Figuren exotude rm huska and brokän rio. - 148 - Aa I g.LU a DA gftbt 2ten Production IX Province, lgg SunfLo- Other . Total Est. Bgar Sugar Breen G. Nanure Cotton soybean Peanet Rep war Od O'sed 0'nsed 0!sed geet Cane Fodder & Vag. Prod Prod Prod Prod Prod Prod Prod iewt Prod Prod Area Area I' oM tonal - r'00 hal - NotnaL JL. L &M &M IAAa 1M 28 J.818 1"g at19 1. 1O-445 Beijing 6 15 U 0 1 2 41 R4 18 0 Is a8 Tiajin a 90 10 0 87 a U 0 e 0 0 a al Hebei 1,050 75 400 g 76 115 875 1.18 I 0 47 87 aban1a 1se a3 6 1e 170 547 810 31 0 93 172 Inner NongoLie 0 380 0 80 481 147 8O3 25 1,410 0 113 112 Liaoning a ON 827 0 53 87 1,055 395 301 0 35 258 Jitin 0 810 10 0 537 51 1,408 390 680 0 7 1N HeLongjiang 0 2,905 3 2 341 10 8,185 534 4,326 0 147 411 Subas.uL a 1IL 601L. 1". M.& 4Z 1IM IM AMD M &AM a &A 1la. Shanghai 106 10 0 19 0 0 1N IN 0 0 i in Jiangeu ON 430 R 454 0 14 1,140 1,236 37 a 81 347 Zh%J iang 188 130 0 337 0 3 459 375 0 707 0 658 Anhui 234 695 190 710 0 SB 1,858 947 0 SO 8 543 Fu.iaon 0 U 147 10 0 1 35 108 0 4,787 10 143 Jiangxi as SO 74 145 0 a ag0 34 0 1,500 22 1318 ahandong 1,72 83 1,800 8 1 13 3,445 8,078 a 0 17 873 Henan aS 970 313 130 0 U 1,495 1,414 a 8 5 401 Hubei' 607 230 344 0 132 774 1,029 0 IS 2 919 HunMn In 25 47 31 1 5 606 413 0 1,079 a 157 Guangdong 0 180 57 6 0 a o9 57 0 17,GO3 67 30 Guengat I 100 808 4 0 7 379 13 0 7,0 46 RMs Sub.Sala E Ait AZl. I."2 LZI R II 10amZ. DIl &M.1. MalAf3 MA ZAA Sichuan 153 820 344 934 4 8 1,500 84 14 2,R47 242 MV Guizhou I I 24 373 a 8 441 0 0 143 14 3i Yunnan 1 100 31 84 7 3 334 GO 3 ,778 42 165 Tibet 0 0 0 1a 0 0 13 8 0 0 5 a Shaenxi 74 I 43 115 14 Is 858 r17 a 0 131 187 emu 8 s0 0 61 30 130 211 115 370 0 306 140 1inghai 0 0 0 78 0 8 78 40 10 0 11 10 Ningxia 0 2 0 2 7 88 .70 a gas 0 37 34 Xi njisa 1eI 0 1 48 AS 80 3M 303 61 0 140 173 MtueLa L 51 dB Al M W1 IM n RL Am dlA la .1i N1 1alt ZA The estimated production of oltesed meL Is basd Chinese estimates of neat production in 190, and are consistent with the folLowing crush and extraction rateso Goybeanse 201 oruahed, 0 extraction rate Repes 151 crushed, 5S extraction rates Peanuts S0 (of product with sheL) crushed, 745 extraction ratel SunfLwaers 0 crushad, 485 extraction ratel Other eitameda O55 orushed, 55 extraction rate (et. fto asme)l Cottoneeda 1.7 times Lint weight, 945 orushed, 05 extraction rate. The extraction rates are high by DeveLoped Country standards, ImpLying a to quaLity product; much of which is used as fertiLizer rather than feed. - 149 - oeilM IÈLLt i LIVESTMC M7i STUDY Nunbers j Liestock aMg Eg Production Dl Province, ISM Totat Total Other To;aL Total Total Human BreedIng Other TotaL Dairy Other TotaL Large Large Shuep & Egg Populatn Sos. Piga Pigu Cattte CattLe Cattle Anteat AniaLe Boate Prod. (MIL Lion) ('0on head) [*VO ti Nationet Ig"g 1.080.5 22,129. 84.072 3092 UM Mai .79 82-I8. se.es 2.e1 106.s88 158.400 4,818 Boij1ng 0.5 148 1,472 1,819 se 57 Sa 171 M07 482 12e Tiarjin 0.0 4s 807 715 14 37 51 179 a0 5ee 72 Nobel 54.9 709 11,319 12,079 se 1,309 1,34 2,784 4,110 7,258 esn Shanxi 90.0 50 3,120 3,37 37 1,253 1,290 1,212 2,02 5,078 S0 Inner lanttole 10.8 2e7 s,MI 4,128 202 3,535 8,787 8,944 8,901 as,773 e Liaoning 88.5 005 8,34 B,WB9 s 1,190 1,939 1,015 2,047 1,939 919 JitIn 9.0 404 s,eu 4,120 29 1,404 1,5M9 1,274 9,790 1,52s 100 HaiLongjiang 3E.9 831 4,227 4,9858 205 1,817 1,522 1,305 2,897 2,4M 170 »btot@L E RA lii Se. 18B 1 Z. 10.798 114Zi 22,5ER 4es10 1.0m8 Shanghai 19.0 lm 1,M 2,0W 44 94 e 0 es 289 se Jiangsu 81.7 1,105 17,097 18,139 97 807 724 10 9M 4,W8 478 Zhejiang 39.9 MO 12,482 13,282 41 743 784 0 784 1,879 1s Anhui 51.0 75 10,80 11,398 O 3,575 3.,588 5M 4,188 2,115 1es Fujian 28.8 448 7,358 7,79W EE 1,127 1,140 2 1,151 027 72 Jiangxi 84.2 773 10,015 11,888 a 2,397 9,335 0 9,335 e5 Ce Shandong 70.4 1,278 15,587 10,815 1a 9,313 9,3a 1,705 4,031 7,539 020 Henan 78.5 1,128 12,142 18,270 a 5,107 5,133 2,814 7,94 8,784 28e Hubel 48.8 1,108 18,383 17,40 10 3,0 3,101 57 3,158 1,99 324 Hunan 55.8 1,574 21,787 2,371 7 8,340 8,847 13 8,860 e9 232 8uangdong 01.7 1,519 18,900 20,475 21 5,011 5,032 0 5,032 304 107 Guangxi 30.1 97 12,527 13,500 5 5,2R 5,927 20E 5,438 717 8 ubtot@L E a 1, j 4d 157.278 10as E11.2.9Z1 U 2aiDi L0i o aa.l 8S.M g1g6 Sichuan 101.1 3,079 M9,720 m,90 97 ,2010 9,27 4.8 9,720 9,381 2M0 Guizhou 9.3 U5 10,084 10,039 8 4,849 4,854 577 5,281 1,821 48 Yunnan 38.8 1,442 15,574 17,018 N 0,990 a,E5 1,39 0,257 7,48 Sa Tibet R.0 94 101 125 103 4,703 4,99s SeU 5,814 10,245 I ShaanxI 29.7 484 0,509 8,9 2s 1,717 1,745 70 2,440 4,128 109 connu 90.9 M51 4,007 4,448 87 9,092 2,059 9,050 4,717 8,934 3 Singhai 4.0 90 884 794 20 5,260 5,90 729 0,009 13,999 a Ningxia 4.1 41 455 4gg g 1g 189 47 g70 g,09 9 Xinjiang 13.4 71 au 057 14 8,874 S,es 2,aW4 5,122 .4,477 s totaBL 1n MZ, ZM Caz7 Ou.OU 88 8a 5 M..0 L.m L. .a - 150 - iM I Eiisat l11 Mg 1hakit Eu@i F"d flegnicad ki Domtice Antasks Total TotaL Other Total Total Total TotaL Basin for Breeding Other Total Dairy Other TotaL Large Large Sheep I Egg Quality Estimates Boes Pia Pis Cattle Cattle CattLa Animals Animals Boats Prod. Food Esll it a eguiremto Tons of Dry Ntter 1.00 0.55 8.00 2.50 2.50 0.40 5.00 Percent Quality Feed 6UN 5s 3O 7Z 71 a s 000 tons) Ntional 121 S3,272 7.285 1I li Hal,3 15.4!t A,596 18..87 L188 H942 13.DR Beijing G 405 482 85 10 45 so 75 10 540 1,128 Tianjin 29 1e R2 18 8 19 81 50 11 824 598 Hebel 457 8,112 8,50 84 229 263 484 747 145 1,044 5,505 Shanxi 154 858 1,012 83 RI9 253 R2 465 101 887 199B5 Inner MongoLim 172 1,0O 1,228 18 619 900 568 1,88 475 806 8,178 Liaoning 868 2,280 2,852 82 209 142 2 524 89 981 4,196 JiLin 278 1,005 1,284 25 81 297 223 510 81 450 2,274 HeiLongjiang 879 1,162 1,541 185 80 415 228 048 50 785 2,999 Jbhilatal 3 hSM1 102071 1i M JZM 2,a4 &Mf A&E M A.MEt Shanghai 114 520 684 40 4 44 0 44 5 887 1,070 Jiangsu 663 4,682 5,845 24 122 14S 80 178 81 2,142 7,745 Zhajiang 480 8,427 8,907 87 130 167 0 167 40 729 4,648 Anhui 455 2,817 8,872 7 8 633 97 730 42 861 5,018 Fuj ion 86 2,012 2,288 10 197 217 0 217 18 824 2,642 Jiangxi 484 2,919 8,868 7 407 414 0 414 1 428 4,26 Shandong 767 4,278 5,089 12 405 416 298 715 151 2,904 8,708 Henan 677 8,889 4,016 28 894 917 492 1,410 186 1,071 6,682 Hubei BR 4,500 5,162 14 540 554 10 584 28 1,458 7,210 Hunan 844 5,994 6,18 a 585 591 2 583 18 1,044 6,56 Guangdong ti1 8,218 6,124 19 677 898 0 86 7 482 7,509 Suangxi 584 8,445 4,09 5 914 918 86 954 14 176 5,178 Subtn&Lk l 3.282 5028. M1A A=ZDi d1. 2l Il Mll 11.918. 185 Bichuan 2,867 14,488 16,685 24 1,821 1,645 77 1,721 18 1,841 20,185 Guizhou 45 2,766 1,268 7 618 620 101 89 8e 216 4,482 Yunnon 865 4,188 5,148 2S 1,207 1,281 238 1,464 148 852 7,018 Tibet 14 Re 42 147 854 980 69 1,048 825 5 1,420 Sheanxi 290 1,790 2,090 25 200 826 128 449 88 481 8,102 Gansu R11 1,127 1,887 88 459 482 850 952 177 1682 2,528 Qinghai 6 183 219 18 821 980 128 1,086 280 86 1,601 Ningxia 25 125 150 5 82 87 8 128 54 41 367 Xinjiang 48 161 204 168 468 684 896 1,030 490 162 1,85 listak Ial 4M . UMI 8838 M JI" Z& IZi LIZA 57ZZZ &ZIL M &M 4 8 - 151 - anE1 UM Idta . Beanom g Aveal Lity AM aumrememnt DL hmality f:g Lr Liva~t Cat~ A a c 0 E F a N x J K L m Caeomations [Ax0l ((9{~1+1 (J-I) (J/K) TotaL Hmmen Par Total ToteL Fina Guality Emt. Eet. 1 *rain' PopuLa- Cpita Human Boybeane 8s&d e ramin 0'ed CereLa Feed L8 Faed LO Feed SeL Prvinom Prod tion Cone. COn. Crumhed Taste For LO Hmal eran For LI Naeded Balana guff. '00 t (NIL) Kg/Cap ('00 t4 11) la /h /o Id la ln /f mtl mlak A7.80 ,3,I MI 257.419 . 0 NE2 1a d8 N.M1 M.i N 180.93 . :Z Beijln 2,175 9.5 175 1,888 3 s28 12 24 128 3e 1,1m -787 20 Ti.jin 1,315 8.0 172 1,878 7 117 -m87 et 57 -14 -98 -70 -CM1 Het 18,7W0 54.9 205 11,278 55 m ,605 4,581 1,518 1,0387 7,116 5,505 1,811 19a Uuni 8,7W0 2.0 215 8,592 88 1,308 1,784 10 am 2,491 1,115 5W7 1271 Inner Hanpatle 5,94 19.9 290 4,545 45 ese 469 m 200 Ws 8,878 -1,491 mes Limiing 14,290 9.5 L19 7,93 17 2,1389 4,01 8m 815 4,710 4,11 514 1125 J1In 18,845 22.9 279 g,8s8 IM 2,452 7,893 220 200 7,80 2,274 5,910 3471 Hateneiang 17,575 N8. 225 7,410 581 2,989 6,948 524 -= 8,105 2,=8 5,108 27 ebMUL E Mg Ri~A ffi U. 1E2 12.755 25.4 m " . & LMI 8L. 22ait LM1 JU hngai 2,525 12.0 20 2,509 2 879 -95 11 18 1E 1,070 -1,059 1% jiangeu 33,595 01.7 28 17,571 95 5,080 10,90 1,285 2,865 14,90 7,745 7,202 193 Zhejiag 18,170 89.9 295 11,767 24 29,78 8,54 375 1,57 5,5M 4,848 748 1153 Anhel 2E,=B 51.0 m 18,655 17 ~,804 mMD 90 1,s41 7,817 5,01 2,90 15 Faj.t 9,505 26.8 e83 7,598 19 1,27 -7o 109 718 442 2,842 -,400 1n Jiangxi 15,490 84.92 312 10,057 97 2,884 2,48 249 1,417 4,142 4, -4 9m andong 80,400 78.4 920 18,018 1 4,580 ,997 8,073 1,790 18,760 8,709 5,041 15 Muman 26,985 78.5 929 17,518 14 4,840 8,989 1,414 2,45M 10,758 6,68 4,117 1es lmbet 2,880 48.6 292 14,238 44 3,895 4,06 1,028 1,998 7,"8 7,10 758 111% Hann 2,130 55.8 810 17,54 45 8,92 4,909 412 2,817 7,630 8,588 -958 8 guangdong 19,725 91.7 249 15,384 24 2,ese 1,178 857 1,702 3,437 7,50 -4,072 4 91ni 12,130 88.1 - B,911 88 1,w0 88 158 1,004 1,55 5,17 -8,848 21 StataL 240.20 EA m E 154=m ZiU 86.180 A5i gIE 1i2I Z8I1 UsE8 9,5 AM M iohuen 40,705 101.1 - 5,9= 94 8,119 0,6m 943 2,81 12,40 0,195 -7,945 - guahou 7,590 29.3 01 8,474 27 1,137 -58 208 491 5m 4,432 -9,852 13% Yunnen 10,050 88.6 24 7,507 20 1,508 Du se 588 1,157 7,018 -5,358 2 Tibet 45 2.0 89 478 0 74 -57 8 19 -80 1,40 1,450 - hsmi 10,585 09.7 240 7,184 s8 1,5=5 1,58 217 905 2,445 3,102 -857 713 Saamu 5,39 20.2 924 4,515 10 80 e1 115 4m9 89 2,528 -1,8 E51 ginghel 1,010 4.0 210 - 0 159 i 48 lm 157 1,001 -1,444 13 Ningla 1,540 4.1 - 915 5 R81 -u 23 141 55 367 to 151 Xt.nji«a 4,70 13.4 IM 2,551 4 745 1,e70 09 445 2,419 1.e5 591 19e =ial g. M.o70 eZn sZ UsEl 2.811 13.7 2076 14 EZZ. MIZE Ma : Z a Frim Annex 3, TebLe 1. mhemd an ruret faily oonsumption eurveys, ruraL/urban popuLation proportion@, und amaumption that urban per capite grain consumption in 901 of rural oonasmption. Figures are for unmiLLed product. 29 5 efumd to be oruhed E Eutmated at 151 of the totaL oropp inoLudes indumtrial ue. a 9a Anne , TabLa 2. g[ es AmeX 8, TabLe 4. - 152 - LIVEST=L SEC.A Comosition JLd Crude Protein tent 91 PoteioL QueLitl Liveetock fJWA kX Provini TotaL Fine Total Protein 01Ltased Total Pro- Grain 01 Laeed Carsals Net LO Grain in in Naal in Protein Bran in Protein tain in Province For LS MeaL Bran Feed Ration Grain Ration in MeaL Ration i Bran Ration Z, LA LA 1k L1 Zd_ Za ['000 tonal -foorcent JNional Tla 8.852 iAli 11 19.U88 MSAS Z.Z 11A. AZM &LA Man MAK Beijing 18 24 123 829 55.41 B.71 7.15 40.71 87.41 13.89 12.1g Tienjin -2W7 61 57 -148 n.a 8.61 n.O. 5.8% n.s. 18.91 n.e. Hebei 4,581 1,518 1,037 7,118 84.11 8.25 21.85 85.91 14.6 14.11 15.M Shanxi 1,784 810 888 2,491 71.5 8.45 12.45 5.0 18.0% 14.4 12.81 Inner MongoLia 482 25 200 887 52.11 8.81 25.8% 84.71 22.61 14.81 16.51 Liaoning 4,001 891 815 4,710 84.91 8.81 8.4 41.25 8.71 12.11 11.85 JiLin 7,3893 290 200 7,83 88.8 8.8 8.71 88.21 2.51 12.81 9.71 Heilongjiang 8,948 524 B38 S,iM 5.75 8.8 9.5% 41.81 7.81 14.91 11.15 Sthetl ldA IK d!!. 1WIZA1 ZAAK i.m .Ze IS IR .11I. UIa Shanghai -3865 191 188 12 n.a. 9.61 n.m. 84.85 n.e. 12.85 n.e. Jiangsu 10,847 1,235 2,865 14,947 72.8 8.05 8.81 86.11 19.11 IS,J% 11.8a ZheJiang 8,654 875 1,557 5,585 a.45 7.11 0.7 5 5.75 27.0 12.3 10.W Anhui 4,930 947 1,941 7,817 83.11 5.11 12.11 87.81 24.81 8.15 1191 Fuj ian -879 I08 718 442 n.s. 8.95 n.e. 48.41 n.s. 12.2 n.e. Jiangxi 2,483 242 1,417 4,142 59.91 4.21 5.15 87.81 84.25 12.11 8.91 Shandong 9,97 3,073 1,780 13,750 64.71 7.91 22.81 87.71 12.91 4.8 14.9 Henan 8,888 1,414 2,458 10,758 84.05 7.85 18.1% 86.51 22.05 14.25 12.75 Hubsi 4,59 1,0286 1,983 7,988 82.21 6.91 12.0 85.51 24.95 12.71 12.0% Hunan 4,902 412 2,817 7,830 64.25 5.41 5.41 38.5% 80.41 12.11 9.11 Guengdong 1,878 857 1,702 8,437 40.11 8.91 10.41 43.8 46.51 12.11 12.1% Guangxi 368 153 1,004 1,525 24.11 7.81 10.11 48.51 85.81 12.11 14.2 Subtotal 4 8l1 iAiM 10JE1 ZL4. U8AI ia .1.l8 M 21.a 85.4 ALK 184 Sichuan 9,888 94 2,881 12,400 69.51 8.8 7.51 87.2 22.91 12.91 10.51 Guizhou -58 208 431 581 n.e. 7.6 n.m. 87.25 n.s. 12.45 n.a. Yunnan 985 88 582 1,657 59.5% 8.0% 5.41 80.75 85.11 12.51 11.81 Tibet -57 8 19 -80 n.e. 9.51 n.e. 86.01 n.e. 14.3 n.s. Shaanxi 1,533 217 685 2,445 682.71 8.0. 