Sri Lanka - Second Rural Development Project

The World Bank FOL OMCIAL USE ONLY N RNsnNM 9905 PROJECT COMPLETION REPORT SRI LANKA SECOND RURAL DEVELOPMENT PROJECT (CREDIT 1079-CE) SEPTEMBER 19, 1991 Agriculture Operations Division Country Department I Asia Regional Office This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  Currency Equivalents Name of Currency : Sri Lanka Rupee (Rs) Rates of Exchange Appraisal (October 31, 1980) : USS 1 - Rs 15.6 Present : US$ 1 - Rs 40.3 Weights and Measures 1 long ton 1.016 metric ton 1 bushel (bu) of paddy = 46 lb 1 acre (ac) 0.405 hectare (ha) Abbreviations and Acronyms CCB - Coconut Cultivation Board DA - Department of Agriculture DAS - Department of Agrarian Services DH - Department of Highways DLG - Department of Local Government DMEC - Department of Minor Export Crops ERR - Economic Rate of Return GOSL - Government of Sri Lanka ID - Irrigation Department IDA - International Development Association IFAD - International Fund for Agricultural Development LD - Livestock Department O&M - Operation and Maintenance PCR - Project Completion Report SAR - Staff Appraisal Report WRB - Water Resources Board Fiscal Year January 1 to December 31  DIC ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~~O C.IILU L THE WORLD BANKON Washington. DC 20433 U.S A. O.-oE dc OsKv.Ce.wtal Opsaeuiw IVAlua September 19, 1991 MEMORANDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT SUBJECT: Project Completion Report on Sri Lanka Second Rural Develonment Proiect (Credit 1079-CE) Attached, for information, is a copy of a report entitled "Pizoject Completion Report on Sri Lanka Second Rural Development Project (Credit 1079-CE)" prepared by the Asia Regional Office with Part II contributed by the Borrower. No audit of this project has been made by the Operations Evaluation Department at this time. Attachment This documer. t 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.  FOR OFFICIAL USE ONLY PROJECT COWLETION REPORT SRI LAA SECOND RURAL DEVELOPHINT PROJECT (Credit 1079-CE) TABLE OF CONTENTS Page No. PREFACE ............................ . i EVALUATION SUMMARY . iii PART I: PROJECT REVIEW FROM BANK'S PERSPECTIVE . . . . . . . . . 1 Project Identity. . 1 Project Background and Sectoral Context . . . . . . . . . 1 Project Objectives and Description . . . . . . . . . . . 2 Project Design and Organization . . . . . . . . . . . . . 3 Project Implementation. . 4 Project Results . 11 Project Sustainability . . . . . . . . . . . . . . . . . 12 Bank Performance . 13 Borrower Performance and Project Relationship . . . . . . 14 Consulting Services . 14 Project Documentation and Data . . . . . . . . . . . . . 15 PART II: PROJECT REVIEW FROM BORROWER'S PERSPECTIVE . . . . . . 17 PART III: STATISTICAL INFORMATION . . . . . . . . . . . . . . . . . 23 Table 1: Related Bank Loans and/or IDA Credits . . . . . 23 Table 2: Project Timetable . . . . . . . . . . . . . . . 26 Table 3: Credit Disbursements . . . . . . . . . . . . . . 27 Table 4: Project Implementation . . . . . . . . . . . . . 28 Table 5: Project Costs and Financing . . . . . . . . . . 33 Table 6: Project Results . .34 Table 7: Record of Compliance with Major Covenants . . . 35 Table 8: Use of Bank's Resources . . . . . . . . . . . . 36 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.  PROJECT COMPLETION REPORT SRILANR& SECOND RURAL DEVELOPMENT PROJECT (Credit 1079-CE) PREFACE This is the Project Completion Report (PCR) for the Second Rural Development Project in Sri Lanka, for which Credit 1079-CE in the amount of SDR 25.6 million (US$33.5 million) was approved on December 9, 1980. The Credit was closed on June 30, 1990, four years behind schedule. Last disbursement was on November 9, 1990 and an outstanding balance of SDR 2.74 million was cancelled. The Preface, Evaluation Summary and Parts I and III were prepared by the Agriculture Operations Division, Country Department 1, Asia Regional Office. Part II was prepared by the Borrower's Ministry of Policy Planning and Implementation. Preparation of the PCR was initiated in early Februr-' 1991. The PCR was prepared based, inter alia, on: (a) the Staff Appraisal Rep, :t (Number 2942-CE); (b) Development Credit Agreement (1079-CE); (c) minutes ur credit negotiations; (d) correspondence between the Bank and the Borrower; (e) internal Bank memoranda; (f) supervision reports; (g) evaluation reports of project components; and (h) discussions with officials of the implementing agencies.  - ili - PROJECT COMPLETION REPORT SRILANA SECOND RURAL DEVELOPMENT PROJECT (Credit 1079-CE) EVALUATION SIMIR Obiectivs i. The main objective of the project was to improve the productivity, income and living standards of people in the two districts of Matale and Puttalam by rehabilitating and improving the economic infrastructure, improving and expanding the provision of inputs and services required by the agriculture sector, stimulating agricultural production, and improving the social infrastructure. The project formulation was based on the underlying premise that the existing economic and social infrastructure in the two project districts were substantially underutilized and that their increased and more efficient utilization, including provision of finance for private investments in the production base, would lead to rapid and significant improvement in the economic and social well-being of the people in the districts. The project demonstrated the appropriateness of that approach. Implementation Experience ii. The project consisted of a wide and diverse range of subprojects covering the economic and social areas. However, it focussed on the development of agriculture, mainly by (a) rehabilitating/improving existing small irrigation schemes to improve/increase rice production; (b) expanding/improving agricultural support facilities and services, including agricultural credit, to provide increased and improved agricultural production inputs; and (c) promoting the development of coconut, minor export crops, and forestry. Complementing these directly productive subprojects, the project also included limited investments to improve rural transportation, health, education, drinking water supply, and rural electrification. Altogether, twenty-five government agencies were involved in implementing the project. iii. Implementation of the project was initially phased over a five-year period from mid-1981 to mid-1986. However, implementation did not progress as anticipated and the project was extended four times, each time for one year, to achieve the set targets and to maximize utilization of the IDA Credit. Actual disbursements amounted to SDR 22.86 million or about 89Z of the total IDA Credit of SDR 25.6 million. Actual project cost in local currency was about 33% higher than appraisal estimate due to domestic inflation and longer project implementation; t-j in US dollar terms actual cost was lower (US$39.3 million  iv compared to appraisal estimate of US$50 million) due to devaluation of the Sri Lankan rupee. iv. The main reasons for the delay in project implementation were: (a) manpower shortages in the initial stage, in some of the implementing agencies; (b) severe budgetary constraints resulting in shortage of local counterpart funda; (c) worsening of the security situation in later years of the project which caused project staff and contractors to stay away from project sites which in turn led to disruptions of project work.; (d) shortage of construction material.; and (e) delays in procurement of project goods. Despite these delays, implementation of the project as a whole had generally progressed satisfactorily in that the physical targets set during appraisal were essentially achieved by the end of 1987, only about a year and a half behind schedule. The latter two extensions of the Closing Date were granted essentially to enable fuller utilization of the IDA Credit, the local currency equivalent of which had increased substantially due to continuous and significant depreciation of the Sri Lankan Rupee. Fuller utilization of the IDA Credit had in turn enabled the actual physical targets achieved by many of the project components to exceed significantly their appraisal estimates. Results v. The overall performance of the project can be considerea a success as it has achieved its main objectives, namely to improve the productivity, incomes and living standards of people in Puttalam and Matale districts. As noted above, the actual results achieved had equalled or exceeded the physical targets set at appraisal for essentially all the project components. A full assessment of the project's benefits and impact was not feasible because much of the required data were not available. However, there are indications, based on available data, which clearly demonstrate that the overall benefits and impact of the project have been substantial. The project has generated sjgnificant increases in yields and production of paddy, coconuts, and minor export crops, which together constituted the major direct benefits (para. 46). It alleviated the overall living conditions of the rural people and their work skills by improving the roads network, health services, rural electrification, drinking water supply, and educational facilities. Although the benefits outlined below cannot be easily quantified, expansion of rural roads has clearly benefitted the rural communities by improving access to markets and non-farm employment opportunities, facilitating the movement of labor and farm inputs, and improving the delivery of other rural services. Bringing electricity to villages which had none before not only improved the well-being of households but also stimulated the expansion of small rural industries and improved their operations. Access to better quality education, medical services, and drinking water improved the health and skill performance of the people. Some diversification of the agricultural production base and income was also promoted by the minor export crops component and to a lesser extent by the fisheries and livestock components. Sustainability