China - Rubber Development Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No. 4521-CHA STAFF APPRAISAL REPORT CHINA RUBBER DEVELOPMENT PROJECT November 7, 1983 Projects Department FIL U East Asia and Pacific Regional Office This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  CURRENCY EQIJIVALENTS Currency TJnit - Yuan (Y) US$1.00 = Y 1.90 (at time of appraisal) Y 1.00 = US$0.53 FISCAL YEAR January 1 to December 31 WEIGHTS AND MEASURES 1 meter (m) = 3.28 feet (ft) 1 kilometer (km) = 0.62 miles 1 hectare (ha) 2.47 acres I ton - 1,000 kg 2,205 pounds 1 kilogram (kg) = 2.2 pounds PRINCIPAL ABBREVIATIONS AND ACRONYMS IJSED CGB - Central General Bureau of State Farms and Land Reclamation (Beijing) MAAF - Ministry of Agriculture, Animal Husbandy and Fisheries PGB - Provincial General Bureau of State Farms and Land Reclamation (Guangdong) SCATC - South China Academy of Tropical Crops, Danxian County, Hainan Island UNDP - United Nations Development Program drc - dry rubber content gdp - gross domestic product ha - hectare ICB - international competitive bidding kcal/sq cm - kilocalories per square centimeter kph - kilometers per hour hp - horsepower km - kilometer kv - kilovolt kw - kilowatt pto - power take off sq m - square meter NR - natural rubber p.a. - per annum SR - synthetic rubber  FOR OFFICIAL USE ONLY CHINA: RUBBER DEVELOPMENT PROJECT STAFF APPRAISAL REPORT Table of Contents Page No. I. BACKGROUND ..... . . . . . . . . . . . . . . . . . . . . . 1 The Agricultural Sector .... . . . . . . . . . . .... 1 The State Farm System .... . . . . . . . . . . . .... 3 Rubber Production and Utilization . . . . . . . . . . . . . 4 Project Formulation .... . . .5 Bank Group Lending for Agriculture . . . . . . . . . . . . . 6 II. THE PROJECT AREA . . . . . . . . . . . . . . . . . . . . . . . 7 Location. 7 Climate ..... . . .. . . . . . . . .. . . . . . . . . 7 Topography and Soils .... . . . . . . . . . . . . . . . . 8 Agricultural Production, Incomes and Labor Force . . . . . . 9 Infrastructure and Social Services . . . . . . . . . . . . . 10 III. THE PROJECT.11 Project Description . . . . . . . . . . . . . . . . . . . . 11 Rubber Establishment .... . . . . . . . . . . . . . . . . 12 Windbreak Establishment .... . . . . . . . . . . . . . . 14 Construction, Agricultural and Transport Machinery . . . . . 14 Wood Processing Facilities .... . . . . . . . . . . . . . 14 Buildings and Utilities .... . . . . . . . . . . . . . . 15 Rural Roads . . . . . . . . . . . . . . . . . . . . . . . . . 16 Training and Research .... . . . . . . . . . . . . . . . 16 Technical Assistance .... . . . . . . . . . . . . . . . . 16 Status of Design .... . . . . . . . . . . . . . . . . . . 17 Implementation Schedule .... . . . . . . . . . . . . . . 17 Cost Estimates ..... . . . . . . . . . . . . . . . . . . 18 Financing ..... . . . . . . . . . . . . .. . . . . . . 18 Procurement ..... . . . . . . . . . . . . . . . . . . . 19 Disbursements ..... . . . . . . . . . . . . . . . . . . 20 This report is based on the findings of a preappraisal mission to China in October 1982 and an appraisal mission in April 1983. Members of the pre- appraisal mission were T.J. Goering, T.B. Wiens, W. Byrd (IDA) and P.L. Lim and J.G. Devitt (FAO). The appraisal mission comprised Messrs. Goering, Wiens, A. Green (IDA), Lim and Devitt. This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  Page No. Accounts and Audit .... . . . . . . . . . . . . . . . . . 20 Environmental Effects . . . . . . . . . . . . . . . . . . . 21 IV. ORGANIZATION AND MANAGEMENT . . . . . . . . . . . . . . . . . 21 Project Management and Coordination . . . . . . . . . . . . 21 Financial Management . . . . . . . . . . . . . . . . . . . . 23 Execution of Project Works . . . . . . . . . . . . . . . . . 24 Management of State Farm-Commune Joint Ventures . . . . . . 26 Monitoring and Evaluation ... . . . . . . . . . . . . . . 27 Agricultural Support Services . . . . . . . . . . . . . . . 27 V. PRODUCTION, PROCESSING, MARKETING AND PRICES . . . . . . . . . 29 Rubber Nurseries .... . . . . . . . . . . . . . . . . . . 29 Rubber Planting and Maintenance . . . . . . . . . . . . . . 30 Intercropping and Cover Crops . . . . . . . . . . . . . . . 30 Windbreaks . . . . . . . . . . . . . . . . . . . . . . . . . 31 Tapping and Production . . . . . . . . . . . . . . . . . . . 31 Processing and Marketing . . . . . . . . . . . . . . . . . . 31 Financial and Economic Prices . . . . . . . . . . . . . . . 33 VI. BENEFITS, JUSTIFICATION AND RISKS . . . . . . . . . . . . . . 34 Production Benefits . . . . . . . . . . . . . . . . . . . . 34 Employment and Income . . . . . . . . . . . . . . . . . . . 35 Financial Analysis . . . . . . . . . . . . . . . . . . . . . 36 Cost and Rent Recovery . . . . . . . . . . . . . . . . . . . 36 Economic Analysis . . . . . . . . . . . . . . . . . . . . . 37 Project Risks .... . . . . . . . . . . . . . . . . . . . 39 VII. AGREEMENTS TO BE REACHED AND RECOMMENDATION . . . . . . . . . 39 TABLES IN TEXT 3.1 Distribution of New Plant and Replant Area by Prefectures . 13 3.2 Summary of Project Costs . . . . . . . . . . . . . . . . . . . 19 4.1 Distribution of Profits from the State Farm System, Guangdong Province, 1980 and 1981 ............ . 25 6.1 Summary of Rate of Return Analysis . . . . . . . . . . . . . . 38  - iii - ANNEXES 1. Background Data 2. Cost Data 3. Data on Production, Processing and Marketing 4. Economic and Financial Analysis 5. Accounting and Disbursement Procedures for Works Executed by Force Account 6. Glossary of Technical Terms 7. Selected Documents and Data Available in the Project File LIST OF FIGURES 3.1 Implementation Schedule 4.1 Organizational Structure of the Ministry of Agriculture, Animal Husbandry and Fisheries 4.2 Organizational Structure of the Provincial General Bureau and Prefectural Bureaus of State Farms and Land Reclamation, Guangdong Province 4.3 Representative Organizational Structure of a State Farm in the Project 4.4 Organization for Project Implementation 4.5 Proposed Layout of the Sawmill and Woodworking Shops on State Farms MAP IBRD 16867 - Project Areas   CHINA RUBBER DEVELOPMENT PROJECT I. BACKGROUND 1.01 The Government of China has requested Bank Group assistance in financing a Rubber Development Project in Guangdong Province, South China. The primary objective of the project is to increase the domestic production of natural rubber and thereby reduce the dependence on imports which in 1979-81 accounted for 50% of total (natural plus synthetic) rubber consumption in China. The project would assist in the new planting and replanting of 40,000 ha of rubber, of which 36,000 would be on state farms and 4,000 on nearby communes. The project would provide for the planting of protective windbreaks in typhoon-prone areas and the expansion of intercropping in immature rubber. It would include expansion of the farm machinery and transport fleets on participating state farms, upgrading of sawmills and timber utilization facilities, construction of housing, administrative and service buildings on state farms, and the construction of road links within state farms. The project would strengthen the training programs for staff and workers of state farms and upgrade research and extension activities. The technical assistance component would facilitate access by China's natural rubber industry to improved technology developed in other rubber producing countries over the last three decades. The assistance to communes provided through the project would support a recent initiative by the state farm system to provide technical and financial help to the poorly developed rubber production activities in the communes. The Agricultural Sector 1.02 Agriculture, including livestock, forestry and fisheries, provides basic sustenance to China's population of just over one billion people. This sector accounts for some 30% of the country's gdp and is the main source of livelihood for the rural population of nearly 800 million. Only about 100 million ha of China's land area of 960 million ha are arable. Farming systems are intensive, with large inputs of labor and chemical and organic fertilizers. Nearly half of the arable land is irrigated and more than half of the arable area grows at least two crops per year. Foodgrains occupy 80% of the cropland and account for 65% of total agricultural output. This intensive system of production allows China to meet the basic food needs of its population (about 22% of the world's total) from less than 8% of the world's arable land. Grain yields are about 50% greater than those in other developing countries. Agricultural production is carried out in communes, state farms and privately managed plots which account for about 87, 5, and 8%, respectively, of the country's arable land and 80, 4 and 16%, respectively, of the gross value of agricultural production.  - 2 - 1.03 Agricultural development has been accorded high priority since the founding of the People's Republic. Investment in agriculture has averaged about 20% of total national investment and recent economic reforms have been designed to increase the efficiency of investment in the sector. Despite declining per capita availability of arable land, efforts to develop China's agriculture have been broadly successful. Food production has expanded more rapidly than population growth and a food security system has been put into place which provides adequate basic nutrients to most of China's population. The agricultural sector grew by about 3.2% p.a. in the 1955-79 period and by more than 5% on average since 1979. Reflecting population and income growth and diversion of larger quantities of grain to livestock feeding, net grain imports have risen in recent years to about 15 million tons p.a., or less than 5% of total consumption. Exports of raw and processed agricultural products contribute about 80% of all foreign exchange earnings. 1.04 The Government controls the pricing and marketing of most agricul- tural products and inputs. It fixes annual production quotas for the major crops such as grains, oilseeds and industrial crops (cotton, sugar and rubber) and purchases them at predetermined prices. Minor crops, such as vegetables, and other products grown in household plots, are sold in rural markets at prices regulated by the provinces or counties. Economic reforms since 1976 have decentralized production decisions, substantially increased prices paid for most agricultural products, increased the maximum area allowed in household plots, and expanded the role of rural markets. Instit- utional reforms in 1982 combined several government agencies involved in agricultural product marketing and input distribution to improve the efficiency of these activities. 1.05 A major task of China's agricultural sector will be to meet the food and agricultural raw material requirements occasioned by population growth and gains in per capita income. The land constraint will require steady increases in yields from already cropped areas and the growing problem of surplus rural labor will make job creation an increasingly important policy objective of Government. To meet these challenges the Government is implementing new policies aimed at giving rural workers more production incentives. Area quotas for particular crops have been relaxed, although minimum grain market- ing quotas are still maintained. Production incentives are encouraging crop specialization in economically suited areas and the production of industrial crops is being fostered by assuring supplies of food grains to farm workers in those areas. Incentives to workers on both communes and state farms are being improved to link reward more closely with work effort. Within the state farm system, financial reforms allow individual farms to retain a large share of any profits earned in excess of stipulated quotas. In addition, it will be necessary to use the few remaining opportunities to expand the cultivated area, increase cropping intensity and yields on existing cultivated areas through relatively quick-yielding land and water resource projects, increase the efficiency of investment, and introduce higher-yielding crop varieties and practices through research and education.  -3- The State Farm System 1.06 State farms, which predate the founling of the People's Republic, were established to provide food to the revolutionary armies and were important sources of employment for demobilized soldiers. Located largely in less-dovelopel borer regions (the provinces of Heilongjiang, Nei Monggol, :4ingxia, Gansu, Xin1iant, Qinghai, Xizang, Yunnan an-I Hainan Island in O,uangdong Province), these farms have become imnortant sources of commercialized agricultural prodruction to meet requirements of the cities and food deficit areas in the countryside. As a national average, more than 35% of grain production on state farms is sold to the state. This contrasts to some 15" delivered to the state from the collective sector. The approximately 2,100 state farms in China, most of which are under the policy direction of the Central General Bureau of State Farms and Land Reclamation (CGB) within the Ministry of Agriculture, Animal Hushandry and Fisheries (MAAF) and its provincial bureaus, manage almost 29 million ha. About 4.5 million ha total are under cultivation, with much of the balance being grass lands, forests or land with limited potential for sustained crop production. Nearly half of the state farms specialize in grain or cotton production, about a quarter in livestock production and the balance in tea, fruit, aquaculture and other specialty crops. Some '3% (170) are involved largely in rubber production. Farm size averages about 12,000 ha, but varies widely from 2 - 3,000 ha in the South to 90,000 ha in the Northeast. 1.07 The population on state farms in 1980 was 11.4 million and state farm employment was just under 5 million (about one worker per cultivated ha, as compared with three to four workers per cultivated ha in the collec- tive sector). Output per worker on state farms is considerably greater than in the communes, reflecting larger per capita quantities of land, machinery and equipment with which to work. NXational average earnings per worker on state farms in 1932 w7ere about US$350 equivalent, including bonuses. Workers on state farms receive a monthl7 wage, annual wage bonuses which depend on the financial performance of the farms, and subsid1ze: social services including housing, education and health. As relatively self-sufficient economic units operating 'n noorl- developed areas, individual state farms must finance, construct and maintain a system of social an1 physical infrastructure (health, education, transportation, communications and noT7er) which is usually the responsibility of central and local governments in other countries. 