9.95 87.11 28.4S 14.11 19.81 Gansu i 115 462 68 9.55 8.11 19.0 87.1 72.5 14.41 17.91 Qinghai 19 48 91 157 12.81 8.5 29.8 88.11 58.11 14.41 20.11 Ningxia 889 23 141 552 70.41 8.89 4.11 88.8 5.5 18.85 11.2% Xinjiang 1,870 302 445 2,419 69.11 9.81 12.51 84.51 18.45 14.15 12.91 Subtotat Vj 1288 li. rUM 7ZE ZI A.M Zia LU .K SLAnK 18.U Il2 /a From Annex 3, Table 5. a Sum of net graina for Livestock, oiLseed meat, and fine cereaL bran. A Assuming 9.45 for tubers, 9.81 for corn, and 0.7t for other coarse grainj and usighting according to provincial production Levels of each. Zd Assuming 34S for anymeat, 441 for peanut meat, 881 for rapeseed meat, 845 for cottonesed meal, 95 for sunfLower eed meet, and 39% for meame and other mealap weighted according to provinciaL production LeveLs. a Assuming 12.11 for rice bran and 14.41 for wheat bran. 1 Weighted average, based on S of ingredient in ration and % of protein in ingredient. - 153 - 2DIA ImJa 1 2UM it" I ReLationshis Between jMg j14 jgaL QueLily Livestock fogd b, Province, jgg CoLumns A B C D E F a H I J K L N CaLulationt [C+0) (C+F+01 (E+F+61 (H/B) (WA) (I/B) (H/A) Traditior- Net OR- Fine Groms Net Gross Gras Net Gross Som Park aL Feed Wheet/ Grains sed CareaL TotaL Total Feeds Feeds Feeds Feeds Province Nos. Prod Brains Lg& Rice A For LS NmeaL Bran LS Feed LS Feed Park Sow Pork son V'000 hdl 1V4 tl It/t l hlasw) It/ ft/sw NltionaL 19.uL 82.12 it4M. 18,I80ilA l3,5 1f4 I0lit5 141sa21 11.25 Za Mn gA Bijing 148 180 1,053 -971 192 24 123 1,200 329 9.2 8.2 2.5 2.3 TiarnJin 48 55 734 -1,001 -!67 61 57 851 -148 15.5 17.7 -2.7 -3.1 Hebei 762 594 9,464 -4,909 4,561 1,518 1,037 12,016 7,116 20.2 15.8 12.0 9.3 Shanxi 258 178 5,018 -3,234 1,794 310 398 5,725 2,491 32.5 22.4 14.2 9.7 Inner MongoLia 287 183 3,794 -3,331 482 225 200 4,218 887 23.0 14.7 4.9 3.1 Lisoning 605 540 9,81 -5,280 4,001 395 315 9,970 4,710 18.5 10.5 8.7 7.8 JiLin 464 253 12,123 -4,730 7,393 290 200 12,613 7,883 49.9 27.2 31.2 17.0 HeiLongjiang 631 291 8,645 -1,687 6,948 524 633 9,902 8,105 33.7 15.5 27.9 12.9 JbIntaL L Lil L88 II2 00. -2..026 e5.LM 3..34 L R 5038 iaZE B 1..7 1.1 LM. Shanghai 10 179 586 -901 -365 191 18S 813 12 5.1 4.8 0.1 0.1 Jiangsu 1,105 1,211 5,913 4,984 10,847 1,235 2,985 10,013 14,947 8.3 9.1 12.3 13.5 Zhj iang 900 678 1,737 1,917 3,654 375 1,557 3,889 5,585 5.4 4.6 8.2 7.0 Anhui 758 548 3,164 1,766 4,930 947 1,941 8,051 7,817 11.1 9.0 14.3 10.3 Fuj im 448 370 815 -1,194 -379 109 719 1,835 442 4.4 3.7 1.2 1.0 Jiangxi 778 508 293 2,190 2,483 242 1,417 1,951 4,142 3.9 2.5 0.2 5.4 Shandong 1,278 967 14,774 -5,878 8,897 3,078 1,780 ig,628 13,750 20.3 15.4 14.2 10.0 Henan 1,128 48 8,141 -1,252 6,888 1,414 2,4586 12,011 10,759 24.2 10.8 21.7 9.5 Hubei 1,103 728 2,684 2,295 4,959 1,028 1,998 5,673 7,966 7.8 5.1 10.8 7.2 Hunan 1,574 1,240 1,2865 3,637 4,902 412 2,817 3,993 7,6830 3.2 2.5 6.2 4.8 Guangdong 1,519 960 1,805 -17 1,378 357 1,702 3,454 3,487 3.8 2.3 3.8 2.3 Guangxi 973 618 1,170 -802 386 153 1,004 2,327 1,525 3.9 2.4 2.5 1.8 StitaL & 1,04 85o0 41,i84 il 4.ll1 ,it 1921 Z13l8 Z801 LA LM A& i Sichuan 3,971 2,340 11,795 -8,108 9,696 943 2,861 15,598 12,400 6.7 3.9 5.3 3.1 Guizhou 825 434 2,714 -2,772 -58 208 481 3,352 581 7.7 4.1 1.3 0.7 Yunnan 1,442 489 3,708 -e,728 985 89 582 4,379 1,657 0.8 3.0 3.3 1.1 Tibet 24 2 815 -672 -57 8 19 342 -30 171.2 14.3 -15.0 -1.2 Sheenxi 484 225 4,178 -e,643 1,588 217 6e 5,06 2,445 22.8 '10.5 10.9 5.1 Sansu 351 150 1,724 -1,863 81 115 482 2,301 638 15.3 8.8 4.3 1.8 ainghai 60 24 302 -29 19 46 I 439 157 18.3 7.3 6.5 2.8 Ningxia 41 19 311 78 389 2S 141 474 552 24.9 '11.8 29.1 13.5 Xiqjiang 71 32 1,391 279 1,870 30 445 2,137 2,416 66.8 10.1 75.5 34.0 SUblat Wtal ZAZZ L 20di -it,20 18827 1.0 laZM 3Hlt 20.l0 Aa Ll i ZR Total production of corn, tubers & other cerests Less 105 for sed and waste. Wheet and rice avaiLable for Livestock after aLLowing for human consumption, seed I waste. - 154 - 2LM a~L LIVESTOCK glSM [[RIAM YiLd Relionships botan fRil Wt _UL 1& Corn Yield Rico Rice Rice Corn Corn Corn As % of Province Area Prod YieLd Area Prod Yield Rice Yield/g 11000 hal('D00 t1 ft/hal 000 half'000 tl ft/hal f%l NationaL Ing,L 83 . I. Z..925 L& I Z. Z,411 &A M1 Beijing 44 260 5.9 208 955 4.6 78.51 Tianjin 24 135 5.6 157 565 3.6 64.01 Hebei 131 750 5.7 1,817 6,400 3.5 61.51 Shanxi 9 65 7.2 607 2,350 3.9 53.1 Inner MongoLia 18 60 3.3 464 1,485 3.2 96.01 Liaoning 432 3,310 7.7 1,231 7,155 5.8 75.91 JiLin 285 1,910 6,7 1,855 11,040 6.0 99.8 HeiLongjiang 278 1,240 4.5 1,920 6,420 33 75.05 SubttaL ZaZ IM Z.B8a A IalIZ AA ZZ'i Shanghai 301 1,700 5,9 5 35 7.0 118.4S Jiangsu 2,516 17,485 , 7.0 431 2,090 4.8 68.41 Zhajiang 2500 15,105 6.0 47 145 3.1 51.11 Anhui 2,169 11,380 5.2 175 480 2.8 53.5% Fujian 1,587 7,310 4.6 2 0 n.e. n.e. Jiangxi 3,37 14,930 4.5 8 10 1.8 27.9 Shandong 108 575 5.3 2,070 9,935 4.8 90.11 Henan 446 2,390 5.4 1,521 5,230 3.4 64.2 Hubsi 2,6821 15,670 6.0 411 1,285 3.1 52.95 Hunan 4,401 24,165 5.5 116 240 2.1 37.71 Guangdong 3,934 17,830 4.8 40 70 1.9 39.41 Buengxi 2,661 10,655 4.0 500 1,040 2.1 51.95 Subtotal III 28 a.I L. 2aA .lUK .5I MA Sichuan 3165 20,340 6.4 1,667 6,250 3.7 58.8 Guizhou 780 4,070 5.2 650 2,175 3.3 64.1% Yunnan 1130 5,005 4.4 975 2,740 2.9 63.4 Tibet 1 5 5.0 2 5 2.5 50.01 Shaanxi 160 935 5.8 951 3,130 a.3 58.3 Bensu 4 15 3.8 227 675 3.0 79.3 Qinghai 0 0 n.e. 0 0 noe. n.e. Ningxia 51 420 8.2 32 115 3.6 43.61 Xinjiang 96 370 3.9 450 1,390 3.1 90.15 SubtotmL YM1 L.W 91i1. L L3A 10.480 1A 39-95 Za Averages by region and nationaL average were weighted by rice production. This indicates the potentiaL change in production from swapping corn for rice in those areas, on average. - 155 - IN& Is.L. I Number. ad4 CitinaL Aggregae Parmeters fa JIgM - No. at year and - Staugh- Pigs Neat Estima- Pork PopuL- Sow Dth Pig Total tared Pork Carcass SoLd/ Pfod/ ted Age Output/ Province tion Nos. Nos. Nos. Nos. Prod. Weight GoWyr Govyr At SaL PopuLn (mill1 ('000 hd] ('000 t) (kg/pig) (no.) (kgo) (moe.) (kg/asp) A. /a .b NotianaL InTatL 1,030.3 2g02 284,07I a00 7 ilZi it.0Z 7 1.1 W IILU 1 . Beijing 9.5 146 1,472 1,818 2,194 180 59 15.0 990 8.1 13.7 Tisqi in 8.0 40 867 715 790 55 71 18.3 1,148 10.3 9.9 Hebei 54.9 762 11,316 12,078 6,85 594 71 11.0 780 16.2 10.8 Shanxi 26.0 256 3,120 3,376 2,887 176 62 111 689 18.2 8.0 Inner NangoLis 19.8 267 8,941 4,128 2,480 188 74 8.6 883 18.6 9.2 Lisoning 88.5 605 8,324 8,929 6,852 540 79 11.3 998 14.6 14.9 Jitin 22.8 464 8,858 4,120 2,995 253 94 8.5 545 14.6 11.1 HeiLongjiang 32.9 9a1 4,227 4,956 3,743 291 78 6.0 461 13.8 8.8 Subt.L L i.A I R ll alB 2 3 .=1 MsME ZzEI ZA &IA 71 AL. I.A Shanghai 12.0 190 1,892 2,002 3,407 178 52 17.9 987 6.7 14.8 Jiangsu 81.7 1,105 17,027 18,132 20,361 1,211 B 19.4 1,09B 10.0 19.6 Zhijiang 39.9 800 12,482 13,282 12,470 878 54 15.8 948 12.0 17.0 Anhui 51.0 756 10,908 11,396 7,407 546 74 9.9 720 17.2 10.7 Fujian 26.8 443 7,353 7,796 5,154 870 72 11.6 a85 17.1 13.8 Jiangxi 84.2 778 10,615 11,888 9,028 50 88 10.4 857 15.0 14.9 Shandong 76.4 1,276 15,537 16,95 12,840 967 75 10.0 757 14.5 12.7 Henan 76.5 1,128 12,142 18,270 6,884 498 75 5.9 440 22.0 6.5 Hubsi 48.8 1,103 16,363 17,466 11,910 726 BI 10.0 660 16.5 14.9 Hunan 55.8 1,574 21,767 23,371 L1,289 1,240 58 13.5 788 12.3 92.8 Guangdong 61.7 1,519 18,958 20,475 18,108 90 78 8.6 882 17.4 15.8 Guangxi 88.1 978 12,527 13,500 7,485 69 as 7.6 g3 20.2 16.2 ubStL a 3eVZ itia J577Iz Jz 9.83 iO.058 a.a . Ma za sA. i. Sichuan 101.1 3,979 52,720 00,699 88,744 2,340 80 9.7 589 16.8 28.1 Guizhou 29.8 925 10,064 10,889 5,916 434 78 7.2 s 20.4 14.8 Yunnon 83.6 1,442 15,574 17,016 7,448 489 7 52 346 25.1 14.9 Tibet 2.0 24 101 15 48 2 42 2.0 aB 25.2 1.0 Sheanxi 29.7 484 6,509 8,993 sem 225 B 8.0 45 20.8 7.6 Sansu 20.2 851 4,007 4,448 2,98 150 50 9.5 7 16.5 7.4 Otnghai 4.0 60 664 724 166 24 IS 8.1 400 21.B 6.0 Ningxia 4.1 41 455 496 831 I9 57 6.1 4683 18.5 4.6 XiuiJisng 18.4 71 5a 657 460 3e 70 6.5 451 15.8 2.4 SubtltoL 19A IZ& Za= 00,ZZ0 M,.4 l47 LZM1 a A MS ML. 1U. Z& Inter-provinciaL figures wiLL be distorted in that those provinces which import Live pigs will appear to have better performance than eatuaLLy achieved while the converse wiLL be true for exporters. & Figure wouLd be correct for a year round production system. Heavy measonat slaughter in early spring may mean numbere are above the yearly average at the year and cenous date thereby causing average @laughter age to be slightLy exagerated. - 156 - ANNI 3 CHLIH TabLsJQl LIVESTOCK SECTO SUD Estimated Physical Parameters for fig Production -Situation For SH & Stats Fams- National -Presant ituatiorr- -Potential- Avg 1984 Local Local Exotic Exotic Local Black x Type Type Black Physical Parameters Breed Exotic Hybrid Hybrid Breed Pigs Weaned per Sow par Litter a a a 9 Litters per Sow per Year 1.6 1.75 1'9 2.1 Age at Weaning (days so so 60 42 80 Weight at Weaning (kge) 10 12 12 12 15 Weight at Sale (kgs Liveweight) 93 97 105 105 BoLa Weight gain from Weening (gme/day) 875 425 485 500 180 Age at Slaughter (months) 9.2 8.5 8.5 7.5 14.7La Mortality Bows: % Deaths per year 2.0% 2.0% 2.0% 2.0% 2.0% Fatteners: % Deaths per pig weaned 2.51 2.51 2.51 2.51 3.51 Feed ina Bows (kg. per sow per year]: Fine Feed Balanced Concentrate B00 750 900 1,000 70 grains, Bran etc. 200 200 100 0 530 Total Fine Feed BOO 950 1,000 1,000 B0o Coarse Feed 180 160 60 160 160 Green Feed 360 360 960 380 1,100 Weeners (kgs per pig to 60 days): Balanced Contentrate 12 12 12 a 0 Fattening Pigs [kgs per pig fattened):- Fine Feed BaLanced Concentrate 250 270 290 300 30 Grains, Bran etc. 120 70 50 0 325 Total Fine Feed 3701 340 340 300 355Z Coarse Feed 27b 25 25 25 iaLk Green Feed 6 b 0 s0 so I,OO01 Indices of Performance Pigs Weaned per Sow per Year 14.4 14.0 15.2 16.9 i0.5a Fattened Pigs Sold per Sow per Year 14.0 13.7 14.9 18.4 10.14a Liveweight Sold per Sow per Year (tons) 1.31 1.32 1.58 1.93 0.91LA Balanced Conc. per Sow per Year (tons) 6.17 5.76 6.22 6.8 4.20 Total Feed per Saw per Year 8.00 7.48 8.02 8.79 18.62 Adjusted Feed per Sow per Year (tonal [green feed on a drymatter basis) 8.9] 6.52 7.02 7.61 7.64 Conversion Factors Total Feed per Unit Live Weight gain 6.1 5.6 5.2 4.5 16.2 Adjusted Feed per Unit Live Weight gain 5.3 4.9 4.5 3,9 B. Fine Feed per Unit Live Weight Gain 4.7 4.3 4.0 3.5 4.6 a Implied by national statistics on 1993 end-year stocks of saws and other pigs, and 1984 n1bers slaughtered and carcass weight. ALL other figures are mission estimates. B Mission estimates. However, national survey data given In Stats Statistical Bureau, Chn Rural Statistical Yearbook JJMl p. 169 suggest considerably batter eff iciencys for fine feed and coarse or green feed respectively, 175 and 731 kgs/pig far a1 . 