Sri Lanka - Smallholder Rubber Rehabilitation Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No. 2827-CE SRI LANKA STAFF APPRAISAL REPORT SMALLHOLDER RUBBER REHABILITATION PROJECT April 15, 1980 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 EQUIVALENTS USSi = Rs 15.5 Rs 1 US$0.065 Rs 1 million = US$64,516 WEIGHTS AND K,ASURES 1 metric ton (mt) = 0.934 long ton = 2,204 lb 1 kilogram (kg) = 2.2 lb 1 hectare (ha) = 2.47 ac 1 kg/ha 0 O.89 lb/ac 1 meter (m) = 3.3 ft. 1 kilometer (km) 0.62 mi 1 sq. km = 0.3:3 sq. mi FISCAL YEAR January 1 - December 31 ABBREVIATIONS AND ACRONYMS ARTI - Agrarian Research and Training Institute ASD - Advisory Services Department CIDA - CanEadian Tnternational Development Agency CFC - Ceylon Fertilizer Corporation CPD - Commodity Purchase Department CRI - Cost Recovery Index drc - dry rubber content DRC - Department of Rubber Control DREO - Divisional Rubber Extelision Officer EO - Extension Overseer ERR - Economic Rate of Return GCE A - General Certificate of Education A Level GCE 0 - General Certificate of Education 0 Level GDP - Gross Domestic Product GMO - Group Marketing Organization GOSL - Government of Sri Lanka GPC - Group Processing Center ICB - International Competitive Bidding IDA - International Development Association IRR - Internal Rate of Return JEDB - Janatha Estates Development Board M - Million mt - metric ton MPI - Ministry of Plantation Industries NII - Net Income Increase NPV - Net Present Value ODM - Overseas Development Ministry PAD - Processing Advisory Division PA - Processing Advisor PCU - Project Coordination Unit PY - Project Year RAO - Regional Advisory Officer  FOR OFFICIAL USE ONLY REO - Rubber Extension Officer RIDA - Rubber Industry Development Authority RPPU - Rubber Policy and Planning Unit RPRC - Rubber Policy Review Committee RRISL - Rubber Research Institute of Sri Lanka RRS - Rubber Replanting Scheme RSS - Ribbed Smoked Sheet Rs - Sri Lanka Rupees SD - Supplies Division SPA - Senior Processing Advisor SPC - State Plantation Corporation SRMC - State Rubber Manufacturing Corporation TSR - Technically Specified Rubber This document has a restricted distribution and may be used by recipients only in the performance their official duties. Its contents may not otherwise be disclosed without World Bank authorization.   SRI LANKA Smallholder Rubber Rehabilitation Project Staff Appraisal Report Table of Contents Page No. I. SECTOR BACKGROUND ...................................... 1 General ................................................ 1 The Role of Agriculture in the Economy ................. 1 The Tree Crop Sector ................................... 2 The Rubber Sub-Sector .................................. 3 IDA's Role in the Tree Crop Sector ..................... 6 II. THE PROJECT AREA ....................................... 7 Physical Characteristics ............................... 7 Socio-economic Characteristics ......................... 8 Social Infrastructure .................................. 9 Supporting Services and Institutions ................... 9 The Replanting Scheme ............................. 9 Research .......................................... 9 Extension ......................................... 9 Department of Rubber Control ...................... 10 Rubber Processing ................................. 11 Rubber Marketing .................................. 12 Government Rubber Estates ......................... 13 III. THE PROJECT ............................................ 13 General Description ............... .. .................. 13 Detailed Features and Project Activities ...... .. ....... 14 A. Replanting Scheme ................................. 14 B. Institutional Support ............................. 19 C. Processing Component .............................. 20 D. Research Component ................................ 22 E. Project Evaluation ................................ 23 F. Technical Assistance .............................. 24 This report is based on the findings of an Appraisal Mission which visited Sri Lanka in October 1979 and included G. Hayes, I. Girardot-Berg (IDA) and C. Brookson and S. Swaminathan (Consultants).  -2- Table of Contents (cont'd) Page No. IV. COST ESTIMATES AND FINANCING ........................... 24 A. Cost Estimates .................................... 24 B. Proposed Financing ........................ 26 C. Procurement ....................................... 28 D. Disbursements and Audits ........................... 28 V. ORGANIZATION AND MANAGEMENT ............................ 29 A. Background ........................................ 29 B. Project Organization and Management .... ........... 31 C. Monitoring and Evaluation ......................... 37 VI. PRODUCTION ASPECTS . ..................................... 38 A. Present Situation ............... .. ................ 38 B. Project Replantings: Inputs and Yields .... ....... 39 C. Intercropping ................. .. .................. 40 VII. DEMAND, PRICES, FINANCIAL ANALYSIS AND COST RECOVERY 41 Demand, Prices and Marketing .41 Financial Returns to Smallholders .43 Financial Return to Government and Cost Recovery 44 VIII. BENEFITS AND JUSTIFICATION .45 Project Benefits .45 Project Beneficiaries .46 Employment .47 Environmental Effects .47 Economic Analysis and Sensitivity Tests .47 Project Risks .49 IX. RECOMMENDATIONS ........................................ 50 ANNEXES 1. Technical Aspects and Project Monitoring 2. Terms of Reference for Technical Assistance 3. Local Procedures for Project Funding 4. Supporting Tables and Charts 5. Selected Documents and Data Available in Project File  -3- LIST OF TABLES AND CHARTS ANNEX 1 Table 1 - Per Hectare Replanting Costs for Rubber Table 2 - Yield Profiles and Tapping Systems Table 3 - Intercropped Bananas: Inputs and Yields Table 4 - Estates Producing Planting Material for 1981 ANNEX 4 Table 1 - Project Participants by Size of Holding Table 2 - ASD Incremental Staffing Costs Table 3 - Training Component Table 4 - Investments and Other Operating Costs (ASD) Table 5 - DRC Regional Offices. Capital Expenditures, Staffing and Other Operating Costs Table 6 - Costs of Project Coordination Unit Table 7 - Costs of Processing, Research, Project Evaluation, and Technical Assistance Components Table 8 - Group Processing Centers (GPC): 8.1 GPC Direct Processing Cost 8.2 Financial Returns to GPC Members from a Successful GPC 8.3 Financial Returns from GPC Processing 8.4 Economic Returns from One GPC Table 9 - Summary Project Costs Table 10 - Yearly Replanting Costs and Proposed Payments Table 11 - Cash Flow of the Rubber Replanting Fund Table 12 - Disbursement Schedule Table 13 - Detailed Disbursement Categories Table 14 - Financial and Economic Rubber Prices Table 15 - Share of FOB Returns from Rubber Exports Table 16 - Financial Returns to Smallholders Table 17 - Replanting Payments per hectare Table 18 - Cash Flow Analysis for Three Types of Farm Table 19 - Government Project Cash Flow Table 20 - Income Levels of Project Participants and Project Impact on Poverty Income Group Table 21 - Economic Benefit and Cost Streams Table 22 - Conversion Factors CHARTS 1. Implementation Schedule 2. Proposed Project Organization 3. Proposed Organization of Advisory Services Department MAP 14733R   SRI LANKA GLOSSARY BUDDED STUMP A green or brown budding which has been pulled from the nursery and which has had the seedling stem trimmed off above the budgraft, the lateral roots trimmed to about seven cm and the tap root trimmed to about 40 cm. Such budded stumps can be packed and transported for planting. BUDDING The operation of grafting a bud onto a stock. Brown and green buds can be grafted onto brown stocks of up to four- teen (14) months; only green buds can be grafted onto green stocks. BUDWOOD The result of growing a bud for multiplication for planting. Budwood can be brown (15 to 18 buds per meter) or green (2 or 3 buds per stick). CLONAL SEED Open pollinated seed collected from monoclone or polyclone plantings. CLONE All plants derived vegetatively from an individual source plant thus having the same genetic composition. CREPE Processed rubber prepared by a number of passes, while washing, through patterned rollers revolving at unequal speeds, which macerate the rubber into laces. Latex crepe is prepared from freshly coagulated latex and is a premium product. It can be laminated into sole crepe. Scrap rubber, off-grades, and clippings can be cleaned in the above manner and made into brown crepe. All crepes are air-dried, either at ambient temperature or arti- ficially. Crepes are visually graded. CRUMB RUBBER A processed rubber prepared by hammer-milling or by extru- ding and chopping, drying, and pressing into 32 kg blocks. The highest grades are prepared from fresh latex, but auto-coagulated rubber, scrap, off-grades, and earth scrap can be cleaned and improved by this process. Prepared from latex, certain of the physical characteristics (such as viscosity) can be altered. Crumb rubber is graded on the basis of its technical characteristics rather than its visual appearance. CULTIVAR Cultivated varieties, either clonal seedlings or clones. CUPLUMP Late drippings of latex that autocoagulate in the cup after collection of the main crop has been made. Collected daily in tapping.  DRY RUBBER The amount of dry rubber that latex is estimated to con- CONTENT tain. A measurement required for payment of tappers (drc) and owners. Usually estimated by a "Metrolac". In some other countries, a "chee" is used. EARTH AND Overflow from the tapping cut onto the bark and the ground. BARK SCRAP Usually contaminated by dust and dirt. METROLAC A hydrometer used to estimate the drc. It is accurately calibrated to only one reading and should the drc (for various reasons) be above or below this, it will exaggerate the difference. NURSERY An area set aside for preparing planting material. Nurseries are used to multiply buds (budwood nursery) or to grow seed- lings for stocks for budding (seedling nursery). RIBBED A processed rubber prepared from latex by dilution, coag- SMOKED ulation with formic acid, milling with smooth and marking SHEET (RSS) rollers, washing, and drying in a smokehouse. The rollers revolve at the same speed; the marking roller impresses ribs to increase the surface area to aid drying. The sheets are dried in a smokehouse. RSS is visually graded. STIMULANT An ethylene-releasing substance which increases yield when applied to the tapping panel. SUCKER Adventitious shoots or clumps of ILeaves used for vegetative reproduction of bananas and pineapples. TECHNICALLY Rubbers, mainly crumb, from which adequate samples are taken SPECIFIED by a competent authority for testing and quality control for RUBBERS specification before shipping. The main parameter is dirt content which, in parts per million identifies the rubber (5, 10, 30, 50). Other parameters are the plasticity reten- tion index (a measure of possible oxidation), viscosity, nitrogen content, volatile matter, ash and magnesium con- tent. Other rubbers (crepe, RSS), if properly prepared and blocked, can be submitted for testing and qualify as a TSR. TREE SCRAP T-he band of latex that coagulates on the tapping cut after dripping occurs. It displays sometwhat different physical properties, when compounded, from the other scrap rubbers.  I. SECTOR BACKGROUND General 1.01 Sri Lanka, with a population of 14.5 M (1978) and a total land area of 65,607 sq. km. has a population density of about 220 per sq. km. Approxi- mately three-quarters of the population live in the wet zone in the southwest quadrant, comprising 23% of the country. The population density reaches 700 per sq. km. in that region, contrasting with only 80 per sq. km. in the dry zone (64% of the country). While output and employment have grown slowly over the past 20 years, the relatively rapid population growth of almost 2% has resulted in slow per capita GDP growth (1% annually), and increasing unemployment. The present per capita GDP of Sri Lanka is estimated at about US$200 and about 22% of the population is estimated to have annual incomes below the absolute poverty level of US$76 per capita. 1/ 1.02 Sri Lanka, nevertheless, compares favorably with neighboring countries in provision of social services for education, health, other social welfare, and rural communications. Consequently, poverty in Sri Lanka does not take the extreme form found in some other South Asian countries. Life expectancy and the literacy rate are high, and nutrition level is generally adequate. The population growth rate is declining aided by emigration; it is currently 1.7% per annum. The Role of Agriculture in the Economy and Government's Development Objectives 1.03 Agriculture plays a major role in the economy of Sri Lanka, account- ing for 39% of total GDP, over 50% of total employment, and almost 80% of export earnings. The processing of agricultural output and supply of inputs provide further substantial indirect contributions to the economy. In rural areas, dependence on agriculture is almost total with over 95% of the rural population either directly or indirectly involved in agriculture. 1.04 Sri Lanka's wide variations in precipitation, topography, and soils permit cultivation of a wide range of crops. Out of some 2.24 M ha under permanent cultivation (34% of total land area), paddy accounts for 0.65 M ha, coconuts 0.50 M ha, tea 0.24 M ha, rubber 0.19 M ha, and other perennial crops 0.04 M ha. The remaining 0.62 M ha comprises mixed rainfed farming, mostly as small gardens around homesteads. An additional 1 M ha are under shifting cultivation. 1.05 Tea and rubber are the major export commodities and jointly account for 60% of total export earnings and 40% of government revenues raised through export taxes and duties. Coconut, after having satisfied the substantial local demand, is also exported. Paddy and most other food crops are grown solely for domestic consumption. Domestic production of rice, sugar, and subsidiary food crops fall short of demand, and are supplemented by imports. 1/ World Bank estimate, 1979.  - 2 - Foreign exchange earnings from tea and rubber exports assist in financing these food imports. 1/ 1.06 The objectives of the Government of Sri. Lanka (GOSL) with respect to agricultural development are based on the recognition of the pivotal role of agriculture in stimulating the country's economic growth. The major objec- tives contained in GOSL's medium-term plans for agriculture are to increase agricultural employment, income, and self-sufficiency in basic food items, and to expand agricultural exports. 1.07 The GOSL strategy for achieving these objectives emphasizes better use of the existing potential through institutional and policy improvements as well as rehabilitation of the production base where necessary. GOSL has given high priority to accelerated programs for the rehabilitation of the tree crops amongst others. The proposed project is consistent with the development objectives and strategy of GOSL. The Tree Crop Sector 1.08 The tree crop sector is second only in importance to the food crops sector within Sri Lanka's agricultural economy. In recent years, tree crops accounted for 772 of total export earnings, provided some 45% of all agri- cultural employment, and contributed about 38% of the value added by the agricultural sector. They also provided some 47% of agriculture's contri- bution to governrient revenue in the form of taxes and levies. The health of Sri Lanka's economy thus reflects to a large degree the economic condition of the tree crop sector. 1.09 Unfortunately, the tree crop sector has been neglected in the past. The neglect has been most obvious in the failure to replant which has led to over-aged low-producing plantations of tea and rubber. The prime causes of past neglect have been the failure to provide adequate incentives to producers and the protracted nationalization of private estates. The potential of the tree crop sector to contribute to the economy has thus been seriously eroded. Bank projections 2/ are that the sector's growth rate for the period 1978-83 would decline to 4.9% p.a. compared with the 9.0% achieved in the period 1970-77. (This high growth between 1970 and 1977 was entirely due to high prices for tea; rubber showed a negative growth averaging -5.4% p.a. over the same period). More importantly, the tree crop sector is expected to provide only 66% of total exports, down 14% on its achievements during 1970-77. The net contribution of the sector to government revenue would fall more than proportionately because of the substantial need for govern- ment funded re-irvestment. 1/ Since 1950, Sri Lanka has maintained a barter agreement with the People's Republic of China, under which rubber has been exchanged for rice. 2/ Sri Lanka: Medium Term Investment Strategy 1979-83. Aide-Memoire to the 15th MeeLing of the Sri Lanka Aid Group. April, 1979.  -3- The Rubber Sub-Sector 1.10 Rubber is the second most important tree crop after tea. In the period 1970-77, rubber accounted for 18% of total export earnings, provided some 8% of all agricultural employment, and contributed about 5% of the value added by the agricultural sector. Duties and cesses levied on rubber exports in 1979 accounted for some 12% of government tax revenue. 1.11 The total area of land currently planted with rubber in Sri Lanka is reported 1/ to be 193,000 ha representing 11% of the total cultivable land and 20% of the tree crop area. The main rubber growing districts (Kalutara, Kegalle, Ratnapura, Galle and Colombo) are all situated within the Wet Zone in the South West quadrant of Sri Lanka. Rubber is also cultivated to a limited extent in districts within the northern margin of the Wet Zone. Official records of rubber area are based on registration of rubber holdings with the Department of Rubber Control (DRC) which indicate an area of some 263,000 ha. The discrepancy in area estimates reflects the fact that registra- tion records are out-dated and have not been corrected for change of land use since initial registration. The discrepancy of some 70,000 ha probably over- estimates the contraction in rubber area which has taken place over the past decade. Nonetheless, at least 16,000 ha of rubber have been removed since 1972-74 2/ owing to perceived higher profitability of alternative land uses (primarily coconuts and tea), uncertainties associated with land reform, and the lack of incentives or assistance for rubber replanting. 1.12 The ownership of rubber land is predominantly private (65%) with 35% owned by the State following the nationalization of private estates between 1972 and 1975. Under the Land Reform laws of 1972 and 1975, private holdings were limited to a maximum of 20 ha. Land in holdings larger than 100 ha is now owned by the state and operated by two government corporations - the State Plantation Corporation (SPC) and the Janatha Estates Development Board (JEDB) 3!. Land in holdings between 20 and 100 ha has also been taken over by the State but most of such land is to be transferred to the Agro- Industrial Corporation. A small proportion of State-owned rubber land has been used for settlement of previously landless people, but most is held and operated as commercial estates. Within the private sector, all holdings are classified as smallholdings. The holding size distribution is considerably skewed as indicated in Table 1.1; 39% of the total land area is owned by only 5% of landowners. Holdings less than 4 ha comprise 98% of the total private holdings and 71% of the total area of private holdings. Holdings less than 1 ha account for 88% of the total but only 46% of the area. Thus the industry is dominated by very small producers for whom rubber is commonly one of several crops grown. 1/ Report on the Rubber Industry Masterplan. Commonwealth Development Cor- poration, September 1979. Area estimate is based on survey results. Throughout this report this document is referred to as the Masterplan. 2/ Masterplan estimate. 3/ Recently, SPC and JEDB have been formed into separate ministries, with regional corporations performing the same functions.  -4- Table 1.1 PRIVATE HOLDINGS: SIZE DISTRIBUTION /a % of Total % of Total Area of Holding Size (ha) Private Holdings Private Holdings 0 - 0.5 67 23 0.5 - 1.() 21 23 1.0 - 2.0 7 15 2.0- 4.0 3 10 4.0 - 10.() 1.4 14 10.0 - 20.0 0.4 8 20.0 + 0.2 7 /a Based orL Masterplan survey results. 1.13 Replanting of over-aged rubber first received direct Government assistance in 1953. In that year, GOSL initiated a Rubber Replanting Scheme (RRS) with an annual target of 6,073 ha (15,000 ac) based on replacement of 3% per year of the then current rubber area. The scheme was initially financed by a GOSL grant supplemented by a cess on rubber exports. The export cess was suspended in 1957 owing to low world prices for rubber; the scheme was funded from general revenue until 1975 when the cess was reintroduced. The scheme continues to be financed by the cess. Under the scheme, replanters receive advice, replanting inputs and some financial assistance to purchase other inputs including labor during the immature period. The replanting payments to smallholders have increased in nominal terms from Rs 2,470/ha in 1953 to the present level of Rs 16,050. 1/ 1.14 The initial operation of the scheme was highly successful. During its first ten years 76,000 ha was replanted island-wide. From 1963 to 1976, the area replanted annually declined continuously from 6,400. ha to 2,550. This decline in replanting rates has resulted in an over-aged national stand which has reduced both present rubber production and the production potential for the next ten years. The total area now requiring replanting is estim- ated 2/ to be 55,000 ha, about 28% of the current rubber area. Some 41,000 ha of this total are in private hands and the balance controlled by the State. Since 1976, the replanting rate has shown some improvement reaching 3,200 ha in 1978 and a projected 4,000 ha for 1979. 1.15 The failure to replant rubber at a rate sufficient to maintain an even distribution over the past decade reflects the lack of producers' incen- tives, particularly in the form of inadequate levels of replanting payments. The level of replanting payments (in constant terms) since 1960 follows a de- clining trend until 1976 when this trend was reversed. 1/ Replanting payments 1/ The payments were Rs 2,470 1953-63, Rs 2,964 1963-67, Rs 3,705 1967-75, Rs 4,940 1975-77, Rs 7,410 1977-79, Rs 9,880 to November 1979, and Rs 16,050 from November 1979. In constant (1979) values, in terms of replanting costs, these payments would be Rs 12,630 (1963), Rs 10,740 (1967), Rs 7,140 (1975), Rs 6,820 (1977), and Rs 16,050 (1979). 2/ Masterplan estimate.  - 5 - accounted for 50% of replanting costs in 1956 and about 25% in 1975. (Since 1975 there have been several increases in the replanting payments and the current level is adequate). The strong simple correlation between replanting rates and replanting payments as well as multiple correlation between replant- ing rates and replanting payments, producer prices (in real terms), and adjustments for the structural change due to land reform as well as the uncertainty preceeding it show convincingly that the level of the replanting payment has been the principal determinant of replanting rates in Sri Lanka. The analysis shows further that the level of producer margins, though less so, also contributed significantly in this respect. Thus, the steady decline in rubber producer prices (in real terms) in Sri Lanka, which generally followed world prices down but improved following the devaluation of the rupee in 1977, contributed further to the decline in replanting activity. In addition, the replanting scheme has been hampered by cumbersome administrative procedures, exacerbated by staff shortages, which have resulted in excessive delays in the processing of requests for assistance, and in failure to provide replanting inputs when required. This has further constrained replanting and has probably encouraged some of the change in land use referred to above (para 1.12). 1.16 Sri Lanka's rubber production is currently about 155,000 metric tons (mt) p.a., down 3% from the record production of 159,000 mt in 1970. In the mid 1950s, Sri Lanka's production of about 100,000 mt constituted 5% of the world total. By 1968, production in Sri Lanka had risen to 149,000 mt reflecting the boost from replanting prior to 1960; Sri Lanka's share of world production was then at its highest at 5.5%. Since 1968, Sri Lanka has become an increasingly minor rubber producer, and now accounts for barely 4% of the steadily increasing world total. Since 1968, production has averaged 148,000 mt annually and despite marked annual savings, overall growth has been negligible at 0.3% p.a. During the past decade, some 97% of Sri Lanka's production has been exported, mainly to the People's Republic of China under a barter agreement which has accounted for some 45% of exports. Since 1970, local consumption of rubber has grown at an average of 10% p.a.; the 1978 consumption was 9,100 mt. Most of the local consumption is used in the manufacture of vehicle and tractor tires which are now exported by Sri Lanka. 1.17 The medium term outlook for the rubber subsector is poor, reflecting the erosion of production potential caused by failure to maintain an adequate replanting rate over the past decade. 