1.03 In 1982 thne gross value (in 1980 prices) of production on state farms was Y 11.6 blllion, of which Y 6.6 billion came from agricultural output and Y 5.0 billion from industrial production. In recent years, state farms have accounted for about 2% of the country's grain production, nearly 6% of sugar outpuit, 3% of the cotton, and 95% of the natural rubber production. In the 1975-31 period, the number of state farms declined by about 10% (largely through consolidation), while the cropped area on these farms rose by 13% and the gross value of output by 50%. After realizing financial losses for many  -4- years, the state farm sector in 1979, 1980 and 1981 reported aggregate profits of Y 397 million, Y 659 million and Y 192 million, respectively (the sharp decline in 1981 reflecting a weather-induced loss of Y 346 million on state farms in Heilongjiang Province). This improved financial performance is due largely to higher procurement prices for state farm output and, possibly, more efficient operations growing out of managerial reforms which decentralize decision-making and enhance production incentives at the family or production group level. 1.09 To meet future requirements for food and agricultural raw materials, the Government plans a major land development program by the state farm system in the current Five Year Plan (1981-86). This calls for the development of 1.1 million ha of unused land for annual field crops, 60,000 ha for tree crops, and 2.2 million ha for improved pasture. Total estimated cost of the program is Y 7.1 billion (US $3.7 billion). Although the major focus is on food crop development in the Northeast, considerable emphasis also is to be given elsewhere to pasture and meat production and production of raw materials such as cotton and rubber. Rubber Production and Utilization 1.10 The rubber tree (Hevea brasiliensis) was introduced into China in 1904, but the planted area expanded very slowly and production in 1949 totalled only about 200 tons. Extensive planting of rubber, largely under the aegis of the state farms, began in 1952 and focused primarily on Hainan Island and Zhanjiang Prefecture in Guangdong Province. Until 1967 virtually all plantings were with unselected, inherently low yielding materials. New planting since the mid 1960s has been rapid, averaging about 10,000 ha annually in 1967-81 in Guangdong, the major rubber producing province. Replanting of over-aged trees began in 1975 and totalled about 4,800 ha by 1981. In 1981 the area under rubber in China totalled 410,000 ha, of which 320,000 were in the state farm sector and perhaps 80,000 ha were in communes, the balance being on farms of Overseas Chinese and in experimental stations. More than 80% of the total planted area is in Guangdong Province. Another 13% is in Yunnan Province and the balance is in Guangxi and Fujian Provinces. An estimated 41% of China's rubber area is immature; 36% is mature seedling rubber; and 23% is mature budded rubber. 1.11 In Guangdong Province, the rubber area on state farms totals 264,300 ha (1981), or about 79% of the province's total area planted to this crop. The area under commune rubber is about 67,000 ha, all of which is planted to low yielding, unselected, generally over-age trees. Some 39% of the provincial rubber area is immature, while 40% is under mature seedling rubber. On state farms in the province, 36,900 ha (14% of total state farm area) are planted to trees which are 29 years of age or older. Yields on state farms in Guangdong Province have grown at 3.4% p.a. (1969-71 to  1979-81) and averaged 670 kg per tapped ha in 1981. This figure can be compared with recent yields of about 1,200 kg/ha in Malaysia, 680 kg in Thailand and 670 kg in Indonesia. State farm yields from old, unselected seedling rubber in China are in the 350-500 kg/ha range, while yields of improved clones planted on a commercial scale on these farms have averaged about 1,130 kg/ha in the first nine years of tapping. Rubber yields on communes in 1981 averaged 106 kg per planted ha. Low yields on communes reflect the dominance of old, unselected trees as well as the virtual absence of chemical fertilizers. Although communes occupy about 20% of China's rubber area, they account for less than 5% of total production. They are in need of technical assistance in rubber production and processing and a systematic program to replant over-age rubber. China's rubber is produced within a range of 18 to 24 degrees north of the equator, or well beyond the latitudes of other major producers of natural rubber. China's major rubber growing areas are subject to typhoons on Hainan Island and low temperature on parts of the mainland but production systems involving windbreaks have been developed to deal with the wind problem and breeding programs have made some progress in producing cold-resistant planting materials. 1.12 Domestic production of natural rubber grew by 9.6% p.a between 1970-72 and 1979-81 to 115,000 tons in 1981 (Annex 1, Table 4) while domestic output of synthetic rubber expanded by 16% p.a. during this period to 124,900 tons in 1981. With total imports of NR and SR averaging 232,000 tons p.a. in 1979-81, total consumption of rubber (NR and SR) is in the range of 464,000 tons yearly. Consumption grew by 4.3% p.a. between 1970-72 and 1979-81. About half of recent consumption came from imports of SR and NR; 27% consisted of domestic production of SR and 23% came from local production of NR. The rapid growth in domestic production of SR and NR, albeit from a relatively small base, has permitted a steady decline in the share of consumption provided by imports. Annual per capita consumption of rubber in China (about 0.5 kg) is less than 20% of the world average. Reflecting population and income growth, the domestic market for natural rubber is expected to grow rapidly over the next several years (para. 5.10). Project Formulation 1.13 A project proposal, prepared by the Guangdong Provincial Bureau of State Farms and Land Reclamation, was reviewed by an IDA mission in April 1982. During that and subsequent missions, discussions involving the Provincial Bureau, IDA staff and consultants led to several refinements of the project's design and technical features. These included: (a) expansion of the new planting share to increase the number of jobs created; (b) introduction of a component to assist rubber production on nearby communes; (c) addition of a technical assistance component to provide for overseas training and technical services;  (d) expansion of local training programs for staff and workers on state farms and communes; (e) expansion of research and trials in rubber and windbreak development; (f) expansion of intercropping in immature rubber and introduction of private intercropping by families on state farms; (g) adjustments in the mix of rubber clones to better match clonal characteristics to weather conditions; (h) adjustments in the types and rates of chemical fertilizer applications to use lower cost nutrient sources; (i) wider spacing of windbreak plantings to increase timber produc- tion without sacrificing wind protection; (j) greater emphasis on intensive tapping and yield stimulation by chemical application in old rubber before felling; (k) lesser emphasis on de-stumping of old rubber lands to reduce land preparation costs in replanting; (1) introduction of improved timber treatment and processing facilities; (m) introduction of sprinkler irrigation systems in rubber nurseries; and (n) organization of preproject study tours in 1983 to Japan and Thailand to study rubber wood processing and technical and administrative aspects of Bank Group-supported projects in rubber development. Bank Group Lending for Agriculture 1.14 Projects are being selected which generate increased output and have significant demonstration and catalytic effects. The first Bank Group operation in agriculture, a $60 million IDA credit for the North China Plain Agriculture Project (Credit 121-CHA), was approved by the Executive Directors on June 15, 1982 and will provide improved irrigation and drainage for 200,000 ha in nine counties. Work on this project, the first large scale approach to problems of soil salinity and water logging in China, is proceeding well. A second project, the Agricultural Education and Research Project (Credit 1297-CHA) for $75 million, was approved by the Executive  - 7- Directors on November 2, 1982 and will support the Government's program of strengthening higher education and research in key agricultural colleges and research institutes. The Heilongjiang Land Reclamation Project (Loan/Credit 2261/1347-CHA), which was approved on April 18, 1983, will develop 200,000 ha on state farms in Northeast China by constructing surface drains and access roads and providing various agricultural machinery and other support services to increase marketable supplies of wheat, maize and soybeans. The proposed Rural Credit Project is to focus on Guangxi and Hubei Provinces and is expected to include assistance for aquaculture, livestock and poultry, tree crops and agro-industrial processing. Discussions with the Government are underway on possible Bank Group assistance in forestry, agricultural research and extension. II. THE PROJECT AREA Location 2.01 The proposed project is in Guangdong Province, South China, located between 18 and 25 degrees north latitude and 109 and 117 degrees east longi- tude. Most of the project farms are within 18 to 22 degrees north latitude. Boundaries of participating farms are shown in the Map (IBRD 16867). The project area comprises parts of Hainan, Tongza and Zhanjiang Prefectures. Hainan and Tongza Prefectures constitute the land area of Hainan Island and are located approximately on the southeastern and northwestern halves of the island. Zhanjiang Prefecture is on the mainland in southwest Guangdong Province. Of the 40,000 ha in the project, 43% are located in Hainan Prefecture, 42% in Tongza and 15% in Zhanjiang. Almost 90% of the area to be new planted is located in Hainan and Tongza Prefectures where unused lands are more readily available. Climate 2.02 Climate in the project area is distinctly monsoonal in character, with marked wet and dry seasons, high solar radiation and abundant rainfall (Annex 1). Mean monthly temperatures are in the 23-250C range, with highs in June and July (280C) and lows in January (16-200C). Zhanjiang has the widest intermonthly temperature variations (12 degrees) and the lowest temperatures, averaging 160C in January and February, but with minimum temperature as low as 2-30C reported in the northern hill regions. Annual mean hours of sunshine varies from about 1,200 in Zhanjiang to just over 2,100 in Hainan, while annual solar radiation is in the range of 110-140 kcal/sq cm. Low temperatures in Zhanjiang Prefecture reduce the number of tapping days (about 215 p.a. in Zhanjiang; 230 on Hainan Island) and slow tree growth by perhaps 10%.  - 3 - 2.03 Average annual rainfall ranges from 1,600 mm in Tongza to 2,000 mm in Hainan and declines moderately from east to west and south to north. About 35%O of annual rainfal.1 occurs in the April-Octnber period, with August and September the wettest months. December and January tend to be the driest months, although the period of sparse rainfall occasionally extends to February, March andl Aeril. Reflecting the area's proximity to the sea, humidity levels are high throughout the year. 2.04 The project area is subject to typhoons, most of which occur between May and November and reach their greatest frequency and severity in the July-September period. Moving in a northwesterly direction, these storms tend to affect Hainan Prefecture most frequently buit occasionally hit Tongza and Zhanjiang. In the 12-year period (1970-81) 27 typhoons struck Hainan Island with peak wind velocities of Force 8 or above on the Beaufort scale. Most of the storms have maximum winis in the 8-11 range (68-94 kph), although some have ratings of up to 12 (125 kph). Tree damage, which tends to be highly localized, may afFect 1-2% of the planted area each year. A few o' the trees in the affected area are totally destroyed, bht most suffer lesser damage. Rubber pro-luction declines most sharply in the period immediately following the storm and may eventuallv largelT recover with subsequent regrowth of the tree canopy. Cultural practices to minimize wind damage include pruning of trees to reduce wind resistance, establishment of win1"reaks around rubber plantings and planting of wind resistant varieties of rubber. These practices, singly or in combination, have substantially reduced the adverse effects of typhoons on China's rubber industry. Topography and Soils 2.0r Ilainan Island is high near its center, with contour lines in approximate concentric circles around lWuzhi Mountain (1,867 m above sea level). The island periphery tends tz be flat to gently undulating and consists largely of coastal and alluvial plains. The relatively large areas between the mountains and the coastal plains are foothills and valleys with soils derived largely from arenaceous and metamorphic rock. The topography of Zhanjiang Prefecture is flat to undulating and slopes downward from north to south. Geologic features of Thanjiang include granitic and gneissic mountains in the north and northeast. The southern coastal area is largely sedimentary plain, while the Leizhou Peninsula in Zhanjiang consists of sedimentary and basaltic platforms. Topography in Elainan and Tongza Prefectures is consider- ably more broken than in Zhanjiang. In Hainan and Tongza 47% and 64% of the project areas, respectively, have slopes of 36% (200) or less. In Zhanjiang all of the project area has slopes of 36% or less. ?.06. Soils in the project area vary widely from site to site, reflect- ing different parent materials and the influence of differing topographical and climatic factors. Major soil types include the granitic red loams, the volcanic laterites, coastal clays and sands and the yellow podzolic soils at higher elevation. The granitic loams are located mainly in the central and  - 9 - southwestern parts of Hainan Island and the northern part of Zhanjiang Prefecture. These soils have deep profiles and are well suited for rubber production. The volcanic laterites derive largely from basaltic rocks, have considerable iron and aluminum content, and are located in the northern and northeastern parts of Hainan Island and on the southern part of Zhanjiang Prefecture. These deep, clayey soils, many of which are in typhoon-prone areas, are suitable for rubber development if appropriate windbreaks are established. The coastal sedimentary soils vary from relatively infertile loamy sands to fertile, well-textured sand and clay loams which are already under cultivation. The yellow soils are generally at elevations above 1,000 m where steep gradients, cool temperatures and limited sunshine preclude rubber production. Most of the soils in the project farms have deep profiles, organic matter content of 2.2-2.7% and pH values in the 4.9-5.6 range. With proper fertilization and land conservation practices, which are already well developed in the project area, these soils are capable of sustaining rubber production at relatively high yields. As of early 1982, the state farm system in Guangdong Province had available 49,600 ha of undeveloped land suitable for growing rubber. Sizeable areas of such lands are also available in the communes, but remain largely undeveloped because of inadequate finance and technical assistance. Rubber is the appropriate crop choice in most of the project area when topography, soils and climate are considered. Agricultural Production, Incomes and Labor Force 2.07 Guangdong is a large province (population in 1982 of 59.2 million) with a relatively diversified agricultural sector. The land area of just over 20 million ha includes nearly 12 million ha in hills and mountains and 8 million in uplands and plains. Grain production is the most important agricultural activity (nearly 5 million ha), although the province also produces large quantitities of industrial crops (rubber, tea, sugarcane, jute and hemp) and animal and other products (pork, beef, fish, silk). Agroindustrial production is well developed and accounts for a substantial part of rural income. Within the Province, average cash incomes on state farms appear to be about 25% higher than on communes. The inclusion of benefits received in kind by state farm workers would raise this differential substantially. 