237 and 874 kge per pig for backyard producers, presumably using Local black breeds. The cor- responding Live weight gains weaning-tor-slaughter were 71 and 89 kg. respectively. - 157 - CHINA LIVESTOCK SECTOR STUDY ANNEX 4: DAIRY CATTLE PRODUCTION 1. Introduction. This annex describes aspects of dairy cattle production not covered in detail in other sections, including management, housing, milk production, feeding, and daily nutrient intakes. Breeding and animal health status are described in the main text, Sections 4-5. Dairy Cattle Management 2. Housing. Adult cattle, replacement heifers on large state farms and calves are normally stall housed and stall fed. Replacement heifers typically occupy separate buildings and are stanchioned and stall fed. In general, dairy houses on state farms are clean, well illuminated and reasonably well maintained. Calf housing areas are often too wet from indiscriminate floor hosing, which predisposes to chilling, calf pneumonia and calf scours. In Tianjin, dairy cow housing on SH farms was often minimal, considering the cold winters. One SH herd was housed in a converted pig house without bedding throughout the year. Such cows are prone to pneumonia and mastitis and use a large proportion of their energy intake to maintain body temperature. Young heifer calves are penned separately, occasionally on raised slatted floors. Water is usually present ad lib. Drinking bowls are rare and water is normally presented in the empty feed troughs. Such troughs were seen empty at times that lactating cattle were present, which would pose a severe constraint to milk production. 3. Adult cattle are let outside into paddocks for exercise between milking and feeding. Most paddocks contain shade roofs, but are dirty with accumulated manure. Cows often lie in contact with such contaminated dirt, which is detrimental to udder hygiene. In times of hot weather the exercise routine would best be adjusted to early mornings or late afternoons, in order to avoid exposure to the hottest times of the day. If cattle must be outside during these times, adequate shading should be provided. Such shade systems were not always present in the Fuzhou area, where exercising cattle were seen to be suffering from exposure to sunlight as early as mid-morning. 4. All fodder and grain concentrates are stall fed, which represents an energy saving to cows of 5-10 per cent, as no additional energy on the exercise of grazing has to be expended. Feeding, milking, cleaning and exercise routines may take up to 18 hours in large state farm herds, providing additional feeding time for cows in which to ingest sufficient fodder volume. Cows appear to take minimal voluntary exercise while outside and quickly lie down to ruminate. Observance of cows by workers for signs of estrus, which is best undertaken while cows are outside and unrestrained, varied by farm. In the smaller herds such observation was carried out more frequently. In several village SH herds, other villagers would undertake to watch cows for signs of estrus in the absence of their owners. - 158 - 5. Bull Calf Raising. Newborn bull calves contribute little to cash flow and are disposed of after birth. Where dairy farms are located near cities (e.g., Tianjin and Fuzhou), bull calves can be sold at birth, at Y50-100 each, to pharmaceutical manufacturers for use in insulin extraction. Otherwise, bull calves were sold at Y15 each to consumers or for beef production. Because of local lack of demand for beef, and for calf meat in particular, the raising of such calves to beef steers or to veal is uneconomical. Of the four beef SH in Dongxiang County involved in raising Holstein steers from calves from the Red Star dairy farm, one SH was losing money while the other three made marginal net profits of Y80 per steer after 15 months of feeding to a 400 kg live weight. Beef exports from China also do not appear economic, due to the lower beef price on the international market. In addition, purebred Holstein bull calves are not ideal for efficient beef production. Alternatives for better utilization of bull calves are few. If dairy farms are located well away from pharmaceutical plants, calf glands could be harvested on the farm, deep frozen in central locations, and shipped in batches for processing. Bull calves could also be collected in batches from groups of nearby farms at 2-3 month intervals, and sold to a local slaughterhouse for rendering into meat and bone meal. 6. Milk Collection. At almost all farms, three times a day hand milking was practiced, with milk being collected from the larger farms once or twice daily. On all state farms, cows were hand milked into pails which were then weighed and recorded to obtain milk as well as milker records. Milkers are assigned bonuses for collecting milk of above average quality and quantity. The state farms were generally equipped with refrigerated bulk tanks. Milk from these tanks was transported directly to the dairy plant in insulated (but not refrigerated) 5-8 t tankers, operated by the farm or by the plant. Milk arriving at the dairy plant or at MCCs is subjected to spot sampling for coliform counts. All milk entering the dairy plant or Milk Collection Centers (MCCs) is tested for acidity and specific gravity by the alcohol test. If milk is rejected at the dairy plant or MCC, the loss is carried by the individual farm or SH farmer. 7. Milk from small SH herds is collected in cans and sent to milk collection centres, two or three times daily. SHs are paid once or twice a month for their deliveries. Producers are issued a receipt noting the milk weight, which can then be used to purchase dairy concentrate at the local feed mill at the rate of 2-2.5 kg milk delivered for 1 kg of grain concentrate, typically consisting of 50% corn, and 25% each of wheat bran and soybean meal. This formulation may vary depending on feedstuffs available and occasional supply bottlenecks develop, where feedmills are unable to supply concentrate to farmers. In Fuzhou these supply problems had been particularly acute and are being worked out in cooperation with the local Dairymen's Association, the feedmill and the AHB. Dairy Cattle Feeding 8. Feed Supply. Dairy . cattle feeding is characterized utilization 'of a great variety of feed grains, fresh and preserved roughages, feed by-products and industrial fermentation by-products. The availability of crop and fodder land to sustain feed supplies for - 159 - dairy herds is often limited, especially to the smaller herds belonging to specialized households (SH) and to farms in the Fuzhou area, where arable land is restricted to the coastal and river basin areas. Seasonal changes in fodder resources, the lack of fodder storage capability, and bottlenecks in feed grain supplies, especially in Fuzhou, result in the periodic supply shortages of feedstuffs for dairy cattle. This can over time result in wide shifts in the quality and quantity of nutrient intakes sufficient to sustain adequate milk production. Of the 10 dairy state farms, bull stations, collective farms and SH herds investigated, the number of feedstuffs used to prepare daily rations varied from 7 to 16 per farm. 9. In the Tianjin area, grain concentrate for lactating cows consisted of wheat bran, corn, soybean cake and occasionally fish meal. Hay, corn silage, and freshly cut grasses represented. the main roughages. Alternative concentrate feedstuffs included cottonseed meal, alfalfa meal, corn starch solubles, sweet potato, soybean (Tofu) waste, beet pulp and carrot. Calcium and phosphorus supplementation, if used, was supplied by bone meal and shell (oyster) meal. In Dongxiang and Nanchang, wheat bran, cottonseed meal, peanut oilcake and soybean cake were commonly used in dairy grain concentrates. Roughages include rye hay, rye grass, corn silage, sweet potato and some corn distillates. In Fuzhou, dairy concentrates consisted of corn, wheat bran, soybean cake and local fishmeal, with as alternatives peanut oilcake, cottonseed meal and sunflower meal. Roughages included sugarcane tops, -leaves and -stalks, rye grass, reed grass, vegetable wastes, sweet potato, green corn and green wheat. Bone meal and shell meal were used to supplement calcium and phosphorus. The ensilaging of roughage for feeding during the cold season was not practiced in Fuzhou, but was commonly practiced in Tianjin and Dongxiang. 10. The availability of cropland ranged from none to over 5,000 ha per farm. Three of the SH farms visited had no land other than that used for buildings, with herds ranging from 9 to 16 animal units (AU)./a In contrast, two dairy state farms contained large cropland holdings, with 852 and 900 AU utilizing 300 and 5,300 ha of cropland, respectively. One medium-sized collective farm containe4 81 AU without cropland. There was no relationship between cropland availability, and milk production performance, however. The availability of sufficient feedstuffs can apparently be independent of farm grown crops and relies on alternative supply arrangements from other sources. Dairy farmers without cropland utilized roadside grasses and riverside reed grasses (largely of low nutrient value) as sources of roughage, which they cut themselves or contracted to be cut by others. In Fuzhou very large yields of rye grass were reported during the April-November hot season, at 90-100 tons per ha from 6-7 cuttings. If more land were made available, ensilaged rye grass would provide a source of fodder during the cold season (see below). 