vi. Having been based on an already existing institutional set-up, project activities, even after their cessation, would be sustainable. The project district level staff, employed for planning and implementation of the  v project, vnuld be absorbed by the planning offices in the respecLive districts. Physica2 assets created by the project activities were harded over to the reopective agencies for continuous operation and maintenance (O&M), and staff recruited for project purposas were absorbed into the cadres of the respective agencies. The agricultural zredit component appears sustainable in view of its satisfactory loan recovery performance. However, funds provided for O&M of irrigation schemes and rural roads have been inadequate. If the shortage of O&M funds were to continue, the economic life and long-run sustainability of these assets would be adversely affected. Findings and Lessons Learned vii. The project confirmed the appropriateness and replicability under Sri Lanka's conditions of the approach taken to achieve a broad-based development of the country's districts which was initiated tunder the Kurunegala Rural Development Project (Part I, para. 4). Other main lessons learned are that! (a) in view of the involvement of a multitude of line agencies in project implementation, the project design should have focussed more clearly on organization and management aspects to ensure effective coordination among these agencies; (b) supervision of th; pcoject would have been easier if it did not include such a wide and diverse range of subprojects; (c) the project period should bave been longer than envisaged given the inevitable delays in implementation when such a large number of line agencies are involved and in view of che shortage of local counterpart funds; (d) allocation of adequate O&M funds, especially for irrigation schemes and roads, should have been incorporated as a requirement under the project, including the assumption by beneficiaries of responsibilities for O&M of the irrigation schemes; (e) greater participation in rehabilitation and O&M of irrigation schemes, rural roads, and water supply schemes by beneficiaries would be required in order to meet their needs; and (f) more emphasis should be given to the implementation of water management programs aimed at more efficient and effective water utilization, including the formation of water user groups for this purpose.  PROJECT COMPLETION REPORT SRILAR SECOND RURAL DEVELOPMENT PROJECT (Credit 1079-CE) PART I. PROJECT REVIEW FROM BANK'S PERSPECTIVE Project Identity Project Name Second Rural Development Project Credit No. 1079-CE Vice Presidential Unit : ASIAG Country : Sri Lanka Sector : Agriculture Sub-Sector Rural Development i Project Background and Sectoral Context 1. Agriculture plays a major role in Sri Lanka's economy. Recognizing the pivotal role of the agriculture sector in stimulating economic growth and rural development, the Government of Sri Lanka's (GOSL) medium-term plans and programs for the sector concentrate on: (a) increasing agricultural and rural employment as well as improving income levels, particularly for the economically disadvantaged people, (b) expanding agricultural production to achieve a high degree of food self-sufficiency and to improve nutritional standards, and (c) increasing export earnings from both traditional export crops (tea, coconut, and rubber) and non-traditional or minor export crops, such as cocoa, coffee, and spices. 2. GOSL strategy for achieving these objectives had emphasized better use of existing resources through institutional arid policy improvements as well as rehabilitation of the production base. Among policies and programs being given high priority by GOSL were: (a) establishment of appropriate price incentives for producers; (b) strengthening agricultural support services such as extension, research, credit, storage, transportation, and marketing; Cc) improving input supply services, especially for seeds, fertilizers and pesticides; (d) expanding and rehabilitating the area under minor export crops; (e) expanding the irrigated area and improving water management; (f) improving and expanding rural roads and electrification; and (g) decentralizing planning and development processes. 3. GOSL started to decentralize the planning and development processes in 1975 by transferring more responsibility and autonomy for carrying out these activities to the country's 24 districts. As part of the decentralization program and to provide an operational content to it, GOSL is undertaking integrated rural development projects in a number of districts. These projects emphasize low-cost, quick-yielding, labor-intensive investments aimed at better utilization of the existing economic and social  2 infrastructures. A major advantage of these projects is their responsiveness to local development needs and priorities. 