2.08 The 137 state farms in the province occupy 1.1 million ha (5.4% of the provincial land area) and support a population of 1.2 million (2% of the provincial total). About 100 farms specialize in rubber production. In 1981 the gross value of agricultural and industrial production on state farms in the province was Y 715 million, of which 79% was accounted for by rubber, 12% by other crops and animal husbandry and 9% by industrial and sideline production. Other items produced in substantial quantities by state farms in Guangdong include sisal fiber (11,000 tons in 1981), tea (2,335 tons), sugar (27,700 tons), paper (2,400 tons) and fiberboard (5,000 tons). State farms in  - 10 - the province also produce cement, chemical fertilizers and various consumer products. Agricultural production in the province is supported by an extensive program of agricultural research and training. 2.09 Most of the project area is too steep for the sustained cropping of annuals. On the more level lands crops such as pineapple, sugarcane, sweet potatoes and groundnuts can be grown. Net economic returns per ha for pineapples appear to be higher than for rubber, although marketing of the fresh fruit is problematic in China and both the domestic and export markets for the tinned product are limited. Returns from most other short term crops are lower than returns from rubber. Other tree crops such as oil palm and coconut have not done well in the project area. Plantation forestry may be a possibility, although forestry would absorb much less labor per ha than does rubber. Taking into account soil, climate, topography, market opportunities and the government's objective of expanding rural employment, rubber is the appropriate crop choice for the project area. 2.10 The work force on the state farms totals 640,000, of whom 232,000 are directly employed in rubber production and processing. The labor supply continues to grow more rapidly than do remunerative work opportunities. In 1978-81 the number of middle school graduates on state farms in the province, the major source of new entrants to the state farm labor force, grew by 5.8%. The surplus labor supply on state farms in the three prefectures in 1982 is estimated to be 95,000 workers (Annex 1). 2.11 The average size of the 50 project farms is 9,500 ha, of which 2,700 ha are in rubber (Annex 1, Tables 10-12). Some of the remaining area is under food or other industrial crops, but much is not developed because of unsuitable topography or inadequate financial resources. Gross value of production in 1981 averaged Y 9.5 million per farm, of which Y 7.1 million came from rubber. Forty-six of the 50 farms realized profits in 1981. Project farms have a total population of approximately 558,000 and a labor force of 290,000. Agriculture, animal husbandry and forestry provide employment for about 200,000 workers, of whom half are in rubber production. The balance of the labor force is in administration and other support services. Infrastructure and Social Services 2.12 Guangdong is served by a relatively well-developed system of physical infrastructure. Major urban areas and state farms are linked by all-weather roads, most of which are hard-surfaced. Deep water ports serve the mainland (at Guangzhou and Zhanjiang) and Hainan Island (Haikou), Zhanjiang and Haikou are linked to the provincial capital (Guangzhou) and Beijing by regularly scheduled air service. Rail lines connect Zhanjiang with Guangzhou and other major urban centers.  - 11 - 2.13 Wit7itn-farm systerms of electric power, road transport and communi- cations are largely the responsib, lity of individual farmq, althlough there are generally good linkages to the provincial and national networks. The state farms of the province have constructed and maintain 8,400 km of trunk telephone lines and 13,400 km of local telephone lines. Every level of management (para. 4.01) above the production team can be reached by vehicle and telephone. Most farms are provided with electric power, either through connection to the provincial grid or through farm-owned power plants and transmission facilities. 2.14 Medical and educational services are available to virtually all state farm staff and workers, although these facilities are generally in need of upgrading. In addition to the eight hospitals under the general super- vision of the three administrative bureaus (essentially covering Ilainan, Tongza and Zhanjiang Prefectures), each state farm operates a hospital and each subfarm or nroduction brigade maintains a clinic with specialist medical personnel. In total, the state farms have available some 12,000 hospital beds ani a slightly larger number oF medically trained personnel. 2.15 The state farm system in the province operates 192 middle schools (102,400 students) 1,060 primary schools (209,200 stuilents), 8 technical sclhools, and 5 staff training centers. In 1932 about 33,200 students gra.Iuated crom lower and upper middle schools on state farms in the three project prefectures (of which 15,500 graduates came from farms participating in the project). III. THE PROJECT Project Description 3.01 The major objective of the project is to increase the production of natural rubber by assisting the Government-s program of new planting and replanting in Guangdong Province. The main features of the project are: (a) new planting of about 21,000 ha and replanting of approximately 15,000 ha on 50 state farms, including intercropping during part of the rubber-s immature phase on a gross area of 9,000 ha; (b) new planting of rubber on about 4,000 ha and intercropping on a gross area of 1,000 ha on nearby communes; (c) establishment of about 12,000 ha of windbreaks on state farms and commnunes; (d) expansion of the equipmnent fleets of participating state farms through procurement of trucks, wheeled and crawler tractors, other  - 12 - agricultural implements, and transportation and construction equipment; (e) upgrading of sawmills on the 19 farms engaged in replanting, and construction of two integrated timber processing facilities in the project area; (f) construction of about 765,000 sq m of buildings, including housing for staff and workers of state farms, offices, service buildings, schools, clinics and training centers; (g) construction or upgrading of bridges and culverts, about 780 km of rural roads, 400 km of power transmission lines and 780 km of telephone lines; (h) strengthening domestic training programs for staff and workers of state farms and upgrading of research and training programs in prefectural research institutes; and (i) provision of overseas training and technical assistance in rubber production and processing and estate management. Rubber Establishment 3.02 The project would replant with high-yielding clones about 15,000 ha, first planted to rubber in the early to mid-1950s with unselected, low-yielding materials, and new plant 21,000 ha of cut-over or new land on state farms. It would also new plant 4,000 ha of commune land, most of which is located in Hainan Island. This would involve plantation establishment, including land clearing and terracing, planting, maintenance and intercropping. These operations are summarized below, and further details on agronomy and cultural practices are given in Chapter V. The distribution of new plant and replant areas is shown in Table 3.1. 3.03 Land Clearing and Terracing. Land clearing would be done with a combination of manual and mechanical methods. Tree stumps would be removed by dynamiting, uprooted by tractors, or treated with chemicals and left to decay, depending on terrain and whether or not intercropping would be practiced. All land with a gradient of over 5% (30) would be contour lined and terraced by hand or tractor, depending on topography. Because of the ready availability of surplus labor, labor intensive methods would be emphasized. Terraces would slope inward at gradients of 23% (130); terrace widths would range from 1.8 m (on lands with slopes of 27% or more) to 2.5 m (slopes of 5-27%). Field roads would be constructed with an average density of 45 m per ha and would be of compacted earth with a width of 3 m.  - 13 - Table 3.1: DISTRIBUTION OF NEW PLANT AND REPLANT AREA BY PREFECTURES New plant Replant Total ----------- 000 ha------- Percent State farms Hainan 8.5 6.3 14.8 37 Tongza 11.2 4.6 15.8 40 Zhanjiang 1.3 4.1 5.4 13 Subtotal 21.0 15.0 36.0 90 Communes Hainan 2.5 0 2.5 6 Tongza 0.8 0 0.8 2 Zhanjiang 0.7 0 0.7 2 Subtotal 4.0 0 4.0 10 Total 25.0 15.0 40.0 100 3.04 Planting. Rubber clones used in the new planting/replanting program would include proven high-yield materials which are widely grown locally and a smaller proportion of promising newer materials grown less extensively in China. About 80% of the area would be planted to budded stumps and 20% to high-stumped buddings. Following land clearing, and beginning in April or May of each year, nursery materials would be transplanted to densities of 600-650 trees/ha. About 16 kg/ha of N and 70 kg/ha of P205 would be applied at planting, along with large quantities of organic manures. Leguminous covers and intercrops would be established on about 50% and 25%, respectively, of the planted area. On lands with slopes greater than about 36% (200) natural cover would be retained to avoid disturbing the soil surface, but would be enriched by planting of legumes. 3.05 Maintenance and Intercropping. Rubber is expected to come into tapping in the eighth (Tongza, Hainan) to ninth year (Zhanjiang) from planting. The fertilization program during immaturity would require N, P205 and K20 at rates averaging 160, 50 and 30 kg/ha/year, respectively. Organic manure (6-15 ton/ha/year) would also be applied, consisting of composted cattle manure and grasses cut from field borders and other marginal land. The  - 14 - manure would be supplied from cattle raised as a self-sustaining side enterprise. During years 1-3 of the planting, about 25% of the gross planted area would be intercropped in pineapple, sugarcane, peanut and sweet potato. The project would provide planting material and fertilizer for intercropping; the labor would be supplied by farm families. Windbreak Establishment 3.06 In high wind risk areas, windbreaks would be established prior to field planting of rubber. Windbreak areas, which on average would occupy about 1 ha for every 3 ha of rubber, would be cleared by bulldozer and the land prepared by disc plowing and by harrowing. Windbreaks would be 10-20 m wide and arranged in a rectangular pattern with the long axis at right angles to prevailing winds. They would divide the rubber area into blocks of 1-3 ha, depending on the nature of the terrain and severity of the winds. Species for the windbreaks would be chosen to produce a three-tiered effect to provide protection at all levels. Planting would be at a spacing of 1.5 x 2 m (3,300 trees/ha). About 125 kg/ha of N, 20 kg/ha of P205, and up to 10 tons/ha of organic manure would be applied at time of planting, followed by additional organic manure in each of years two and three after planting. Commercial thinning would begin in the 10th year of planting. Clear felling would coincide with the felling of the old rubber before replanting. Construction, Agricultural and Transport Machinery 3.07 The project would expand and replace equipment fleets on project area farms through procurement of 350 trucks (6-8 ton capacity), 10 crawler tractors (140-150 hp), 60-wheeled tractors (45-50 hp), 10 utility vehicles, and ancillary transportation, agricultural and construction equip- ment. This equipment would serve the needs of present production activities on project farms, as well as those of project construction. Trucks would be used for on-farm and long-distance (100-200 km) transport of rubber, chemical fertilizers, logs and sawn timber. Crawler tractors, equipped with tree pushers, dozer blades and plows, would be used in land clearing, site prep- aration and road building, to supplement labor-intensive methods. Wheeled tractors would be used for land preparation and on farm transport of rubber latex, chemical and organic fertilizer and other inputs. Utility vehicles, vans and automobiles would meet personnel transportation needs of farms, training and research centers and project management. Wood Processing Facilities 3.08 The project would rehabilitate sawmills and construct timber processing complexes to improve exploitation of timber resources, including wood from old rubber and windbreaks. Old rubber trees felled in the replan- ting program would yield about 30 cu m of sawlogs per ha and some fuelwood. Regular thinnings of the windbreaks would yield another 30 cu m of poles and  - 15 - fuelwood in year 10 and more thereafter. In the past, these resources have been poorly utilized, due to lack of proper treatment of rubber wood after felling and poorly equipped sawmills and woodworking shops on state farms. The project