11. Nutrient Intakes. Indexed nutrient intakes from 'Lactating cow rations used on nine state farms, SH farms and collective farms visited are summarized (Table 1; for details, see Appendix). The nutrient parameters chosen are digestable protein (DP), total digest- /a FAO definition of Animal Units. - 160 - Table 1. Nutrient Intake Indices Za of Lactating Dairy Cattle on Various Farms Dairy Farm DP/c TDN/b Calcium/c Phosphorus/c Gaoguanzai Coop, Jinan 84 111 75 82 Miao SH, Tianjin 76 88 29 33 Panzhuang, Tianjin 57 80 32 26 Shuanglin, Tianjin 65 115 120 128 Dairy Bull AI Stn, Tianjin 161 142 235 321 Dairy SH Nanchang 85 98 75 109 Red Star, Dongxiang 93 68 67 89 Northern Suburb, Fuzhou May - November Season 154 164 241 224 November - April Season 185 277 92 108 Nong Xiang, Fuzhou May - November Season 91 169 116 149 November - April Season 166 180 159 149 Wu Chang-Mei SH, Fuzhou 114 190 61 98 ---- --------------------- --------------------------------------- /a Index 1Q0 is equal to required daily nutrient intake at average herd production level at NRC standards. Index above or below 100 indicates nutrient requirements above or below standard daily requirement set at 100. /b DP - digestible protein; TDN - total digestible nutrients. /c Taken as 40% bioavailable from plant sources; 90% available from bone meal and shell meal. able nutrients (TDN), calcium and phosphorus intakes. TDN represents the digestable nutrients available in crude protein, ether extract, crude fibre and N-free extract. These data are only to be interpreted as rough estimates, as feed stuff compositions, and therefore nutrient intakes, change frequently. The feeding patterns indicated regional, rather than individual farm differences in nutrient intakes. Rations of two farms in Tianjin Municipality and one farm each in Nanchang and Dongxiang demonstrated significantly lower nutrient intakes. Diffe- rences in seasonal nutrient intakes on the two farms in Fuzhou, while generally at standard, demonstrated the shifts in nutrient intakes due to seasonal changes in feed composition. Dairy cattle on farms in the Fuzhou area appeared to receive higher nutrient intakes than those in Tianjin and Dongxiang, even though shortages of land, grain and avai- lable fodder were cited as major production constraints in Fuzhou. (Of course, these findings may be biased by the better management of farms that were visited). 12. Dry Matter intake by lactating cows was generally high in or above the recommended range of 2.5-3.5 per cent of body weight, and - 161 - averaged 3.0 per cent. This is indicative of the low nutrient values of most fodders, requiring greater quantities of feed intake in order to satisfy nutrient requirements. DM intakes ranged from 2.0 per cent of body weight to 4.0 per cent, whereas with higher quality feed these intakes should not have to go much beyond 2.8 per cent for the actual amount of milk produced. 13. Mineral Intakes. Calcium:phosphorus (Ca:P) ratios in dairy rations investigated ranged from 0.9:1 to 1.8:1, versus the ideal ratio of 1.4:1, and averaged 1.45:1. These are reasonably acceptable ratios in view the limited sources of calcium and phosphorus availa- ble. An incidence of post-parturient hypocalcemia (milk fever) of 3-5 per cent on one state farm with an average milk production of 5,000 kg per lactation is perhaps indicative of the need for additional Ca and P supplementation where milk production begins to result in. signifi- cant withdrawals of these minerals from the body. In most herds, calcium and phosphorus drains will only become a constraint when rising milk production resulting from better stock, feeding and ma- nagement creates greater demands. 14. Trace mineral and multivitamin supplementation in dairy ra- tions leaves a great deal to be desired. On only two state farms could the use of these feed components be verified. The consensus amongst farm managers was that sufficient Vitamin A and other vitamins are present in the green fodder. This is not always the case where hays, silages and mature grasses are concerned. The optimal supply of Vitamin A to dairy rations is very important in providing optimal reproductive activity. Vitamin D3 plays a role in calcium and phos- phorus metabolism, whereas Vitamin E is involved in a wide range of body functions ranging from fertility to muscle activity. The trace minerals iron, zinc, copper, cobalt, manganese and iodine are all essential to general health and need to be supplemented regardless of possible sources in fodders and grains. Low trace mineral and vitamin intakes have been implicated in a range of body dysfunctions related to fertility, such as anestrus, slow uterine involution after calving, the repeat breeder cow syndrome, retained placenta and low calf birth weights. 15. Calf Feeding. Newborn heifer calves receive fresh or fermen- ted (sour) colostrum at birth, and are continued on whole milk for up to 90 days of age. During this time they also receive grass or hay and grain concentrate. The average whole milk consumption per calf ranges from 350-400 kg and represents an on-farm use of product at the expense of sales. At a procurement price of YO.56/kg milk, this feeding practice costs Y200 per calf raised, and about 8% of each calf's dam's annual production. Alternative, milk-saving calf feeding practices could include a shorter milk feeding period (40 - 50 days), the use of diluted whole milk and calf starter rations or the use of a calf milk replacer, made from low cost imported skim milk powder and butter fat. 16. Feed Storage and Reserve Supplies. The need for developing longer term feed storage systems to provide a reserve supply of higher quality fodder during the winter months in the Tianjin area and during the cold season in Fuzhou is urgent, represents a constraint to in- creasing milk production. Corn silage is already being fed on three - 162 - larger farms in Dongxiang and Tianjin at a percentage of total feed intake of 11 per cent on two farms, to 32 per cent on a third farm. The need for additional silage production was voiced on most of these farms. Efforts towards this goal consisted of planting more corn, usually at the expense of other crops, and converting more grass, fodder and by-products from fresh feeding to silage feeding. Ensila- ging on smaller farms with little or no land would have to rely on supplies of roadside grasses, reeds, vegetable and other offals and by-products. Common silos could be prepared and shared by neighbour- ing farms to allow for the ensilaging of sufficient quantities of feedstuffs to provide a critical fermentation mass in such silos. Such silos could be of the horizontal type, covered with heavy plastic sheets to seal in the contents. 17. In the Fuzhou area, ensilaging is not now used as a method of fodder storage, as grasses continue to be harvested during the cold season. However, hay production is hampered by high humidity and heavy rain fall. The ensilaging of fodder abundantly available during the hot season could help to alleviate cold season shortages. Field research and farmer education in the use of small,. horizontal silage technology, the selection and use of fodder, by-products and grasses, and the use of urea, proprionic acid and other non-protein nitrogen agents to upgrade silage nutrients could all contribute towards sol- ving the fodder gap during the cold season. 18. Investigations into the use of brewer's grains, vegetable wastes and pulps and other industrial by-products should continue. A ration of 8 kg of wet brewer's grain is adequate for cow maintenance plus the production of 4 kg of average quality milk per day. Treatment of low quality reed grasses, other roadside grasses and rice straws in the Fuzhou area with sodium hydroxide to improve digestability for use as cattle fodder should be investigated in areas where fodder resour- ces and fodder croplands are limited. These pre-digestion techniques can also be combined with simple ensilaging methods, where additional non-protein-nitrogen compounds can again be included to upgrade over- all nutrient quality. 19. A considerable extension effort will be involved to teach the techniques involved. Pilot silos on selected farms with typical fodder procurement problems would best be installed first, based on an economic analysis of the additional cost of inputs required versus additional milk output gained. Moist by-products are usually highly perishable when exposed to open air and must be fed quickly and diffi- culties in transporting wet brewer's grain are quoted as a constraint to its use on dairy farms. If the quantities generated by the local breweries are sufficiently large, on-site drying and processing faci- lities could be established near breweries. The present use of brewe- ry wastes on pig, poultry or duck farms would have to be evaluated in the light of its alternate use on dairy farms. 20. Feeding Costs. Total monthly feed costs on farms in the Fuzhou and Tianjin areas were similar for calves, heifers and adult cows (Table 2), although calf feeding costs for concentrates in Tian- jin were 2.5 times higher and double the percentage of total produc- tion cost as in Fuzhou. Relative green fodder and roughage costs for all classes of cattle and concentrate costs for heifers and cows were - 163 - Table 2. Average Dairy Cattle Feed Intakes and Feed Costs Tian'in and Fuzhou Farms Calves Heifers/a Adult Cows Average (0 - 12 mth) (13 - 24 mth) (over 24 mth) Price % % % (Yuan/kg) kg Prod Cost kg Prod Cost kg Prod Tianjin Farms (2): Milk 0.448 18 14 -- -- -- -- Grain Concentrate 0.26 81 35 107 33 293 44 Green Fodder/b 0.03 305 16 610 32 702 11 Roughage 0.03 92 3 138 21 183 11 Total Feed 68 86 66 Fuzhou Farms (4) Milk 0.522 20 21 -- -- -- -- Grain Concentrate 0.28 32 18 90 29 208 47 Green Fodder 0.02 398 16 900 28 937 19 Roughage 0.055 86 9 300 22 255 11 Total Feed 64 79 77 /a Heifers on only 1 farm in Fuzhou. /b Green fodder fed on only 1 farm in Tianjin. otherwise in similar ranges. Cows in Tianjin received higher grain intakes but less green fodder than in Fuzhou. Milk production on farms visited in Tianjin averaged 18 kg per day versus 14.2 kg in Fuzhou, but the municipal average in Tianjin, at 18.6 kg, was very much higher than the 8 kg produced in Fuzhou. Obviously, concentrate and fodder intakes on other farms in Fuzhou not visited must be Lower, resulting in the lower milk production average. - 164 - ANNEX 4. APPENDIX: DAIRY CATTLE RATION ANALYSES Farm: MIAO SH Dairy, Tianjin Dairy Cattle: 500 kg Average Body Weight Average Milk 16.6 kg @ 3.5% BF/day ----------------------------------------------------------- -----As % of Dry Matter----- Daily Ration/a kg DM DP TDN Ca P ------------------------------------------------- Wheatbran 2.3 88 11.5 63.2 0.15 1.23 Corn 2.0 89 6.6 79.7 0.02 0.36 Soybean Cake 2.0 89 28.7 69.5 0.2 0.73 Corn Silage 4.0 20 0.52 13.2 0.31 0.23 Ryegrass, fresh 16.0 20 2.0 10.2 0.62 0.34 Ryehay 2.5 85 2.6 31.6 0.68 0.9 Soybean Waste 6.0 15 0.3 10.0 -- -- Fishmeal, imported 0.05 92 50.7 74.0 3.7 2.4 Shellmeal 0.05 90 -- -- 38.2 -- Bonemeal 0.05 93 9.3 24.0 32.0 15.0 kg DM DP TDN Ca P kg kg kg g g Total Daily Intake 35.0 16.0 1.37 7.72 18/b 14/b Index/c -- 76 88 29 33 Ca:P ratio: 1.2 1. DM Intake: 3.2% of body weight. /a DM - dry matter, DP - digestable protein, TDN - total digestable nutrients, Ca - calcium, P - phosphorus /b Ca and P taken at 40% bioavailability of quantities listed /c Index of daily nutrient requirements as per NRC standards, based on cow weight and production listed above, equals 100. - 165 - Farm: Panzhuang State Dairy, Tianin Dairy Cattle: 500 kg Average Body Weight Average Milk: 15.2 kg @ 3.5% BF/day -----As % of Dry Matter----- Daily Ration/a kg DM DP TDN Ca P Corn .2.3 89 6.7 79.7 0.02 0.36 Wheatbran 2.3 88 11.5 63.2 0.15 1.23 Soybean Cake 1.2 85 28.7 69.5 0.2 0.73 Cottonseed Meal 0.9 92 36.4 71.3 0.22 1.34 Shellmeal 0.25 95 -- -- 38.2 -- Hay 3.7 85 2.6 31.6 0.68 0.9 Green Cut Grass 16.0 20 3.0 9.2 0.62 0.34 kg DM DP TDN Ca P kg kg kg g g Total Daily Intake 26.7 13.4 1.34 7.38 21/b 12/b Index/c -- 57 80 32 26 Ca:P ratio: 1.8 f1. DM Intake: 2.7% of body weight. /a DM - dry matter, DP - digestable protein, TDN - total digestable nitrients, Ca - calcium, P - phosphorus /b Ca and P taken at 40% bioavailability of quantities listed /c Index of daily nutrient requirements as per NRC standards, based on cow weight and production listed-above, equals 100 - 166 - Farm: Shuanglin State Farm Tianjin Dairy Cattle: 500 kg Average Weight Milk: 22.3 kg @ 3.5% BF/day -----As % of Dry Matter----- Daily Ration/a kg DM DP TDN Ca P Wheatbran 2.1 88 11.5 63.2 0.15 1.23 Corn 2.0 89 6,6 79.7 0.02 0.36 Soybean Cake 1.2 89 28.7 69.1 0.2 0.73 Corn Chaff 0.25 85 1.0 66.0 0.05 0.06 Cottonseed Meal 0.3 93 36.4 71.3 0.22 1.34 Fishmeal, local 0.12 92 30.0 74.0 3.7 2.4 Shellmeal 0.023 95 -- -- 38.2 -- Bonemeal 0.11 93 -- 24.0 32.0 15.0 Corn Dregs 3.6 15 0.54 6.7 2.63 1.11 Hay 1.7 85 2.6 -31.6 0.68 0.9 Green Cut Grass 2.0 20 1.0 9.2 0.62 0.34 Corn Silage 11.0 35 0'.52 13.2 0.31 0.23 Alfalfa Meal 2.5 85 9.8 30.6 0.3 0.4 Beetpulp, dry 2.0 85 4.4 69.9 1.09 0.16 Sweet Potato , dry 4.0 85 0.6 64.4 -- -- Carrot, dry 1.4 85 7.9 58.0 -- -- kg DM DP TDN Ca P Total Daily Intake 34.3 19.8 1.48 12.08 92/b 69/b Index/c -- 65 115 120 128 Ca:P ratio: 1.3 f1. DM Intake: 4.0% of body weight. /a DM - dry matter, DP - digestable protein, TDN - total digestable nutrients, Ca - calcium, P - phosphorus /b Ca and P taken at 40% bioavailability of quantities listed /c Index of daily nutrient requirements as per NRC standards, based on cow weight and production listed above, equals 100 - 167 - Farm: Dairy SH, Nanchang Dairy Cattle: 500 kg Average Body Weight Milk at 13.1 kg @ 3.5% BF/day --------------------------------------£- ----- I- - - I- -----As % of Dry Matter----- Daily Ration/a kg DM D TDN Ca P Wheatbran 2.0 88 11.5 63.2 0.15 1.23 Cottonseed Meal 1.0 93 36.4 71.3 0.22 1.34 Fishmeal 0.75 92 30.0 74.0 3.7 2.4 Sweet Potato , dry 1.0 85 0.6 64.4 -- -- Ryegrass, green cut 1.3 20 3.5 7.7 0.6 0.3 Ryegrass hay 3.0 85 6.4 27.2 0.6 0.9 Corn 0.75 89 6.6 79.7 0.02 0.36 kg DM DP TDN Ca P kg kg kg g g Total Daily Intake 20.75 10.0 1.29 7.57 40/b 41/b Index/c -- 85 98 75 109 Ca:P ratio: 1:1. DM Intake: 2.0% of body weight. /A DM - dry matter, DP - digestable protein, TDN - total digestable nutrients, Ca - calcium, P - phosphorus /b Ca and P taken at 40% bioavailability of quantities listed /c Index of daily nutrient requirements as per NRC standards, based on cow weight and production listed above, equals 100 - 168 - Farm: Red Star Dairy, Dongxiang Cty. Dairy Cattle: 500 kg Average Body Weight Average 16.7 kg Milk @ 3.5% BF/day -----As % of Dry Matter----- Daily Ration/a kg DM DP TDN Ca P Corn 1.1 89 6.6 79.7 0.02 0.36 Peanut Oilcake, extr. 0.3 93 43.2 63.2 0.11 0.74 Cottonseed Meal, shelled 0.3 92 36.4 71.3 0.22 1.34 Soybean Cake 0.25 85 28.7 69.5 0.2 0.13 Wheatbran 1.7 88 11.5 63.2 0.15 1.23 Cut Grass 12.0 20 2.5 9.2 0.62 0.34 Starch Residue, dry 5.0 80 4.6 17.9 -- -- Corn Silage 3.0 35 2.6 13.2 0.31 0.23 Corn, Hominy Feed 0.7 85 7.2 72.7 0.03 0.27 Hay 2.0 85 3.6 31.6 0.68 0.90 kg DM DP TDN Ca P kg kg kg g g Total Daily Intake 26.4 13.0 1.67 6.00 41/b 39/b Index/c -- 93 68 67 89 Ca:P ratio: 1.1 il. DM Intake: 2.6% of body weight. /a DM - dry matter, DP - digestable protein, TDN - total digestable nutrients, Ca - calcium, P - phosphorus /b Ca and P taken at 40% bioavailability of quantities listed /c Index of daily nutrient requirements as per NRC standards, based on cow weight and production listed above, equals 100 - 169 - Farm: Northern Suburb Dairy, Fuzhou Dairy Cattle: 500 kg Average Weight Average 12.6 kg Milk @ 3.5% BF/day -----As % of Dry Matter----- Daily Ration/a kg DM DP TDN Ca P Wheatbran 4.9 88 11.5 63.2 0.15 1.23 Peanut Oilcake, extr. 1.4 93 43.2 63.2 0.11 0.74 Soybean Cake 0.2 85 28.7 69.5 0.2 0.73 Fishmeal 0.2 92 30.0 74.0 3.7 2.4 Shellmeal 0.2 90 -- -- 38.2 -- Rice Straw 2.0 85 6.0 70.0 0.21 0.08 Reed Grass, Fresh 40.0 15 1.0 12.7 0.5 0.2 Corn Pulp 10.0 20 4.6 16.3 0.1 0.5 Sweet Potato 3.0 85 0.6 26.4 0.2 0.2 Reed Hay 6.0 85 4.2 50.1 0.2 0.2 Rootcrop 3.0 15 8.0 78.0 0.3 0.2 Cabbage Leaves 15.0 20 3.0 70.0 -- -- kg DM DP TDN Ca P kg kg kg g g Total Daily Intake: a) May-November 58.9 15.8 2.27 12.38 123/b 83/b b) November-April 33.9 17.3 2.72 20.90 47/b 40/b /c Index:a) 154 164 241 224 b) 185 277 92 108 a) May-November Season; b) November-April Season Ca:P Ratio: a) 1.2 t1. b) 1.5 *1 DM Intake: a) 3.2% of body weight. b) 3.5% of body weight. /a DM - dry matter, DP - digestable protein, TDN - total digestable nutrients, Ca - calcium, P - phosphorus /b Ca and P taken at 40% bioavailability of quantities listed /c Index of daily nutrient requirements as per NRC standards, based on cow weight and production listed above, equals 100 -170 - Farm: Nong Xiang, Fuzhou Dairy Cattle: 500 kg Average Weight Average 14.7 kg Milk @ 3.5% BF/day -----As % of Dry Matter----- Daily Ration/a kg DM DP TDN Ca P Corn 2.1 89 6.6 79.7 0.02 0.36 Wheatbran 3.5 88 11.5 63.2 0.15 0.23 Soybean Cake 0.35 85 28.7 69.5 0.20 0.73 Sunflower Meal 1.4 93 31.0 70.0 0.46 1.12 Bonemeal 0.14 95 -- -- 32.0 15.0 Fishmeal, local 0.07 92 30.0 74.0 3.7 2.4 Shellmeal 0.035 95 -- -- 38.2 -- Ryegrass, fresh 3.6 15 3.5 7.7 0.6 0.3 Reed Grass, fresh 25.0 20 1.5 12.7 0.5 0.2 Green Wheat 15.0 15 18 73.0 0.42 0.4 Sugarcane Stalk 19.0 20 0.5 7.4 -- -- Sweet Potato , dry 6.0 85 0.6 64.4 -- -- kg DM DP TDN Ca P kg kg kg g g Total Daily Intake: a) May - November 36.2 12.3 1.48 13.84 65/b 61/b b) November - April 47.6 18.0 3.39 21.41 89/b 61/b /cIndex a) -- 91 169 116 149 b) 166 180 105 103 a) May-November Season; b) November-April Season a) Ca:P Ratio: 1.1 t1. DM Intake: 2.5% body weight b) Ca:P Ratio: 1.5 t1. DM Intake: 3.6% body weight /a DM - dry matter, DP - digestable protein, TDN - total digestable nutrients, Ca - calcium, P - phosphorus /b Ca and P taken at 40% bioavailability of quantities listed /c Index of daily nutrient requirements as per NRC standards, based on cow weight and production listed above, equals 100 - 171 - Farm: Wu Chang Mei SH Fuzhou Dairy Cattle: 500 kg Average Weight, 16.6 kg Milk @ 3.5% BF/day -----As % of Dry Matter----- Daily Ration/a kg DM DP TDN Ca P Wheatbran 5.3 88 11.5 63.2 0.15 0.15 Corn, Hominy Feed 2.9- 85 7.2 72.7 0.03 0.27 Soybean Cake 0.95 85 28.7 70.0 0.2 0.73 Shellmeal 0.1 95 -- -- 38.2 -- Fishmeal 0.1 92 30.0 74.0 3.7 2.4 Ryegrass 16.0 20 3.5 77.0 0.6 0.3 Sugarcane Leaves 4.0 20 1.5 12.0 -- -- Cabbage Leaves 4.0 20 3.0 70.0 -- -- kg DM DP TDN Ca P kg kg kg g g Total Daily Intake 33.4 16.3 3.13 17.2 37/b 43/b Index/c -- 174 196 61 98 Ca:P Ratio: 0.9 fl. DM Intake: 16.3 kg or 3.3% body weight. /a DM - dry matter, DP - digestable protein, TDN - total digestable nutrients, Ca - calcium, P - phosphorus /b Ca and P taken at 40% bioavailability of quantities listed /c Index of daily nutrient requirements as per NRC standards, based on cow weight and production listed above, equals 100 - 172 - Farm: Dairy Bull AI Station, Tianjin Bulls - 800 kg Average Body Weight ------------------------------------------------------------ -----As % of Dry Matter----- Daily Ration/a kg DM DP TDN Ca P --------------------------------------------------- Soybean Cake 1.95 85 28.7 69.5 0.2 0.73 Corn 0.08 89 6.6 79.7 0.02 0.36 Barley 0.08 89 10.0 73.0 0.05 0.37 Wheatbran 1.6 88 11.5 63.2 0.15 1.23 Corn Chaff 1.7 85 1.5 66.0 0.05 0.06 Fishmeal, local 0.4 92 3.0 74.0 3.7 2.4 Shellmeal 0.15 95 -- -- 38.2 -- Hay 10.0 85 3.5 31.6 0.68 0.9 Carrots, fresh 1.5 20 7,.9 58.0 0.37 0.34 ------------------------------------------------ kg DM DP TDN Ca P kg kg .kg g g ------------------------------------------------ Total Daily Intake 17.5 13.9 1.37 7.94 61/b 61/b Index/c -- 161 142 235 321 ------------------------------------------------ Ca fratio: 1:1. DM Intake: 1.7% of body weight /a DM - dry matter, DP - digestable protein, TDN - total digestable nutrients, Ca - calcium, P - phosphorus /b Ca and P taken at 40% bioavailability of quantities listed /c Index of daily nutrient requirements as per NRC standards, based on cow weight and production listed above, equals 100 - 173 - Gaoguanzai Town Dairy Cooperative, Jinan Dairy Cattle: 500 kg Average Body Weight Production 13.3 kg Average/Day @ 3.5% BF/day -----As % of Dry Matter----- Daily Ration/a kg DM DP TDN Ca P Corn 4.5 89 6.6 79.7 0.02 0.36 Wheatbran 1.9 88 11.5 63.2 0.15 1.23 Soybean Cake 1.1 89 28.7 69.5 0.2 0.73 Fishmeal 0.5 92 30.0 74.0 3.7 2.4 Bonemeal 0.05 93 -- 24.0 32.0 15.0 Shellmeal 0.05 90 -- -- 38.2 -- Green Corn 3.3 25 5.0 61.6 0.01 0.35 Cut Grass 6.6 30 2.0 10.2 0.62 0.34 kg DM DP TDN Ca P kg kg kg g g Total Daily Intake 18.0 10.0 1.28 8.64 40/b 40/b Index/c -- 84 111 75 82 Ca:P Ratio: 1:1 DM Intake: 2.0% of body weight /a Dm - dry matter, DP - digestable protein, TDN - total digestable nutrients, Ca - calcium, P - phosphorus /b Ca and P taken at 40% bioavailability of quantities listed /c Index of daily nutrient requirements as per NRC standards, based on cow weight and production listed above, equals 100. - 174 - CHINA LIVESTOCK SECTOR STUDY ANNEX 5: PIG PRODUCTION 1. Introduction. This annex describes aspects of pig production not covered in detail in other sections, including management, housing, production costs, feeding and nutrition. Breeding and animal health status are described in the main text, Sections 4-5. 2. Feeding Practices and Feeds. In all areas visited, feeding was the weakest point in pig production. The following problems generally occur: (a) because of an insufficient grain proportion within the total feed, energy,levels were low; (b) total protein and protein quality were insufficient; (c) often too much green feed was fed in combination with compound feed; (d) insufficient attention was paid to supply of minerals and trace elements and the Ca/P ratio, even in compound feeds; (e) amounts of feed used were often too low and nutrient levels therefore below standard; (f) animals (particularily younger ones) suffer from the practice of producing a single form of compound feed. 3. It is common for farmers to supplement their own grain, middlings and green feed with purchases of compound feed which, however, has been formulated as a complete feed rather than a balancer, as would be appropriate. (Only one mill visited - Tawenkou feed mill in Tai'an, Shandong - produced balancers.) The "dilution" of complete feeds by combination with farm-available feedstuffs tends to reduce the density of nutrients, minermals, vitamins, and other additives. 4. There is, in addition, reason for concern about the quality of complete compound feeds distributed by some Chinese feed mills. For example, a pig finisher produced by one mill visited used a formulation consisting of 24% whole rice, 17% broken rice, 15% rice bran cake, 10% rice bran, 7% wheat bran, 4% soybean meal, 9% rapeseed meal, 4% fishmeal (62%), 8% wheat flour, and 2% minerals. The claimed analysis of this feed, and our estimated analysis based on the reported composition, are compared below: Met.En. %CP %CF %Ca %P. .. %Lys. %Meth. (cal/kg) Claimed 2,900 14.6 6.03 1.02 0.8 0.500 0.6 Estimated 2,422 17.1 6.90 0.890 - 175 - It is noteworthy that estimated metabolic energy is considerably below the claimed level, and below requirements. Protein exceeds claims, but this may be illusory, as many different protein ingredients are substituted, some with lower protein levels than assumed (43% was assumed for soymeal, whereas 37% is not uncommon in China). Estimated fiber 'ontent exceeds the claimed level, which is already too high for pigs and a potentially limiting factor on performance (levels < 5% are optimal). Lysine content exceeds claims, which is likely to be the case in most Chinese feeds. Methionine, rather than lysine, is likely to be a limiting factor in these feeds. 