1/ Assuming that replanting rates in the private sector can be increased to 8,200 ha annually and sustained at that level from 1984 to 1990, thereafter declining to about 3,000 ha/year by 2005, private sector production will decline about 40% to 55,000 mt by 1987. There- after, production will increase fairly steadily at an average annual rate of 5% to reach 143,000 mt by 2005. The current private sector production level (90,000 mt), will not be achieved again until the mid 1990s. The outlook for the estates sector is relatively better since much of its replanting backlog has been removed by concerted action over the past few years. Nonetheless, production from this sector is also projected to decrease from its present 1/ This outlook is based on Masterplan estimates.  - 6 - 42,000 mt to a low of 39,000 mt in 1982. Thereafter, production would increase at an average annual rate of 3% to reach 84,000 mit by 2005. The total rubber production in Sr-i Lanka is thus expected to decline about 35% to 101,000 mt in 1986 and thereafter to increase at an average annual rate of about 4% to reach 227,000 mt by 2005. The record 1970 production (159,000 mt) is unlikely to be surpassed before the mid 1990s. These production projections are based on rather crude data and may overstate the magnitude of the production decline by inter alia undlerestimating the capacity for ccntrolled slaughter tapping of old trees. Nonetheless, a sharp drop in qut-pubt w-ill occur during the next five years. 1.18 Although rubber prices are projected to remain buoyant during the next decade (para 7.02), it is unlikely that price increases would completely offset the declining production. The rubber subsector will thus contribute much less to exports and to government revenue than it has done in the past. The proposed project replantings comprise almost two thirds of what is con- sidered as a feas,ible national target for the private sector island-wide for the period 1981-85 1/. Without this early acceleration of replanting and the enhanced capacity for future replanting which the proposed project would create, the outlook for the rubber subsector would be bleak. IDA's Role in the Tree Crop Sector 1.19 IDA has financed a total of ten projects in Sri Lanka's agriculture sector since 1968. The majority of these has been directed towards increasing food-crop production, mainly paddy, to reduce Sri Lanka's dependence on food imports. IDA's interest in providing support to Sri Lanka's main export crops (tea and rubber) developed as the need for significant investments to arrest the decline of the tea and rubber industries became evident. However, in the absence of a policy framework for the development of these industries, IDA's direct involvement materialized in two relatively small and urgently needed projects for tea rehabilitiation and tea diversification (Credit 818-CE, US$21.0 M and Credit 819-CE, US$4.5 M approved in May 1978). The progress in implementation and general performance of both these projects has been highly satisfactory. Meanwhile at IDA's suggestion, GOSL requested assistance for two policy studies. CIDA initiated its tea masterplan study in January 1978 and ODM its rubber masterplan study in January 1979. Following the release of the Tea Masterplan in July 1979, an IDA subsector mission visited Sri Lanka in August/September to help identify an investment program for the tea industry which is now being considered by GOSL. The Rubber Masterplan has only recently been released (December 1979). Because of this delay, the proposed project, the first which IDA would finance in the rubber subsector, has only a limited scope and is intended to meet the most urgent needs of the industry. It is nevertheless envisaged that the Masterplan will iclentify a substantial invest- ment program for the rubber subsector which will provide further opportunities for external assistance. 1/ The Masterplan suggests that a feasible replanting target over the period would be 30,000 ha constituting some 73% of the 41,000 ha already over-aged.  -7- II. THE PROJECT AREA 2.01 The project area is located in the lowland wet zone of south western Sri Lanka and consists of the three administrative districts of Kalutara, Ratnapura and Kegalle (Map 14733R). Rubber is its most important crop, accounting for 49% of the total agricultural area of about 290,000 ha. The balance comprises coconuts 18%, paddy 16%, tea 15% and minor crops including cinnamon, cardamom, cocoa and coffee, 2%. Physical Characteristics 2.02 The project area includes some of the best rubber growing land in Sri Lanka, and 68% of the country's total rubber. The climate is monsoonal with 2,000-5,000 mm of annual rainfall associated largely with the southwest monsoon (April-June) and the northeast monsoon (October-December). Depending on location, between two and four months of the year have a 75% probable rain- fall between 25 and 75 mm. Since the soils are not excessively shallow or sandy, this relatively brief dry season does not restrict rubber yields signi- ficantly. The rainfall distribution conforms with the requirements for rubber cultivation. The main adverse effect of rainfall is the restriction of the number of tapping days to between 160 and 200 per year. Monthly mean tempera- tures range from 26 0C to 30 C with no cold season and relative humidity averages 70% throughout the year: both are ideal for rubber cultivation. 2.03 The terrain varies from rolling to steep with more extreme slopes in Ratnapura district. Although rubber grows best on flat or only moderate slopes, it can be grown satisfactorily on slopes of up to 180 provided suit- able soil conservation measures are taken. More than half the rubber island- wide and within the project area is grown on hilly to very hilly land. Most rubber is restricted to elevations below 370 m above sea level since yields are constrained by low temperatures and disease at greater altitudes. 2.04 The soils of the rubber growing areas in Sri Lanka are generally stony, often with visible rocky outcrops; even the better loamy soils commonly contain many rocks. Fortunately rubber trees can adapt well to such soils. The soils of Sri Lanka have been mapped at the reconnaissance level and the rubber growing soils provisionally classified into six series. 1/ These rubber growing soils vary in texture from silty to gravelly loams. The Rubber Research Institute of Sri Lanka (RRISL) has developed fertilizer recommenda- tions specific for each soil series; in practice, one broadly based fertilizer mixture is used successfully on all of the soils. Despite the steep terrain and high rainfall over much of the project area, soil erosion is not a major problem since there is widespread recognition of the need for soil conservation measures, with silt pits and bunds commonly used where required. In areas with lower erosion potential, the widely established leguminous ground cover reduces soil disturbance and the many rocky outcrops serve both to hold soil and to impede runoff. 1/ Handbook of the Soils of Sri Lanka. J. Soil Sci. Soc. of Ceylon, II, 1972-73.  Socio-Economic Characteristics 2.05 The total population of the project area is about 2 M or some 14% of Sri Lanka's total. Fifty percent, or about 1 M are over 14 years old and potentially economically active. Of the estimated 0.6 M people regularly employed, about two thirds work in agriculture, mainly on rubber and tea holdings. Apart from agriculture, employment opportunities are limited primarily to the tertiary sector, there being little secondary industry. Rubber cultivation entails a labor input of about 260 workdays/ha/year; each hectare of rubber thus provides approximately one full-time job. On small- holdings, the workforce generally comprises the husband and wife assisted by one of their children. In many cases the husband may find employment as a laborer off the smallholding and the wife manages the rubber enterprise. Women are also employed as tappers on large smallholdings and on Government estates. Ownership of rubber holdings is predominantly private. The Master- plan survey data indicate that some 32% of all rubber land in the project area is state owned or controlled, with 68% in private 'hands. Some 90% of the many small landowners have clear title to their land. 2.06 Estimates of holding size are available from a number of sources 1/ and differ somewhat according to their bases. Mission estimates 2/ for the project area suggest that some 46% of the private holdings are less than 0.8 ha and account for about 14% of the private rubber area. Holdings between 0.8 and 4.0 ha account for a further 49% of holdings and 57% of the private rubber area. Thus smallholdings below 4 ha account for some 95% of all private holdings and 71% of the private rubber area. 2.07 Land use within the project area is relatively intense and diver- sified, especially on the smaller holdings. Very few 3/ producers grow only rubber; other crops grown in order of decreasing importance include coconut, paddy, tea, spices, and vegetables. 2.08 Various surveys have collected data on smallholder income sources which, while showing variation between surveys, indicate that many small- holders earn significant off-farm income. Data from the Masterplan survey of 126 holdings indicate that 38% of owners of holdings less than 4 ha reported off-farm earnings and the corresponding figure for holdings above 4 ha was 63%. The relatively high proportion of smallholders having off-farm employ- ment explains in part why some 47% of smallholders with less than 4 ha hire some labor. Within the 0-4 ha group, the smaller Smallholders generally do not hire labor and it can be surmized that many such smallholders have few off-farm employment opportunities. 1/ MPI survey of 1,389 holdings in Kalutara and Ratnapura 1978; Inventory study and socio-economic survey by Masterplan team, 1979; and DRC records. 2/ Details in Project File. 3/ Some 15% according to Masterplan survey.  -9- Social Infrastructure 2.09 The project area is adequately served by all weather roads. Rail service connects the three administrative centers to Colombo and there are frequent aiid cheap bus services throughout the area. Postal services are prompt and reliable. The literacy rate is high, newspapers are readily avail- able, and radio broadcasts can be received throughout the area. Government health and educational facilities are adequate. Produce for local consumption is sold from roadside markets ("fairs") on a regular schedule. Supporting Services and Institutions The Replanting Scheme 2.10 The GOSL Rubber Replanting Scheme has been described previously (para 1.13). Replanting within the project area has accounted for about 70% of island-wide replanting since 1953, but only 36% since 1970. Research 2.11 The Rubber Research Institute of Sri Lanka (RRISL) is responsible for all aspects of research into rubber husbandry, production and processing. Its main station is situated at Aglawatta within the project area. It also operates a sub-station in Ratnapura District and carries out field experiments throughout the rubber growing area and has maintained some contact with research institutes of other rubber producting countries. RRISL has provided a good service to the rubber industry in the past but requires strengthening to better service the smallholder sector. Whilst improvements in research capability are both feasible and desirable, and are being considered by bilateral donors, the present level of research support available from RRISL is sufficient for the successful implementation of the proposed project. Extension 2.12 The Advisory Services Department (ASD) is responsible for the com- munication of research findings to smallholders, and to the medium size estates. ASD officers also assist in the operation of GOSL's replanting scheme by inspecting rubber for registration purposes and for authorizing replanting payments at the request of DRC. In addition, its staff are responsible for a program to establish and supervise Group Processing Centers (GPCs). ASD is currently operated as a department within RRISL and has its headquarters in Colombo. For administrative purposes, the rubber growing areas have been divided into three regions each of which is the responsibility of a Regional Advisory Officer (RAO) stationed in the field. Each region comprises three to four divisions, there being 11 divisions covering the entire rubber area. A Divisional Rubber Extension Officer (DREO) is responsible for ASD services within each division. Finally, each division is sub-divided into six or seven ranges which are each the responsibility of a Rubber Extension Officer (REO). The 72 REOs are thus nominally responsible for some 196,000 private holdings with 134,000 ha of rubber. In fact, ASD staff devotes most of its attention to smallholders' replanting or new planting. Even with this reduction in responsibilities, ASD is grossly understaffed: each REO deals with 1200 - 2500 holdings accounting for 900 - 1300 ha of planted rubber. A reasonable  - 10 - target would be 900 - 1000 holdings per extension officer which implies a 300% staff increase if all 196,000 smallholders were to be adequately serviced. 2.13 Within the project area, there are six DREOs and 41 REOs whose work- loads tend to be higher than the country-wide averages presented above. The understaffing is exacerbated by lack of transport for REOs and by the distance many officers must travel from their homes to their range. The work of the REOs is primarily directed towards assistance to smallholders planting rubber and the major components of this work are as follows: (a) recording details of holdings for registration purposes and certification that replanting is justified; (b) contour lining and design/specification of soil conservation works; (c) advice to replanters on replanting practices and general husbandry of immature rubber; and (d) inspection of work carried out by smallholders (weed control, girth increments, stand density, etc.) as the basis for authorization of staged replanting payments. Very little assistance is directed towards improving the husbandry of mature smallholder rubber. The inspection activities (d) institutionally are a responsibility of DRC which has its own staff of inspectors. However, since the workload has f-ar exceeded DRC capabilities, DRC has requested REOs to carry out inspect:ions for a fee (Rs 30 per visit). This inspection work is normally outside t:he control of ASD and its irregularity leads to problems of work:Load scheduling. In cases where DRC officers retain their responsibility for authorizing replanting payments, REOs have difficulty in convincing smallholders to act on their technical advice on replanting procedures. Department of Rubber Control 2.14 DRC was established in 1934 to control the production and planting of rubber. In 1953, DRC undertook responsibility for the Rubber Replanting Scheme (RRS) (para 1.13). The major functions of DRC are: registration of rubber holdings and producers; licensing of dealers, manufacturers, and nurseries; collection of industry statistics; and operation of the RRS includ- ing distribution of planting materials and fertilizer. DRC is headed by a Rubber Controller and its operations under the RRS are guided by a Rubber Replanting Advisory Board. 2.15 The work of DRC is organized on a regional basis with the exception of replanting payments. 1/ Each of the seven regions is the responsibility of an Assistant Controller who reports to the Senior Assistant Controller. Within 1/ The regions do not correspond to those of ASD nor to administrative districts.  - 11 - each Assistant Controller's region, work is functionally organized (registration, replanting and new planting). In 1977, DRC opened a regional office in Kalutara and has plans to decentralize operations further by opening an office in Kegalle and Ratnapura. To date, replanting payments are still made from the head office in Colombo but it is the intention to decentralize this responsibility to regional offices also. 2.16 The total staff of DRC is about 220 including 17 inspectors, 13 of whom carry out inspections for replanting payment authorization. DRC relies heavily on ASD to conduct most of the inspections for replanting payments. The current procedures followed by DRC in administering the RRS are described in the Project File. The operation of the RRS has been justly criticised by smallholders for its cumbersome and slow moving administrative procedures, and for its failure to provide necessary replanting inputs when and where required. DRC operations have been subject to a thorough study conducted as part of the Masterplan which has recommended some new procedures to improve operation of the RRS. It is understood that DRC would receive assistance from the UK gov- ernment to implement these Masterplan proposals. Rubber Processing 2.17 The project area is generally well served with facilities for pro- cessing smallholder latex (or coagulum) virtually all of which is processed in the form of Ribbed Smoked Sheet (RSS). In periods of high prices for crepe rubber such as have been experienced during 1979, the few smallholders close to crepe factories deliver latex to these factories. 2.18 Independent private operators process some 50% of the output on a custom basis for smallholders. The quality of RSS produced by private pro- cessors is quite high by smallholder standards. 1/ The private processor is patronized primarily by smallholders with less than 2 ha. Some 40% of small- holder latex is processed on holdings using the smallholder's own rollers most of which were purchased secondhand from estates or acquired long ago. The qua- lity of output tends to be low. 2/ An estimated 8% of latex or field coagulum is sold directly to factories for manufacture into crepe, crumb rubber or finished products. There is one crumb rubber factory in operation at Mawanella which takes latex from about 2,500 smallholders within a 20 km radius. In addition, many of the Government estates have surplus processing capacity and accept smallholder latex. Finally, 2% of smallholder latex is processed in Group Processing Centers (GPCs). 2.19 About 100 GPCs have been established including 67 within the project area. The GPCs have been established under the supervision of ASD staff using grant and loan funds provided by GOSL through RRISL. They offer a potentially useful means of improving smallholder returns in areas inadequately served by latex processing factories or private processors. The principle of the GPC is that 20 to 100 smallholders join together to collectively process and market their latex as RSS using modern equipment and techniques. Each GPC has a daily processing capacity of some 225 kg drc. The GPCs have not been operating 1/ Averaging 50% of RSS1, 25% RSS2 and 25% RSS3. 2/ Averaging 50% RSS2, 25% RSS4 and 25% RSS5.  Y - 12 - satisfactorily primarily because of low throughpuc-, poor management, and absence of a legal basis for operation. GPCs with low throughput must charge members a high processing fee to defray fixed staEf costs and this discourages new members. Poor management reflects the fact that office bearers are unpaid and often without managerial skills -- the employed 'managers' are in effect laborers since the position pays too little to atltract capable managers. The lack of legal status leads to problems of securitv of assets and to general distrust amongst members; it also means that there are no generally accepted rules of conduct. 1/ Despite these problems, some 25% of GPCs operate effi- ciently and profitably in the smallholders' interests. GPCs do provide a valuable focus for smallholder development by providing a forum which facilitates group extension activities and enables the smallholders' views to be noted. The quality of output from well-managed GPCs is comparable to or better than that from private processors. The processing fee GPCs charge members ranges between Rs 0.28 and Rs 0.33 per sheet and includes a nominal Rs 0.04 to repay the loan which GPC members have t:aken. Many, but not all GPCs, gain further benefits for their members by selling relatively large consignments (4-6 mt) of graded, high quality RSS. Rubber Marketing 2.20 Some 90% of smallholder production is marketed as RSS and scrap. GOSL operates a monopoly in the marketing of RSS (grades 1-3) and has, since the early 1950s, 'bartered this rubber for rice with the People's Republic of China. 2/ The Ministry of Trade and Shipping administers the China Contract through its Commodity Purchase Department (CPD). Although CPD is the only agency empowered to market RSS internationally, most of the purchasing, handling and shipping is carried out on behalf of CPD by registered private sector individuals in return for handling charges. CPD also operates 51 buying depots at which it buys directly from smal]holders. Within the project area, there are about 24 depots and two new depots are scheduled for con- struction in 1980. By regulation, CPD depots are required to accept all RSS offered to them for sale but often intake is limited by lack of storage space. Furthermore their staff rules, which make field managers bear the cost of any accidental overgrading, put their buyers under strong pressure to undergrade. Although many GPCs sell their RSS to CPD, this is by no means universal since some GPCs have found they could obtain better gradings from private dealers. 2.21 CPD fixes its prices of all RSS grades daily and these are widely reported. Private dealers generally pay the same prices as those announced by CPD. One price is given for Colombo and another f'or outside Colombo, with a difference of Rs 0.15 per kg to cover the handling and transportation cost to Colombo. CPD bases its prices on the Singapore daily price and deducts cesses, duties and handling charges to arrive at producer prices. Export duty 1/ Most GPCs have no legal status except for a f'ew registered as cooper- ative societies. The cooperative society ordinance is unsuited to a production enterprise and a more appropriate legal basis is being sought. 2/ Under this so-called China Contract, Sri Lanka receives a premium of some Rs. 0.1/kg over the prevailing world price.  - 13 - is fixed by the Customs Department once a week, based on the London Weekly Average CIF Quotation for RSS1. Duty is then levied on the basis of a slid- ing scale. Smallholders receive the RSS price announced for that day for the grade to which their crop is allocated. At current prices, smallholders receive about 50% of the fob value of rubber. 2.22 There are more than 2,000 licensed rubber dealers in the project area and a significant but unknown number of unlicenced dealers. Most small- holders sell their RSS to the nearest dealer who transports it to Colombo to the authorized shippers where it is sorted and packed. Some 75% - 80% of RSS is marketed in this way. CPD purchases between 15 and 20% of the total RSS production and ships it to packing centers in Colombo and Galle for export. About 4% of RSS is sold to unauthorized middlemen and then resold to dealers. About 1% of RSS is sold by cooperatives. The private dealers also purchase lower grade sheet (RSS 4 and 5 and unsmoked) and scrap which are freely marketed outside CPD and generally processed into crepe. 2.23 The price smallholders receive for their output is a function of quality and the extent to which buyers unfairly downgrade the quality of their output. The only effective remedy to downgrading is to strengthen the position of the seller by group marketing of relatively large lots. The prices received by smallholders using private processors, GPCs, or selling latex are about the same. Prices received by home processors are some 30% lower owing both to poorer quality and buyer downgrading. 2.24 In summary, the smallholder rubber producer faces few difficulties in marketing. The operation of CPD, though much criticized because of its limited capacity and tendency to downgrade, serves a valuable purpose in establishing a floor price. The coexistence of CPD depots and a widespread private dealer network has resulted in relatively efficient marketing without the dangers inherent in a completely government managed purchasing scheme. Given some improvement in grading practices, CPD could continue to ensure an efficient and fair marketing system. Government Rubber Estates 2.25 Within the project area, SPC and JEDB control some 32% of the rubber area. GOSL has recently encouraged greater integration between estates and smallholders and many estates now accept latex from neighboring smallholders which is then processed into crepe at the estate's factory. Other opportuni- ties for closer integration exist and are being developed although SPC and JEDB are to remain as commercially operated corporations. Under the project, selected estates within the project area would supply planting material for smallholders at commercial rates. III. THE PROJECT General Description 3.01 The proposed six year project would increase Sri Lanka's future rubber production through its support of an on-going GOSL scheme to replant  - 14 - the backlog of over-aged, low-yielding smallholder rubber. The project would accelerate replanting under this scheme within the project area. It would achieve this obj(ective by reorganizing and strengthening the agencies respon- sible for replanting and by the provision of adecLuate replanting inputs and incentives. Concomitantly, the quality of smallholder rubber output would be improved and the capacity for future replantirng expanded. The major components of the project would be: (a) replanting of 18,800 ha with high yielding rubber on about 27,000 smallholdings (below 20 ha) over five years to replace about 70% of the project area's, or 45% of the island-wide total of over-aged privately owned rubber; (b) institutional support for the two major implementing agencies (Department of Rubber Control [DRCI and Advisory Services Department [ASD]) comprising reorganization and strengthening through recruitment, training, and provision of equipment and civil works, and the creation of a Project Coordination Unit w:ithin MPI; (c) improvement of smallholder processing standards through establishment of a Processing Advisory Division within ASD which would service private processors and 67 existing Group Processing Centers (GPCs). And, following the attain- ment of- satisfactory operating norms for existing GPCs, establishment of 50 new GPCs; (d) support for research directed towards the smallholder through the provision of vehicles and equipment for the Rubber Research Institute of Sri Lanka; (e) support: to the Agrarian Research and Training Institute to conduct surveys and case studies aimed at evaluating the project's impact on smallholders; and (f) provision of technical assistance for training, project evaluation, development of improved procedures for admin- istration of the replanting scheme, and for the estab- lishment of the Processing Advisory Division. An implementation schedule for the project is presented as Chart 1 (Annex 4). Detailed Features and Project Activities A. Replanting Scheme 3.02 Sri Lanka has permitted the replanting rate to decline dramatically over the past ten years resulting in an over-aged national stand and declining rubber production (para 1.17). Failure to replant has been caused by lack of incentives owing to low rubber prices and inadequate replanting payments, and to an inefficient replanting scheme which failed to provide sufficient re- planting inputs when required and which was hampered by staff shortages and  - 15 - complex administrative procedures. The project would directly address each of these constraints other than price. The problem of low producer prices would be addressed indirectly by proposing that a mechanism be established for maintaining adequate producer margins through variation in tax and cess levels (para 5.25). The specific improvements to the replanting scheme to be in- corporated in the proposed project are described below (para 3.03-3.13). 