would introduce proper chemical treatment to preserve rubber logs after felling and would equip one sawmill on each state farm to increase throughput and quality and reduce waste and work hazards. Each sawmill would include a vertical bandsaw, a band resaw, log conveyor belts and secondary woodworking equipment capable of handling about 35 cu m of round logs per day on an 8-hour shift. Output would include sawn timber and simple furniture for use on the farm. In addition, two integrated timber complexes would be constructed in the project area. Each complex would include a sawmill, a furniture plant, and a plant to produce wood-based panels. It would utilize each year about 10,000 cu m of sawlogs in the furniture plant and up to 20,000 cu m of smallwood in the panel plant. Annual output per complex is expected to be about 2.5 million pieces of furniture, or furniture compon- ents, and 15,000 cu m of panels. The two complexes together would consume only about 20% of the rubber wood from the felling of old trees in the project and would continue to be supplied by logs from the ongoing rubber replanting program after the project period. Buildings and Utilities 3.09 As relatively self-sufficient, independent economic units, the state farms are responsible for providing a wide range of physical infrastructure, including housing, schools and clinics, as well as buildings for storage, training, research and administration. Project construction would include 648,000 sq m for housing, 72,000 sq m for service buildings, 39,000 sq m for schools and clinics, and 6,000 sq m for training centers. Construction of housing would be limited to that required to accommodate additional workers needed for the project. About 88% of the housing would be built in the residential site to be established in the new plant areas and 12% would be constructed in the replant areas. Construction standards would allot an area (gross) of about 10 sq m/capita for housing, 0.2 sq m/student for schools and 0.17 sq m/capita for clinics. This housing standard is modest, costing about US$500 equivalent/capita, and can be compared with the present national average of over 11 sq m/capita in rural China. Technical and managerial staff would be allocated slightly more space than other workers, although basic housing designs for all groups are similar. Some 80% of the housing would be single storey, multiple family dwellings, while 20% would be two storey buildings, all constructed of timber, brick and concrete. Housing designs have been examined by the Association and are considered to be reasonable. Fifteen regional centers for training and applied research, each serving about five nearby state farms, would be created from existing facilities, by adding 6,000 sq m of floorspace for laboratories, classrooms, dormitories and small libraries. About 400 km of 10 kv transmission lines and 780 km of telephone lines would be added, primarily to serve farms engaged in new planting.  - 16 - Rural Roads 3.10 The state farms would construct or upgrade about 780 km of rural roads to provide access to state farms from public highways and link residential centers within state farms. About 50% of this road construction would take place in Tongza Prefecture, another 40% in Hainan Prefecture and the balance in Zhanjiang. Roads would be designed and constructed to the standards of the Guangdong Provincial General Bureau of State Farms. Roads within the state farms, accounting for about 90% of rural road construction in the project, would be of compacted earth and six meters in width. Roads linking state farms to public highways, constituting about 10% of rural road construction, would be seven meters wide. A central strip of 3.5 meters width would have a rock fill base of 15 cm and would be gravel surfaced. Road standards have been examined and are acceptable to the Association. Training and Research 3.11 The project would equip fifteen regional extension and training centers with audio-visual and basic laboratory equipment. These centers would provide training in topics related to rubber planting and production, wind- breaks, plant disease control, soil testing and fertilizer use. The project would provide a total of about 31,000 mandays of training by these centers, primarily for technicians and administrative personnel employed by project farms and the joint ventures. The project would also support the costs of on-farm training of the existing work force, including technical and administrative staff, and of newly employed workers. This training, largely in the form of on-the-job instruction by state farm technical and managerial staff, would include some 30,000 training opportunities, totalling about 500,000 training days, of which 50,000 would be provided to commune members participating in joint venture rubber production. The project would provide scientific equipment to upgrade the prefectural rubber research and testing institutes, which are responsible for applied research in rubber breeding, soil management and fertilization, intensive tapping systems, disease and weed control, chemical stimulation, cover cropping and rubber processing. Technical Assistance 3.12 Higher level training for prefectural, provincial and CGB staff would be expanded to include overseas short courses and study tours (in rubber tapping, stimulation and processing, soil and leaf analysis, production and management of fast growing tree species, estate management, audiovisual techniques in training and extension, processing of rubber wood and furniture manufacture) and degree programs (training to the BSc or MSc levels) in forestry (1), genetics and breeding (1), soil chemistry (1), and agricultural business and economics (1). Candidates for the degree program would be selected from managerial and technical staff of the PGB and the prefectural administrative bureaus.  - 17 - 3.13 China's natural rubber industry has been relatively isolated from research and development activities in other rubber producing countries for much of the past three decades. The staff development program would facili- tate the flow of technical information from abroad. In addition to the overseas training program, the project would provide 26 staff months of technical assistance through visits and consultations by experts who will examine specific technical and managerial areas of China's natural rubber industry. This assistance has been costed at US$10,000 per man-month, including travel, per diem and fees. Areas of emphasis would include assist- ance in treatment and utilization of rubber wood, production of wood-based panels, data management and computer training and in the formulation of an applied research program in windbreak establishment and management. The technical assistance component would include annual visits to China by an expert who would carry out a comprehensive review of rubber development generally, and project implementation specifically, to identify problems and discuss findings with project and government authorities. Assurances were obtained that all consultants engaged under the project would be employed in accordance with the Bank Group guidelines on consultants. Status of Design 3.14 All areas to be new planted or replanted have been surveyed and estimates of the earthwork required have been made. Alignments of rural and field roads have been selected and standard designs for farm, training and research buildings have been completed. Technical specifications for chemical fertilizers, farm and construction machinery and vehicles have been prepared. Contract documents would follow closely those used for the Heilongjiang Land Reclamation Project. Implementation Schedule 3.15 The project would be implemented over a five-year period beginning in late 1983 (Figure 3.1). Most project activities are under way as part of the PGB's on-going program of rubber new planting and replanting. Some felling and land preparation for windbreaks in parts of the project area vulnerable to wind damage were started in 1983. The established rubber nursery area in 1983 included about 1,200 ha and contained 21 million plants of appropriate materials for field planting in 1984 and 1985. Nurseries for the production of additional stumped buddings and seedlings for windbreaks are being expanded to meet full project requirements. The 125 ha in wind- break nurseries in 1983 are adequate to meet 1984 planting requirements. Similar areas of windbreak nursery would be established in each subsequent year. Land clearing and preparation of the project area, including felling of old rubber trees in replant areas, began in October 1983. Rubber planting would be completed in years 1984-87. The proposed phasing is 11,300, 13,400, 11,600, and 3,700 ha, respectively in those years, or about 85% of the total area to be replanted or new planted to rubber in the province during  - 18 - 1984-88. A detailed work plan for the first year has been prepared by the PGB; assurances were obtained from the Government that by September 30 of each year a detailed work program and financing plan for the following year would be prepared for Association review. Procurement actions for goods and equipment would begin in late 1983, with first deliveries expected in mid-1984. Cost Estimates 3.16 Total project cost in mid-1983 prices is US$301.0 million (including $2.7 million in taxes and duties), with a foreign exchange component of $79.5 million, or 26% of the total.. Unit costs for rubber replanting and new planting, windbreak establishment, roads and building construction are based on expenditures by the state farms for carrying out similar activities in the project prefectures. Prices for fertilizers and locally available equipment are based on data provided to the PGB by local suppliers. Prices for large agricultural machinery, related equipment and transport vehicles are based on list price obtained from manufacturers. Physical contingencies (6% of project base cost) equal 10% of base costs of construction (expenses on buildings, wood processing facilities, transport and communication) and 5% of base costs of rubber and windbreak establishment. Assumed price increases amount to 20% of the base cost plus physical contingencies and are based on annual price escalation factors for goods and services of domestic and foreign origin in US dollar equivalents of 7.5% in 1984, 7.0% in 1985, and 6% annually in 1986, 1987 and 1988. A summary of project costs is shown in Table 3.2 and details are given in Annex 2. Financing 3.17 Bank Group assistance would be US$100 million (34% of the total project cost net of duties), consisting of an IDA credit of SDR 37.8 million (US$40 million equivalent) and an IDA Special Fund credit of SDR 56.8 million (US$60 million equivalent). The balance of US$201.0 million equivalent would be met from a combination of local sources, viz.: (a) funds from the central government budget totalling US$84.2 mil- lion equivalent in 1984-88; (b) proceeds from the rubber development cess, totalling US$66.6 mil- lion equivalent, which is levied against participating state farms at the rate of US$182 (Y 346) per ton of rubber produced; and (c) contributions from profits retained by the state farms totalling US$50.2 million equivalent. Details of the financing plan are given in Annex 2, Table 4.  - 19 - Table 3.2: SUMMARY OF PROJECT COSTS For- For- Foreign Local eign Total Local eign Total exchange --- (Y million) --- -- (US$ million) - (%) Rubber establishment 196.8 37.5 234.3 103.6 19.7 123.3 16 Commune development 21.3 4.0 25.3 11.2 2.1 13.3 16 Windbreak establishment 22.9 2.6 25.5 12.1 1.3 13.4 10 Buildings and utilities 64.1 21.4 85.5 33.7 11.3 45.0 25 Rural roads 14.6 3.4 18.0 7.7 1.8 9.5 19 Equipment and vehicles 0.0 31.7 31.7 0.0 16.7 16.7 100 Sawmills and wood- processing 4.6 17.1 21.7 2.4 9.0 11.4 79 Training and research 4.0 2.1 6.1 2.1 1.1 3.2 34 Technical assistance 0.0 2.1 2.1 0.0 1.1 1.1 100 Base cost (mid-1983 prices) 328.3 121.9 450.2 172.8 64.1 236.9 27 Physical contingencies 20.4 6.4 26.8 10.7 3.4 14.1 24 Price contingencies 72.2 23.0 95.2 38.0 12.0 50.0 24 Total Project Cost/a 420.9 151.3 572.2 221.5 79.5 301.0 26 /a Includes duties of Y 5.0 million (US$2.7 million equivalent) on imported fertilizers. Total project cost net of duties is US$298.3 million. Procurement 3.18 The IDA credit would finance agricultural and construction machinery and vehicles (US$15.6 million), a timber complex (US$6.0 million), training and research equipment (US$1.2 million), plantation and windbreak establish- ment (US$15.5 million) and technical assistance (US$1.3 million). Inter- national competitive bidding (ICB) in accordance with Bank Group guidelines would be followed in the procurement of the first three categories above. The IDA Special Fund credit would finance additional agricultural and con- struction machinery and vehicles (US$3.3 million), a second timber complex (US$5.9 million), sawmills (US$1.6 million), chemical fertilizers (US$22.4 million) and plantation and windbreak establishment (US$23.5 million). Under the IDA Special Fund credit, procurement of all items in categories except the last would be through ICB limited to suppliers from Part II countries and those Part I countries eligible for Special Fund procurement. Domestic manufacturers would receive a preference of 15% or the applicable customs duty, whichever is less, in bid evaluation. Small items of equipment cost- ing not more than US$50,000 each and US$1 million in total for the Special  - 20 - Fund and US$3 million in total for IDA, would be procured on the basis of three price quotations under procedures acceptable to the Association. Establishment of rubber and windbreaks on state farms (US$143.9 million) and in the joint ventures (US$13.4 million), construction of buildings (US$54.2 million) and roads (US$11.3 million) would be carried out by force account employing the workers on participating state farms and communes. Because these works would be relatively small, dispersed geographically and carried out over the implementation period of the project, they would not be suitable for ICB. Further details of the procurement arrangements are provided in Annex 2, Table 7. Disbursements 3.19 Disbursements under both the IDA credit and the IDA Special Fund credit against expenditures for equipment, vehicles and goods would be at a rate of 100% of the foreign exchange costs of imported items or 100% of the ex-factory price of locally manufactured items. Disbursements for overseas training and expert services would be 100% of total costs. Disbursement for rubber and windbreak establishment on state farms, excluding chemical fertilizer, would be 27% of total expenditures (11% from the IDA credit and another 