5. The same mill takes samples of finished feeds for analysis six times per year by the provincial Feed Company. The following is the latter's analysis of two feeds which, according to the analysis sheet, "have reached the national standard:" (analysis in percent) H20 CF CP Ca P P1g feed 10.9 3.0 16.3 2.1 0.17 Layer feed 10.1 2.3 17.3 2.1 0.16 Note that the analyses are so close that both feeds could come from the same batch. Ca and P content for pigs should be 0.7% and 0.5% respectively, and the Ca/P ratio about 1.5:1. At the same time, the layer feed lacks sufficient Ca, which will result in poor egg quality. In fact, the reported analysis of P-content is improbably low, in view of the listed ingredients, which calls into question the analytical techniques. The example indicates the general lack of attention to supply of minerals. Although this mill might not be representative, it was one of a few cases where both ingredients and laboratory analysis of finished feed were available. 6. Lack of availability of particular protein or energy feeds results in frequent variations in ingredients. It appears that insufficient attention is given, and adjustments made, for the wide variation in protein and energy levels among these partial substitutes. Also, some raw materials are used which are not considered typical pig feeds. This includes-industrial waste products which may be better fed to cattle, and oilseed meals such as rapeseed and cottonseed meal which, given present varieties and processing techniques, should be carefully restricted components of pig rations. 7. Housing and Management. Housing was generally adequate, with well-constructed pens with concrete floors. However, investment costs could be considerably reduced, as too much space was allowed for fattening pigs and sows during pregnancy and lactation. Major differences in production levels observed in different areas derive mainly from differences in feeding practices combined with variable management. In general, both the condition of pigs and general hygiene were poor in Tianjin compared to the south or even Shandong; breeds and breeding management also seemed weaker. In most areas, due to low protein content and quality, carcasses were excessively fat, although producers for the export trade were producing excellent carcasses with more protein, better quality, and higher energy feed. - 176 - On many farms, it was observed that water was only provided two or three times per day at feedings; more and steadier water supply should be made available, particularily for lactating sows. 8. Production Costs. Tables 1-4 indicate the range of production costs for backyard producers, specialized households, and two state operations. It appears that, for most categories of pig, unit costs of production for backyard producers are lower than for specialized households, which in turn are lower than for state farms. The differences in cost are due to (a) differences in feed quality and quantity (which, of course, affect product quality and length of production cycle), and (b) different levels of management and other overhead costs (which are high for state farms). 9. Pig Performance. Accurate figures on average performance were difficult to obtain. Reported figures were generally -too optimistic, and have required adjustment to estimate average performance. For example, nearly all producers keeping sows claimed to obtain 2 litters per year, which is not possible on average when piglets are weaned after 60 days./1 Typical litter sizes for local sows were often reported as 15, but actual observation suggests 7-9 as normal and 12 as high. In view of the long suckling period and true average litter .size, a total. of 14 piglets per sow per year is realistic. 10. Mortality rate during suckling and weight at weaning present similar problems. Usually losses of 10- 2% and weaning weights of 15 kg were cited, which again seems a bit o*timistic. Farms which kept production records provided more precise data: (a) Tianjin Red Flag Farm: 9.7 pigs born alive per litter (24 litters), and 7.9 pigs weaned per litter (8 litters) imply 18.6% mortality; weaning weight at 57.5 days average 9.05 kg; (b) Dongxiang Red Star Breeding Farm: 1985 average 10.3 pigs born alive, 9.0 weaned at 60 days, implying 12.6% mortality, and 14.6 kg per pig weaned; 1986 January-May 10.4 pigs born alive, 9.67 weaned (7.0% mortality), and 14.4 kg liveweight at weaning. As these are relatively well managed farms, under average conditions, mortality should probably be about 15-20% and weaning weight about 12 kg. 11. Fattening time for collective units was reported in some cases to be 6-7 months from birth to sale. However, these units bought in their piglets supposedly at 2 months, at weights ranging from 22-26 kg, compared to an estimated average weaning weight of 12 kg. Adding 10-14 kg to the piglets weight would require an additional /l 115 days pregnancy plus 60 days suckling period plus 7 days until next heat equals 182 days, or exactly 6 months. However, not all sows will come into heat 7 days after weaning, especially after the long and stressful 60 day suckling period, and not all sows will become pregnant after first service. - 177 - month, so most likely the bovght-in pigs were 3 months old. Moreover, under the low energy, low protein feeding systems observed in these operations, it is likely that 6 months is required to fatten pigs from 10-25 kg up to 100 kg, making a total of 9 months from birth to sale. This would give a daily gain over the whole 9 months of 370 g and for the fattening period (3-9 months) of 444 g, which seems to be realistic for collective units. 12. Feed conversion rates are difficult to calculate when large amounts of coarse and green feed are fed, but not.weighed,and ignored in feed conversion calculations. Reasonable est imates can be derived from farms which provide only small amounts of green feed as an appetizer, and otherwise depend on compound feed or grain. For examples, Dongxiang's Red Star Breeding Farm estimated feed conversion of 1:4.14 (liveweight basis), with 600 g daily gain to market weight of 90-100 kg on compound feed. Tianjin Hangshengzhuang's Red Flag Collective Pig Farm reported 1:4.3-4.5, with daily gains of 500-600 g to slaughter weight of 91.5 kg on various compound feeds with some green feed. Dongxiang County Breeding Station reported 1:3.6-3.8, 600 g daily gain, and 90-100 kg slaughter rate on compound feed. In general, 600 g daily weight gain (fattening period) with compound feed is good under Chinese conditions, requiring only 4.5 months to fatten from 20-100 kg. However, most production units will not reach this standard, despite claims to the contrary, and would require 5-6 months with 400-500 g daily gain and feed conversion of 1:4.5-5 (counting green feeds at their nutritional equivalence to grain or compound feed). This ignores differences among breeds: with local breeds, such as Shandong's Laiwu pig, reportedly 350-450 g daily gain is attainable, whereas exotics and cross could give figures approaching 650 g. The best performance observed was that of a Fujian Putian County specialized purebreed farmer, who finished pigs from 25 kg to sale for export at 90-100 kg in about three months, giving a daily gain of 700-800 g. This was obtained with use of five different feeds - two for sows, three for fatteners. Good quality protein (fishmeal) was combined with high energy (corn) in the ration. -'178- AMi I LIVESTOCK SWT[O g[Q Fig Production Coeate Backyard Producers IR foral (Yuan per animaL Unit Soms RevLace. Sons /c Parker_ PioLets Cost Am't Cost Am't Cost Am't cost Am't Cost (Y/kg) (kg/day) (Y) (kg/day) (Y) (kg/day) (Y) (kg/day) (Y) Concenirate Le 0.28-0.32 1.0-1.7 115-174 1 92.4 0.3-0.6 19-29 .25 4.2-4.9 Coarse feed Z_ 0.16-0.18 0.3-1.5 18-86 1 57.6 0.2-0.5 5.4-7.2 - Green feed La 0.02-0.03 2.5-3.0 22-27 3 20.0 0.7-0.8 3.0-4.2 0.5 0.6 Labor 39-42 24.0 8.5-9.0 0.6-3.5 Veterinary 4-7 4.0 1.D-4.5 0.8-1,2 Breeding 3-5 - - - Water/eLect. 1 1.0 0.2-0.6 0.3 TooLa,repaire 1-4 1.0 0.5-0.7 0.2 Other /b - - - TOaL cost 210-Z7 2O.0 7a2-4.. Lq Besides compound feed, rice, eweet potatoes (roots and vines) and cabbage were fed Z IncLudes management and other overheeds; excLudes depreciation. Lc One farm onLy. - 179 - IJL 2 LIVESTOCK Jil0lQ A1M1 Pig Production Coste: SDeciaLized HousehoLds [A fmae] (Yuan per animaL) Unit saws Porkers Piglets Cost AIt coat Am't Cost Am't Cost IY/kg] (kg/day) (Y) [kg/day) (Y) (kg/day) (Y] Concentrate Za 0.2-0.3 2.0 144-216 0.4-0.5 14-27 0.25-0.8 3.2-4.5 Coaree feed /a 0.1 0.4-0.5 14-18 0.2-0.7 3-13 - - Green fed La .015-.024 3.9-4.1 22-88 0.85-0.5 1-2 0.25-0.3 0.5 Labor 3&-42 8.5-7.5 3-5 Veteri nary 5-7 0.-0.8 0.5 Breading 5 - ater/eLect. 1 0.4-0.6 0.2-0.5 TooLs repai re 1.2-1.4 0.2-0.8 0.2 Other L .B 0.4 0.1 TotaL cost RM:al B& 1 9.2-10.1 /a Besides compound feed, rice, umeet potatoes (roots and vines) and cabbage we Ai IncLudes manegment and other overheads; excLudes depreciation. - 180 - LIVESTOCK SfECTO JMjQ Vg Production Coats Donaxiana fig Breading Station (Yuan per animaL) Unit SMm RepLace, Sams Porkers PigLete ost Am't Cost Am't Cost Am't Cost Am't Cost [Y/kg] (kg/day] (Y] (kg/day) (Y) (kg/day) [Y) (kg/day) (Y) Concentrate 0.5 2.6 465.0 1.5 142.5 2.4 128.0 0.17 7.7 Coarse feed - - - - Green feed .045 1.0 16.2 0.8 8.8 1.2 7.2 Labor 68.4 11.0 4.4 Veterinary 15.0 1.0 2.0 Breeding 8.0 - - - Water/eLect. 1.0 0.3 0.3 - TooLs,repairs 1.0 0.3 0,3 - Other /a 7.0 38.3 36.3 - TotaL cost 581.6 Allji 1ZLA Z52l Za IncLudes management end other overheads; excLudes depreciation. - 181 - Stillia : LIVEMTCK SEC[[TOR[gg EJA Production CRIL Donoxiana Count1 EW fjLE (Yuan per animal) UMt ae .oars RepLace. Some Prkers- Piglete Cost A't Cost A't Cost N't Cost Am't Cost Au't Coat (Y/kg) (kg/day) (Y) lkg/day) (Y) (kg/day) (Y) (kg/day) I[Y) (kg/day) (Y) Concentrate 0.35 2.0 252.0 8.0 893.8 1.8 189.0 2.0 120.0 0.5 10.5 Coarse feed 0.14 1.0 50.4 - - - - - - - - Green feed 0.04 2.0 28.8 8.0 48.2 1.7 24.0 2.0 14.4 .075 0.6 Labor 87.5 50.0 25.0 7.5 2.5 Veterinary 2.2 1.2 1.0 3.0 0.3 Breeding 4.0 - 2.0 - - Water/eLect. 7.2 7.2 0.6 1.8 0.2 TooL,repairs 4.0 4.0 0.7 2.0 - Other /a 21.6 21.6 1.8 5.4 0.9 1.aL cstl A072 lim IM. Li 1-o L. Includes management and other overheada excLudes depreciation. - 182 - CHINA LIVESTOCK SECTOR STUDY ANNEX 6: POULTRY PRODUCTION 1. This annex compute requirements for pure line grandparent broiler or layer stock in a modern poultry breeding program; it also describes the current standards for poultry vaccination in China. 