3.03 Selection of Replanters. Participation in the project would be open to any smallholder having title to registered rubber which is either over 20 years old or particularly low yielding because of disease, damage, or inherent genetic potential. These conditions already apply for smallholder participation in GOSL's replanting scheme. It is expected that some 27,000 smallholders or about 30% of the project area total, would replant 18,800 ha with high-yielding rubber over the project period. Table 3.1 provides a summary of proposed replanting under the project. Details of the assumed holding size distribution of project participants are presented in Annex 4, Table 1. Table 3.1: PROJECT PLANTING SCHEDULE Project Year Area Replanted (ha) Proposed District Share (%) Kalutara Ratnapura Kegalle 2 1,600 50 25 25 3 2,800 42 28 28 4 4,000 50 25 25 5 4,800 46 25 29 6 5,600 36 28 36 Total 18,800 44 26 30 3.04 Publicity. Many smallholders are currently unaware of the extent and nature of the replanting assistance available and some postpone replanting believing old restrictions still apply. The project would include a publicity campaign to ensure that smallholders are aware of the assistance available and the new procedures. In each of the three administrative districts within the project area, a publicity unit would visit about 220 villages each year following a schedule prepared by the RAO. The mobile unit would comprise a long wheel base vehicle equipped with a portable generator, public address system, still and movie projectors, screen, films and slides. The local REO would advise smallholders of the date each village would be visited. Demonstration films, prepared by the RRISL film unit would be shown together with a local commercial film to attract a wide audience. In addition, leaflets describing the assistance available for replanting would be widely distributed and notices would be periodically carried in local newspapers. 3.05 Issue of Replanting Permits. Smallholders must be issued with a replanting permit by DRC before becoming eligible to receive assistance under the replanting scheme. Apart from meeting the requirements outlined above  - 16 - (para 3.03) the smallholding must be registered with DRC prior to issue of a replanting permit. In the past, the issue of permits has been a slow process with delays caused both by failure of the smallholder to understand his obligations and by DRC's inadequate registration data base. The system of registration and issue of permits would be maintained under the project since it provides a check against fraud and valuable statistical information but it would be facilitated by a number of changes. The smallholder would be made aware of the procedure and his specific obligations by the publicity campaign. Application forms for permits would be readily available from ASD staff who would assist the smallholder with the replanting permit application. DRC would streamline its procedures for checking registration details and would process permit applications promptly (Annex 2, Appendix 1). 3.06 Provision of Replanting Inputs. The inputs required by the small- holder replanter include technical advice, high yielding planting material, fertilizers and cash payments for purchase of other inputs (including labor). Under the project, all of these inputs would be provided in a form and manner convenient to the replanter. In the past this has not been the case. 3.07 Technical advice has not always been provided in the past because of staff shortages within ASD. The capacity of ASD r:o provide technical advice to replanters would be greatly expanded under the project. Firstly, ASD would benefit from much closer integration of its activities with those of DRC, which would share its regional offices with ASD staff. ASD would use DRC's records on replanters to schedule field staff activities and to ensure that all replanters were regularly visited and invited to group discussions on various aspects of replanting. ASD staff would direct particular attention towards replanters lagging behind in their replanting program as indicated by non-authorization of annual replanting payments. Secondly, ASD staff would use the opportunity presented by their involvement: in distribution of planting material and fertilizer to conduct group discussions for replanters at the various distribution points. Thirdly, and most importantly, ASD professional staff would be increased from the present 54 to 154 and would be better trained and equipped to carry out their responsib:ilities. This staff increase would mean that on average each field level extension officer would deal with a total of 175 new replanters over the six year project. 3.08 Planting material has been supplied by DRC but often the supply was insufficient to meet demand and smallholders had to postpone planting or seek alternative and often unreliable sources of supply. The supply and distribution of planting material would be improved under the project. Smallholders with replanting permits would automat.ically receive planting material without having to make prior application, The clones to be used are discussed in Chapter VI (para 6.05). DRC would estimate planting material requirements based on replanting targets which would be set by the Rubber Policy and Planning Unit of MPI (para 5.24). These requirements would be esti- mated 30 months before expected delivery. All planting material would be produced by SPC and JEDB estates under contract to DRC (para 6.04). ASD would distribute planting material using its own vehicles and contract carriers as necessary. Within each range (the administrative zone of a REO), about ten input distribution points would be selected after consultation with small- holders. These distribution points would include the office of the REO, CPD  - 1 7 - depots, GPCs, Agrarian Service Centers, Cooperative Stores and Gram Sevaka offices. Each smallholder would nominate one distribution point for all his inputs. ASD would arrange schedules for delivery of planting material such that the REO would spend one day at each distribution point. Each REO would distribute material to about 10 smallholders per day throughout the 3 week planting period. 3.09 Fertilizer supply has been the responsibility of DRC in the past but supplies have been insufficient, partly because of difficulties in obtain- ing fertilizer from the Ceylon Fertilizer Corporation primarily due to import restrictions existing at the time, but also because of inadequate forward planning. Import liberalization policies and DRC's ability to also acquire fertilizer through direct imports has essentially estimated this problem. DRC uses the CPD depots as distribution points for fertilizer and for plant- ing materials. In cases where DRC is unable to supply the specially mixed fer- tilizer, 1/ the smallholder is authorized to purchase such fertilizers locally using a cash advance from his replanting payment. Unfortunately, supplies of the appropriate fertilizer mixture are often unavailable locally. It has been estimated that only 20% of the required quantities of fertilizer has been applied to replanted rubber in recent years. This under-fertilization has prolonged the immature period of replanted rubber and reduced the ultimate production potential. Fertilizer distribution would be improved under the project. Total fertilizer requirements for all replantings would be calcu- lated by DRC and orders would be placed with the two major GOSL fertilizer companies some six months in advance, for delivery over the year or imported directly by DRC. DRC would provide ASD with lists indicating the fertilizer entitlement of individual replanters. ASD's Supplies Division (SD) would use its own trucks plus contractors where necessary to transport fertilizer from Colombo to three regional fertilizer stores where it would be stored under ASD supervision before distribution to smallholders by SD twice per year. All replanters would receive fertilizer in kind and there would be no cash pay- ments in lieu. Details of application rates are given in Annex 1, Table 1. The distribution points used for planting material would also be used for fertilizer as would be the system for scheduling distribution. Each REO would be required to distribute fertilizer to about 35 smallholders per day over the fertilizer distribution period. The maximum quantity of fertilizer to be distributed would be about 5,700 mt per distribution in project year (PY) 8 which would be equivalent to about 7 mt/range/day. 3.10 The inspection of holdings for authorization of replanting payments and supply of replanting inputs has been the responsibility of DRC. The 13 inspectors in DRC have been unable to cope with the volume of inspections and DRC has authorized ASD staff to assist by carrying out inspections for a fee. This has led to various problems described previously (para 2.13). Under the project, the inspection of holdings for authorization of replanting payments would be the sole responsibility of ASD. The schedule of inspec- tions and the criteria for authorization of payment and further supply of inputs is discussed in Chapter VI. Staff of ASD would be provided with a list of all replanters in their range by the regional office of DRC/ASD. Based on 1/ The fertilizer most commonly used is R462 Mg. The analysis is 9:5:3:1::N:P:K:Mg.  - 18 - this list, each REO and EO would prepare an inspection program which would ensure that each replanter was visited annually ior inspection during the period May to mid October to keep the interval between payments as close as possible to 12 months. The smallholder would be notified in advance of the proposed date of the inspection visit, and could request an earlier or a later inspection under exceptional circumstances. 3.11 Cash payments are disbursed by DRC following satisfactory inspection and debited to each replanter's replanting payment account which is also debited with expenditures on planting material and fertilizer. These payments are for the purchase of inputs not supplied in kind such as agro-chemicals and fencing material, and for the smallholder's labor inputs. Although the cash payments may be made either by cheque or money order and mailed to the small- holder, in practice the smallholder has often had. to visit DRC in Colombo to secure the release of the payment. Under the project, the disbursement of cash payments to replanters would be fully decentralized and procedures streamlined. Smallholders would nominate their preferred form of payment (money order or cheque) and their local bank or post office when applying for a replanting perrnit. Payments would be made promptly following receipt of a satisfactory inspection report from ASD indicating that specified standards had been achieved (Annex 1). The number of inspections of the immature rubber would be increased to eight to ensure that the trees were not opened for tapping prematurely. 3.12 Ombudsmen The operation of the replanting scheme would be further improved under the project by the creation of an ombudsman position in each of the three administrative districts of the project area. These ombudsmen would receive ancd investigate complaints from smallholders relating to the functions of DRC and ASD. Where complaints were found to be justified, the ombudsman would refer these to the Secretary of the Ministry of Plantation Industries for action. Further details are given below (para 5.28). 3.13 Intercropping. In the past, DRC has prohibited the cultivation of any crop other than leguminous covers between the rows of immature rubber trees. This prohibition reflected past industry bias toward estate-type production goals where the objective was to maximize rubber production per se. The objective of the smallholder, and particularly the smaller smallholder, is more to maintain adequate cash income levels and to maximize returns per hectare. Intercropping offers the opportunity to do this during the immature period, and is therefore attractive to the smallhDlder. Intercrops can be grown successfully with rubber with little or no adverse effect on long term rubber production provided that the crop is adequately fertilized and certain other precautions are observed (para 6.12). DRC has recently permitted intercropping in suitable circumstances. The project would encourage wider adoption of intercropping through various means. Firstly, the publicity campaign would include information on intercrops to create smallholder interest and awareness. Secondly, ASD would distribute leaflets describing appropriate cultivation techniques for the major approved intearcrops (bananas, passionfruit and pineapples). Thirdly, smallholders would be given the opportunity to order planting material at the time of applying for a replanting permit. DRC would aggregate the planting material requirements so identified and pass these estimates on to local Department of Agriculture farms which have agreed  - 19 - to supply the planting material at cost. Finally, ASD staff would stress the need to apply adequate quantities of fertilizer to the intercrops and would indicate to the Agrarian Service Department the expected demand for intercrop fertilizer from each of the project area's Agrarian Service Center. In view of the high and prompt cash returns from intercrops, and the adequate replant- ing payments provided to replanters, it would not be necessary to provide intercrop inputs on a grant basis as had been originally considered. An agree- ment was obtained that adequate supplies of planting material would at all times be available to meet project requirements. B. Institutional Support 3.14 In order to execute the replanting scheme outlined above, and to expand the capacity for replanting in the future, the project would assist in the strengthening and reorganization of the two major implementing agencies, ASD and DRC. In addition, the project would establish a Project Coordination Unit (PCU). Support for ASD, DRC and PCU 3.15 Details of the responsibilities of ASD, DRC and PCU, and of the recruitment and training of their staff, are given in Chapter V. The project would provide civil works and equipment to improve the functioning of these implementing agencies. 3.16 Civil Works. The project would construct a permanent training center for ASD staff at Nivitigalakele near RRISL. The center would comprise a lecture hall, classrooms, office, dormitory, kitchen and living facilities. It would accommodate 30 residential trainees in addition to teaching and support staff, and would be constructed early in project year 1 to facilitate training of incremental staff. (Cost details in Annex 4, Table 3). Three regional offices would be constructed, one in each district to be used both by DRC and ASD staff. Each office would accommodate about 50 DRC staff and 4 ASD staff. Plans for these offices would be prepared by architects commis- sioned by PCU. Construction would commence mid PY1 and be completed by end PY2. (Cost details in Annex 4, Table 5). Three regional fertilizer stores would be constructed adjacent to the regional offices. Each store would hold about 2,000 metric tons, sufficient for the peak fertilizer demand from re- planters in the project area. A total of 78 houses would be provided for all field level ASD staff above the level of EO, to minimize the considerable staff time currently lost in travelling to work. Each house would be centrally located within the officers' administrative area. The houses for REOs and DREOs would include a small attached office and store. Designs for the houses would conform with GOSL standards. (Cost details for houses and fer- tilizer stores are in Annex 4, Table 4). 3.17 Equipment. The project would provide 108 bicycles for EOs and 92 motorbikes for REOs and PAs which would meet incremental staff needs and provide necessary replacements during the implementation period. ASD officers would be required to purchase bicycles and motorbikes and would receive a loan for this purpose under the standard terms currently applying. The loans would be repaid from the proceeds of the travelling allowances  - 20 - paid to these staff. An agreement was obtained that GOSL would periodically review travel allowances in relation to vehicle operating costs and would undertake to ensure that project vehicles are fulLy utilized. Four wheel drive vehicles would be provided for: RAO (3); SPA (3); DREO (9); Assistant Controllers (3); the PCU (1); the training center (1); and for the ASD head office (3). Since these vehicles are expensive in relation to staff salaries, they would remain the property of the implementing agencies and would be driven and maintained by the agencies' drivers. Three mobile publicity units would be provided to ASD for use in the project area. Three passenger cars would be provided for the ASD head office and one for the PCU. Three microbus type vehicles would be attached to the training centers to transport trainees to and from courses and on tours. Ten trucks would be provided to distribute fertilizer and planting materials. 3.18 The project would also provide required office equipment for the PCU and for all ASD and DRC offices in the project: area. 3.19 ASD staff would be provided with field equipment consisting of pro- tective clothing including motorcycle helmets, ancL measuring tapes and levels for contour lining. Each range would be provided with three tree pullers (small winches) which would be loaned to smallholclers to facilitate uprooting thus permitting more smallholders to earn substantial cash income through the sale of old rubber trees. 3.20 Further details including the costs of equipment for ASD, DRC and PCU for training and for other activities, are provided in Annex 4, Tables 3, 4, 5 and 6 respectively. C. Processing Component 3.21 By providing technical advice the project would improve the quality of output from both private processors and GPCs arid, by strengthening and expanding the GPCs, would provide alternative processing channels to reduce the proportion home processed. This action would be in support of ASD's general approach to processing which is firstly tco encourage smallholders to sell latex for crepe or block production wherever geographically feasible; secondly to improve processing standards in the private sector; and thirdly to establish GPCs where there are no factories or private processors. The cost of the processing component is shown in Annex 4, Table 7. 3.22 Processing Advisory Division. A Processing Advisory Division (PAD) would be established within ASD to provide technical and managerial advice to both private processors and GPCs. Details of the staffing and training of PAD staff are given in Chapter V. 3.23 Each Processing Advisor (PA) would supervise all the existing GPCs in his division, taking over from the REO who has had this responsiblity in the past. On average, each PA would be responsible for about 8 GPCs which he would visit on a regular basis. The frequency of visits would depend on the need of each GPC. Initially the PAs, with the assistance of the Senior Proces- sing Advisors (SPA), would survey the existing GPCs to define and record their  - 21 - characteristics (membership, throughput, financial status, etc.) and to iden- tify their problems. After completion of this survey, ASD would develop a strategy to improve the performance of the GPCs. It is envisaged that this strategy would include: provision of a sound and appropriate legal basis for operation of the GPCs; development and institution of financial controls based on a revised record keeping system; expansion of membership through recruitment drives and offer of incentives such as daily advance payments; improvement in quality of output through attention to technical procedures and improvement of facilities; and improvement in sale prices through greater use of group marketing. 3.24 PAD would also provide advice to private processors. All private processors within the project area would be contacted by mail and requested to fill in a questionnaire concerning their equipment, processing methods and quality achievements. The questionnaire would be accompanied by a leaflet introducing the new service and explaining its potential benefits to processors. Processors would be invited to contact their local PA for advice on processing methods likely to improve quality. In addition, staff of the PAD would compile a list of the larger private processors and visit these processors routinely. It is envisaged that the major thrust of the advice to private processors would concern improvements in the design and operation of smoke- houses since much of the quality deterioration is caused by incorrect smoking. 3.25 The Group Processing Centers. The GPCs offer a potentially useful means for improving smallholder returns in areas not adequately served by latex processing factories or private processors. In view of their potential, the project would assist in their improvement, and, following the achievement of satisfactory operating norms, in their expansion. To date, only a small proportion of existing GPCs have realized their potential for reasons outlined above (para 2.19). The project would address the technical and managerial weaknesses through the efforts of the processing advisors. 3.26 The project would also support physical improvements to existing GPCs. These improvements would include a tiled bulking tank, a dripping rack to prevent uneven enzyme darkening, a working table for coagulation pans, and an additional supply of 100 coagulation pans. Existing GPCs would be eligible to borrow up to Rs 6,000 from ASD to meet the full cost of these improvements. Loans would be available only to GPCs meeting the following conditions: (a) that they had an adequate legal basis for their operations including satisfactory operating by-laws (MPI is currently studying alternative forms of legal association and is expected to propose an adequate basis in the near future); (b) that they obtain pledges from members to collectively provide a satisfactory minimum input of latex not less than 100 kg/day; and (c) that a satisfactory system of accounting and auditing was followed. The project would provide for the rehabilitation of the 67 existing GPCs in the project area once these conditions were met.  - 22 - 3.27 Once existing GPCs have become financiaLly viable following the improvements referred to above, the project would support an expansion of GPCs in areas where existing processing facilities were inadequate. During negotiations an assurance was obtained that: (i) GOSL would establish a legal basis for the operation of GPCs satisfactory to II)A not later than March 31, 1981, and (ii) that no new GPCs would be established in the project area until the potential members pledged to collectively supply not less than 100 kg of latex per day of operation. ASD has already idenl.ified 50 potential sites 1/ in the project area however detailed studies would be required before sites are finally chosen. The new GPCs would be similar to existing GPCs but would incorporate various improvements. Details are given in Annex 2, Appendix 3. The cost of each GPC is estimated at Rs 50,000 including Rs 5,000 working capital. It is recommended that ASD should re-examine the design and con- struction standards to identify the scope for reducing capital and operating costs. It is proposed that Rs 20,000 would be provided as a grant, with the balance in the form of a loan at 10% interest over 12 years. For GPCs oper- ating at 75% of capacity, the loan repayments would amount to 10 cents/kg. Details of the financial and economic returns expected from GPCs are presented in Annex 4, Table 8. The operators of GPCs would be employed by the members and would receive training from ASD in addition tc) the advisory service provided by PAD. 3.28 The project would also encourage group miarketing of output. Group sales are currently used by most GPCs but smallho]ders outside the GPCs almost invariably sell their output individually. If such smallholders formed a group marketing organization (GMO) which could offer large lots of graded rubber to buyers it would tend to reduce downgrading and thus provide a better return to GMO members. The GMO could also be used as a nucleus for the purchase of latex for crepe or block manufacture. The staff of PAD would investigate the feasi- bility of establishing GMOs which would require very little capital since they would use existing processing facilities. PAD would also investigate the marketing of scrap rubber all of which is currently sold at a low price by individual producers. GPCs and GMOs could accept, dry, segregate, and grade, tree scrap, cup lamp, and earth scrap and sell in bulk lots. Such sales would provide additional income for members. D. Research Component 3.29 The Rubber Research Institute of Sri Lanka (RRISL) carries out research which is of potential benefit to smallholders. Work on cultivar test- ing, disease control, nutrient requirements, husbandry, intercropping, and small scale processing are of immediate importance to smallholders. In addi- tion, officers from RRISL serve as subject matter specialists for the extension staff of ASD. The conduct of field trials and frequency of advisory visits by RRISL staff is constrained by a lack of vehicles within RRISL. In view of the wide range of soil types and micro-climates represented within the rubber 1/ Twenty each in Kalutara and Ratnapura, ten in Kegalle. The Mawanella block rubber factory in Kegalle already collects smallholder latex from a 20 km radius thus reducing the need for new GPCs in Kegalle.  - 23 - growing area of Sri Lanka, it is important that RRISL have the mobility to conduct field trials over a wide area. RRISL laboratory work is constrained by a shortage of support equipment. The recently completed Masterplan has identified major investment requirements for RRISL and it is understood that the UK government would provide financial assistance for such investments. The project would therefore limit its support for RRISL to the provision of transport and some ancillary equipment to enable the Institute to better serve the smallholder sector of the industry. 3.30 The project would provide RRISL with 16 small conventional vehicles and two four wheel drive vehicles. It would also provide fifteen 1.5 HP air conditioners and six 0.5 HP dehumidifiers. These would be installed in recently completed laboratories and would provide a suitable environment for the operation of costly existing equiment such as an autoanalyzer, and atomic absorption spectrophotometer and direct recording balances. The lack of air-conditioned laboratories has restricted the use of such equipment in the past. Finally, to assist the publicity and training units within ASD, the project would provide equipment for the photographic unit of RRISL. The equipment would include slide and movie cameras, projectors and ancillary items. Provision would also be made for specialized aspects of film making which would be carried out under contract by the established cinematographic industry in Sri Lanka. These provisions would be sufficient for the production of two, 20 minute demonstration films and would save foreign exchange while providing extension material tailored to the particular circumstances of the smallholder in Sri Lanka. Details of the cost of the research component are presented in Annex 4, Table 7. 3.31 The project would also provide for five 12 month post-graduate fellowships for selected RRISL staff. These fellowships would assist RRISL in countering the heavy attrition of its better staff through promotion to other organizations. The five fellowships offered would include Soils Chemistry, Plant Pathology, Genetics and Plant Breeding, and two other fields to be selected. The cost of these fellowships has been included in the training component (Annex 4, Table 3). Present rules require persons receiv- ing fellowship training to enter into a bond to serve the Government for a stipulated period. This period often amounts to one year of service for every month of fellowship training received with a maximum of 7 years. During negotiations agreement was obtained that persons receiving training under the project would remain with their current employer at least for the period of the bond. E. Project Evaluation 3.32 The proposed project would attempt to reverse the trend towards declining replanting rates over the past decade by addressing the major apparent causes of low producer prices, inadequate replanting payments, and a poorly administered replanting scheme. In view of the many factors which are likely to influence the smallholder's decision to replant, it is probable that a better understanding of the smallholder's viewpoint would reveal oppor- tunities to make the replanting scheme more attractive to smallholders and  - 24 - more efficient. It is therefore proposed that the project be evaluated during its implementation in the belief that the results of such an evaluation would permit fine-tuning of the replanting scheme so as to best serve the small- holders and the economy. 3.33 To avoid the risk of partiality, the evaluation would be carried out by the Agrarian Research and Training Institute (ARTI). ARTI is an estab- lished national research organization with considerable experience in evalua- tion of agricultural projects. It is currently conducting similar evaluations for projects financed by the Asian Development Bank and IDA. The evaluation study has been discussed with ARTI which has agreed to the proposal that the project would meett ARTI staff and operating costs associated with the conduct of the evaluation. A consultant would assist ARTE in the design and initial implementation of the study (para 3.34). It is envisaged that the evaluation would comprise an initial baseline survey of smalLholders in the project area which would be ccmpleted together with a report by December 1980. In subse- quent years, case studies would be conducted as indicated in the terms of reference for the consultant (Annex 2, Appendix 4) which give further details of the nature of the studies. ARTI would consult with IDA on the details of project evaluation. Details of the cost of project evaluation are presented in Annex 4, Table 7. F. Technical Assistance 3.34 Whilst Sri Lanka has considerable experience and expertise in most aspects of the rubber industry, the project would introduce several new approaches not previously undertaken by the project agencies. The project therefore would provide about 15 manmonths of consultant input to supplement resources within the implementing agencies. Details of this technical assist- ance are provided in the draft terms of reference (Annex 2, Appendices 1, 2, 3 and 4). An assurance was obtained that the project consultants would be retained for the above assignments in accordance with the implementation schedule and on terms and conditions satisfactory to IDA. The cost of the technical assistance component is presented in Annex 4, Table 7. The average cost of consulting services is estimated at US$6000 per man-month 1/. IV. COST ESTIMATES AND FINANCING A. Cost Estimates 4.01 The total cost of the six year project is estimated at US$28.0 M of which US$7.4 M or 26% would be foreign exchange and US$0.25 M duties and taxes. Details of costs, summarized in Table 4.1 below, are in Annex 4, Table 9. 1/ Consisting of salary US$3,500, allowances, US$1,200, travel US$800, and miscellaneous, US$500. The consultants are expected to be nationals of South East Asian countries.  - 25 - Table 4.1: PROJECT COST ESTIMATES -----Million Rs------ ----Million US$---- % of Local Foreign Total Local Foreign Total Base Cost Replanting Costs - Budded stumps 22.1 1.1 23.2 1.43 0.07 1.50 - Fertilizer 12.8 51.2 64.0 0.82 3.30 4.12 Labor and other inputs 104.9 2.2 107.1 6.77 0.14 6.91 Total replanting costs 139.8 54.5 194.3 9.02 3.51 12.53 73.8 Institutional Support Advisory Services Dept. (ASD) Investments - Civil works 8.8 3.0 11.8 0.57 0.19 0.76 - Equipment 0.6 7.7 8.3 0.04 0.50 0.54 Staffing 15.8 - 15.8 1.02 - 1.02 Operating Costs 2.8 5.0 7.8 0.18 0.32 0.50 Total ASD 28.0 15.7 43.7 1.81 1.01 2.82 16.6 Dept. of Rubber Control (DRC) Investments - Civil works 0.6 0.2 0.8 0.04 0.01 0.05 - Equipment 0.5 0.6 1.1 0.03 0.04 0.07 Staffing 5.2 - 5.2 0.34 - 0.34 Operating Costs 0.8 0.6 1.4 0.05 0.04 0.09 Total DRC 7.1 1.4 8.5 0.46 0.09 0.55 3.3 Training Investments - Civil works 0.9 0.5 1.4 0.06 0.03 0.09 - Equipment 0.2 1.1 1.3 0.01 0.07 0.08 Training Courses 1.7 2.3 4.0 0.11 0.15 0.26 Total Training 2.8 3.9 6.7 0.18 0.25 0.43 2.6 Technical Assistance 0.2 1.2 1.4 0.01 0.08 0.09 0.5 Processing Component 1.7 1.2 2.9 0.11 0.08 0.19 1.1 Research Component 1.3 2.2 3.5 0.08 0.14 0.22 1.3 Project Coordination 1.2 0.6 1.8 0.08 0.04 0.12 0.7 Project Evaluation 0.2 - 0.2 0.02 - 0.02 0.1 Total Base Cost 182.3 80.7 263.0 11.77 5.20 16.97 100.0 Physical Contingencies 6.2 2.0 8.2 0.40 0.13 0.53 3.1 Price Contingencies and rounding 131.2 31.6 162.8 8.46 2.04 10.50 61.9 Total Expected Project Cost 319.7 114.3 434.0 20.63 7.37 28.00 165.0  - 26 - Cost estimates for equipment, civil works, staffing and technical assistance, are based on recent actual costs for comparable items or services in Sri Lanka adjusted to November 1979 levels. Replanting costs represent the full cost of all replanting inputs provided to smallholders by GOSL under the Rubber Replanting Scheme (Annex 4, Table 10). Staffing and other operating costs for ASD and DRC include those recurrent costs directly attributable to project activities. The physical contingencies included in the estimates amount to 3.1% of base costs and include 15% for planting material, 10% for civil works and 5% for fertilizer. Contingencies included for subsequent price escalation over the project period amount to 60.0% of base costs plus physical contingen- cies. Price contingencies for local costs have bieen estimated on the basis of the following estimates of domestic inflation Eor the years 1980 to 1985 respectively: 21%, 15%, 12%, 10%, 10%, and 10%. Price contingencies for foreign costs have been calculated using the following estimates of inter- national inflation for the years 1980 to 1985 respectively: 10.5%, 9%, 8%, 7%, 7% and 7%. B. Proposed Financing 4.02 Project Financing. The proposed IDA credit of US$16.0 M would finance about 58% of the project cost excluding duties and taxes. It would cover the full foreign exchange cost (US$7.4 M) and about US$8.6 M equivalent of the local costs. GOSL would finance the balance of project costs including taxes and duties (US$12.0 M). Funds would be made available to the implement- ing agencies in the form of budgetary allocations, The proposed project financing is summarized below: Table 4.2: FINANCING PLAN (US$'000) Total Project Costs Components GOSL IDA including Contingencies Replanting Costs /a - Replanting Payments 2,207 13,271 15,478 - Fertilizer Subsidies 5,559 - 5,559 ASD 2,967 1,551 4,518 DRC 848 139 987 Training 158 427 585 Technical Assistance - 96 96 Processing 48 237 285 Research 17 253 270 PCU 168 26 194 Project Evaluation 28 - 28 Total 12,000 16,000 28,000 /a The replanting cost includes fertilizer subsidies which would be funded by GOSL. These subsidies together with the replant- ing payments made under the Rubber Replanting Scheme, are fully recovered from rubber growers by GOSL (para 7.10).  - 27 - 4.03 Retroactive financing is proposed for pre-project expenditures required to achieve the 1981 rubber replanting target. These include training of newly recruited extension staff, early construction of the training center, and procurement of some urgently needed vehicles and equipment. The amounts involved would be small and IDA's share of the retroactive financing would not exceed US$0.2 M. 4.04 Financing of the Rubber Replanting Fund. Rubber replanting payments are financed by a Replanting Fund which is financed by a Replanting Cess levied on rubber exports. The Rubber Replanting Fund is administered by DRC. At the end of 1979, the fund had accumulated cash reserves of about US$3.1 M. Financial constraints have not restricted the rubber replanting program in the past but the proposed rapid acceleration of replanting over the 1980-85 project period would lead to a substantial cash deficit in the fund. As shown in Annex 4, Table 11, the replantings both within and outside the project area would entail cash disbursements from the fund of US$39.1 M 1/ while total revenues over the same period would amount to only US$18.6 M assuming the cess remained at its current level of Rs 0.35/kg. 1/ Expenditure on replantings would fully exhaust cash reserves of the fund by end PY2; thereafter the fund would accumulate a deficit estimated at US$17.4 M by PY6. During the subse- quent five-year period, the annual deficit of the fund would be of the order of US$5.5 M. 1/ IDA's proposed direct contribution to the Rubber Replanting Scheme of US$13.3 M would meet 47% of the expected accumulated fund deficit of US$28.4 M in current dollars (equivalent to US$17.4 M in constant 1979 dollars) over the project period. The shortfall of US$15.1 M, as well as the annual deficits thereafter, would have to be funded from government resources. GOSL's commitment to finance the fund's impending deficits during and beyond the project period would thus be crucial to the implementation of the accel- erated replanting program. 4.05 The deficits could be funded through budgetary provisions as indi- cated in Annex 4, Table 11. Alternatively, the replanting cess could be increased to generate additional revenue for the fund, in which case a Replant- ing Cess of some Rs 0.80/kg would be required. The latter approach, however, would reduce the producers' margin unless the export duty were reduced by an amount corresponding to the increase in replanting cess. A decision on the method by which the fund deficits would be financed would be reached during negotiations. In view of the importance of funding the accelerated replanting scheme adequately, and of the danger of eroding producer margins through indis- criminate increases in the cess, the following assurances were obtained at negotiations. GOSL would (i) provide the Replanting Fund with the resources necessary to finance deficits involved in implementing the project and sustain- ing the island wide replanting scheme at the optimum replanting rate of about 8,100 ha per year; and (ii) annually review and submit to IDA the financing requirements of the Fund to carry out the replanting scheme: such review would include an analysis of the impact of financing arrangements on producer margins. 1/ In constant 1979 prices.  - 28 - C. Procurement 4.06 Replanting activities would be carried out by individual smallholders using family or hired labor. Replanting payments for such work would cover the costs of (i) planting materials (US$1.50 M) ard fertilizer at the subsi- dized rate applicable to all fertilizer sales in Sri Lanka (US$1.51 M) which would be provided in kind to the planters; and (ii) the cost of other required inputs and of hir:Lng labor (or otherwise compensating the smallholder for his own labor) (US$6.91 M). Nursery works to provide planting materials would be carried out by force account on State rubber estates (JEDB and SPC). This approach is preferred both on grounds of quality and cost; the estates have the expertise necessary to produce high quality planting materials, and they have a large, under-utilized work force. Fertilizer would be purchased by the project at government rates from either of the two government corporations dealing with fertilizers or imported directly by the DRC. Separate tenders for the project requirements would not be practicable since rubber fertilizer requires mixing of several inputs (themselves ingredients of other types of fertilizer); nor would it be in the interest of economy since the government corporations, by tendering on much larger quantities, would obtain more competitive prices. Contracts for civil works (US$0.99 M) would be small and widely dispersed (one training center for US$0.09 M, three regional office buildings for US$0.06 M, three fertilizer stores for US$ 0.06 M, 78 houses for US$0.70 M and 50 GPCs for US$0.08 M). Such works would not be suitable for ICB contracts and would be let following local competitive bidding amongst the numerous local contractors. Contracts for vehicles and equipment (US$1.02 M) costing more than US$100,000 would be let following ICB: a preference limited to 15% of the cif price of imported goods or the import duty, whichever is lower, would be extended to local manufacturers in the evaluation of bids. All other contracts would be let following local competitive bidding in accordance with local procedures acceptable to IDA. There is an adequate representation of foreign suppliers in Sri Lanka. Items purchased under contracts costing less than US$5,000 and specialized research equipment, would be procured through prudent shopping. The balance of project costs would comprise staff salaries and other administrative expenditures (US$ 2.08 M), technical assistance and overseas training (US$0.23 M), local training (US$0.12 M), fertilizer subsidies (US$2.6 M) and contingencies (US$11.03 M). D. Disbursements and Audits 4.07 Disbursements from the proposed credit are expected to be completed by June 30, 1986, approximately six months after the end of project implemen- tation. The estimated schedule of disbursements is presented in Annex 4, Table 12. Disbursements under the Credit would cover: (a) 80% of replanting payments made to rubber replanters (US$11.20 M); (b) 80% of expenditures for civil works (US$'.15 M); (c) 100% of foreign expenditures for imported equipment, (d) 100% of local expenditures (ex-factory) ior locally manufac- tured equipment and materials following competitive bidding procedur es,  - 29 - (e) 80% of local expenditures for vehicles and equipment locally procured without competitive bidding procedures (US$1.10 M); and (f) 100% of the cost of technical assistance and overseas training (US$0.25 M). The unallocated portion of the credit (US$2.30 M) would be reallocated to the above categories as required. No disbursements would be made against salaries and related administrative expenditures or against local training. Details of disbursements for each component are given in Annex 4, Table 13. 4.08 Disbursements against replanting payments referred to in para 4.07 would be made against quarterly certified statements of expenditures. All payments under small civil works, vehicles, and equipment procured through local contracts, the value of which would not exceed Rs 100,000, would be made against statements of expenditures. Documentation for these expendi- tures would not be submitted to IDA for review but would be retained by DRC, ASD, and RRISL, and made available for inspection by IDA on request. Full documentation would be required for all other disbursements. Although the implementing agencies' accounting procedures are adequate, the accounting units require additional staff which the project would provide to enable the agencies to prepare adequate project accounts and statements of expenditures. During negotiations an assurance was obtained that the additional staff required would be appointed not later than January 1, 1981. Project accounts, including those related to statements of expenditures would be maintained in the manner described in Annex 3 of the report. The estimated schedule of disbursements is presented in Annex 4. 4.09 Assurances were obtained that DRC, ASD and RRISL would keep separate accounts of expenditures made under the project irrespective of whether these would be eligible for IDA disbursements. These accounts would be audited annually for each fiscal year in accordance with sound auditing principles consistently applied. Accounts would be submitted for audit within four months 1/ of the end of the financial year, audited not later than ten months 1/ after the end of such year, and the audit report submitted to IDA within twelve months. Statements of expenditure would be audited at least once every six months and the reports furnished to IDA. Such reports would show, inter alia, that the funds withdrawn were used for the purposes intended, that funds have been received or work performed, and that payments have been made. V. ORGANIZATION AND MANAGEMENT A. Background 5.01 Two ministries within GOSL are directly concerned with the rubber industry. The Ministry of Plantation Industries (MPI) is responsible for 1/ These requirements follow GOSL standing regulations for the routine audit of government agencies' accounts.  - 30 - all aspects other than trading, while the Ministry, of Trade and Shipping regulates and participates in rubber trading. A third ministry, Industries, is indirectly involved with rubber through its responsibility for industrial development which includes rubber processing. ThE. key ministry for the rubber industry is MPI. Agencies responsible to MPI include the Department of Rubber Control (DRC), the Rubber Research Institute of Sri Lanka (RRISL) and its subordinate Advisory Services Department (ASD), and the recently established State Rubber Manufacturing Corporation (SRMC) which implements the processing development program of GOSL. The State Plantation Corporation (SPC) and the Janatha Estates Development Board (JEDB), formerly under MPI, have now become two separate ministries and together operate public sector estates. Details of the operations of DRC, ASD and RRISL have been presented previously (para 2.11 et seq). 5.02 The Ministry of Trade and Shipping regulates the export trade through its Commoclity Purchase Department (CPD). CPD exercises a monopoly on the marketing of RSS grades 1, 2 and 3 (para 2.20), licenses shippers,issues export permits, and supervises the grading of shipments. 5.03 A fourth ministry, Finance and Planning, is peripherally involved with the rubber industry through its financing functions which include the collection of cesses and duties on rubber exports by the Department of Customs. 5.04 The present system of management of the rubber industry has a major weakness in the area of policy formulation and planning. Although these are responsibilities of MPI, it does not have the capability to carry out these key functions effe!ctively with the result that there has been no consistent policy for rubber development. A further weakness related to the policy area, is lack of clearly defined responsibilities for the concerned institutions with some areas of overlap in both regulatory and executive functions. This latter weakness suggests the need for some changes in the industry's organizational structure. 5.05 The Masterplan has proposed extensive reorganization of the industry. The report was released in December 1979 and its recommendations are currently under consideraticn by GOSL. The Masterplan has proposed that a Rubber Indus- try Development Authority (RIDA) be established within MPI with responsibility for all aspects of the industry, including planning. It proposes that RIDA operate through two executive boards; one for research and development, the other for rubber manufacturing and marketing. RIDA would be headed by a Director-General with executive power over all agencies within RIDA. He woulc be responsible to and guided by a Board with members drawn widely from the industry and the concerned ministries. 5.06 The Masterplan was originally scheduled for completion by mid 1979 and its delayed release has meant that the overall organizational structure of the rubber industry has yet to be formulated. IDA proposes to continue to discuss this aspect with GOSL and has meanwhile indicated to GOSL what it regards as the critical elements of the overall organization. These critical elements comprise a minimum package of organizationial changes which would be required in any overall organizational restructuring and which at the same time, would be sufficient to ensure successful implementation of the project. They consist of the establishment of a Rubber Policy and Planning Unit (RPPU), a Rubber Policy Review Committee (RPRC), and a Project Coordination Unit (PCU),  - 31 - and the separation of ASD from RRISL. MPI has agreed in principle with these changes which are in general conformity with the Masterplan proposals. B. Project Organization and Management 5.07 The project would be implemented by three existing agencies: ASD, DRC and RRISL. Overall coordination of the project would be the responsibility of a Project Coordination Unit to be created under the project. Project evaluation would be carried out by ARTI (para 3.32). The overall project organization is presented graphically in Annex 4 Chart 2. Project Coordination Unit (PCU) 5.08 PCU would be established within MPI and located in the MPI offices in Colombo. 