16% from the IDA Special Fund credit). Such disbursements would be against statements of expenditure supported by progress reports showing physical quantities and unit prices. Agreement has been reached at negotia- tions with the Government on unit prices, as indicated in Annex 5, Table 1. Unit prices would be updated annually by the PGB and reviewed by the Association. No disbursements would be made against expenditures for land development in the joint ventures, construction of rural roads, buildings and utilities, and cost of domestic training other than costs of equipment and vehicles. It is estimated that disbursement would be completed by about June 1989. In view of the PGB's extensive experience with this type of work, the projected disbursement schedule is shorter than the disbursement profile for Bank Group Operations in perennial crop development in the region. Estimated schedules of expenditures and disbursements are given in Annex 2, Tables 2 and 5. Accounts and Audits 3.20 Accounting systems are adequate to control project finance and disbursements. The PGB maintains a comprehensive system of financial administration and management in which accounting units at each administra- tive level (production brigade, farm, prefectural administrative bureaus, PGB, and central general bureau) submit aggregated financial accounts cover- ing capital construction and production expenses to the next highest level. These records are paralleled by similar accounts which are maintained within the banking system and which permit cross checking at each administrative level. Thorough audits are conducted by higher level units if discrepancies are detected. Financial accounts of prefectural administrative bureaus are aggregated by the PGB and sent to the MAAF, the Ministry of Finance, and provincial offices of disbursing banks. The PGB would maintain a separate project account in sufficient detail to record expenditures on the major  - 21 - expenditure categories indicated in Table 3.2. Assurances were obtained that the PGB would maintain such accounts and that they would be audited annually by independent auditors acceptable to the Association and submitted within six months of the close of each financial year. The audit review would examine and determine that funds disbursed against statements of expenditure (SOE) had been used for the purpose for which they were provided. It is expected that the central government's recently-established General Auditing Office would audit the PGB project accounts when that office becomes capable of taking on this responsibility. Meanwhile, it is expected that annual audits would be carried out by the Ministry of Finance. Environmental Effects 3.21 The project is expected to have a positive environmental impact. Improved techniques in windbreak establishment and utilization are expected to enhance timber and fuelwood supplies and thereby reduce depletion of existing forests. The planting of additional windbreaks will help to maintain ground cover and reduce surface runoff and the destructive effect of typhoons. It is not expected that any environmental problems would arise from land clearing and planting of rubber and windbreaks because of the sound conservation practices which have been developed over some 30 years of experience. These include proper terracing, retaining natural cover wherever possible, restricting rubber cultivation to lands with slopes of less than about 47% (250) and establishing leguminous cover crops or intercrops. Appropriate fertilizer programs would be established to maintain soil tilth and fertility. Other agricultural chemicals, largely herbicides, would be used carefully and in limited quantities. Assurances were obtained that provisions would be made by the PGB for proper treatment of effluents from plants processing rubber produced in the project. IV. ORGANIZATION AND MANAGEMENT Project Management and Coordination 4.01 The administrative structure of China's state farm system includes the Central General Bureau (CGB) within the MAAF, the provincial general bureaus, the prefectural administrative bureaus and the state farms (Figures 4.1, 4.2). A state farm in Guangdong Province is typically divided into an administrative headquarters, operational districts (or subfarms) and specialized production brigades (Figure 4.3). The CGB has responsibility for helping formulate and implement national policies regarding state farms and is a conduit for investment funds from the state budget. It is respon- sible for overall investment planning in the state farm system, for allocat- ing production and sales quotas to provincial state farm bureaus and ensur- ing the availability of materials needed to implement planned investments in the provinces. General bureaus at provincial level-are responsible to the  - 22 - CGB on matters of general policy and finance and to respective provincial governments on matters of personnel and local policies. Provincial bureaus provide policy guidance and administrative and technical support to the prefectural bureaus which, in turn, exercise control over individual state farms through well-developed lines of authority. 4.02 Guangdong Provincial General Bureau (PGB). The PGB is administra- tively a part of the Guangdong Provincial Government, but is also responsible to the CGB for carrying out the CGB-approved investment program and for returning to the center a portion of profits earned by the provincial state farm system. The PGB allocates to prefectural bureaus the capital and operating funds from the state budget and approves the investment program to be undertaken by prefectural bureaus. The PGB also sets production and sales quotas for prefectural bureaus which in turn allot these quotas to individual farms. The PGB comprises 13 divisions or offices which provide, through the prefectural bureaus, a wide range of technical and administrative services, including research, marketing, health, education and culture. 4.03 Prefectural Bureaus and State Farms. The four local administrative bureaus in Guangdong correspond geographically to Hainan, Tongza, Zhanjiang and Shantou Prefectures. There is very little rubber in Shantou Prefecture and this administrative bureau is not participating in the project. The prefectural bureaus are the links between the PGB and individual state farms and provide them with policy guidance and technical and administrative support. The organizational structure of the prefectural administrative bureaus includes specialized crop production support units, construction units and other service units including research, processing, marketing, health and education (Figure 4.2). State farms in the province are organized similarily. Support units at farm headquarters are available to service the specialized production brigades in each of the operations districts. The production brigade, consisting on average of about 150 workers, is the basic management and accounting unit on state farms. 4.04 The project would be implemented by the Guangdong Provincial General Bureau of State Farms and Land Reclamation, although the Planning Office of the CGB would exercise an oversight function and would have primary responsibility for liaison with the Association. Within the PGB a Project Management Committee, comprised of the heads of key divisions, has been established to provide policy guidance and coordinate project implementation (Figure 4.4). A Project Management Office, made up of administrative and technical staff of the PGB, would be responsible for day-to-day activities in executing the project. Similar groups would be established within the prefectural bureaus and on participating state farms and communes. An assurance was obtained that the Project Management Office would be maintained during project implementation and be adequately staffed with qualified personnel.  - 23 - Financial Management 4.05 Funds for capital investment and operating expenses of the state farm system in Guangdong are provided through a combination of grants from the Central Government, retained profits from the state farms, replanting and depreciation charges and bank loans. Grants from the Central Government are used for capital construction, circulating capital and special purposes such as small-scale water conservancy projects, social services and adminis- trative expenses of the prefectural bureaus and the PGB. Annual appropria- tions from the Central Government to the Guangdong state farm system averaged Y 168.5 million in 1980 and 1981. Bank loans include short-term funds and medium-term loans for equipment purchase and tree crop development. State farm profits retained within the province in 1981 totalled Y 111.7 million, an increase of 33% from 1980. Some of the funds for replanting originate in a cess which had been levied at the rate (1982) of Y 400 per ton of rubber produced, but which is to be reduced to Y 346 in 1983. Seventy percent of this charge is retained for subsequent use in the individual state farm's "modernization and renewal" account at the local branch of the Agricultural Bank; 20% is passed upward to the prefectural administrative bureau; and 10% goes to the PGB. 4.06 Upward financial flows from the state farm system in the province include, in addition to the rubber development cess, an agricultural tax assessed at the rate of 3.5% of the value of rubber production, a 7% tax (1982) on net profits of rubber producing farms, and an industrial- commercial (turnover) tax on activities other than rubber production. Rubber is exempted from this latter tax because it is a key producer good. All taxes are collected at the level of individual state farms and go directly to the local tax bureau, or the local government in the case of the 7% tax on rubber profits. The agricultural tax is included in provincial budgetary revenues and the industrial-commercial tax is shared between the province and the Central Government, Rubber production is exempt from the agricultural tax during the first three years of tapping. Of the 7% tax on net profits of rubber producing state farms, 4% is earmarked for commune development and 3% is retained for use by the county. 4.07 State farms in Guangdong operate within a well-developed financial system characterized by increasing profits (Y 221.7 million in 1981, up 14% from 1980) and a robust balance sheet indicating a small amount of short and long term debt. With the introduction of the economic responsi- bility system in 1979, financial management has become increasingly decentralized. Each administrative unit (state farm, prefectural bureau, PGB) is assigned a fixed target for profits to be remitted to the next higher level (or subsidy to be received from that level). If a state farm exceeds its profit remittance target, it is allowed to keep 70% of the surplus but must remit 30% to the prefectural bureau. Prefectural bureaus retain 20% of excess profits earned in their respective jurisdictions but remit 10% to the PGB. In Guangdong annual profit targets have been established through 1985.  - 24 - The PGB remitted Y 110 million of system profits to the Central Government annually in 1980 and 1981; this is to increase to Y 140 million in 1985. In 1981 financial flows from the Central Government to Guangdong's state farm system (about Y 165 million) exceeded the total remitted upward (Y 145 mil- lion) by the provincial state farm system to government budgets at all levels, (county, provincial, center). Details of the distribution of profits from the provincial state farm system in 1980 and 1981 are indicated in Table 4.1. Notable features include the increased retentions by production brigades and state farms, increased contributions to local governments and communes and the smaller relative distribution to the Central Government. In 1981 about 50% of total profits were remitted to the Central Government and 32% were retained by production brigades and state farms. 4.08 The domestic financing of the proposed project is not expected to place an inordinate burden on the state farm system. Funds from the re- planting cess have been running at about Y 80 million annually, or more than three times the expected average annual financial contributions to the project from this source. The Central Government grant in 1981 and 1982 to the provincial system averaged about Y 168 million, as compared with Y 32 million annually expected from this source in the project. Profits retained in the province in 1980 and 1981 averaged Y 98 million, compared with Y 19 million annually of project finance from this source. The 50 farms in the project account for 35% of all state farms in Guangdong and 55% of the state farm rubber area in the province. Assurances were obtained that the Government will provide the funds, services and other resources necessary to bring the project plantings into production. Execution of Project Works 4.09 All agricultural development and construction activities in the project would be carried out by specialized units within participating state farms and communes. Project buildings, roads, and communication facilities would be constructed by farm-level construction brigades using designs and standards developed by the technical and engineering divisions of the PGB. Construction activities would be supervised by PGB staff. Agricultural development, including land clearing and planting, nursery operations, intercropping, and establishment of windbreaks would be carried out by production brigades on participating farms and communes. Participating farms would enter into annual contracts with individual brigades on the farms to carry out specific development tasks within a given level of expenditure. From this budget the brigade would pay wages of its members and finance the costs of production inputs. Project inputs such as fertilizers, fuels and building materials would be supplied through a well-developed system of supply and marketing units at the state farms. These units are linked to government supply departments which maintain offices at the prefectural and provincial bureaus. In land clearing and planting, targets of brigades would relate to the amount of land cleared, the quality of work performed and the survival rate of the young rubber.  - 25 - Table 4.1: DISTRIBUTION OF PROFITS FROM THE STATE FARM SYSTEM, GUANGDONG PROVINCE, 1980 and 1981 (Y million) 1980 1981 % of % of Receiving Unit Amount total /a Amount total /a Rural Collectives & Local Governments 21.6 11 30.7 14 Expenditures to aid communes /b (10.9) (6) (19.5) (9) Special profits tax (estimate)/c (10.7) (6) (11.2) (5) State Farms & Production Brigades 40.3 21 71.6 32 State farms (25.9) (13) (44.9) (20) Production brigades (14.5) (8) (26.7) (12) Other Enterprises 3.5 2 5.3 2 Industrial enterprises (2.2) (1) (2.2) (1) Construction enterprises (0.1) (negl.) (0.1) (negl.) Supply and marketing enterprises (1.2) (negl.) (3.0) (1) Prefectural Administrative Bureaus 10.9 6 6.8 3 Provincial General Bureau 7.9 4 -2.8 /d -1 Central Government Budget 110.0 57 110.0 50 Total 194.1 100 221.7 100 /a Percentages may not add to the total because of rounding. /b This item is viewed as a cost in the accounts of the PGB. Hence, the PGB statements of profits are lower than the figures reported here by this amount. /c Estimated at 7% of the operating profits of state farms. /d In 1981 the profits received by the PGB were insufficient to meet its target for profit remittances to the Central Government. The PGB made up the shortfall with its accumulated reserves (retained profits from previous years).  - 26 - Maintenance tasks specified by the contract would relate to terrace upkeep, weeding, fertilization and mulching. Bonuses depend in part on satisfactory growth rate of the young rubber. Payment to tappers would be determined by the volume of latex produced and the quality of the tapping (absence of damage to the tree, etc.). 4.10 Intercropping during Years 1-3 of the rubber planting would be practiced where soils and topography are suitable. About 2,000 ha would be intercropped to pineapple and sugarcane using paid labor of the state farm workers and other production workers supplied by the state farms. Another 4,000 ha would be allocated to individual families for the production of groundnuts and sweet potatoes, or other crops which are agronomically suitable and profitable. State farms would assist production on these privately-operated plots by providing technical assistance and production inputs. State farms would fell, clear, destump, plough and harrow the land at no charge to the intercropper. Each participating family would receive up to one ha for the entire intercropping period and would agree to proper cultural practices, including weeding and care of the young rubber. The application of fertilizers to the rubber plants would be done separately by workers of the state farms. At the end of the intercropping period (generally Year 3), participants would plant cover crops using seed and fertilizer provided by the farms. State farms would recover the costs of materials provided and all net returns would accrue to participating families. Management of State Farm-Commune Joint Ventures 4.11 The 4,000 ha to be developed through joint-venture arrangements would be located primarily on commune lands adjacent to the proposed 16 parti- cipating state farms (5 farms in Hainan Prefecture with joint ventures total- ling 2,500 ha; 7 in Tongza on 800 ha; and 4 in Zhanjiang on 700 ha). Commune production brigades would supply their land and labor, while state farms would provide finance, production inputs (largely planting materials and chemical fertilizers) and technical guidance. A written contract would govern each joint venture and would remain valid for the economic life of the rubber planting. Individual joint ventures would be implemented under the direction of a Project Management Group comprised of managerial and technical staff of participating communes and state farms. Technical standards of production, specified in governing documents, would help ensure effective use of land and financial resources. Ownership of land under joint ventures would continue to reside with the commune production brigade, while rubber trees and windbreaks would be jointly owned by the two partners in the venture. 4.12 Net benefits would be allocated in accordance with each party's contribution to production. Financial resources would derive in large part from the retained profits of participating state farms, although  - 27 - supplementary funds would be provided by the provincial administrative bureaus and the PGB. Rubber production from joint ventures would be processed on state farm facilities and marketed through established state marketing facilities. Decisions regarding the processing and marketing of other production from joint ventures would be made jointly by participants. Accounting and management systems would be developed by technical personnel of participating state farms in accordance with established standards. Monitoring and Evaluation 4.13 The state farm system in Guangdong province has in place a well- developed system which provides timely and detailed technical and financial information to management at all administrative levels. Benchmark data on yields, production, employment and incomes are generally available on participating state farms. Data collection systems and benchmark information on participating communes are likely to be less comprehensive. Each state farm and commune would provide information on project implementation on a regular basis to the Project Management Office of the PGB, primarily to assist in the efficient implementation of the project. The Project Management Office in turn would prepare twice-yearly reports on the progress of the project and submit these reports to the Association within two months of the end of each project semester. At the end of the disbursement period, the Project Management Office, together with Central General Bureau of State Farms and Land Reclamation, would prepare an assessment of project results and compare these with appraisal estimates. Agricultural Support Services 4.14 Agricultural research. Basic research on rubber and most other tropical tree crops in China is carried out by the South China Academy of Tropical Crops (SCATC), which is attached to the Ministry of Agriculture, Animal Husbandry and Fisheries (MAAF) and jointly operated by MAAF and the Guangdong Provincial Government. The SCATC, located on Hainan Island, comprises six research institutes and four experimental farms covering 6,220 ha. The Academy employs some 400 technical personnel, of whom 27 are senior research staff. A major thrust of SCATC research has been the search for high-yielding, cold resistant rubber clones through selection and plant breeding, including tissue culture techniques. The SCATC would play a supportive role in the project by providing research results, planting materials and technical guidance to the prefectural research institutes. 4.15 Rubber research of an applied nature is largely the responsibility of the three research and one testing units attached to the prefectural administrative bureaus. The Rubber Research Institute, Wencheng County, Hainan Administrative Bureau, employs 33 senior research staff, and operates an experimental station of 1,820 ha (of which 616 are in rubber). A major research emphasis has been the development of wind resistant rubber clones. A branch station on a state farm is exploring new techniques of tea and  - 28 - rubber inter-cropping in typhoon-prone areas. The Tropical Crops Research Institute, Baoting County, Tongza Administrative Bureau, operates an area of 4,000 ha (of which 1,060 are in rubber) and is staffed by 30 senior research workers. Major research responsibilities include the development of disease resistant, high-yielding rubber clones and work on new tapping techniques. The Rubber Research Institute, Huazhou County, Zhanjiang Administrative Bureau, is developing high-yielding planting materials which are cold tolerant. The 20 senior research staff have available about 630 ha, of which 100 are experimental plots. A branch station on the Leizhou Peninsula is carrying out work on wind tolerant planting materials and windbreak improvement. The Rubber Specification Testing Center, Haikou City, Hainan Administrative Bureau, with 20 senior technical staff, consists of laboratories to test dry rubber, latex and rubber products and provide soil and plant tissue analyses to state farms. 4.16 Research at the SCATC is funded largely by grants from the state budget through the MAAF. Research at the prefectural research institutes is financed through the replanting cess and depreciation funds provided by the regional administrative bureaus (38% of the total in 1982), funds from the PGB (48%) and a small grant (accounting for 13% of total research expediture by the prefectural institute) from MAAF for high priority projects. In 1982 the costs of operating the prefectural-level research institutes was about Y 4 million, or less than one percent of the gross value of rubber produced in Guangdong during that year. The project would assist the prefectural-level research institutes by supplying laboratory and other research equipment and by providing consulting services on specialized research topics. 4.17 Training. Specialized training related to rubber production and processing is provided through training programs operated by the state farms, by prefectural administrative bureaus and by the PGB. The PGB and SCATC are responsible for training state farm directors and providing refresher courses for agronomists and engineers. The prefectural bureaus and their subordinate research institutes train other state farm staff and junior technical personnel. State farm extension stations, which have been established on 120 farms in the province, are responsible for training of farm workers. These extension stations, which employ about 500 technical personnel and manage 2,500 ha of experimental and demonstration plots, have the major responsibility to transfer the technology developed at the prefectural research institutes to state farms. Operating expenses of the stations have been financed from the replanting cess and depreciation charges levied against individual farms. 4.18 The project would upgrade 15 of the existing extension stations to regional extension centers in different agro-climatic areas. Administrative support for the centers would be provided by the farms on which the centers are located while technical support would be the responsibility of prefectural research institutes. These centers would train primarily the technical and skilled staff of state farms and joint ventures in specialized  - 29 - topics related to propagation of planting materials, weed control, soil and foliar sampling techniques and rubber processing. Training opportunities would be offered to technical staff of non-project farms and communes when facilities are available. The project would assist in the construction and equipping of these centers and support them indirectly by providing specialized consulting services and advanced-level training opportunities for technical staff. V. PRODUCTION, PROCESSING, MARKETING AND PRICES Rubber Nurseries 5.01 The choice of clones for the project area reflects the need to balance high yield attributes with resistance to disease, wind and low temperatures. About 38% of the project area would be planted to RRIM 600, 27% to PR 107, 9% to GT 1, 8% to Haiken 1 and 18% to other less well-known, but promising materials (Haiken 2, Hongxing 1, Tafeng 95, Nanhua 1 and others). The RRIM 600 plantings would be concentrated in Tongza Prefecture and southwestern Hainan Prefecture where winds and low temperatures are less problematic, while the more cold-resistant Haiken 1, GT 1 and PR 107 are better suited for Zhanjiang and northern Hainan Prefectures (Annex 3, Table 1). The newer clones would be widely dispersed through establishment of clone observation plots of two ha per clone, or multiple thereof, on each participating farm. This would permit an early assessment of their general suitability for future large-scale plantings. Nurseries are being developed to supply this mix of clones in a planting program in which about 80% of the area would be planted to bare root budded stumps and at least 20% to high stumped buddings. A small number of polybag buddings would be made available for use as replacements and in limited plantings in Tongza Prefecture. The record with stumped buddings is good - over 95% survival rate has been achieved - and state farms would be encouraged to increase the area planted to this type of material beyond 20% of the total. Incremental nursery requirements for the project total about 180 ha, with the largest increase needed for the stumped buddings. Rubber Planting and Maintenance 5.02 All land with a gradient of over 5% (30) would be contour lined and terraced and no rubber would be planted on land with slopes of over about 47% (250). Felling of old trees would be done largely with the use of chain saws and hand tools. Where root disease is not a problem, and intercropping is not practiced, tree stumps would be left in the ground and treated with appropriate chemicals to hasten the kill and natural decay. Root disease surveys would be carried out in all fields scheduled for replanting. Removal of stumps would be necessary in areas to be mechanically terraced, intercropped or devoted to nurseries and some types of windbreaks. Terraces would be cut by hand or, where topography permits,  - 30 - by tractor. Planting densities of rubber would range from 600-650 trees per ha. Tree spacings would be not less than 2.0 m between trees and not more than 8.0 m between rows. 5.03 The project would continue present practices of using both organic and chemical fertilizers, although trials would be carried out to determine whether alternative fertilizer programs are preferable. Organic manure consists of animal dung and vegetative matter of weeds (Eupatorium odoratum) and cultivated leguminous bushes (Tephrosia candida). Chemical fertilizer would include nitrogen, phosphorus and potassium and, in few areas, magnesium. Application rates would vary over the tree life and would be based on soil and leaf analysis. In general, applications would include about 16 kg of N and 70 kg per ha of P205 at planting and 160, 50, and 30 kg per ha of N, P205 and K20, respectively, in each of the years 2-7 of the rubber planting and 130, 50 and 40 kg of N, P205 and K20 annually during maturity. Some 6-15 tons/ha of organic fertilizer also would be applied annually. Intercropping and Cover Crops 5.04 In the rubber's immature phase, the planted area would be intercropped, planted to leguminous cover crops, or left under natural covers. Intercropping would be restricted to planted areas with slopes of 150 or less. The steepest lands would be kept under natural cover which would be enriched by planting of legume covers. It is expected that about 10,000 ha (25% of the area to be replanted or new planted) would be intercropped (gross intercropped area) and that the net intercropped area would be about 6,000 ha. Of the net intercropped area, the state farms are expected to plant 1,000 ha to pineapple (two crops in three years with a yield of 22.5 tons/ha/crop) and 1,000 ha to sugarcane (two crops in three years with a yield of 37.5 tons/ha/crop). The remaining 4,000 