2. Pure Line Breeding Program. In order to produce 200,000 broiler chicks yearly or 4,000 per week, annual placements of two flocks of 1,200 breeder females are required. To produce the two breeder flocks, grandparent flocks of 1,100 female-line females and 200 male-line females are required. Pureline flocks of the following size, after selection, are required to produce the grandparent flocks: Sex Line Males Females ---------------------------------------------------- A-Male Line Male Line 30 300 B-Male Line Female Line 30 300 C-Female Line Male Line 30 300 D-Female Line Female Line 130 1,100 ---------------------------------------------------- Improvement of the genetic stock requires placing approximately five times as many females of each line and 15-20 times as many males of each line. This selection intensity will increase growth rate between 40 and 60 grams per generation. Several more grandparent flocks could be produced from the pureline flock as the need for grandparents increases. With complete utilization of the grandparents eggs, appro- ximately 39 breeder females can be produced from each grandparent female-line female housed. 2. In order to produce 100,000 layer chicks yearly or 2,000 per week, annual placement of two flocks of 1,300 parent females are required. To produce the two parent stock flocks, pureline flocks of 1,100 female-line females and 200 male-line females after selection are required. If there is no desire to select the purelines and improve egg production, egg weight and egg quality, pureline flocks of 1,100 female-line females, 120 female-line males, 200 male-line fe- males and 50 male-line males need to be placed. However, if the genetic stocks are to be improved, the pureline flocks need to consist of approximately five times as many females and 10 to 15 times as many males. The selection intensity will increase egg production by appro- ximately 2.5 to 3.5 eggs per generation. Several more parent flocks could be produced from this size of pureline flock as the need for parents and commercial layers increases. 3. The following table describes the current vaccination program for poultry in China. Not all vaccines are administered to every flock. - 183 - Table 1: Poultry Vaccines and Vaccination Protocols Vaccine Type & Route Vaccination Farmer Disease of Administration Program. . Price/Dose Newcastle a) LaSota strain, intra- 7 days of age Yuan 0.02 Disease conjunctival, and 35-40 days - 0.03 intra nasal or water b) F1 strain intramuscular, or at 60 days Yuan 0.02 wing web at 120 days - 0.03 Fowl Cholera a) Killed bacterin, at 60 and 70 intramuscular days b) Modified live, at 60 days Yuan 0.03 duck and poultry strains, intra- muscular Fowl Pox Skin scratch, check at 21 days Yuan 0.03 for take 7 days later Infectious For adults: subcutaneously, at 60 days Yuan 0.012 Bronchitis for chicks: drinking water. at 3-7 days antibiotic cover given simultaneously Marek's Breeders and layers only day 1 Yuan 0.02 Disease Gumboro Broilers only day 1 Not quoted Disease - 184 - ANNEX 7 Table 1 AIMA IVESTOCK .SECTO1 STUDY Prices of feed Inaredients. China 1985 Negotiated Prices (Yuan per JinaJ Location: S Inland SE Coet NE Coast S Inland NE Coast SE Coast Unweighted County Capital Capital County Capital Capital Average Name of Unit: Red Star Fd miLL Fd miLL Fd miLL Dairy Fm PLtry Fm ALL China Product Hatchery Fuzhou Tianjin Dongxiang Tianjin Fuzhou Corn 0.250 0.250 0.210 0.275 0.210 0.250 0.241 Wheat 0.250 0.140 0.195 UnmiLLed Rice 0.245 0.170 0.208 Broken Rice 0.200 0.200 Rough Wheat Bran 0.130 0.130 Wheat Bran 0.150 0.150 Wheat Pollard 0.170 0.170 Rice Bran 0.150 0.150 0.150 Rapeseed Meal 0.150 0.150 Corn Protein MeaL 0.185 0.185 Groundnut MeaL 0.220 0.380 0.275 FishmeaL (imported] 0.800 1.000 0.800 0.887 FishmeaL (Local) 0.500 0.560 0.580 Soybean MeaL 0.310 0.360 0.335 Sesame Meal 0.210 0.210 Meat MeaL 0.700 0.700 Bons MeaL 0.280 0.280 SheLl Meal 0.060 0.080 Corn Chaff 0.040 0.040 Cottonseed Meat 0.160 0.160 "1 including hulls 0.075 0.075 Soybean Pulp 0.150 0.150 Dry Carrots 0.270 0.270 9weet Potato (fresh) 0.050 0.050 Local Reeds 0.013 0.013 Alfalfa (fresh) 0.025 0.025 Hay (from NE) 0.100 0.100 Bone Meal 0.200 0.200 Been Curd Dregs 0.150 0.150 List Prices (Yuan per Jirta) Location: S Inland SE Coast NE Coast -NE Coast NE Coast NE Coast Urnsighted County Capital Capital Capital Capital Capital Average Name of Unit: Feed MiLL Feed MiLL Feed MiLL Feed MiLL Feed Bur Dairy Fm ALL China Product Dongxien Fuzhou Tianjin TiarjJin Tianjin Tianjin Corn 0.180 0.095 0.115 0.138 0.123 0.128 0.130 Wheat 0.090 0.090 UnmiLLed Rice 0.116 0.116 Broken Rice 0.156 0.156 Wheat Bran 0.054 0.060 0.057 Wheat Pollard 0.110 0.110 Rice Bran 0.042 0.055 0.055 0.095 0.062 Rice Bran Cake 0.037 0.037 Rapeseed MeaL 0.090 0.090 FishmeaL 0.500 0.750 0.780 0.800 0.708 Soybean Meal 0.220 0.175 0.310 0.300 0.115 0.224 Bone MeaL 0.190 0.190 Shell MeaL 0.035 0.05 0.050 Cottonseed MeaL 0.160 0.160 " including huLls 0.075 0.075 Salt 0.110 0.110 La One Jin equals 0.5 kge. - 185 - ANNEX 7 CdilNA lable LIVESTOCK SECTOR STUDI Financial Results Sfor Poultry: Growers (Specialized Household, 140 days/cycle, 2.4 cycles/year) Unit Total Units Quantity Cost Cost ( T ( Y) Capital Investment Poultry Housing sq. m. 250 100 25,000 Other Buildings sq. m. 10 100 1,000 Equipment N.A. N.A. 5,000 1otal Investment 31.000 Ooerating Costs Labor manday 140 2.8 392 Day Old Chick number 1,050 0.85 893 Compound Feed kg 9,000 0.6 5,400 Vetrinary costs bird 1,000 0.1 100 Fuel bird 1,000 0.1 100 Water & Elec bird 1,000 0.15 150 Transport bird 1,000 0.1 100 Maintenence (1% Buildings per year) 108 Equipment Deprec. (10% Capital cost per year) 208 Building Deprec., (5% Capital cost per year) 542 Management & Miscl. Costs bird 1,000 0.1 100 TIal 0perating Cats 88093 Sales Point of Lay Pullets bird 1,000 9 9,000 Manure Sales kg 7,860 0.04 314 Profit Z=t Batch 1,22 Profit Zt I=a 23Z La=b Elia Years 2-_ 72. Internal hI tf Return OveC 11 1ears - Residual Valu Y13,000) 11.61 - 186 - ANNEX 7 CHINA Table1 LIVESTOCK SECTOR STD Financial Results =fr Poultrx Lavers (Spectalized Household, 365 days/cycle, 1 cycles/year) Unit Total Units Quantity Cost Cost (Y) (Y) Capital Investment Poultry Housing sq. m. 371 100 37,100 Other Buildings sq. m. 10 100 1,000 Equipment N.A. N.A. 7,420 Total Investment 4,.9;20 OeratIng osts Labor manday 365 2.8 1,022 Day Old Chick number 1,000 9 9,000 Compound Feed kg 41,000 0.55 22,550 Veterinary Costs bird 1,000 0.1 100 Water & Elea bird 1,000 0.4 400 Transport bird 1,000 0.1 100 Maintenence (1% Buildings per year) 381 Equipment Deprec. (105 Capital cost per year) 742 Building Deprec. (5% Capital cost per year) 1,905 Management &,Miscl. Costs bird 1,000 0.25 250 Tntal GOeratin Costsl 16.4r.0 Eggs (240/bird/year) no. 1,000 0.1625 39,000 Cull Hens (1.7/bird/year) kg 880 2.6 3,890 Manure Sales (41.4/bird/year) kg 880 0.04 1,457 Profi.t Z=r Bate 70897 Profit Zga J= 7897 LAh E1QW Iar 2-11 10544 Internal hA Af Return LOver 11Ia -urs Residual Value: Y19f005) 207 - 187 - ANNEX 7 CHINA lable . LIVESTOCK SECTOR SDE Financial Results S= Poultry* Layina Duks (Specialized Household, 730 days/cycle, 0.5 cycles/year) Unit Total Units Quantity Cost Cost (Y) fY) Capital Investment Poultry Housing sq. m. 250 100 25,000 Other Buildings sq. m. 10 100 1,000 Equipment N.A. N.A. 2,500 Total Investment .28,500 Operating Costs Labor manday 730 2.8 2,044 Point of Lay Bird number 1,000 10 10,000 Compound Feed (0.17 kg/day) kg 124,100 0.55 68,255 Veterinary Costs -bird 1,000 0.1 100 Water & Elec bird 1,000 0.3 300 Transport bird 1,000 0.1 100. Maintenence (1% Buildings per year) I I 520 Equipment Deprec. (10% Capital cost per year) 500 Building Deprec. (5% Capital cost per year) 2,600 Management & Miscl. Costs bird 1,000 0.25 250 Ital 00erating C:sts 84,669 Eggs (15/kg, 36/bird) no. 1,000 2.55 91,800 Cull Hens (1.8/bird/year) kg 900 2.6 4,212 Manure Sales (10/bird/year) kg 900 0.04 360 Tta1 Salea 96372 Profit f.= Bate 110704 Profit Z= Isa 8S2 fUh la" Years 2 -1 1i02 Internal ja±& DI Return.(ve 11Iea e- aiual Value* n1I.000) 24.0t - 188 - ANNEX 7 CHINA Tl 5 LIVESTOCK SECTOR STD Financial Results f= Poultry: Laying Duckso Semi-scavenging Syste (Specialized Household, 730 days/cycle, 0.5 cycles/year) Unit Total Units Quantity Cost Cost Y ) (Y) 1 1 Capital Investment Poultry Housing sq. m. 50 80 4,000 Other Buildings sq. m. 5 80 400 Equipment N.A. N.A. 400 Total InvestMent 4j80O Operating Costs Labor manday 730 2.8 2,044 Point of Lay Bird number 200 10 2,000 Compound Feed (0.12 kg/day) kg 17,520 0.36 6,307 Veterinary Costs bird 200 0.1 20 Water & Elec bird 200 0.2 40 Transport bird 200 0.1 20 Maintenence (1% Buildings per year) 88 Equipment Deprec. (10% Capital cost per year) 80 Building Deprec. (5% Capital cost per year) 440 Management & Miscl. Costs bird 200 0.2 40 Toal Operating Costs 119079 Sales Eggs (15/kg, 30/bird) no. 200 2.1 12,600 Cull Hens (1.8/bird/year) kg 180 2.6 842 Manure Sales (5 /bird/year) kg 180 0.04 36 1ota1 alAA 11,478 PrftfAr ad43aa Profit fAr It=r 1,200 Lub Eow Years 2-11 1.460 Internal jAtg.t Retur LQr 11 Years - Residual Value:, Y2,200 28. - 189 - ANNEX 7 LIVESTOCK SECTOR STUD Financial Results fQr Poultry: Broilers (Specialized Household, 56 days/cycle, 5 cycles/year) Unit Total Units Quantity Cost Cost (Y) (Y) Capital Investment Poultry Housing sq. m. 100 100 10,000 Other Buildings sq. m. 10 100 1,000 Equipment N.A. N.A. 2,000 TotaEl,Investment 11 0000 Operating Costs Labor manday 60 2.8 168 Day Old Chick number 1,040 1 1,040 Compound Feed (Cony. 2.5:1) kg 4,500 0.7 3,150 Veterinary costs bird 1,000 0.1 100 Fuel bird 1,000 0.1 100 Water & Elec bird 1,000 0.1 100 Transport bird 1,000 0.1 100 Maintenence (1% Buildings per year) 22 Equipment Deprec. (10% Capital cost per year) 40 Building Deprec. (5% Capital cost per year) 110 Management & Miscl. Gosts bird 1,000 0.05 50 Ital 00erating Costs _,a8 Sales Broilers (1.8/bird) kg 1,000 3.2 5,760 Manure Sales (3.2/bird) kg 1,000 0.04 127 Total Sales908 Profit Jpr Bc %a Profit LUr Ia 4,S11 Lub EloH Years 2-.11 28 Internal IaIg f Return LQer 11 Iars Residual Value:, Y5,600 19.61 - 190 - ANNEX 7 CHINA Table Z LIVESTOCK SECTOR STUD Financial Results f=r Poultry: EgAt Ducks (Intensive) (Specialized Household, 60 days/cycle, 5 cycles/year) Unit Total Units Quantity Cost Cost (Y) Y ) Capital Investment Poultry Housing sq. m. 150 80 12,000 Other Buildings sq. m. 10 100 1,000 Equipment N.A. N.A. 2,400 Total Investment 1S.400 ORerating Costs Labor manday 65 2.8 182 Broiler Chicks number 1,040 1.2 1,248 Compound Feed (Cony. 3.0:1) kg 7,500 0.65 4,875 Veterinary'costs bird 1,000 0.1 100 Fuel bird 1,000 0.1 100 Water & Elec bird 1,000 0.1 100 Transport bird 1,000 0.1 100 Maintenence (1% Buildings per year) 26 Equipment Deprec. (10% Capital cost per year) 48 Building Deprec. (5% Capital cost per year) 130 Management & Miscl. Costs bird 1,000 0.05 50 Ital 02erating Costs 6.9r. Sales Broilers Ducks (2.5/bird) kg 1,000 3.2 8,000 Manure Sales (3.4/bird) kg 1,000 0.04 136 Profit fTr Batc 1,177 Profit 2= la 5i g88t; L.ah Zoy Years 2-11 6071i Internal 2&U nf Return LOve 11 1ears Residual