5.09 Responsibilities. The PCU would coordinate all aspects of project implementation and would serve as the contact point between the implementing agencies and IDA. Its specific responsibilities would include: (a) assistance to ASD and DRC in the preparation of annual work plans and budgets for project activities. These would be prepared at least six months prior to the start of each financial year; (b) assistance to ASD and DRC in project procurement including preparation of tender documents, issue of tenders, evalu- ation of bids and recommendations for awards; (c) design of progress reporting procedures for ASD, DRC and RRISL in conjunction with staff of these agencies; (d) monitoring of project implementation through formal (progress report review) and informal means (field visits and personal contacts); (e) preparation and submission of applications for with- drawal from the proposed credit on behalf of ASD, DRC and RRISL; (f) collation of progress reports and project accounts from ASD, DRC and RRISL, and preparation and submission to IDA of Quarterly Progress Reports and Annual Project Accounts and; (g) liaison with the Agrarian Research and Training Insti- tute to coordinate and facilitate the evaluation study. 5.10 Staffing. PCU would have a professional staff of three with nine support staff (Annex 4, Table 6). It would be headed by a Project Coordinator (PC) who would be responsible to and report to the Chairman of the Steering Committee or its successor, the Rubber Policy Review Committee (RPRC) following its formation. The RPRC would provide the forum for decisions on project implementation. A satisfactory Project Coordinator has already been designated  - 32 - and agreement is being reached on his terms of reference. An agreement was obtained during negotiations that the position of Project Coordinator would at all times be filled by a person whose qualifications and experience were satisfactory to IDA. Advisory Services Department (ASD) 5.11 ASD would be transferred from RRISL and would become a department in its own right. During negotiations an assurance was obtained that ASD would be established as a separate department and fully staffed not later than January 1, 1981. 5.12 Responsibilities of ASD would be revised under the project and would comprise: (a) provision of advice to smallholders and estates on all aspects of rubber husbandry, exploitation practices, disease control, and the transmission of research findings; (b) provision of technical and managerial advice to RSS producers comprising both private processors and group processing centers; (c) all field level contacts with smallholder replanters including inspection of holdings for issue of replanting permits, inspection of replantings for authorization of replanting payments and inputs, and distribution of replanting inputs; and (d) systematic quadrennial field checks on rubber holding registration details to update the industry records on extent, age, and condition of rubber on smallholdings. Details of the monitoring system ASD would use are given in Annex 1. 5.13 Staffing. To enable it to carry out these activities, ASD would be reorganized and its staff increased substantial:Ly. The proposed internal organization of ASD is indicated in Annex 4 Chart 3. ASD would be headed by a Director who would be assisted by a Deputy Director for Field Services and a Deputy Director for Administration and Training. Field Services would comprise three divisions: Extension, dealing with replanting and general advisory work; Processing, dealing with advice to processors; and Supplies, dealing with the distribution of replanting inputs to smallholders. The Training Division would be located at the training center. Administrative staff would be located at ASD headquarters in Colombo and would include a new accounts section. ASD professional field staff serving the project area would be increased from 47 to 139 by recruitment and promotion over a period of four years. Head office professional staff would be increased from 8 to 15 includ- ing the staff of the Training Division. Details of incremental staff are presented in Annex 4, Table 2.  - 33 - 5.14 Staff of the Extension Division would be increased threefold. The three new RAOs would be either experienced graduates or senior field staff without university qualifications and the three new DREOs would be promoted from REO ranks. The existing cadre of 41 REOs would be expanded to 54 by recruitment of high school graduates with GCE A level qualifications. The position of REO is eagerly sought after and there would be no difficulty selecting suitable staff. The bulk of the incremental staff in ASD would comprise 108 Extension Overseers (EO) who would be locally recruited. The qualifications for this new cadre would be high school graduate with GCE 0 level. Competition for these positions is expected to be intense. 5.15 The Processing Advisory Division (PAD) would be established within ASD under the project. Its staff would comprise three Senior Processing Advisors (SPA) located at the regional offices, and nine Processing Advisors (PA) who would be located in the field. Each SPA would supervise all PAs in one region and would assist them in field activities. The SPAs and PAs would be recruited in project year 1 and would participate in a one month training course to be given at the ASD training center early in year 2 after the more urgent training of new ASD recruits had been completed. All recruits would be required to have had extensive experience in RSS production and should pre- ferably be experienced in bookkeeping and accounting. Academic qualifications of SPAs and PAs would be the same as for REOs. It is envisaged that some of these positions would be filled from DREO/REO ranks and the rest by recruit- ment outside ASD. The Supplies Division (SD) would be a new division within ASD with responsibility for supply of replanting inputs to smallholders. Its head would be located at the head office and he would operate through the RAOs at regional offices. The Training Division would also be established within ASD under the project. It would have a professional staff of seven including the publicity unit and its staff would operate from the training center. The Administrative Division would comprise an administrative officer, an accoun- tant, and 14 clerks, and would include a new accounting unit. 5.16 Training. The project would support a greatly expanded training pro- gram for ASD staff which is summarized in Annex 4, Table 3. All formal train- ing courses would be conducted at the ASD training center under the supervision of the Training Division. Teaching staff would also be drawn from RRISL, DRC and SPC/JEDB as required. Newly recruited REOs would undergo a three month induction course and EOs a two month induction course. Refresher courses for existing REOs would last one to two weeks. Courses would also be conducted in rubber processing: the new cadre of processing advisors would be trained in year 2 and regular courses conducted twice yearly from year 2 onwards for private sector processors and GPC members. These regular courses would cover both the technical aspects of processing, and the managerial and financial aspects of GPC operations. The training courses in processing would be designed by senior ASD staff working in conjunction with the processing con- sultant. A consultant would assist ASD in developing curricula for the various courses. Two senior ASD staff would participate in a six week pre-project study tour to Malaysia, Thailand and Indonesia, to review the operation of replanting programs in those countries. Two officers from DRC and the Project Coordinator would also participate in the study tour. Three RAO's from ASD would join three Assistant Controllers from DRC for a three week study tour to Malaysia in year 1. From year 2 onwards, three selected ASD field staff  - 34 - would visit one country to be decided for a three week study tour. The project would also provide two fellowships for postgraduat:e studies for ASD staff. 5.17 Funding. ASD would be funded by Budgetary allocation following normal GOSL procedures as outlined in Annex 3. Department of Rubber Control (DRC) 5.18 DRC would be strengthened under the projlect and would revise some of its procedures for the administration of the replanting scheme. These revised procedures would be necessary to continue its decentralization to regional offices, and to increase its processing efficiency to enable it to service increased numbers of replanters. DRC would establish three regional offices in the project area which would also accomodate the regional staff of ASD, to improve inter-agency coordination. One regional office has already been established in Kalutara, and the other two offices would be occupied as soon as construction is completed. 5.19 Responsibilities. DRC would relinquish t:o ASD its present responsi- bilities for inspection of holdings, authorizatioa of replanting payments, and distribution oDf replanting inputs. It would also hand over its respon- sibility for production of planting material to SFIC/JEDB. Its previous responsibility for setting replanting payments anc, cess levels, would be assigned to the Rubber Policy and Planning Unit in MPI (para 5.24). An agree- ment on a satisfactory timetable for the transfer of these responsibilities has been obtained. Within the project area, DRC operations would be fully decentralized with one regional office servicing each region. The regional offices would register holdings, issue replanting permits and make cash pay- ments on receipt of authorization from ASD. DRC would continue to collect and maintain industry statistics relating to rubbe!r area, production, process- ing and marketing. DRC would estimate replanting input requirements and procure these inputs. It would retain its inspectors to verify a sample of the inspections carried out by ASD officers. 5.20 Procedures. DRC would develop simpler procedures for the adminis- tration of the replanting program (Annex 2, Appendix 1). These procedures would be fully developed under the project after DRC staff had studied pro- cedures in Malaysia and Thailand. A consultant would assist in the develop- ment of such improved procedures (para 3.34). It is envisaged that DRC would reach a decision on procedural changes by end PYI and would implement these changes gradually over the subsequent years of the project. In the interven- ing period, DRC would continue to follow its present procedures but the staff- ing proposals outlined below would enable it to speed up its handling of replanting assistance. The Masterplan includes de!tailed recommendations for procedural and organizational changes within DRC. It has proposed that future DRC operations makie extensive use of a computer to speed up its work but it has also proposed an alternative manual system. Ihe UK Government is expected to provide additional assistance, 1/ to DRC to further improve its capacity to I/ Such assistance is expected to include computer facilities and technical assistance.  - 35 - service replanters and to maintain industry statistics. There is a clear need to ensure conformity between the responsibilities of DRC under the project, and those that may be envisaged for DRC under the anticipated UK- financed project referred to in the Masterplan. Therefore, during nego- tiations, an agreement was obtained that GOSL would consult with IDA concern- ing any proposed changes to its procedures for the administration of the replanting scheme. 5.21 Staffing. Despite the reduction in its responsibilities, DRC would require some additional staff to reduce the time taken to process the paperwork for replanting assistance. The total personnel required to staff the regional offices would be 162, of whom all but 11 would be clerks or support staff. The staffing has been based on work norms suggested by senior DRC officers and would be subject to revision after details of improved administrative proced- ures are decided. Staff requirements would be reduced if DRC proceeds with a proposal to use a computer-based data retrieval system. Details of the staff requirements are presented in Annex 4, Table 5. 5.22 Training. Two senior DRC staff would participate in a pre-project study tour to Malaysia, Thailand and Indonesia, to review the operation of replanting programs in those countries. In addition, the three Assistant Controllers in charge of the regional offices would study the administrative procedures of one national replanting program for a period of three weeks in year 1. All other training of new and existing staff would be carried out on the job in Colombo and the regional offices. 5.23 Funding. In the past, DRC administrative costs have been met from a number of cesses with resulting complications in maintaining accounts which allocate costs to the appropriate cess. The increased workload of DRC under the project would result in increased administrative costs necessitating increases in the cesses. However, since DRC's operations on behalf of rubber producers are comparable to those of ASD which are funded by budgetary alloca- tion, it has been proposed that DRC operations would in future also be funded by budgetary allocaton from MPI, and only replanting payments themselves be met from the replanting cess. Rubber Policy and Planning Unit (RPPU) 5.24 RPPU would be established within MPI to develop appropriate policies for the rubber industry and to plan their implementation. It is envisaged that RPPU would have a staff of about 10 comprising a senior economist, an econome- trician, two graduate assistants and the staff of the Economic Research Unit currently under RRISL. In addition, RPPU would co-opt the services of special- ists in particular areas requiring technical knowledge beyond that of its staff. 5.25 RPPU would serve in an advisory role and would submit its policy and planning recommendations via the Secretary, MPI to the Rubber Policy Review Committee whose compostion and functions are described below (para 5.26). The specific concerns of RPPU would include the following:  - 36 - (a) Replanting Targets -- these would be fonnulated by RPPU based on the age structure of the national stand, exploit- ation rates, and projected demand prospects. Data for this analysis would be drawn from DRC registration records, aerial surveys, and specific studies as required; (b) Replanting Payment Levels -- the level of replanting payments would be adjusted in future according to movements in the costs of labor and material inputs which RPPU would monitor; (c) Producer Margins -- RPPU would monitor the costs of produc- tion of rubber for producers of differing holding sizes in order to recommend appropriate levels of producer margins and thus to define the surplus available for export duties, taxes and cesses; (d) Processing Capacity -- RPPU would review market prospects for alternative forms of processed rubber and recommend levels of processing capacity appropriate to such prospects. These recommendations would form the basis on which licences would be issued for capacity expansion. RPPU would also recommend what level of incentives, if arLy, would be required to induce private enterprise to expand processing capacity; and (e) International Organizations -- RPPU would represent Sri Lanka in its dealings with the Association of Natural Rubber Producing Countries (ANRPC), the International Rubber Study Group (IRSG) and other similar organizations. An assurance was obtained during negotiations that a Rubber Policy and Planning Unit would be established within MPI not later than January 1, 1981, with staffing and terms of reference satisfactory to IDA which would include, inter alia, responsibility for advising the RPRC on levels of industry cesses, duties, and replanting payments. Rubber Policy Review Committee (RPRC) 5.26 RPRC would review all policy proposals produced by RPPU. Its function would be to ensure that policy recommendations to the Minister MPI reflected adequately both the views of the various interests in the rubber industry and those of the three other ministries concerned with rubber. The membership of RPRC would thus necessarily be broad. The existing Steering Committee, established within MPI to coordinate the preparation of this pro- posed project and to guide the Masterplan team, would form the basis of RPRC. Additional members would be drawn from the other ministries and from rubber producers and processors. The proposed composition of RPRC is indicated in Chart 2 of Annex 4. 5.27 RPRC would meet as required to review policy proposals from RPPU. Those proposals endorsed by the Committee would be conveyed in the form of  - 37 - recommendations to the Minister MPI, by the Chairman. Proposals not receiv- ing the Committee's endorsement would be either submitted to RPPU with sug- gested amendments, or passed on to the Minister for consideration together with a report of the Committee's views. The Committee would also review the proposed workplan of RPPU and suggest areas worthy of detailed analysis. A further function of the RPRC would be to receive reports from the PCU and to reach agreement on details of project implementation. At the discretion of the Chairman, this latter function would be carried out by a working group comprising those members of RPRC directly involved in project implementation. During negotiations an assurance was obtained that a Rubber Policy Review Committee would be established within MPI with powers and composition accept- able to IDA, not later than January 1, 1981. Ombudsmen 5.28 The ombudsmen proposed under the project would be appointed by the Secretary MPI. It is envisaged that the ombudsmen would be respected local officials such as district court judges and that they would devote up to two days per month reviewing written or verbal complaints from smallholders con- cerning the operation of the replanting scheme. The ombudsmen would receive an honorarium from MPI for their work. One ombudsman would be appointed in each district for a three year term. Smallholders would be advised of the services available from the ombudsman and each leaflet produced by ASD and DRC would include a statement recommending that smallholders with complaints concerning the services of ASD or DRC should contact their local ombudsman for action. C. Monitoring and Evaluation 5.29 The project would strengthen existing internal monitoring systems in both ASD and DRC and would support the development of a project-specific monitoring system within PCU. Details are given in Annex 1. 5.30. The project's achievements would be evaluated throughout the project period by ARTI. The evaluation would be carried out from the viewpoint of the smallholder and would seek to identify any potential changes in the replanting scheme which would make it more attractive to smallholders. It would also seek to identify any means whereby the work of RRISL could be modified to better service smallholders. Further details of the evaluation component are given in Chapter III (para 3.32 et seq) and in the terms of reference for the evalua- tion consultant (Annex 2, Appendix 4).  - 38 - VI. PRODUCTION ASPECTS A. The Present Situation 6.01 About 30% of smallholder rubber is either currently over-aged or will be due for replanting within five years. 1/ 'While the average yield of smallholder rubber is fairly satisfactory (750 kg/ha 1/), the yield from aged and over-aged smallholder rubber is very low (100-500 kg/ha 1/). 2/ Much of the old rubber is unselected seedling material with low genetic yield potential. Taken over all age groups, about 70% is budgrafted with relatively high yielding clones (of which PB86 accounts for more than 70%); 20% is seedling material, and 10% clonal seedling (mainly Tj 1). Stands are high (about 440 trees/ha) reflecting both high planting densities and low losses due to disease or wind damage. Runts are common since thinning is not normally practiced. Some 70% of smallholder rubber has beea assessed as acceptable or better in respect to evenness, vigor and weed control. Ground covers are generally good with adequate leguminous covers established at planting and satisfactorily maintained. Soil conservation works are also generally ade- quate; these usually take the form of small platforms at the planting points rather than continuous terraces. Fertilizer is rarely applied by smallholders except during the immature period. Even then, only 20% of the recommended fertilizer is being used. Disease incidence is geaerally low: oidium out- breaks occur in some years, particularly at higher elevations; panel diseases are uncommon despite high rainfall; white root disease is however, common, particularly on estates. 6.02 Exploitation standards are variable and could be considerably improved. Smallholders normally tap daily but not continuously since tapping is carried out according to convenience rather than system. Thus there is no tapping during festivals, paddy planting, fruit harvesting or wet days. Tap- ping may be discontinued during periods of low rubber prices. On average, smallholders tap about 180 days per year. Although high level tapping of older trees is common, it is usually of poor standard with excessive bark consumption. Unplanned slaughter tapping causes considerable loss of potential crop and has hastened the need for replanting. Poor exploitation practices, including premature opening, reduces the life of many plantings by as much as one third to about 20 years. 1/ Estimates based on surveys conducted as part of the Rubber Masterplan Study, 1979. 2/ The average yield for smallholder rubber in Malaysia is about 950 kg/ha.  - 39 - B. Project Replantings: Inputs and Yields 6.03 The projected yield from replanted rubber would average about 1,060 kg/ha over the 24 years of tapping with a peak yield of 1,350 kg/ha. The aver- age yield thus represents a 400% increase of the present yield of over-aged rubber which itself could only be sustained for about five years. Complete details of replanting inputs and projected yields are presented in Annex 1. The major points concerning replanting procedures are described below. 6.04 Planting Materials. All replanters would be provided with bare root brown budded stumps produced by selected SPC/JEDB estates. The cost of production is currently Rs 1.85 per budded stump and the ex-estate price has been set at Rs 2.50 per stump for 1981. DRC and ASD have selected 33 estates for planting material production based on the likely geographical distribution of replanting: the locations are indicated on Map 14733R. Nurseries for 1981 replantings have already been satisfactorily established. An agreement has been obtained that the terms and conditions of the contract for the production of planting material by SPC and JEDB would be satisfactory to IDA. A draft outline of contract conditions for planting material supply has been provided to MPI. 6.05 The clones used for project replantings would be those recommended by RRISL. At present, these are PB 86, RRIC 100 and RRIC 103. 1/ Clone PB 86 has been extensively planted and, although not among the highest yielders in major rubber producing countries, has proved successful under conditions in Sri Lanka. Clones RRIC 100 and 103 have been tested, but are not yet planted extensively. It is probable that there would be a considerable smallholder preference for PB 86. It is strongly recommended that RRISL should consider using GT1 for part of the project plantings. The Rubber Masterplan has also recommended the gradual introduction of higher yielding varieties and these suggestions are being considered by GOSL. A conservative projected yield profile for project replanting is presented in Annex 1, Table 2. RRIC 130 is currently being tested and provided its potential by RRISL is proven, it would be recommended for use by smallholders. 6.06 Land Preparation and Planting. All felling, uprooting and sawing of old rubber trees would be manual. The only tools used would be hand winches and cross-cut saws. Wherever possible, timber would be sold for firewood or higher use purposes, 2/ otherwise it would be windrowed and burned. Sufficient branch timber would be retained to obtain an adequate burn of the roots and a reburn would be necessary. White Root disease is an increasing problem in replanting, and every effort would be made to remove and destroy all sources of nutrients for it by following prescribed uprooting practices. 6.07 Planting distances would be 7.3 m x 3.6 m (24 x 12 feet) on flat land and 9.1 m x 2.4 m (30 x 8 feet) on slopes over 80 or where intercrops are 1/ RRIC refers to Rubber Research Institute of Ceylon. 2/ If properly and promptly treated to resist decay, rubber wood is a useful lumber for furniture, flooring and structural purposes. The project would endeavor to expand this use of old rubber trees by notifying potential users of the availability of supplies.  - 40 - planted. All replantings on slopes of over 8 0, wou]d be either on rehabili- tated terraces or on platforms; both would be aligned on the contour. Drains would be dug between the rows. The location and frequency of these and other soil conservation works would be decided by the REO during lining. Cover crops would be established as early as possible to prevent erosion; if burning permits, covers would be established in the pre-plarting period. Pueraria phaseloides and Desmodium sp. are currently used and would be recommended under the project. Covers would be established using cuttings and/or fresh seed, and establishraent aided by the addition of rock phosphate. 6.08 Although planting can be carried out in both May/June and October/ November, the May/June plantings are preferable since the risk of subsequent drought is much lower than later in the year. The project would therefore aim to complete all plantings in June although inevitably some plantings would be delayed until October/November. Supplies (11%) wDuld be made from budded stumps planted in trenches on smallholdings at the time of the main planting. 