net intercropped ha would be planted to annual crops (groundnuts, sweet potatoes, etc.) under arrangements in which state farms allot the land for private use to workers and staff of the farms. Crops other than sweet potatoes and peanuts may prove superior and experimentation with other feasible crops would be encouraged by farm managers. In the project, assumed yields for sweet potatoes and groundnuts are 15 and 0.75 tons/ha/crop, respectively, with two crops each year. 5.05 Where intercropping is not practiced or natural cover retained, leguminous covers would be established. The preferred cover is a mixture of Pueraria phaseoloides, Calopogonium mucunoides and Centrosema pubescens. Experimental plantings also would be undertaken to determine the feasibility of commercial production and export of high value leguminous seeds (e.g. Pueraria phaseoleides and Calopogonium caeruleum) which are in great demand in other natural rubber producing countries as covers but which set seed poorly in those countries. No net financial returns have been assumed in the project analysis from production of cover crop seeds. Successful production of commercial seed from leguminous covers would enhance financial  - 31 - flows in the early years of the project. Cover crops would receive 30 kg/ha of N and 30 kg/ha of P205 at planting and another 30 kg of P205 in each of the years 2-4 of the planting. Windbreaks 5.06 In wind-prone areas, windbreaks generally would be established one to two years before field planting of rubber. Windbreak areas would be clean cleared and the land prepared by disc ploughing to 30-35 cms depth, followed by harrowing. Camellia oleosa would continue to be the dominant species for the lowest tier while Acacia confusa would be planted for middle-tier protection. Although Eucalyptus exserta is now widely planted to provide upper-tier protection, the project would support field trials to identify other promising varieties for this purpose. At present plant spacing is 1 m x 1 m (10,000 plants/ha) or 1 m x 1.5 m (6,700 plants/ha). Under the project, most of the windbreaks would be planted at a density of 1.5 x 2 m (3,300 trees/ha; 2 m between rows and 1.5 m between plants), although trials would be undertaken to assess windbreak performance (wind protection and timber production) at wider spacing such as 2.5 m x 2.5 m (1,600 plants/ha) which also offer reduced costs of establishment. Chemical and organic fertilizers would be applied during the first three years after planting. Under these cultural practices, commercial thinning can be carried out at ages 10 and 18 years and clear felling at 32 years. Total wood production (fuelwood, poles, logs) over the life of the planting is projected to be about 275 cu m/ha of windbreak. Tapping and Production 5.07 Tapping on rubber plantings would commence when 50% of the trees in the planting have a girth (measured 150 cm above ground level) of at least 50 cm. It is expected that this condition would be largely met in the eighth year from planting in Tongza and Hainan Prefectures and in the ninth year in Zhanjiang. Where large stumped buddings are planted in Hainan and Tongza, the trees are expected to reach tappable size in the seventh year (i.e., after six years of immaturity). Reflecting different soil and weather conditions and the mix of clones to be planted, rubber yields on state farms are expected to be highest in Tongza Prefecture (averaging 1,460 kg/ha over 28 years of tapping and peaking at 2,050 kg/ha in the thirteenth year of tapping) and lowest in Zhanjiang (an average of 1,120 kg/ha and a peak yield of 1,530 kg/ha). Yields on the state farm-commune joint ventures are projected to be about 6% lower than yields in Zhanjiang (Annex 3). Yields in the project area have been adjusted to reflect weather conditions (cool temperatures in Zhanjiang, possible wind damage in Hainan and Tongza). In general, half spiral, alternate daily tapping would be practiced, although for clones susceptible to brown bast (RRIM 600, Haiken 2, Hongxing 1, Tafeng 95, Nanhua 1), the half spiral, every third day system would be followed in the first three years of tapping. Chemical stimulants would be used on slow yield starters (PR 107, GT 1) during the first three years of tapping. Farms would be encouraged to employ upward tapping with stimulants in the old rubber to be felled for replanting and in the rubber replanted and  - 32 - newplanted in the project in the last five years of tapping. At peak production (about Year 2003), the yearly rubber output from the project is expected to be about 72,000 tons. Processing and Marketing 5.08 China produces a relatively high grade of rubber. In 1981 55% of production was block rubber, 25% was smoked sheet, 17% was latex concen- trate and the remaining 3% was in the form of crepe rubbers. About 75% of 1981 production graded No. 1 on the basis of grading standards similar to those used by the Malaysian rubber industry. State farms in the project prefectures have a total installed rubber processing capacity in 186 fac- tories of 892 tons per day, or (at 220 operational days/year) about 200,000 tons/year. This capacity, though approximately double 1982 prefectural output, is widely dispersed and generally in uneconomically-small factories (average daily capacity of about 4.8 tons). Under the project, 85% of the output is to be processed into standard block rubber, 9% as latex concentrate and 6% as crepes. Experience elsewhere suggests that the minimum size factory for efficient operations in block rubber production is about 20 tons/day in two shift operations. Additional block rubber processing capacity would be required in the project by the early to mid-1990s. In the interim, a study would be carried out by the PGB to determine the optimum size and phasing of factory construction, taking into account the location of participating farms and communes and the economies of scale in modern processing facilities. Project costs do not include investment in rubber processing capacity, but these costs are included in the financial and economic analysis. Assurances were obtained that, upon completion of the project, the Government would construct adequate processing facilities. 5.09 All rubber produced in China is purchased by the state through the National Federation of Supply and Marketing Cooperatives (NFSMC) which maintains offices on most state farms. Rubber production on communes is usually processed by nearby state farms and marketed by the farms' marketing units. Prices for natural rubber vary slightly according to quality, but the basic price of Y 6,090/ton, ex-Haikou or Zhanjiang, has been maintained for several years to stimulate production and is uniform throughout China's rubber-growing areas. 5.10 Future population growth in China (perhaps 1.2%/yr) and annual gains in per capita income (say 4%) would be expected to translate into growth of domestic consumption of rubber of about 4.0% p.a. This compares with the 4.3% annual growth registered in the 1970-72 to 1979-81 period. Official projections of domestic supply for NR and SR are not available, although an additional 160,000 tons/year of SR capacity reportedly is under construction. In the next decade expansion of SR capacity may be limited by tight domestic crude oil supplies, although this constraint could be eased in later years if large supplies of crude materialize from new sources. Four percent annual growth in total rubber consumption in China implies rubber usage by the year 2000 of nearly one million tons. Increased domestic supplies of NR can be expected from the 165,000 ha of immature  - 33 - rubber (1980) which will come into tapping in the next several years and production from the new planting and replanting planned for the next 10 years. A plausible supply scenario would include little additional SR production beyond the 160,000 tons noted above to about 300,000 tons in total. The NR demand in the Year 2000 of 700,000 tons (1 million tons total consumption less 300,000 tons domestic SR capacity) can be compared with 1981 NR production of 115,000 tons and annual incremental production from this project at peak output of 72,000 tons. It is expected that the domestic market will easily absorb project output. 5.11 Marketing of other project output should present no difficulties. Incremental supplies from the intercrops would be relatively small and would be limited to the early years of the project. Pineapple production of 22,500 tons annually in 1986 and 1987 would be processed by nearby canneries. Sugarcane production of 37,500 tons/year in 1985 and 1986 would be milled on nearby state farms which have ample spare processing capacity. Sugar output would be sold in the local market where consumption is growing at more than 4% annually. Groundnut and sweet potato production, averaging 2,550 tons and 69,000 tons per year, respectively, in 1984-86, would be consumed by producing families or sold through local markets. Growing shortages of timber products (particularly poles and sawlogs) in the project area imply ready markets for wood from old rubber trees and for the timber produced in the windbreaks. Most of the output from the timber complexes (furniture and panels) is expected to be marketed domestically to supply needs of the expanding housing and construction industries. Financial and Economic Prices 5.12 Farmgate financial prices generally are projected to remain constant in mid-1983 values, in view of the Government's commitment to maintain agricultural production incentives and avoid undue inflationary pressures. Economic prices for rubber and chemical fertilizers are based on Bank Group price forecasts, adjusted for location and quality differences and expressed in mid-1983 values. Conversion factors are used to derive economic prices for items which do not enter into international trade. Details on economic and financial prices and conversion factors are in Annex 4. Financial and economic costs of construction are estimated to be roughly equal. Although the financial cost of labor in construction is substantially higher than its economic costs, the financial costs of construction materials tend to be lower than their economic costs. 5.13 The economic cost of labor is considered to be in the range of Y 0.50/man-day. This relatively low economic value for labor in the project area reflects the existing large element of underutilized labor and the likelihood that the state farm labor force may continue to grow more rapidly than remunerative work opportunities. Authorities estimate that 95,000 workers on state farms in the three prefectures (15% of the work force) are surplus in the sense that they could be withdrawn with little or no impact on production; the comparable figure for the project farms is 35,000 workers, or 12% of their work force. Underemployment on neighboring  - 34 - communes also is substantial. Possibilities for migration of excess labor to other areas are severely limited by official policy and lack of job opportunities elsewhere. About 20,000 middle school graduates enter the labor force of the Guangdong state farms each year. At this rate, the increase in the state farm labor force during project implementation (1984-88) would be more than twice the number of net full-time jobs expected to be created by the project in rubber production and processing at full development (para. 6.02). The financial wage of Y 1.50/man-day paid in the project area is considered to substantially exceed the value of labor's marginal product and contains welfare elements. The assumed economic cost of labor of Y 0.50 may be compared with the estimated cost of about Y 0.40 per day of additional consumption (largely food) associated with physical labor. 5.14 The current financial price for rubber is more than twice the economic (world) price projected for 1983. This reflects efforts to stimu- late rubber production and the need for state farms to finance social over- head costs and physical infrastructure. In view of expected increased domestic demand for rubber and, given China's potential for expanded pro- duction at satisfactory economic rates of return, a favorable economic encironment for rubber production should be maintained. Financial prices are assumed to remain greater than economic prices, although this differ- ential is expected to decline as world prices for rubber are projected to rise in real terms. An understanding was reached that the Government would examine over the next two years its production and management systems in natural rubber, with a view to further increasing efficiency. This exam- ination would cover major cost categories such as labor, material inputs, and infrastructural costs, the possibilities for cost reductions, yield pros- pects and their effects on production costs, and recommendations for cost reductions. Continued productivity gains should permit some reduction in producer prices, although this would be best done over time as part of a general price reform. A reduction in the financial price in the short-run would reduce or eliminate the flow of profits which rubber farms now remit to the central government. Over time lower prices probably would encourage the use of less labor in rubber, although employment for displaced labor would be difficult to find. For fuelwood, poles and sawlogs, financial prices are only about half as high as economic prices. Among the intercrops, financial prices in 1983 are equal to or greater than economic prices. VI. BENEFITS, JUSTIFICATION AND RISKS Production Benefits 6.01 New planting and replanting of 40,000 ha of rubber would add an annual average of 54,000 tons of rubber production, equivalent to at least 5% of China's total rubber consumption in the year 2000. Windbreaks on 12,000 ha would provide 3.3 million cu m of timber, of which 1.3 million cu m  - 35 - would be sawlogs, over a 32-year period. An additional 0.45 million cu m of sawlogs would result from the felling of old rubber trees prior to replant- ing. Improved wood processing technology is expected to increase the value added from old rubber trees and windbreaks. Intercropping during three years of the rubber's immature phase would yield 45,000 tons of pineapple, 75,000 tons of sugarcane, 7,650 tons of groundnuts and 207,000 tons of sweet potatoes for local markets. Employment and Income 6.02 The traditional concept of farm income derived from an individually operated plot is not applicable in the project area where land is owned by the state and managed by state farms which employ wage labor. Pre-project per capita incomes of families on state farms in Guangdong Province averaged (1982) about Y 370 (US$195) p.a. State farm workers receive an annual basic wage of about Y 550, various subsidies and bonuses averaging Y 215 p.a., and other benefits such as housing, medical care and retirement averaging Y 112. Income per worker thus totals about Y 879 p.a., or Y 3.25/man-day, assuming 270 workdays/year. Total income (cash and in kind) per worker on state farms is comparable to other state enterprises in China, but appears to be substantially greater than incomes of commune workers in the project area. Because incomes of state farm workers are already relatively high, despite a continuing labor surplus, further increases in real wages on state farms are not foreseen. The income benefits of the project, therefore, are likely to take the form of some income gains for commune members in the joint ventures and greater employment opportunities for the unemployed or underemployed. This will be important in both the state farm and the collective sector. 