Value. X6,500) 41 - 191 - ANNEX 7 CHINA Table& . LIVESTOCK SECTOR STUDY Financial Results r ia Fattening al Negotiatedf rices (Specialized Household, 180 days/cycle, 2 cycles/year, 100% purchased feed) Unit Total Units Quantity Cost Cost Y) ( Y) Capital Investment Pig Housing sq. m. 60 100 6,000 Other Buildings sq. m. 10 100 1,000 Equipment N.A. N.A. 600 Iotal Investment 7p600 02eratine Costs Labor manday 180 2.8 504 Weaner Pigs (20 kg) number 62 60 3,720 Compound Feed (Cony. 3.8:1) kg 18,240 0.45 8,208 Veterinary Costs pig 60 3 180 Other Direct Costs pig 60 2 120 Maintenence (1% Buildings per year) 35 Equipment Deprec. (10% Capital cost per year) 30 Building Deprec. (5% Capital cost per year) 175 Management & Miscl. Costs pig 60 1 60 oa1 0perating Cots 134,032 Pigs for Slaughter (100 kgs) kg 60 2.14 12,840 Manure Value (400 kgs) kg 60 0.02 480 Iotal Sales 14,320 Profit 2= ate 2Uh Profit Lr I= k21 Lamb E1w Years 2-11 91k Internal ILa Rugrn L_Qfr R ( 111ar Residual Val ue:3 i21 - 192 - ANNEX 7 LIVESTOCK SECTOR STUD Financial Results fQr Pigs: Fattening At Negotiated Prices (Specialized Household, 1 cycle/year,crossbred sows mostly purchased feed) Unit Total Units Quantity Cost Cost (Y) (Y) Capital Investment Pig Housing sq. m. 80 100 8,000 Other Buildings sq. m. 10 100 1,000 Equipment N.A. N.A. 600 6 300 1800 Total Investment 11,400 Operating Costs Labor manday 365 2.8 1,022 Replacement Sows number 2 300 600 Compound Feed kg 26,640 0.45 11,988 Bran kg 6,940 0.25 1,735 Coarse Feed kg 3,130 0.08 250 Green Feed kg 7,490 0.04 300 Veterinary Costs pig 6 25 150 Other Direct Costs pig 6 30 180 Maintenence (1% Buildings per year) Equipment Deprec. (10% Capital cost per year) 60 Building Deprec. (5% Capital cost per year) 450 Management & Miscl. Costs pig 6 15 90 Ioal Operating Costs 16,91r Pigs for Slaughter (100 kgs) kg 82 2.14 17,548 Cull Sows (130 kg l.w.) 2 1.6 416 Manure Value (500 kgs/pig) kg 82 0.02 820 Iotal Sales 18,784 Profit 2=r Batch 1o86 Profit Lar I186 Lahb Elw Yas 2-11 Internal IIg _f Return (Over 111Years - Residual Value: Y6,048) 18.61 - 193 - ANNEX 7 CHINA Table 1 LIVESTOCK SECTOR SIUDY Financial Results S=t Pigs: Fattening at LW& 2ri:L= (Specialized Household, 1 cycle/year,crossbred' sows mostly purchased feed) Unit Total Units Quantity Cost Cost (Y) (Y) C3pital Investment Pig Housing sq. m. 80 100 8,000 Other Buildings sq. m. 10 100 1,000 Equipment N.A. N.A. 600 Breeding Sows number 6 300 1800 Tota1 Investment 11,400 Operating Costs Labor I manday' 365 2.8 1,022 Replacement Sows number 2 300 600 Compound Feed (neg. price) kg 14,596 0.45 6,568 Compound Feed (list price) kg 11,844 0.275 3,257 Bran kg 6,940 0.25 1,735 Coarse Feed kg 3,130 0.08 250 Green Feed kg 7,490 0.04 300 Veterinary Costs sow 6 215 150 Other Direct Costs sow 6 30 180 Maintenence (1% Buildings per year) 90 Equipment Deprec. (10% Capital cost per year) 60 Building Deprec. (5% Capital cost per year) 450 Management & Miscl. Costs sow 6 115 90 Total Operating Costs 14,725 Pigs for Slaughter (100 kgs) kg 82 2.1 17,220 Cull Sows (130 kg l.w.) 2 1.6 416 Manure Value (500 kgs/pig) kg 82 0.02 820 Iotal Sals 18,496 Profit Lar Bh 1I17O Profit LUr I= 3170 Lahb El2W lears 2-1 214 Internal Zatg Df Return (Over 11 Years- Residual Value: l(L.048) 19.2% - 194 - ANNEX 7 .CHINA Tle1 LIVESTOCK SECTOR SIUD Financial Results f= Dairy Cattle: Specialized Household (with purchased roughages) Unit Total Units Quantity Cost Cost , (Y) (Y) Capital Investment Cow Housing sq. m. 60 100 6,000 Other Buildings sq. m. 10 100 1,000 Equipment N.A. N.A. 1,000 Purchased Heifers number 6 3,000 18,000 Iotal Investment 26,000 Operatina Cnts Labor manday 365 2.8 1,008 Replacement Cows number 1 3,000 3,000 Compound/Fine Feed (neg. pr.) kg 9,000 0.27 2,430 Roughages kg 21,000 0.05 1,050 Green Feed kg 72,000 0.025 1,800 Veterinary/AI Costs cow 6 50 300 Other Direct Costs cow 6 20 120 Maintenence (1% Buildings per year) 70 Equipment Deprec. (10% Capital cost per year) 100 Building Deprec. (5% Capital cost per year) 350 Management & Miscl. Costs cow 6 20 120 Total 00erating Costs 10,348 Milk Sales (3,200 1/cow) liters 6 0.56 10,752 Female Calves calf 2 600 1,200 Male Calves calf 3 20 60 Cull Cows (500 kg l.w.) kg 1 2.00 1,000 Manure Value (5,000 kg/cow) kg 6 0.015 450 1otal Sale 13462 Profit Psx I= 011 Internal 2,-Le of Return (Ovr 11 Years -Residual Value:.Y1r5,r00) 10,gs - 195 - ANNEX 7 CHIN Table 12 LIVESTOCK SECTOR STUD Financial Results fr Dair Cattle: State farm (with purchased roughages) Unit; Total Units Quantity Cost, Cost (Y) Y ) Capital Investment Cow Housing sq. m. 3,600 100 360,000 Other Buildings sq. m. 900 150 135,000 Equipment N.A. N.A. 100,000 Purchased Heifers number 300 4,000 1,200,000 Iotal Investment 1l79)5,000 ODerating Costs Labor manday 60x300 2.8 50,400 Replacement Cows number 20%x300 4,000 240,000 Compound/Fine Feed (neg. pr.) kg 2,500x300 0.27 202,500 Roughages kg 2,800x300 0.05 42,000 Green Feed kg 10,000x300 0.025 75,000 Veterinary/AI Costs cow 300 50 15,000 Other Direct Costs cow 300 20 6,000 Maintenence (1% Buildings per year) 4,950 Equipment Deprec. (10% Capital cost per year) 10,000 Building Deprec. (5% Capital cost per year) 24,750 Management & Miscl. Costs cow 300 20 6,000 Iotal Operating Costs 676,600 Milk Sales (4,500 1/cow) liters 300 0.56 756,000 Female Calves calf 120 600 72,000 Male Calves calf 120 20 2,400 Cull Cows (500 kg l.w.) kg 54 2.00 54,000 Manure Value (5,000 kg/cow) kg 300 0.015 22,500 1otal Sa1es 906,900 Profi 2 4' 240,300 LAht Elgx Years 2=11 2650-Q';0 Internal iaU _Qf Return (Ove 11 iears Residual Value: Y997,500 11.9% DISTRIBUTORS OF WORLD BANK PUBLICATIONS ARGENTINA FRANCE KENYA SPAIN Carlos Hirsch, SRL World Bank Publications Africa Book Service (E.A.) Ltd. Mundi-Prensa Libros, S.A. Galeria Guemes 66 Avenue d'lena P. 0. Box 45245 Castello 37 Florida 165, 4th Floor-Ofc. 453/465 75116 Paris Nairobi 28001 Madrid 1333 Buenos Aires GERMANY, FEDERAL KOREA, REPUBLIC OF SRI LANKA AND AUSTRALIA, PAPUA NEW REPUBLIC OF Pan Korea Book Corporation THE MALDIVES GUINEA, FIJI, SOLOMON UNO-Verlag P. 0. Box 101, Kwangwhamun Lake House Bookshop ISLANDS, AND VANUATU D-5300 Bonn 1 Seoul P.O. Box 244 Info-Line Simrockstrasse 23 KUWAIT 100, Sir Chittampalam A. Gardiner Overseas Document Delivery GREECE MEMRB Mawatha Box 506, GPO KEME P.O. Box 5465 Colombo 2 Sydney, NSW 2001 24, [ppodamou Street MALAYSIA SWEDEN AUSTRIA Athens-11635 U rstyof Malaya Cooperative ABr Fnte Kt ngts A-rldl Win o GUATEMALA Booksh op, Limited Hovbokhandel A-1011 Wien Librerias Piedra Santa P. 0. Box 1127, Jalan Pantai Baru Regeingsgatan 12, Box 16356 Graben 31 Centro Cultural Piedra Santa Kuala Lumpur S-103 27 Stockholm BAHRAIN 11 calle 6-50 zona 1 MEXICO For Subscription orders: MEMRB Information Services Guatemala City INFOTEC Wennergren-Williams AB P.O. Box 2750 HONG KONG, MACAU San Fernando No. 37 Box 30004 Manama Town 317 Asia 2000 Ltd. Col. Toriello Guerre S-104 25 Stockholm BANGLADESH 6 Fl., 146 Prince Edward Road, W, Thalpan, Mexico D.F. Micro Industries Development Kowloon SWITZERLAND Assistane Society (MIDAS) Hong Kong Soete dEtudes Marketing 6 Rue Grenus G.P.O. Box HUNGARY Marocaine Case postal 381 Dhaka Kultura 2 Rue Moliere, Bd. d'Anfa CH 1211 Geneva 11 BELGIUM P.O. Box 139 Casablanca Publications des Nations Unies 1389 Budapest 62 THE NETHERLANDS Oxford University Press Av. du Roi 202 INDIA . Medical Books Europe, BV (MBE) P. 0. Box 5299 1060 Brussels For single titles Noorderwal 38, Dar es Salaam BRAZIL UBS Publishers' Distributors Ltd. 7241 BL Lochem Publicacoes Teccas Post Box 7015 A rtment Store Internacionais Ltda. New Delhi 110002 NEWl ZLaND Centramtin06Slo Depatetd tr Rua Peixoto Gomide, 209 10 First Main Road Hri LirrBn nomto 0aRoad 01409 Sao Paulo, SP Gandhi Nagar Private Bag CANADA Bangalore 560009 New Market TRINIDAD & TOBAGO, Le Diffuseur Apeejay Chambers, P.O. Box 736 Auckland ANTIGUA, BARBUDA, 5 Walace SreetBARBADOS, DOMINICA, C.P. 85, 1501 Ampere Street 5 alc tetNIGERIA GRENADA, GUYANA, Boucherville, Quebec Bombay 400001 Uiest rs iie OTERT T IT CPu585, meee8tee1-B BCh righee Lane Niriy Press Limited MRTSRRT ST KCT,S Calcutta 700016 Three Crowns Building Jencho AND NEVIS, ST. LUCIA, CHILE Private Mail Bag 5095 ST. VINCENT & GRENADINES Editorial Renaciniento 7/188, 1(A), Swarup Nagar Ibadan Systematics Studies Unit Miraflores 354 Kanpur 208001 NORWAY 55 Eastern Main Road Santiago Sivaganga Road Tanum Karl Johan, A.S. Curepe COLOMBIA Nungambakkam P. O. Box 1177 Sentrum Trinidad, West Indies Enlace Ltda. Madras 60003 Oslo 1 TUNISIA Carrera 6 No. 51-21 A 0 Ra ra Nagar PAKISTAN Societe Tunisienne de Diffusion Bogota D.E. Mirza Book Agency 5 Avenue de Carthage Apartado Aereo 4430 For subscription orders 65, Shahrah-e-Quaid-e-Azam Tunis Cali, Valle Universal Subscription Agency P.O. Box No. 729 TURKEY Pvt. Ltd. Lhr ae iaeiAS COSTA RICA 18-19 Community Centre Saket Lahore 3 Haset Kitapevi AS. Libreria Trejos New Delhi 110 017 PERU 469, Istiklal Caddesi Calle 11-13 Editorial Desarrollo SA Beyoglu-Istanbul Av. Fernandez Guell INDr mited Apartado 3824 UGANDA San Jose Jl. Sam Ratulangi 37 Lima Uganda Bookshop COTE D'IVOIRE Jakarta Pusat THE PHILIPPINES P.O. Box 7145 Centre d'Edition et de Diffusion P.O. Box 181 National Book Store Kampala Africaines (CEDA) IRELAND 701 Rizal Avenue UNITED ARAB EMIRATES 04 B.P. 541 TDC Publishers Metro Manila MEMRB Gulf Co. Abidjan 04 Plateau * 12 North Frederick Street PORTUGAL P. 0. Box 6097 CYPRUS Dublin 1 Livraria Portugal . Sharjdh a MEMRB Information Services ISRAEL Rua Do Carmo 70-74 UNITED KINGDOM P.O. Box 298 The Jerusalem Post .1200 Lisbon Micronfo Ltd. Nicosia' The Jerusalem Post Building SAUDI ARABIA, QATAR ,. P. 0. Box 3 DENMARK P.O. Box 81 Jarir Book Store Alton, Hampshire GU 34 2PG SamfundsLitteratur Romena Jerusalem 91000 P. 0. Box 31% England Rosenoe drs Alle 11 ITALY Riyadh 11471 VENEZUELA DK-1970 Frederiksberg C. Licosa Commissionaria SINGAPORE, TAIWAN, Libreria del Este DOMINICAN REPUBLIC Sansoni SPA BURMA, BRUNEI Aptdo. 60.337 Editora Taller, C. por A. Via Lamarmora 45 Information Publications Caracas 1060-A Restauracion Casella Postale 552 Private, Ltd. WESTERN SAMOA Apdo. postal 2190 50121 Florence 02-06 1st Ft., Pei-Fu Industrial Wesley Bookshop Santo Domingo Bldg., 24 New Industrial Road P.O. ox 207 EGYPT, ARAB REPUBLIC OF Eastn Singapore Apia Al Ahram 37-3, Hongo 3-Chome, SOUTH AFRICA YUGOSLAVIA Al Galaa Street Bunkyo-ku 113 For single titles Jugoslovenska Knjiga Cairo Tokyo Oxford University Press YU-11000 Belgrade Trg Republike FINLAND JORDAN pSouthern Africa FILN ODNP.O. Box 1141 ZIMBABWE Akateeminen Kirjakauppa Jordan Center for Marketing Cape Town 8000 Textbook Sales Pvt. Ltd. P.O. Box 128 Research Box 3799 SF-00101 P.O. Box 3143 For subscription Orde Harare Helsinki 10 Jabal International Subscription Service Amman P.O. Box 41095 Craighall Johannesburg 2024  The World Bank Headquarters European Office Tokyo Office 1818 H Street, N.W. 66, avenue d'lna Kokusai Building Washington, D.C. 20433, U.S.A. 75116 Paris, France 1-1 Marunouchi 3-chorne Telephone: (202) 477-1234 Telephone: (1) 47.23.54.21 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