6.09 Upkeep to Maturity. ASD officers would ensure that the RRISL recom- mendations for disease control, weeding frequency, cover crop and soil conser- vation measures would be followed. Details of the operations to be carried out for the life of the plantation are given in Annex 1, Table 1. Fertilizer would be distributed twice per year, but ASD officers would encourage smallholders to retain half their allocations to make four applications in the planting and second years; thereafter, two applications per year would be satisfactory. 6.10 Inspection Procedures. ASD field officers (REOs and EOs) would conduct regular inspections of replantings as a basis for authorization of replanting payments. During the seven year period from uprooting to opening, eight standard inspections would be carried out; additional inspections would be required for those replantings failing to meet prescribed standards. Re- planting payments would be made to smallholders by DRLC following satisfactory inspection. The inspections would ensure that prescribed replanting procedures had been followed. Further details of these procedures and standards are presented in Annex 1. An agreement was obtained that: not later than January 1, 1981, DRC and ASD would jointly issue a revised Advisory Leaflet for small- holders detailing the revised procedures to be followed for replanting under the project which wouald be satisfactory to IDA. 6.11 Exploitation. Tapping of replanted rubber would commence in the seventh year after planting and all trees would be brought into tapping by the eighth year. The suggested tapping system would be alternate daily although some smallholders would continue to tap daily. Yield stimulants would be used only for the last six years of exploitation. Further details of exploita- tion practices are given in Annex 1. C. Intercropping 6.12 The project: would encourage the cultivation of intercrops together with rubber in those areas technically suited to intercropping. The major risks of intercropping are that it would compete with immature rubber for  - 41 - nutrients, light, labor and capital and increase soil erosion. To minimize these risks, intercropping under the project would be carefully supervised and permitted only for certain crops which would be adequately fertilized and cultivated so as to minimize soil erosion. 6.13 The intercrops to be encouraged under the project would have good market prospects, give a quick return, be shade tolerant, non-competitive, and require minimal soil disturbance. In addition, they would be crops which RRISL had researched and satisfied itself as to their potential. All rubber to be intercropped would be spaced at 9.1 x 2.4 meters (30 x 8 feet) which is standard for the sloping land generally found in the project area. ASD officers would indicate the intercrop most appropriate to the smallholder's circumstances although the final choice of intercrop would rest with the smallholder. Bananas could be cultivated almost universally throughout the project area and are expected to be the most common intercrop. A single row would be planted at the center of the inter-row with 3.65 m spacing within the row to give 296 plants/ha. The two favored cultivars would be edible types equivalent to Gros Michel (synonym Ambon) and Cavendish. Planting material (suckers) are readily available locally. Smallholders would establish a nursery in the year prior to planting equivalent to 10% of the final area, to reduce the cost of planting material. ASD staff would ensure that small- holders applied recommended levels of fertilizer (15:15:5) and controlled weeds to minimize competition. Banana is subject to few diseases although there is some Panama disease and bunchy top and a possibility of nematode build-up after four years cultivation. No plant protection chemicals would be used. Details of operations, inputs, yields are given in Annex 1, Table 3. Annex 4 Table 16 indicates the financial returns to smallholders. Details of operations, inputs, yields and returns for alternative intercrops (pineapples and passionfruit) are available in the Project File. For the purpose of economic analysis, it has been conservatively assumed that less than one third of smallholders would grow intercrops on an area equivalent to 11% of the total area of holdings. VII. DEMAND, PRICES, MARKETING, FINANCIAL ANALYSIS AND COST RECOVERY Demand, Prices and Marketing 7.01 Demand. The demand outlook for elastomers, including both natural and synthetic rubber is bright, and the competitive position of natural rubber compared with synthetic rubber is now favorable. It has been strengthened by recent technical and economic developments, such as the strong market penetra- tion of radial tires that require more natural rubber, and the improvement in the price competitiveness of natural rubber compared to petroleum-based products. Data on projected production from the world's existing area of rubber and on the planting intentions of producing countries, indicate a future decline in the rate of expansion of natural rubber output. Unless investment in natural rubber production increases within the next few years, growth in consumption of natural elastomers will be severely constrained by supply. If this supply shortfall is not filled by natural rubber, the synthetic rubber industry will expand its capacity to produce polyisoprene rubber instead. The proposed project investments are thus most timely and Sri Lanka would have no  - 42 - difficulty in exporting the additional project output. This output of about 25,000 tons/year would represent some 5% of the incremental world demand. 1/ 7.02 Prices. The rubber prices used for financial and economic analyses were derived from the most recent forecast for RSS1 prices in 1990, and used throughout the project period. Even though the world market demand for elastomers is highly insensitive to price changes, the incremental project output will not affect world market prices due to its small size relative to total world rubber production. The incremental project output represents only 5% of the projected increase in world demand and only 0.4% of the 1980 pro- duction of natural rubber (or alternatively 0.13% of the total 1980 elastomer production). Consequently, the possible depressive effect on (projected) world market prices, due to the project would be insignificant. In the analysis the economic price for rubber is the world price for RSS1 corrected for freight costs and quality differentials and expressed as the FOB Colombo price (Rs 19.45/kg). The farmgate financial price within the project area (Rs 7.60/kg) was estimated by deducting projected levels of cesses and duties as well as handling charges and profit from the financial FOB price. 2/ The basis for estimates of the project's financial and economic prices for rubber are presented in Annex 4, Table 14. 7.03 Price Structure. The forecast rubber price of Rs 19.45/kg FOB Colombo would be shared as follows: GOSL 55% (52.5% from export duty, and 2.5% from cesses;) intermediaries 5%; rubber producers 40% (Table 15 of Annex 4). Duties and cesses levied on rubber would thus ba quite high, in fact the highest of any ANRPC 3/ country. For example, at the price level forecast for 1990 total government deductions would be only 27% in Malaysia, and about 25% in Thailand. On the other hand, the Sri Lankan rubbar producer would benefit from a more generous package of government assistance in the form of replant- ing payments and fertilizer subsidies -- which are indirectly financed by levies on rubber. GOSL deductions of 55% of the FOB price wot'd be applied against replanting payments (22%) and rubber fertilizer subsidies (14%), leaving GOSL with 19% of which 8% would be used to administer the Rubber Replanting Scheme. Thus GOSL's surplus would be only 11% of the FOB price. In practice therefore the Sri Lankan rubber producer would receive 76% of the FOB price (40% directly, and 36% in the form of government incentives to rubber production). Provided that the replanting payments are maintained at the prevailing rate in real terms (Rs 16,050 in 1979 prices), this would constitute an appropriate package by providing adequate incentives for the producer (para 7.07), while making a substantial contribution to Govrernment General Revenue for investment in other sectors (para 7.10). 1/ On conservative assumptions, an annual supply shortfall of about 0.5 M tons is expected to develop by 1985. Source: The World Rubber Economy, Structure, Averages, Prospects: World Bank/FAO 1978. 2/ This approach reflects the price setting mechanism used by CPD (para 2.21). 3/ Association of Natural Rubber Producing Countries.  - 43 - 7.04 Rubber Marketing. The bulk of project output would be processed as at present (details in para 2.18). Under the project, a slightly increased share of the output (4-5%) would be processed by GPCs with a corresponding reduction in processing by smallholders. All project output would be marketed through existing channels (para 2.20) which provide an efficient and fair market to smallholders. As indicated, only small quantities of rubber are expected to be sold as latex to crepe or block factories. In the longer term, it is possible that latex sales would increase in response to changes in world demand for RSS compared with other forms of rubber. Such a change is too uncertain to warrant investment in latex collection centers at this stage. 7.05 Intercrop Marketing. No difficulty is foreseen in the marketing of bananas for which there is a strong local demand. The crop is traditionally grown in home gardens throughout the rubber area and there is a well estab- lished network of small middlemen who go on foot from farm to farm to contract the purchase of bunches before maturity. They later collect the bunches and transport them by hired truck to the local fair where larger middlemen from Colombo purchase their weekly supply. The rapidly expanding tourist industry in Sri Lanka, and increasing demand from the Middle East is expected to main- tain the recent trend of strong local prices. Hence, for the financial analy- sis, current prices for bananas of average grade were projected throughout the period. (Rs 20/bunch at farmgate, resold at about three times that amount in Colombo.) Colombo prices less estimated costs for transport and handling (Rs 20/bunch) were used in the economic analysis (Rs 40/bunch). 7.06 Marketing of the alternative intercrops (pineapples and passion- fruit) would present no difficulty: there is a strong demand for these fruits which are purchased by GOSL's efficient Marketing Department and by private processors. Details of marketing arrangements for pineapples and passionfruit are available in the Project File. Financial Returns to Smallholders 7.07 An analysis of financial returns from replanting was carried out to ascertain whether the proposed replanting scheme would be attractive to smallholders. Two extreme cases are presented: firstly the smallholder who uses only his family's labour, and secondly the smallholder who hires labour. It was felt that in the latter case, which applies to the larger smallholdings, the smallholder would be mainly interested in increasing his return to capital, and hence the financial internal rate of return (IRR) provides an appropriate measure of incentive. In the former case however, the smallholder does not contribute capital and the IRR is less relevant. Such smallholders would be interested in increasing annual income and the net return to family labor. The appropriate measures of incentive for such smallholders are thus the net income increase (NII) 1/ and the net return per man-day (NR/MD) at full devel- opment. Values for these indicators are presented in Annex 4, Table 16. The 1/ NII = Present Value of Incremental Net Income Present value of Without Project Net Income It represents the increase in the smallholder's average annual income as a result of the investment, and is expressed in percentage terms.  - 44 - results indicate that, without intercrops or financial assistance, the proposed investment has a marginal NII of 18% and an unsatisfactory IRR of 11% and hence would provide little incentive to either category of smallholder. How- everD with financial assistance in the form of replanting payments, replanting would be attractive with a corresponding NII and IRR of 35% and 32%, respec- tively. 7.08 Replanting Payments and Smallholder's Cash Flow. Smallholders parti- cipating in the project would receive replanting payments to offset the costs of replanting. These replanting payments comprise both the supply of inputs in kind (planting material and fertilizer) and payments in cash. The cash payments provide for the purchase of inputs not supplied in kind including payment for labor inputs. For smallholders using only family labor, part of the cash payments represents compensation for their own labor. The replanting payments constitute an advance to replanters since they are fully recovered through taxes on rubber exports (para 7.10). The replanting payments would be spread over eight installments as shown in Annex 4, Table 18. (A comparison with actual replanting costs is presented in Annex 4 Table 10) 7.09 These replanting payments, totalling Rs 16,050, would provide an adequate cash flow for the smallholder throughout the immature period as shown in Annex 4, Table 18. They would fully compensate the cash deficit 1/ of a smallholder using only family labor (Case I of Table 18). Smallholders able to sell their rubberwood, would earn a net surplus of about Rs 2,500 per ha, which would provide an additional incentive to replant. Smallholders who intercropped their immature rubber (Case II) would earn incomes in excess of their pre-project income during the immature period. Intercropping would thus provide a further incentive to replant. The larger smallholder who uses all hired labor on his rubber plantation (Case III) would receive identical replant- ing payments. This is reasonable in view of the administrative difficulties involved in providing differential replanting payments, the lack of long-term credit for investments in agriculture, the marginal return on replanting with- out the replanting assistance (IRR of 11%), and in view of past low replanting rates among this category of producers in times of low replanting payments. Furthermore, some 57% of the replanting payment would comprise payment for labor which would be largely drawn from the under-privileged landless group. Finally, the overall cost recovery by Government wDuld be most satisfactory, even excluding additional revenue from increased income tax levied on the larger smallholders. Financial Return to Government and Cost Recovery 7.10 While providing smallholders with adequate remuneration and incen- tives, the replanting scheme would also provide Government with an adequate financial return. Government contribution to the replanting scheme takes the form of replanting payments, administrative overheads and fertilizer subsidies. 1/ Defined here as cash outlay on inputs plus inzome foregone from uprooted old rubber.  - 45 - Its revenues from the scheme include cesses and export duties. An illustra- tive Government Project Cash Flow is shown at Annex 4, Table 19. High replant- ing payments and overhead costs, plus loss of revenue from uprooting would result in a cumulative deficit of Rs 350 M (US$22.6 M) over the project implementation period. Government expenditure on fertilizer subsidies would further increase the total cumulative deficit to Rs 391 M (US$25.2) by PY6. An additional Rs 299 M (US$19.3 M) would be required between PY7 and PYIO to bring all project plantings into production, resulting in a peak cumulative deficit of Rs 690 M (US$44.5 M) in PY10. Beyond PY10 there would be annual surpluses in net GOSL revenue and all project costs would have been recovered (without interest) by PY15 when the cash flow shows a cumulative surplus of Rs 0.35 M (US$0.2M). Annual surpluses from that year on would progressively increase from Rs 200 M (US$12.9 M) to Rs 251 M (US$16.2 M) in PY20 when peak yields would have been reached. The financial rate of return to Government from the project is estimated to be about 14%. The project cost recovery index (CRI) 1/ at the estimated opportunity cost of capital (12%) would be 1.10. The CRI for replanting payments to smallholders would be 2.2 (i.e., excluding administrative overheads and subsidies). 7.11 Although the replanting cess is earmarked to finance the national replanting program, cess revenue would contribute only part of the revenue required to meet all replanting payments as they became due. As indicated in Annex 4 Table 11, cess revenue would meet 71% of total replanting payments in PY1 but only 36% in PY6. The balance would come from the export duty which is credited to Government general revenue. If the replanting cess were to fully meet replanting payments as they became due, thus maintaining the liquidity of the Replanting Fund, it would have to be set at a level of about Rs 0.80/kg 2/ (see paras 4.04 and 4.05). VIII. BENEFITS AND JUSTIFICATION Project Benefits 8.01 The project's direct benefits would result from increased production and improved quality of rubber most of which would be exported thus improving Sri Lanka's balance of payments. The benefits would take the form of addi- tional Government revenue from export taxes levied on the project's rubber output and increased earnings for some 33,000 smallholders, half of whom would be drawn from the poverty income target group. The project would also streng- then the national institutions responsible for smallholder rubber development, thus yielding benefits much beyond the immediate project. 8.02 At full development in 1999, successful project implementation would lead to incremental annual production of about 25,000 tons of rubber, a 15% increase over Sri Lanka's current annual production. At projected world market prices, the gross value of the annual incremental output would decline 1/ CRI = NPV Government Revenue NPV Government Expenditures 2/ In 1979 dollars.  - 46 - from a peak of US$30 M 1/ in 1999 to about US$23 bN 1/ at the end of the project life in 2009. Allowing for the added imports of fertilizers and agro- chemicals which would be necessary to obtain the projected yields, the project would represent a net annual foreign exchange savings of about US$28 M 1/ in 1999. The processing component would yield additional foreign exchange earn- ings of about US$0.25 M by 1990, US$0.70 M in 1999, decreasing to US$0.6 M by the end of the project life. Intercrops would provide an additional US$1.6 M 1/ annua'lly between PY2 and PY9 as a result of the increased avail- ability of fruits which would make it possible to release part of the con- strained domestic output for exports and for sale within Sri Lanka to the local tourism industry. In terms of net present value at the estimated opportunity cost of capital (12%), total net foreign exchange savings from rubber production alone would be about US$30 M ovEr project life, or approx- imately 6 times the foreign exchange cost. These foreign exchange benefits would, however, bes slow to materialize due to the long gestation period of rubber trees. External assistance support to GOSL's balance of payment in the medium term would therefore be warranted. Project Beneficiaries. 8.03 Some 27,000 smallholders would benefit f'rom the project by replant- ing. Another 6,000 smallholders would benefit from the processing component of the project. 'In an effort to determine project impact on the target group (income below US$76 per capita per year or Rs 1,178), pre-project household income levels of project participants were estimated using results of previous surveys. Data remain uncertain since household incomes were influenced markedly by off-farm earnings for which only average figures were available. Mission estimates of per capita income for the four representative holdings of 0.4, 0.8, 2.6 and 12 ha are shown in Annex 4 Table 20. It shows that pre-project household income of the two smaller holdings is below the poverty income level. Given that these fEigures are averages, one can expect a proportion of these smallholders to be absentee owners with possibly substantial off-farm earnings. Hence, to be conservative, it has been assumed that only 80% of the 0.4 ha and only 70% of the 0.8 ha project participants would be drawn from the target group. Under these assumptions, at least 47% of direct project beneficiaries would belong to the target group and would share 26% of total project bene- fits. 2/ At present the income distribution in Sri Lanka is such that the target group (lowest 22% income group) receives some 10% of total private income. Thus, by allocating 26% of total project benefit to this group, the project will have a positive impact on income distribution. At full develop- ment, the total income of the first category of smallholders would increase by more than 50% on average but would still be below the poverty income level. The incomes of the next larger smallholders (0.8 ha) would increase 70% which would bring them above the poverty income level. These findings are sum- marized in Annex 4 Table 20 which indicates that as a result of the project, the proportion of project participants within the poverty income group would decrease from 47% to about 18%. 1/ In 1979 dollars. 2/ This analysis has been restricted to beneficiaries of the replanting component and does not inc'Lude some 6,000 smallholders expected to bene- fit marginally through improved processing services under the project.  - 47 - Employment 8.04 Under the assumption that without the project, plantations of already over-aged rubber would be abandoned after five years, the project would result in a net increase in labor demand of 5 million workdays/year at full develop- ment, or approximately 19,000 full-time jobs. In practice, however, it is likely that some of this land would be replanted using poor cultural techniques and low yielding seedlings even without assistance from a replanting scheme. The project would therefore provide fewer jobs than indicated above. Assuming that half the project rubber area would have been replanted anyway, the project would provide an additional 9,500 full-time jobs. The project's impact on labor productivity would be more certain: the net return to labor would increase fourfold to reach Rs 32 at full maturity (15 years after planting). Since many hired tappers are paid on the basis of output, it can be expected that part of the incremental earnings from rubber would be passed on to these laborers, most of whom would certainly belong to the poverty income group. Thus, the project would have a significant but unquantifiable impact on the incomes of many poor laborers in the project area. Environmental Effects 8.05 The project would have no adverse effects on the environment. Pro- ject replantings would generally improve ground cover and reduce soil erosion. Agro-chemical use with rubber is minimal and the chemicals are non-toxic. The small scale processing supported by the project has no associated effluent disposal problems. Economic Analysis and Sensitivity Tests 8.06 The economic analysis has been carried out for the project as a whole. Summary economic costs and benefits are presented in Annex 4, Table 21 and details are included in the Project File. The following assumptions have been used in the analysis. (a) benefits and costs are expressed in constant end-1979 prices, net of identifiable local taxes and subsidies; (b) the incremental project output of rubber and bananas is valued at forecast world prices as described in paras 7.02 and 7.05; (c) the exchange rate is taken at a constant Rs 15.5 = US$1.0, which was the prevailing official exchange rate (OER) at time of appraisal; (d) incremental farm labor is valued at a seasonally adjusted wage rate ranging from 50% to 100% of the peak market rate,  - 48 - in order to reflect the prevailing unemployment and under- employment in the project area. 1/ (e) The economic prices for traded goods (ruhber, bananas, fertil- izers and equipment) are based on their 'order prices at the OER. The domestic cost components, i.e. transport and market- ing charges, are adjusted by conversion factors (CF) respec- tively 1.11 (CF for road transport) and 0.91 (Standard CF). The specific conversion factors applied to all major non-traded inputs are indicated in Annex 4, Table 22. (f) Expected benefits from upgrading of rubber through better processing as a result of the processing component have been included; (g) Production foregone as a result of uprooting and replanting of rubber is fully costed at an average yield of 265 kg/ha for the estimated remaining productive life of the plantation (five years); (h) Full costs of the institutional support components (DRC, ASD, technical assistance, research and project administration) have been included, without allowance for residual values on buildings and equipment; and (i) The period of analysis is 30 years. Using the assumptions given above, the best estimate for the ERR is 23%, and for the NPV (at the estimated opportunity cost of capital of 12%) Rs 484 M. Sensitivity tests indicate that the project is remarkably insensitive to all important parameters and would therefore remain viable under a variety of adverse circumstances (Table 8.1). 1/ Rs 12 per day during peak periods of agricultural activity corresponding to land preparation, sowing and harvesting of the maha paddy crop (20% of project labor requirement); Rs 10 per day (luring medium peak periods corresponding to land preparation, sowing and harvesting of the yala crop (20% of project labor requirement); and Rs 6 per day at all other times (60% of project labor requirement. Based on these assumptions, an average shadow wage rate of Rs 8/day (or 66% of the market rate) was obtained.  - 49 - Table 8.1: SENSITIVITY ANALYSIS Test Cases ERR (%) NPV (Rs M) Base case 22.8 483.7 No return from processing component 22.0 448.9 No return from intercropping 19.3 383.8 No sale of rubber-wood 21.6 458.2 Farm labor at market rate 22.4 451.6 20% decrease in rubber yields or price /a 21.5 344.0 20% decrease in rubber yields and price /a 20.0 232.4 50% increase in fertilizer prices 20.4 398.4 20% increase in rubber production costs 21.3 425.7 20% increase in all overheads 22.3 473.7 20% increase in all costs 20.8 415.7 1 year lag in rubber production 21.4 398.1 2 year lag in rubber production 20.4 321.5 /a Switching Value for Rubber Yield at OCC (12%) is 69% below that projected or 420 kg/ha at full maturity instead of 1,350 kg/ha. Rubber Price switching value is Rs 5.9/kg compared to the base estimate of Rs 18.9/kg. Project Risks 8.07 The project faces no major risk. The economic rate of return is relatively insensitive to variations in the key parameters such as rubber yields, prices and production costs. The techniques and inputs for replanting and maintenance are all proven. It is possible that the rate of uptake of the GPCs may be less than projected in the project. However, even if no improve- ments in rubber quality were achieved under the project, the ERR would be reduced by less than one percentage point (Table 8.1). There is, however, a risk that the project may over-extend the implementation capacity of the concerned institutions. Should this occur, replanting targets would not be met and both benefits and costs would be less than projected. The project would, nonetheless, remain viable under such circumstances. The problem of constrained implementation capacity has been directly addressed in designing the project. Firstly, it has been accepted that some 30% of the over-aged rubber now awaiting replanting would have to be excluded from this first project. Secondly, the annual replanting targets build up slowly giving time for the institutions to recruit staff and to implement the new procedures. Thirdly, the institutions themselves would be strengthened under the project and their activities would be coordinated and monitored by PCU. Finally, some 70% of project disbursements are directly tied to actual replanting achieve- ments, providing a strong incentive for GOSL to ensure that implementation capacity does not become a constraint.  - 50 - IX. RECOMMENDATIONS 9.01 During regotiations, the following principal assurances were obtained from GOSL: (a) (i) GOSL would establish a legal basis for the operation of GPCs satisfactory to IDA not later than March 31, 1981; and (ii) no new GPCs would be established in the project area until the potential members pledged to collectively supply not less than 100 kg of latex per day of opera- tion (para 3.27); (b) GOSL would (i) provide the Replanting Fund with the resources necessary to finance deficits involved in implementing the project and sustaining the replanting program at the optimum rate of about 8,100 hectares per year; and (ii) annually review the financing requirements of the Fund to carry out the project and thereafter to continue the replanting scheme, and submtit to IDA the results of that review (para 4.05); (c) GOSL would ensure that the additional accounting staff re- quired by ASD and DRC would be appointed and posted not later than January 1, 1981 (para 4.08); (d) ASD would be established as a separate department and fully staffed not later than January 1, 1981 (para 5.11); (e) an adequately staffed Rubber Policy and Planning Unit would be established within MPI not later than January 1, 1981 with terms of reference satisfactory to GOSL and to IDA which would include inter alia responsibility for advising the RPRC on levels of industry cesses, duties and replanting payments (para 5.25); and (f) a Rubber Policy Review Committee would be established within MPI whose powers and composition would be mutually acceptable to GOSL and IDA, not later than January 1, 1981 (para 5.27). 9.02 During negotiations, the following major agreements were obtained from GOSL: (a) MPI would periodically review travel allowances in relation to operating costs and would undertake to ensure that project vehicles are fully utilized (para 3.17); (b) staff receiving post-graduate training under the project would remain with their current employer for at least the period of the bond as stipulated under the present govern- ment rules (para 3.31);  - 51 - (c) the position of Project Coordinator would at all times be filled by a person whose qualifications and experience were satisfactory to IDA (para 5.10); (d) GOSL would consult with IDA concerning any proposed changes to its procedures for the administration of the replanting scheme (para 5.20); 9.03 With the above assurances, the project would be suitable for an IDA credit of US$16.0 M on standard IDA terms. The borrower would be the Govern- ment of Sri Lanka.  - 52 - ANNEX I Page 1 SRI LANKA RUBBER REHABILITATION PROJECT Technical Aspects and Project Monitoring A. Replanting Procedures Replanting Operations and Costs 1. The replanting operations have been described in Chapter VI and their estimated costs together with further detailed are presented in Table 1. 2. Planting Materials have been described in Chapter VI (para 6.04). The project would use brown budded stumps initially although it is suggested that green budding should be further investigated by RRISL since it offers considerable time savings. 1/ The use of green budding by smallholders should also receive consideration from RRISL. To produce brown budded stumps, fresh seed would be collected in July/August, germinated and planted in seedling nurseries. Budwood nurseries would be cut back in November/December of the same year and budding carried out 14 to 16 months later. The stumps would be pulled in May about: twelve months after budding, trimmed, labeled and packed for dispatch. The budded stamps would be packed in bundles of 25 for transport to the replanting areas. 2/ 3. Fertilizer use has been described in Chapter VI (para 6.09). GOSL is currently considering the use of urea-based fertilizers for smallholders rather than those based on sulfate of ammonia. In view of the poor handling characteristic of urea, it is recommended that urea-based fertilizer be care- fully evaluated in a small but practical field trial involving smallholders, before introduction of such fertilizers nationwide. 4. Exploitation Practices have been briefly mentioned in Chapter VI (para 6.11). Table 1 provides further details. Tapping would commence in a replanted stand when 50% of the trees reach a girth of 50 cm at 90 cm above the union. ASD officers would provide stencils anc, demonstrate the marking of panels; instruction would also be given in tapping, bark conservation and pro- cessing. The system recommended would be s/2 d/2, 3/ but owners of very small 1/ Green budding may be carried out with 5-6 mont:h old seedlings which could be delivered to smallholders one month after budding. 2/ Per hectare distribution would be 494 plants; 445 for field establishment and the balance planted in trenches for later infilling of vacancies. 3/ Tapping systems are by convention indicated by the symbols s/2 (for half spiral cut) and d/2 for alternate daily tapping.  - 53 - ANNEX 1 Page 2 areas would probably still tap daily. Other owners would usually tap s/2 d/2, but after long periods of rainy weather, would recover crop by daily tapping. ASD officers would stress the need to avoid a prolonged period of daily tapping. Tapping would be stopped for a month during wintering (February). Some trees are expected to reach opening standards in the seventh year and tapping would commence in April of that year with the balance brought into production in the eighth year. Annex 1, Table 1 indicates the operations, inputs and costs from the seventh year over the normal s/2 d/2 tapping on virgin and first renewed bark to the 24th year. 5. Stimulation would only be employed in the six final years of exploit- ation. Yield stimulation during the seventh to 24th years would not be practised: the necessity for large nutrient inputs would not be understood by smallholders and under-fertilization would reduce the productive life. In addition, since smallholders commonly use family labor with few alterna- tive employment opportunities, the increased labor productivity resulting from stimulation and reduced tapping frequency has no particular appeal. Under the project, stimulant would be applied three times per year in the 25th and 26th years, and the tapping system would be amended to 2 s/2 d/3 (25th year) and 2 s/2 d/2 (26th year). In the 27th and 28th years, stimulant would be applied four times per year on a 2 s/2 d/2 system. The 2 s/2 cuts would be on opposite sides of the tree, separated by about 30 cm to avoid any girdling effect. Stimulant would be applied four times a year for the final two years with a 4 s/2 d/2 system, opening two upward cuts above the existing downward cuts. The use of stimulant for final exploitation would require controlled tapping (not what is locally termed "slaughter tapping") to obtain worthwhile results. ASD officers would offer courses and advice on the methods to be employed in specific cases. It would be advisable if RRISL were to reinvest- igate such controlled final exploitation and also the possibility of use of virgin bark with stimulant above the existing 90 cm tapping panels. With the use of a stool (ladders being impractical on steep terrain), it is possible that high yields would be obtained. Yields 6. The yield projections for replanted rubber under the project are shown in Table 2. A separate projection has been given for rubber grown with intercrop which is assumed to yield slightly less during the first nine years resulting in an average lifetime yield some 2% below that of rubber grown without an intercrop. Stimulation is assumed to start in the nineteenth year of tapping and to be continued for six years although some smallholders may have insufficient bark reserves for a full 24 year period of exploitation. The projected yields are modest and are based on the performance of PB 86. Since it is likely that clones with higher yields than PB 86 would be planted within the project period, the projections understate the likely incremental rubber production. Inspection Procedures and Replanting Payments 7. Under the project, the procedures for replanting would be modified and an Advisory Leaflet describing the new procedures would be issued to small- holders. The Advisory leaflet would incorporate the new schedule of inspec- tions, and replanting payments as described below.  - 54 - ANNEX I Page 3 8. The first instalment (Rs 1,000/ha) would be paid when clearing and uprooting had been satisfactorily completed. All trees and other vegetation must be uprooted and either removed from the holding or burnt. The inspection to authorize this payment would be conducted between February and April. Lining, holing, and specification of soil conservation works would be completed during this inspection. 9. The seccnd instalment (Rs 3,550/ha) would be paid when planting had been completed and the area securely fenced with all conservation works constructed and covers established. To be eligible for payment there must be a full stand of trees (445/ha) with stumps for supplies planted in trenches. All points must be circle weeded. This payment would normally be authorized between July and September of the planting year. 10. The third instalment (Rs 2,500/ha) would be paid following satis- factory inspection 12 months from planting. The replantings should have been regularly weeded and fertilized, covers should be well established, intercrops if planted should conform with requirements (see para 6.12 et seq) and soil conservation works maintained. 11. The fourth, fifth, sixth, seventh and eight instalments would be paid following achievement of satisfactory growth as measured during annual inspectons around the planting anniversary. (Replanters may request earlier inspection and authorization). The growth as measured by girth increment would be compared with norms to be established for each clone. The table below shows the norms for PB 86. The minimum stand should be 370/ha. Instalment Required girth at 90 cm above union Number and Areas above Area below Expected Years Amount (Rs) 2,500 mm rainfall 2,500 mm rainfall from Planting 4 (2,500) 10 cm 10 cm 2 5 (2,500) 16 cm 14 cm 3 6 (2,000) 26 cm 24 cm 4 7 (1,000) 36 cm 32 cm 5 8 (1,000) 50 cm /a 50 cm /a 6 /a 60% of trees should have achieved this girth before opening. Replant- ings in which trees were opened before 60% ac'hiieved 50 cm girth would not be eligible for the final payment. Arrangements for Supply of Planting Material 12. As indicated in Chapter VI, SPC and JEDB estates within the project area would produce budded stumps for DRC on a contract basis. The 33 estates currently contracted to supply budded stumps for 1981 plantings are indicated in Table 3. Based on current nursery areas, these estates would produce 960,000 budded stumps for 1981, some 170,000 or 22% over present estimated requirements. It is anticipated that any oversuppLy could be used outside the project area.  - 55 - ANNEX I Page 4 13. In view of the importance of ensuring adequate supplies of plant- ing materials, formal contracts for the production of planting material between DRC and the estates would be established. The contract should include inter alia: (a) An estimate of numbers and types of budded stumps required 24 months in advance of delivery. (b) An agreement on delivery price (ex-estate on trucks) with an escalation clause based on, perhaps, an index of labor costs. (c) A statement of required quality standards and the basis for acceptance rejection. (d) Acceptance of a system of quality control based on three visits per year by a Visiting Agent with a report to be submitted to both Buyer and Seller; the Seller undertaking to implement Visit- ing Agents' recommendations. (e) Guaranteed access by the Buyer to the nursery provided notice is given. (f) An agreement that delivery shall only be made to the Buyer or his authorized representative. (g) Payment within 30 days subject to arbitration of disputes. (h) Arbitration by representatives of both sides with expert opinion to be provided by the RRISL. B. Project Monitoring 14. Monitoring within ASD would be based on a series of work plans and systematic work scheduling which would be subject to regular checks for each level of field staff. Indicative annual workplans would be prepared for each field officer based on DRC records of registered holdings and replanters within each range. These plans would cover all ASD functions outlined above (para 5.11). Each officer would use the annual workplan as the basis for preparing weekly workplans which would indicate the functions to be performed and holdings to be visited in the forthcoming week. The weekly plans would be sent to the supervisor prior to the start of the week together with actual work achievements for the previous week. These work records for each level of field staff would be monitored by the immediate supervisors. The supervisor would visit subordinates weekly to check whether the officer was following the current work program, and, by visiting holdings already reported to have been serviced by the officer, to check whether the work had been carried out adequately. Minor infringements would be dealt with by the supervisor but all major deviations from the work plan would be reported to the next highest level within ASD for disciplinary action.  - 56 - ANNEXI Page 5 15. Monitoring within DRC would comprise both internal and external systems. The internal system would monitor the processing of DRC's admin- istrative work in relation to established work norms and processing times. Each functional unit within DRC would be required to report its work achieve- ments weekly to the Assistant Controller in charge of each regional office. DRC would also set indicative annual workplans on the basis of projected workloads from existing and anticipated replanting activities. These workplans would specify target dates by which various critical actions were to be completed. The head office of DRC would monitor the work of the regional offices on the basis of monthly reports from the Assistant Controllers, supplemented by regular visits to each regional office. The external monitor- ing system of DRC would consist of field checks on replanting inspections by ASD and would include all observations required in the ASD inspection. The DRC inspectors would be provided with a list of holdings to visit by the Assistant Controller for each region but would not be informed as to the outcome of the ASD inspection. Any discrepancy between the recommendation of ASD staff and those of DRC inspectors would be reported to the Rubber Con- troller who in turn would take the matter up with the Director of ASD. 16. PCU would monitor overall project implementation. Together with the staff of the implementing agencies, the PCU would prepare annual workplans against which project implementation would be assessed. PCU would assist the implementing agencies in the preparation of progress reporting formats includ- ing the selection of appropriate key progress indicators. The implementing agencies would submit monthly progress reports tc PCU which would provide the basis for the formal monitoring of implementation. They would be supplemented by reports on procurement and disbursement activities which would be prepared by PCU. Project accounts would also be reviewed by PCU as part of its formal monitoring role. PCU would monitor implementation informally through field visits and personal contacts with the implementing agencies to provide early indications of potential implementation problems which PCU would attempt to resolve. PCU would bring outstanding problems to the attention of RPRC for their resolution. PCU would also monitor the adequacy of budgetary allocations from the Treasury to the implementing agencies.  SRI LANKA I' Per Hectare Replanting Costs for Rubber (Labor valued at Rs 17.5 /MD) Per Hectare Replanting Costs for Rubber Smallholdings Second Year Upkeep CALENDAR YEAR OPERATIONS Work Days Material Labor Total 1) Weeding (7 months by 12 man days, Per Hectare (Rupe) (Rupees) (Rupees) 1 month X 5 man days) 89 1557 1557 2) Pests and diseases: Pre-Planting Year Materials 65 65 (1) Felling, rooting, cutting, stacking, Labor 5 88 88 burning ( 2 1/2 trees per man day) 140 2,450 2,450 3)Supplying labor (44 plants to vacancies) 3 53 53 Total 140 - 2,450 2,450 4) Fertilizer: 960gm of R462 Mg to 445 plants in 4 Planting Year applications (427kg at Rs 1.103/kg) 471 471 (1) Reburning 3 52 52 Labor for application 10 175 175 (2) Rehabilitation of plantforms, 5) Covers: Reseeding 10% 26 26 terraces and drains (silt pits) 37 648 648 Labor 1 18 18 (3) Lining, including cutting pegs 5 88 88 TOTALS 108 562 1891 2453 (4) Holing, 462 holes, 60cm X 60cm X 75cm (12 holes per man day) 37 648 648 (5) Filling holes (45 holes/man day) 10 175 175 Third Year Upkeep (6) Fencing material 1,390 1,390 1) Weeding (7 months X 8 man days, Labor for erection 15 263 263 1 month X 4 man days) including (7) Covers: Seed, 12kg, Rs 22/kg 264 264 pruning 60 1050 1050 Rock phosphate, 50kg, Rs 0.69 35 35 2) Pests and diseases: Labor, drilling seed and Materials 65 65 fertilizer 10 175 175 Labor 5 88 88 (8) Weeding inter-rows and covers. One 3) Fertilizer: round at 20 man days, one at 15 man 1.44kg of R462 Mg to 432 plants in days, eight at 12 man days 131 2,293 2,293 2 applications (622kg at Rs 1.103/kg) 686 686 (9) Planting (including supplies) Labor for application 10 175 175 Purchase of 494 budded stumps at 4) Soil Conservation, terraces and drains 5 88 88 Rs 2.50 1,235 1,235 5) Fencing repair 3 53 53 Transport to holding at Rs 0.2 99 99TOTALS 83 751 1452 2205 Planting (30 stumps/man day) 17 298 298 T (10) Fertilizer: 120gms rock phosphate/plant Fourth Year Upkeep (60kg at Rs 0.69/plant) 41 41 1) Weeding (7 months X 6 man days 120gms R462 Mg/plant including 1 monthsXX2 man days) supplies (60kg at Rs 1.103/kg) 66 66 1 month X 2 man days) including 480gm R462 Mg to field stand of 445 2) Pesrs and diseases: plants in 4 applications 35ad s (214kg at Rs 1.103/kg) 236 236 Materials 35 35 240 gm application of crushed dolomite 3) 4e70izr to field stand (107kg at Es 0.18/kg)199 3) Fertilizer: to fiel stand(107kg t Rs 018/kg)19 19 1.44kg of R462 Mg to 420 plants in Labor for application 12 210 210 14k fR6 gt 2 lnsi (11) Pests and diseases: 2 applications (605kg at Rs 1.103/kg) 667 667 Materials 50 50 Labor for application 10 175 175 3s4) Soil Conservation, maintaining terraces Labor for application 3 52 52 and drains 5 88 88 TOTALS 280 3,435 7,352 8,337 TOTALS 63 702 1103 1805 1/ See Annex 4, Table 10. -s H4  Fifth Ylear UFkcg-p Fifth Yeear ( kmonchs X 6 man days, Eighth Year (12 Months Taping) 1 month X 2 man days) 44 770 770 1) Weeding (6 months X 2 man days) 12 210 210 2) Pests and diseases: Materials 35 35 2) Pests and diseases: Materials 35 35 Labor 4 70 70 Labor 4 70 70 3) Fertilizer: 3) Fertilizer: 1.44kg of R462 Mg to 415 plants in 2 1.44kg of R462 Mg to 408 plants in 2 applications (598kg at Rs 1.103/kg) 660 660 applications (588kg at Rs 1.103/kg) 649 649 Labor for application 9 158 158 Labor for application 9 158 158 4) Soil conservation, maintaining terraces 4) Tree girth census and marking for opeining 1 18 18 and drains 3 53 53 5) Tracing tapping panels on 200 trees 2 35 35 TOTALS 60 695 1051 1746 6) Clearing tappers paths 2 35 35 7) Tapping equipment (as year seven) 190 190 8) Tapping and processing labor for s/2 d/2; Sixth Yea U k 2 tasks, 180 tappings each at 5 1/2 Weedn months X 4 man days, hrs./task 248 4340 4340 1 month X 2 man days) 30 525 525 Processing 620kg rubber 19 333 333 2) Pests and diseases: Materials 35 35 9) Soil conservation. Upkeep of terraces Labor 4 70 70 and drains 2 35 35 3) Fertilizer: TOTALS 299 874 5234 6108 1.44kg of R462 Mg to 408 plants in 2 applications (588kg at Rs 1.103/kg) 644 644 Labor for application 9 158 158 Ninth to Twenty-Fourth Years 4) Soil conservation, maintaining 1) Weeding (6 months X 2 man days) 12 210 210 terraces and drains 3 53 53 2) Pests and diseases: Materials 35 35 TOTALS 46 679 806 1485 Labor 4 70 70 3) Fertilizer: 1.44kg of R462 Mg to average 312 Seventh Year (9 months Tap .im) plants in 2 applications 1) Weeding (8 months X 2 man days) 16 280 280 (450kg at Rs 1.103/kg) 496 496 2) Pests and diseases: Materials 35 35 Labor for application 7 123 123 Labor 4 70 70 4) Clearing tappers paths and removing 3) Fertilizer: dead trees 2 35 35 1.44kg of R462 Mg to 408 plants in 2 5) Tapping equipment (every 4th year), applications (588kg at Rs 1.103/kg) 649 649 average cost 50 50 Labor for application 9 158 158 6) Tapping and processing labor: 1 1/2 tasks /5 Tree gftth cene'iF ond -srkl-n, fr- n-e,I 2 35 35 180 tappings each at 5 1/2 hrs./task 214 3745 3745 5) Tracing tapping panels on 204 trees 4 70 70 Frocessing average 1l0kK -ubLei 2. 43e A25 6) Cleaning tappers paths 2 35 35 7) Soil conservation. Upkeep of terraces 7) Tapping equipment: Spouts 40 40 and drains 1 18 18 Hangers 40 40 TOTAL AVERAGE YEAR 265 581 4639 5220 Half shells 20 20 TOTAL 16 YEARS 4240 9246 74,224 83,520 Buckets (2) 50 50 Knives (1) 40 40 8) Tapping and processing labor at one 200 tree task (Tapping 2 1/2 hrs., drip- ping 1 hr., collection 1 1/2 hrs., transport 1/2 hr., total 5 1/2 hrs. day) for 135 tapping days s/2 d/2 93 1628 1628 Processing 240kg rubber 17 298 298 TOTALS 147 874 2574 3448  Twenty-Fifth Year 1) Weeding (6 months X 2 man days) 12 210 210 2) Pests and diseases: Materials 35 35 Twenty-Ninth and Thirtieth Years Labor 4 70 70 1) Stimulant: Materials ta4U trees X 4 3) Clearing tappers paths and removing applications at Rs 0.50) 480 480 dead trees 2 35 35 Application labor (100 trees/man day) 10 175 175 4) Stimulant: Materials: ethylene releasing 2) Tapping and processing labor for 4 s/2 d/2, (300 trees X 3 applications at Rs 0.50) 450 450 1 1/2 tasks at 80 trees a task, Application labor (100 trees/man day) 9 158 158 6 1/2 hrs. 439 7683 7683 5) Tapping and processing labor for 2 s/2 d/3, Processing 870kg rubber 21 368 368 one task of 133 trees (Tapping 3 hrs., TOTAL 470 480 8226 8706 dripping 1 1/2 hrs, collection 1 1/2 hrs, TOTAL 2 YEARS 940 960 16,452 17,412 transport 1/2 hr, total 6 1/2 hrs) 293 5127 5127 Processing 1225kg rubber 26 455 455 TOTAL 346 485 6055 6540 Notes to Table 1. All wages at Rs 17.50 per man day. Twenty-Sixth Year 1) Weeding (6 months X 2 man days) 12 210 210 2. Since replanting is assumed, no underbrushing is included. 2) Pests and diseases: Materials 35 35 Labor 4 70 70 3. Pruning of stock shoots is included in the earlier weeding rounds. 3) Clearing tappers paths and removing dead trees 2 35 35 4. R462 Mg is a 9:5:3:1 mixture. 4) Stimulant: Materials (290 trees X 3 appli- cations at Rs 0.50) 435 435 5. Fertilizer costs including delivery charges at subsidized prices are: Application labor (100 trees/man day) 9 158 158 Rock phosphate Rs 690/metric ton 5) Tapping and processing labor for 2 s/2 d/2, Crushed dolomite Rs 180/metric ton 1 1/3 tasks of 120 trees, 6 1/2 hrs. 390 6825 6825 R462 Mg Rs 1103/metric ton Processing 1070kg rubber 23 403 403 TOTAL 440 470 6601 8171 6. Task size is a mission estimate of 200 trees per tapper for s/2 d/2 tapping, 120 trees for 2 s/2 d/3 tapping, 100 trees for 2 sf2 d/2 tapping and 80 trees for 4 s/2 d/2 tapping. The latter assumes two upward cuts. Labor productivity is based on Twenty-Seventh and Twenty Ei hth Years reported task sizes and is 20 to 35X below that of other rubber producing countries. 