6.03 The state farm-commune joint ventures are expected to provide supplementary employment opportunities for commune members equivalent to 3,400 full-time jobs per year during project implementation at a wage of Y 1.5 per man-day. At full development, the tapping and processing of rubber from 40,000 ha would require an additional 12.1 million man-days annually, excluding managerial and support labor. Assuming 270 work-days/ year/worker, this would create about 44,800 job equivalents, of which 10% would be among commune members engaged in joint venture rubber production. The incremental work requirements on state farms are likely to be met by new entrants to the labor force, transfers from other farms and, possibly, additional labor time from those already working, but underemployed. The joint venture jobs would go to underemployed commune workers. In view of the considerable degree of underemployment on both state farms and communes, the number of net new full-time jobs created in rubber production and processing at full development may be less than the 44,800 job equivalents estimated for the project. However, additional employment also would be created in intercropping, in the forestry and wood processing activities and in managerial and technical support services.  - 36 - 6.04 The major objective of the project is to increase the supply of natural rubber and its effect on income distribution would be limited. The most important distributional element of the project is likely to be its support of the recent state farm initiative to assist rubber development on nearby communes. In addition to the production, income and employment benefits to communes from this component, the technology provided to the communes in the form of improved planting materials and cultural practices would be helpful in rubber development elsewhere in that sector. Average rubber yields in the joint ventures are expected to be at least five times current yields on communes in Guangdong Province. A second distributional element of the project would be the expected increase in financial assistance regularly provided from profits of state farms to assist commune agriculture. Financial Analysis 6.05 The financial rates of return (FRRs) for rubber and timber production in Hainan, Tongza, Zhanjiang Prefectures and the joint ventures are in the range of 13 to 16%. The effects on the FRR of lower rubber yields assumed in Zhanjiang are approximately offset by early revenues from the relatively large area under intercrops made possible by more level terrain in that prefecture. Relatively low financial costs of labor in the joint ventures (Y 1.50 per day as compared with Y 3.25 on state farms) help to bolster their FRRs. The FRR for a timber complex which produces furniture and particle board is about 22%. 6.06 The cash flow analysis (Annex 4, Table 5 ) suggests that income from felling of old windbreaks and rubber, from the timber complexes, and from intercrops would make a sizeable contribution to revenue flows during implementation (1984-88) of the project. The projected cash deficit during these early years would be financed from retained profits of state farms and funds from the rubber development cess. The cumulative cash deficit of about Y 145 million (US$76.4 million) in 1989-92 would require continued funding from state farm earnings and the development cess. Beginning in 1993, project income from rubber production and wood processing would regularly exceed costs. Cost and Rent Recovery 6.07 Recovery by the Central Government of capital grants and proceeds from the IDA and Special Fund credits would be made through a combination of agricultural taxes, profit remittances and the replanting cess. At full development farms would pay (in constant 1983 values) Y 19.3 million annually through the development cess, debt service obligations to the Central Government of Y 4.0 million and taxes on rubber production at the rate of 3.5% of gross value. In addition, participating farms and joint ventures would  - 37 - remit about 50% of profits to the Central Government (based on current rates) and 7% to local governments and adjoining communes. In present value terms, with all annual values discounted at 12% over the 35-year project life, these charges would be equivalent to 78% of project rent (Annex 4, Table 6). On this basis the cost recovery index for contribution from the Central Government (Y 160 million) and the IDA and Special Fund credits (Y 190 million) would be 166%. These rent and cost recovery indexes are relatively high in large part because high financial prices for rubber permit sizeable profits, a substantial portion of which is remitted to the Central Government. These indexes also could be calculated to reflect the additional cost which the Central Government pays to obtain project output locally rather than from foreign sources. This amount, measured on the basis of differences between foreign and domestic prices for project output, could be viewed as a negative project charge (i.e., project revenue foregone by the Central Government). On this basis, the rent and cost recovery indexes (in present value terms) would fall to about 30% and 70%, respectively. These figures are considered reasonable. Economic Analysis 6.08 All economic values were converted to local currency at the official exchange rate at appraisal of Y 1.9 = US$1.00. Project costs of US$36.2 million equivalent for construction of housing, schools and health facilities were excluded from the economic analysis on grounds that these are normally provided to meet social objectives and that resulting benefits are likely to at least equal costs. The economic analysis also excludes estimated project costs of $1.6 million (and possible benefits) for the farm sawmills. Benefits from the upgrading of these sawmills are expected to substantially exceed costs because of likely productivity gains in sawmill operations and the strong local demand for sawn timber and simple furniture. In the economic analysis the total investment cost of US$156 million includes physical contingencies, but excludes price contingencies. 6.09 Using the foregoing assumptions and discounting costs and benefits over 35 years, the project's economic rate of return (ERR) is 17% (Table 6.1). Estimated ERRs for project activities in the three prefectures and the joint ventures are at least 14%. The ERR for the timber complexes is about 37%. At a discount rate of 12%, which is considered to approximate the opportunity cost of capital in China, the net present value (NPV) of the project is Y 201 million (US$106 million equivalent). The economic return of 17% is relatively high for long-gestation tree crops such as rubber and reflects the contribution from intercropping and timber procesing in the early years of the project.  - 38 - Table 6.1: SUMMARY OF RATE OF RETURN ANALYSIS Economic net Economic rate present value rate of return @ 12% Subarea or variation (%) (Y million) Base Case Subarea Hainan prefecture 17 /a 57 Tongza prefecture 17 /a 74 Zhanjiang prefecture 14 /a 10 State farm-commune joint ventures 14 Ia 7 Timber complexes 37 /a 53 Total Project 17 /a 201 Variations from Base Case Total project A 20% increase in project costs 14 100 A 20% decrease in rubber yields 15 116 a 20% decrease in rubber prices 15 98 Tapping advanced to eighth year through- out project area /b 17 203 Tapping delayed to tenth year throughout project area 17 182 Economic labor cost of Y 1 per man-day /c 16 164 Economic labor cost of Y 3 per man-day 12 21 /a FRRs are: Hainan 15%; Tongza 16%; Zhanjiang 13%; joint ventures 14%; timber complexes 22%. /b Base case assumes initial tapping in the eighth year in Tongza and Hainan and in the ninth year in Zhanjiang and in all joint ventures. /c Base case assumes economic labor cost of Y 0.5/man-day. 6.10 Sensitivity of the ERR was tested to cost overruns, implementation delays, shortfalls from projected yields or prices and variations in the economic value of farm labor (Table 6.1 and Annex 4, Table 9). Switching values (value of the variable tested at which the NPV of the project, with benefits and costs discounted at 12%, is reduced to zero) and NPV elastici- ties (the percentage change in NPVs due to a one percent change in the key variable) were calculated. No single variant proved crucial to the project's economic viability. An assumed economic wage of Y 3/man-day, or about twice the reported market wage for casual labor, would reduce the ERR to 12%. Economic prices of both rubber and timber at full development  - 39 - would have to fall simultaneously by about 39% to reduce the ERR to the opportunity cost of capital. The ERR is moderately sensitive to cost increases or rubber yield shortfalls. Switching values are, respectively, 39% above or 47% below appraisal estimates. Project Risks 6.11 The project faces no significant technical and organizational risks. The major clones to be used in rubber planting have been proven through local cultivation; the newer clones have shown promise in smaller plantings in China and are limited to less than 20% of the area. Basic cultural practices in the windbreaks have been used for a number of years in the project area. Improved practices in rubber and windbreaks have been proven in other countries and will be introduced through project-supported research and training activities. The implementing agency (PGB) has considerable experience in rubber cultivation and the annual new planting and replanting program foreseen in this project is within the actual replanting and new planting achieved by the PGB in 1977-81. The PGB's industrial experience is less extensive, although it manages a number of industrial operations, including a hardboard factory near Zhanjiang. A possible uncertainty relates to full implementation of the state farm- communes joint ventures where experience is limited. It is expected that this uncertainty would be kept within acceptable bounds through control and close supervision of the joint ventures by state farm management. VII. AGREEMENTS REACHED AND RECOMMENDATION 7.01 Agreements were reached with the Government on the following: (a) consultants for the project's technical assistance component would be employed in accordance with the Bank Group guidelines on consultants (para. 3.13); (b) by September 30 of each year a detailed work program and financing plan for the following calendar year would be prepared for review by the Association (para. 3.15); (c) the Guangdong Provincial Bureau of State Farms and Land Reclamation would maintain a separate account for the project. This account would be audited annually by independent auditors acceptable to the Association. The audited account would be submitted to the Association within six months of the close of each financial year (para. 3.20); (d) the Government would ensure that effluents from the plants which process the rubber produced in the project would be treated in accordance with sound environmental practices (para. 3.21);  - 40 - (e) the Project Management Office of the PGB would be maintained during project implementation and adequately staffed with experienced and qualified personnel (para. 4.04); and (f) the funds, services and other resources necessary to bring the project plantings into production, including the installation of processing facilities would be provided (para. 4.08). 7.02 With the above assurances, the project would be suitable for an IDA credit of SDR 37.8 million (equivalent to US$40 million) and an IDA Special Fund credit of SDR 56.8 million (equivalent to US$60 million) on standard IDA terms.  ANNEX 1 Table 1 - 41 - CHIINA RUBBER DEVELOPMENT PROJECT State Farms -- Selected Statistics Item Unit 1975 1981 Percent change, 1981 over 1975 Number - 2,320 2,094 -10 Total land area thousand ha 29,907 28,647 -4 Cropped area (" 3,967 4,409 +11 Production of: Grain thousand ton 5,853 6,100 +4 Natural rubber (") 65.7 115.2 +75 Cotton (") 62.2 111.0 +78 Meat ( ) 145.9 261.5 +79 Milk (") 253.6 351.4 +39 Wool C") 14.1 14.8 +5 Aquatic products (") 19.8 31.3 +58 Source: Ministry of Agriculture, Animal Husbandry and Fisheries, Beijing.  ANNEX 1 -42 - Table 2 CHINA RUBBER DEVELOPMENT PROJECT Areas Planted to Natural Rubber, 1980 (000 ha) Mature Total Province Seedling Budded Immature Area % Guangdong Province State farms Hainan 29.6 21.2 45.8 96.6 Tongza 12.4 24.3 46.5 83.2 Zhanjiang 19.7 21.2 28.1 69.0 Shantou 0.2 0.9 2.9 4.0 Overseas Chinese farms 1.5 1.5 1.5 4.5 Communes 66.7 - - 66.7 SCATC - 0.4 0.4 0.8 Subtotal 130.1 69.5 125.2 324.8 81 Yunnan Province State farms 3.4 17.0 27.7 48.1 Communes - 1.3 2.7 4.0 Subtotal 3.4 18.3 30.4 52.1 13 Guangxi Province State farms 0.7 5.9 6.7 13.3 Communes 4.0 - - 4.0 Subtotal 4.7 5.9 6.7 17.3 4 Fujian Province State farms 1.1 - 3.0 4.1 Communes 4.0 - - 4.0 Subtotal 5.1 - 3.0 8.1 2 Total 402.3 100 All Provinces State farms 67.1 90.5 160.7 318.3 79 Communes 74.7 1.3 2.7 78.7 20 Overseas Chinese 1.5 1.5 1.5 4.5 1 SCATC - 0.4 0.4 0.8 negl. Total 143.3 93.7 165.3 402.3 100 Source: MAAF.  ANNEX 1 -43- Table 3 CHINA RUBBER DEVELOPMENT PROJECT Age Profile of Rubber Trees on State Farms, Guangdong Province, 1981 Hectares Percent of total Years of age (000) planted area 29 or older 36.9 14.0 24-28 16.4 6.2 19-23 38.2 14.4 14-18 31.8 12.0 9-13 67.6 25.6 4-8 32.7 12.4 3 or younger 40.7 15.4 Total 264.3/a 100.0 /a Commune lands under rubber in Guangdong Province total another 15-20% of this figure. Rubber on these lands is of indetermi- nate age. Source: Provincial Bureau of State Farms and Land Reclamation.  - 44 - ANNEX 1 Table 4 CHINA RUBBER DEVELOPMENT PROJECT Domestic Production, Imports and Apparent Consumption of Rubber, 1965-81 Domestic produc- Domestic tion of natural production Imports of rubber /a of natural and Apparent Guangdong synthetic synthetic total Year Total Province rubber rubber consumption