1) Weeding (t montls A mang7 12 210 210 The major factors appear to be the steep topography over much of the area, un- 2) Pests and diseases 2 35 35 warranted care in making shallow tapping cuts, and lack of routine in tapping/ 3) Removing dead trees 2 35 35 collection procedures. 4) Stimulant: Materials (260 trees X 4 appli- cations at Rs 0.50) 520 520 7. Stimulant costs are an estimate of the cost of purchase of the proprietary stimulant Application labor (100 trees/man day) 11 193 193 "Ethrel" and consequent dilution. Prices may fall. 5) Tapping and processing labor for 2 s/2 d/2, one task of 120 trees, 6 1/2 hrs. 293 5128 5128 8. Application costs for stimulant can very according to the method employed. Processing 955kg rubber 22 385 385 Processing 955kg rubber 22 385 385___________________________________________________ 9. It is unlikely that sufficient bark will remain for the 29th and 30th years explof- TOTAL 342 555 5951 6506 tation in many smallholdings. Commitment of final exploitation would depend on TOTAL 2 YEARS 684 1120 11,902 13,012 advice from the local extension officer. 10. The large labor Inputs for final exploitation may deter some smallholders. 11. Processing labor costs were computed at a standard rate for diluting, coagulating and smoking and a variable rate, depending on crop, for rolling. December 17, 1979. > D m ' '  - 60 - SRI LANKA ANNEX 1 Table 2 Smallholder Rubber Rehabilitation Project Yield Profiles and Tapping Systems, Smallholder Budded Rubber YEAR FROM TAPPING EXPLOITATION FREQUENCY OF YIELD IN KG./HECTARE PLANTING YEAR SYSTEM STIMULATION W/ COVER CROPS W/INTER CROPS 1. 2 3 4 5- 6 7 1. (9 mos.) sf2 d/2 300 240 8 2 II-675 620 9 3 it-840 785 10 4 if-955 900 11 5 It.- 1070 1010 12 6 - 1180 1025 13 7 - 1235 1235 14 8 1290 1235 15 9 1350 1290 16 10 1350 1350 17 11 - 1350 1350 18 12 - 1350 1350 19 13 2- 190 1290 20 14 2- 35 1235 21 15 " - i180 1180 22 16 "- 1.125 1125 23 17 - 1070 1070 24 18 - 1010 1010 25 19 2 s/2 d/3 3 1225 1225 26 20 2 s"2 d/2 3 1070 1070 27 21 4 1070 1070 28 22 4 900 900 29 23 4 s/2 d/2 4 955. 955 30 24 4 785 785 December 17, 1979.  - 61 - ANNEX 1 Table 3 SRI LANKA Smallholder Rubber Rehabilitation Project Intercropped Bananas: Inputs and Yields (Per Hectare) LABOR OPERATION MAN DAYS MATERIAL Rs 12/ YIELD MAN DAY (BUNCHES) YEAR 1 1) Holing (including lining) 15 180 2) Planting Material (Rs 1/point) - 296 - 3) Transport (estimated) - 150 - 4) Planting 12 - 144 5) Weeding (2 rounds/annum) 18 - 216 6) Fertilizer (1 kg/clump/annum) - 370 - (296 kg at Rs 1.25/kg 7) Fertilizer application (1) 7 - 84 8) Harvesting 6 - 72 9) Tools - 100 - TOTALS 58 916 696 150 YEAR 2 1) Weeding (2 rounds/annum) 20 - 240 2) Fertilizer (2 kg/clump/annum) - 740 - (592 kg at Rs 1.25/kg) 3) Fertilizer application 14 - 168 4) Supply vacancies 2 _ 24 5) Removing suckers 4 - 48 6) Transport - 300 - 7) Harvesting 12 - 144 TOTALS 52 1040 624 290 YEAR 3 As for Year 2 except: 1) No supply -2 - -24 2) Extra harvesting 6 - 72 3) Extra transport - 50 - TOTALS 56 1090 672 290 YEAR As fo Year 3 except: 1) Re uction in transport - -50 - 2) El mination of plants 9 - 108 - TOTALS 65 1040 780 250 i I  - 62 - ANNEX 1 Table 4 SRI LANKA Smallholder Rubber Rehabilitation Project Estates Producing Planting MaterialE for 1981 DISTRICT RESPONSI'3E ESTATE ]NA1E LOCATION Nl?MERS OF AUTrHORIT` BUDDED STUMPS Kalutara SPC Sorana S. P'. Horana 50,000 Perth S. P. " 40,000 Mirishena S.P. Mahagoma 20,000 Frocester S. P. Govinna 40,000 Halwatura S. P. Ingiriya 40,000 Yatadola S. P. Matugama 40,000 St. George S. P. " 40,000 Eladuwa S. P. Paiyagala 40,000 Miriswatte S. P. Walipenne 40,000 Pimbura S. P. A,alawatte 40,000 Hillstrean S. P. Bilathsinhala 10,000 SUBTOTAL 400,000 Ratnapura SPC Mahawela S. P. Ritnapura 45,000 Pussella S. P. Parakadewa 45,000 Peenekande S. P. Uda-Karawita 30,000 Durampitiya S. P. Gatehat'he 45,000 Galatura S. P. Ktriella 45,000 Matuw.agala S. P. " 30,000 SUBTOTAL 240,000 Klegalle JEDB Udapolla Estate Doraniyagala 20,000 Panawatte Estate Yatiyantota 20,000 Avissawella Estate Pilwakpitiya 20,000 Moralioya Estate Riiwanwella 20,000 Sunnycroft Estate Waharaka 20,000 Clunes Estate Dehiowita 20,000 Atale Estate AI:ale via Kegalle 20,000 Kiripiruwa Estate Yontiyantota 20,000 Pallegama Estate MRyandurapola 20,000 Ederapolla Estate Biilathkohupitiya 20,000 Urumiwella Estate " 20,000 Halgolla Estate Yctiyartota 20,000 Parambe Estate UTdugoda 20,000 Ambadeniva Estate Aionayake 20,000 'Hatbawe Estate R mbukkama 20,000 Etana Estate Wcrikarola 20,000 SUBTOTAL 320,000 1/ TOTAL 960,0C0 2/ 1/ *Estate and subtotal figures liable to revision. 2/ 1620 hectares are die for replanL.ing in 1981, at 494/1-ectarc (including s Lpi,li.s) which requires some F00,000 units; there wotuld arpeur to be a potential oc-r s;pply of lf0,000 unlits (3 '4 hi. or 20. at a possible ccst of Ks 400,000). Althougl this MAight he taken up b 'ncreased azpplications or b1 e thclc sales, it points tc thlt requlrernct for foril l arrangements in the 1980 cc der.  - 63 - ANNEX 2 Appendix 1 Page 1 SRI LANKA RUBBER REHABILITATION PROJECT Draft Terms of Reference for Technical Assistance Replanting Scheme Administration 1. Background. The Department of Rubber Control (DRC) plans to improve its procedures for the administration of a rubber replanting scheme. The major objectives of the intended improvements would be: (a) to make the scheme more attractive to smallholders by reducing to a minimum the various actions required of the smallholder in applying for, and being provided with, replanting assistance; (b) to simplify procedures for the granting of replanting permits and supply of replanting inputs to minimize delays and staff costs and to maximize cost effectiveness; (c) to ensure that the administration of the scheme can be easily and continually monitored to give early warning of problems and to simplify audit procedures; and (d) to incorporate a simple system of data retrieval to permit current information on the status of individual replanters to be promptly provided to field staff responsible for advice to replanters. 2. DRC officers would study the operations of replanting schemes in Thailand, Malaysia and Indonesia, to provide information on alternative proce- dures. A consultant would be required to work with DRC staff for about six months to develop and implement the new procedures. Some of the major features of the anticipated changes in DRC procedures are: (a) information on registration of rubber holdings would be revised to reflect the present situation and thereafter regularly updated. Registration details would thus permit early identification of smallholders whose rubber is due for replanting; (b) replanting targets would be set by a Rubber Policy and Planning Unit based inter alia, on age statistics pro- vided by DRC from its registration of holdings;  -64 - ANNEX 2 Appendix 1 Page 2 (c) processing of all paperwork associated with registration and replanting would be hastened by (a) use of simplified forms and procedures; (b) increased staff in regional offices, and (c) introduction of a monitoring system which would facilitate daily checks on progre!ss and comparison with established norms; (d) responsibility for checking and authorization of routine work would be delegated to staff at lower levels to avoid delays; and (e) DRC would identify holdings due for inspection well in advance and would pass this informatior. to ASD. 3. The recesntly completed Rubber Masterplarn for Sri Lanka contains considerable information on the existing procedures used by DRC and makes recommendations for an improved system. The consultant and DRC would ensure that due attention is given to these proposals in developing any new scheme. 4. Qualifications. The consultant would have at least ten years experi- ence with a rubber smallholder replanting program and be familiar with the details of replanting programs in Thailand, Malaysia and Indonesia. Although there are no specific discipline requirements, the consultant would be required to have competence in accountancy, bookkeeping, auditing, monitoring and data processing including design of forms, vouchers, etc. Familiarity with elec- tronic data processing would be required although no decision has yet been taken to base the improved procedures on the use of a computer. 5. Specific Responsibilities. The consultant would be required to: (a) provide detailed descriptions (including copies in english of forms/vouchers) of the replanting schemes in Thailand, Malaysia and Indonesia; (b) review and comment on the adequacy of the existing proce- dures followed by DRC; (c) review with DRC staff the proposals for revised procedures recommended in the Rubber Masterplan anc. establish what assistance is to be provided in this fie!ld from the UK Government. (It would be important to ensure that there was close coordination between inputs under this consul- tancy to DRC and any other consultancy cealing with related aspects); (d) propose a revised set of administrative procedures together with supporting documentation (forms, vouchers, records, etc.); and (e) assist :DRC in the implementation of the revised procedures for administration of the replanting scheme.  - 65 - ANNEX 2 Appendix 2 Page 1 Draft Terms of Reference Training 1. Background. The Advisory Services Department (ASD) has undertaken new responsibilities in the provision of assistance to smallholders to increase replanting rates. In order to carry out this task, its staff would be increased by recruitment of some 154 new officers. All existing and new staff would require some training which is to be provided under the project. ASD has begun work on a residential training center and it is expected that the first training courses would commence about mid 1980. The specific courses envisaged under the project include: (a) an annual three month induction course for 15 new Rubber Extension Officers which would prepare high school graduates for extension work by providing the rudiments of rubber husbandry and replanting technology in addi- tion to training in extension methodology; (b) refresher courses for ASD staff; two courses per year, two week duration with 30 participants per course. Course to cover changes in the replanting scheme proce- dures, recent advances in rubber husbandry, and extension methodology; (c) induction courses for Extension Overseers; two courses per year, two month duration with 30 trainees per course. Similar to induction course for REOs but simplified and condensed and with more emphasis on replanting activities including contour lining; (d) an induction course of one month duration for a new cadre of ASD staff (Processing Advisor) for 15 trainees. This course and the course described below (e) would be prepared by the consultant on processing; and (e) twice yearly courses of one week duration on processing for private processors and smallholders. 2. Qualifications. The consultant would be qualified as a lecturer with specific experience in the preparation of curricula and course material for training of extension staff. The consultant would have had field experi- ence in an extension service for rubber smallholders. 3. Specific Responsibilities. In conjunction with the staff of ASD and RRISL, the consultant would: (a) prepare detailed curricula and outline lecture notes for the induction courses for REOs and EOs, and for the refresher courses;  - 66 - ANNEX 2 Appendix 2 Page 2 (b) assist in the preparation of course material for the courses; (c) assist in the presentation of the first course for REOs; (d) prepare detailed proposals for the publicity campaign to be launched under the project; and (e) provide necessary specific training to the publicity and training officers within ASD.  - 67 - ANNEX 2 Appendix 3 Page 1 Draft Terms of Reference Processing 1. Background. The Advisory Services Department (ASD), plans to create a Processing Advisory Division (PAD) to provide private processors and Group Processing Centers (GPCs) with advice on rubber processing. The new division would take over and expand the advisory activities directed towards processors. The total staff strength of PAD is expected to be about 12 all of whom would be located in the field. Each processing advisor would closely monitor the operation of about 8 GPCs providing technical, managerial and financial advice where required. In addition, PAD staff would contact private processors to provide advice on processing technology with the objective of raising the quality of rubber output. ASD seeks the services of a qualified consultant for a period of about three months to assist in the development of PAD. The project activities directed towards improvement in smallholder processing are described in the Staff Appraisal Report (para 3.21-3.28). Additional suggestions for processing improvements are given below: (a) Smallholders would be encouraged to use a 50/60 double mesh monel sieve to remove impurities from latex rather than the commonly used pad of grass. Subject to a field evaluation of its practicality, estimation of drc at GPCs or latex collec- tion centers (LCC) would be improved by use of the 'chee' system. If the 'chee' system did not prove practicable, the Metrolac would continue to be used but with greater care taken in making readings. (b) The milling of the coagulum would be improved to produce thinner sheets by adjustment to rollers and by minimizing the handling of coagulum prior to rolling. Sheets would be washed and allowed to drip from racks to remove serum and to minimize uneven enzyme discoloration. (c) Improvements would be made in the drying process. Wet sheets would be loaded onto the lowest racks of the smokehouse (the coolest part), to avoid the formation of a skin which hinders subsequent drying. Hangers would be rotated after 24 hours to avoid hanger marks and the formation of bubbles. Rather than the present technique of excessive stoking at long intervals, the fire would be stoked about every three hours and the furnace doors would be fitted with adjustable air inlets. When dry, the rubber would be stored off the ground prior to sale to prevent sweating and consequent mould growth. (d) The improvements in processing techniques outlined above would be promoted by PAD staff for existing GPCs, private processing centers, latex collection centers, and for home  - 68 - ANNEX 2 Appendix 3 Page 2 processors. In addition, the project would support the expansion of new GPCs subject to the existing GPCs improving their performance. The new GPCs would include the following specific improvements: - a tiled bulking tank; - ia tiered wooden rack for dripping sheets before smoking; - aE two-chambered smokehouse (each chamber 680 kg capacity) with flue gas entry system such that one chamber could be closed when intake is low, to reduce expenditure and to maintain an even temperature; - the furnace door will be cut to provide for expan- sion of metal without buckling and will have a row of air inlets. - internal drainage in the smokehouse to remove serum water. 2. Qualifications. The consultant would have had at least five years experience in the operation of group processing and/or marketing schemes for rubber smallholders. In addition to having technical competence in all aspects of Ribbed Smoked Sheet production, the consultant: would have had experience in the provision of managerial advice to smallholder processing groups. The consultant would also be familiar with record keeping and accounting proce- dures appropriate to the operation of GPCs. 3. Specific Responsibilities. The consultant would work closely with ASD and would inter alia: (a) assist in the formulation and presentation of a one month training course for PAD staff; (b) assist in the formulation and initial presentation of a one week course for GPC imembers and privatE: processors; (c) assist in the preparation of material for the above two training courses; and (d) assist in the formulation of an appropriate strategy for improving smallholder processing which may include specific actions to strengthen and expand the GPCs if appropriate.  - 69 - ANNEX 2 Appendix 4 Page 1 Draft Terms of Reference Project Evaluation 1. Background. The Agrarian Research and Training Institute proposes to evaluate the impact of a project to accelerate replanting of rubber by smallholders. A replanting scheme has been in operation in Sri Lanka since 1953, but its performance over the past decade has been disappointing with the result that the national stand of rubber has become over-aged. The project would directly address the major apparent constraints to smallholder replant- ing which are low producer prices, inadequate compensation for income foregone and a cumbersome replanting scheme. However it is likely that there are less obvious factors which influence the smallholder's decision to replant and that a better understanding of the smallholder's viewpoint would reveal oppor- tunities to make the replanting scheme more attractive to smallholders. It is therefore proposed that the project be evaluated during its implementation to identify opportunities for improvement. ARTI plans to carry out a baseline survey covering a representative sample of the 90,000 smallholders in the project area. This survey would be followed by case studies. Through these investigations, ARTI would try to: (a) identify why some smallholders with over-aged rubber fail to replant; (b) identify what changes in the replanting scheme are required to induce such smallholders to replant; (c) establish how effective the services of ASD, DRC and RRISL have been in improving production of smallholder rubber; and (d) identify the changes required in the operation of ASD, DRC and RRISL to better service smallholders in a cost-effective manner. ARTI proposes to recruit a qualified consultant to assist in the evaluation study for a period of about three months. 2. Qualifications. The consultant would have had at least five years experience in the design and conduct of surveys and case studies of smallholder producers, preferably rubber producers. He would be a University Graduate with training on/in survey design and analysis, rural sociology, agricultural economics and statistics. 3. Specific Responsibilities. The consultant would work with ARTI staff consulting with the Project Coordinator and other staff of the implementing agencies as necessary. He would: (a) assist in the design of the baseline survey of project area smallholders including sample selection and questionnaire formulation;  - 70 - ANNEX 2 Appendix 4 Page 2 (b) advise on the supervision of the baseline survey as required; (c) assist in the design of the case studies including preparation of specific terms of reference for about: four or five studies which are likely to provide the requirea. information; and (d) prepare a suitable format for the report on survey findings and the case studies.  - 71 - ANNEX 3 Page 1 SRI LANKA Rubber Rehabilitation Project Local Procedures for Project Funding Project Funding 1. Government funding for a project is based on a Project Report prepared by the implementing agency (the PCU for the proposed project). The Project Report must be approved by the Development Committee of Secretaries before sanctioning by Cabinet. All implementing agencies keep a separate head for the project in their annual budget. Initial budget proposals are submitted to the Treasury by the various departments through the relevant ministry by May 31st of the preceeding year. Draft budget estimates proposed must be submitted to the Treasury by June 12 for inclusion in the annual budget, which in turn is submitted to Parliament for approval in November/ December. 1/ 2. Release of funds to the departments by the Treasury is on a monthly basis, against the annual budget as sanctioned by Parliament and approved in detail by the Ministry of Finance. A monthly return of expenditures must be submitted to the Treasury for comparison with the annual budget and for release of funds in the succeeding month. Annual estimates of expenditures must be revised by May 31st in the light of expenditures made during the first quarter. After sanction of the revised estimates by the Treasury, requisition of funds are made against these revised estimates. 3. Annual budgets for the project would be prepared by each implement- ing agency (DRC, ASD and RRISL) with the assistance of the PCU which would ascertain that the proposed estimates match those of the Project Report. The PCU would consolidate the project annual budget and submit it to the Treasury through the Ministry of Plantation Industries. 4. Funds for capital and operating expenditures would be released by the Treasury directly to the respective implementing agencies, while staff would be paid directly by Treasury. Replanting payments to smallholders (including fertilizer and planting materials to be delivered in kind) would be met from the Rubber Replanting Fund administered by DRC. The revenues of the Fund would initially comprise the Replanting Cess which is paid directly into the Fund by the Customs Office,- and from end of project year two the shortfall in cess revenue would be met by annual Government contribution to the Fund. Accounting Records and Annual Controls 5. Adequate books of accounts are maintained at the headquarters of DRC and RRISL. Internal control is adequate within both agencies. The Chief 1/ This procedure does not apply to DRC since it is currently funded by cesses. However, it would apply under the project if, as envisaged, the capital and operating costs in administering the Replanting Fund were now funded by normal budgetary provisions.  - 72 - ANNEX 3 Page 2 Accountant's office of the RRISL is well run and has the capacity to maintain accounts efficiently; the accounts of the RRISL are subject to regular checking by the RRISL internal auditor. The Accounting Section of DRC is headed by a competent Chief Accountant. However, his office is understaffed with the result that DRC accounts have been systematically late. Disbursements from the Replanting Fund are subject to a severe, and possibly unnecessarily thorough, system of checking with multiple layers of control before payments are made. 6. The annual accounts of both agencies are subject to an external audit which is carried out yearly by a team of Public Auditors under the direction of the Auditor General. The finance Act No. 38 of 1971, requires that accounts be prepared and submitted for audit by April 30th of the fol- lowing year, and the audit completed by October 31st. At the time of the mission's visit (September 1979), the 1978 accounts of RRISL were available and the mission was advised that their audit was in progress. The 1978 accounts of the DRC were also available but the latest audited accounts were those of 1976. Accounting Arrangements for the Project 7. Under the project, a new accounting unit would be set up within ASD which would become an independent department. In addition, accounting sub-units would be established in each of the three DRC regional offices; this would considerably lessen the workload of the Chief Accountant's office. Project accounts would be kept under separate heads by DRC, ASD and RRISL and submitted quarterly to the PCU for purpose of monitoring and preparing disbursement applications. Disbursements Against Statements of Expenditures 8. For the purpose of Credit disbursements against statements of expenditures, the budgeting, accounting and auditing procedures of DRC and RRISL are satisfactory, and with the proposed strengthening of their staff, their institutional capability would be adequate. The Auditor General would audit the statements of expenditures biannually and GOSL would furnish the reports to IDA. Supporting documents and vouchers retained by the Auditor General would be made available for inspection by Supervision missions. 9. Credit disbursements against statements of expenditures would be for petty contracts for civil works and equipment (individually less than Rs 100,000) and for replanting payments. In view of the large volume of replanting payments, (US$9.8 M), DRC would be required to maintain a sub- sidiary Project Replanting Fund account relating to the project replantings. Disbursement under the Credit would be against a statement of expenditure from this Project Replanting Fund account submitted quarterly. At the end of each fiscal year, these statements would be reconciled with the annual audited DRC project accounts. The DRC project accounts would be supported and reconciled with the annual audited accounts relating to the overall Replanting Fund Account.  - 73 - ANNEX 3 Page 3 10. Records of progress of replanting work would be systematic, and up to date. Both for management and audit purposes, DRC would be required to submit Quarterly Progress Reports to the PCU giving, for each year of permit issue, particulars of: (a) area approved for replanting; (b) replanting payments committed; and (c) installments paid. 11. ASD would be responsible for the issue of fertilizer and planting materials to smallholders against permits approved by DRC. Adequate stocks financed out of the Replanting Fund would be carried by ASD for this purpose. After being issued to replanters, the materials would be charged to the Replanting Fund Account, against which IDA disbursements would be made.  SRI LANKA Smallholder Rubber Rehabilitation Project Project Participants by Size of Holding I. Representative Farm Models Holding Total Paddy Rubber Replanted Intercrop Category Holding Area Rubber