4. The IDA-supported Kurunegala Rural Development Project (Cr. 891- CE) was the first attempt by GOSL to develop an entire district in an integrated manner. The project closed in 1986 and, as a whole, can be considered a success because it achieved its main objectives of increasing agricultural production, creating employment, and improving the income and living standards of the rural population in the district. It has also served as a model to be replicated in otier districcs, and many of the lessons learned during implementation were incorporated in subsequent rural development projects. To date, 14 projects have been or ara being implemented (three financed by IDA, two by IFAD, and nine by bilateral donors). GOSL intends to have such projects implemented in all the districts. Project Objectives and Description 5. Objectives. The project sought to improve the productivity, income and living standards of people in the two districts of Matale and Puttalam by rehabilitating and improving the economic infrastructure, improving and expanding the provision of inputs and services required by the agriculture sector, stimulating agricultural production, and improving the socia. infrastructure. 6. Components. The project components as appraised were as follows (proportion of total baseline costs indicated in parenthesis): (a) Economic Infrastructure (46Z), including: (i) rehabilitation of existing rrigation systems and improved water management; (ii) improvement of existing roads and construction of new roads; (iii) bringing electricity to villages; (iv) installation of wells in areas deficient in drinking water supplies; and (v) groundwater exploration; (b) Inputs and Services (23Z), including: (i) provision of agricultural credit; (ii) improvement of existing facilities for fertilizer retail outlets, storage of paddy and pesticides; (iii) improvement. of livestock services, and (iv) provision of additional staff and supporting equipment; (c) Agricultural Production (20Z), including: (i) development of minor export crops and coconut; (ii) reforestation/afforestation of degraded land; (iii) improvement of landing and shore facilities at fishing centers; (iv) development of sericulture; and (v) provision of staff and office facilities; (d) Social Infrastructure (7Z), including: (i) construction and equipping of hospital facilities (operating theater, surgical ward, operating clinics, and staff quarters), and provision of ambulances; and (ii) improvement of education and vocational training through the provision of classrooms and workshops, teaching aids and other supporting equipment, and staff quarters; and  3 (e) Project Coordination and Management (1'?), including provision of additional staff, office equipment, vehicles, and consultant services tc assist in project implementation. Ppv1ect Design and Organization 7. Desig. Project design had used as a model the Kurunegala Rural Development Project and the experience gained from its implementation. The main assumption underlying the concept and formulation of the project was that the existing economic resources and social infrastructure in the two project districts (Matale and Puttalam) were substantially underutilized and that their increased and more efficient utilization, mainly through the upgrading of institutional services and infrastructure and provision of finance for private sector investments in the production base, would lead to rapid and significant improvement in the economic and social well-being of the people in the districts. The pro4ect covered a wide range of development aspects and attempted to complement directly productive investments, such as rehabilitation of irrigation schemes and development of coconut and minor export crops, with limited investments in agricultural support services, transportation, health, education, etc., to improve the utilization and effectiveness of existing facilities, to reduce regional disparities, and to develop the social infrastructure for growth in the long term. 8. The project was generally well prepared. The timing of the project was appropriate because it came at the outset of the liberalizatioa of economic policy in the country and helped to stimulate further private investments in the production base. Additionally, the project supported the newly introduced decentralization program at a time when it was critical to demonstrate that this approach would lead to increased respunsiveness of development efforts to local needs and priorities and would promote a more balancea regional development. Perhaps the weakest design issue was the planned implementation period which was too short. 9. Organization. Due to its diverse development operations and the wish to use existing agencies rather than creating a new institutional setup for this purpose, the project was implemented by about 25 agencies (including two banks) from 12 ministries. To achieve project coordination, a Project Steering Committee was established at the national level and a Project Coordinating Committee in each of the two project districts. In each district, a project office, headed by a Project Director, was set up to monitor, supervise and coordinate implementation of work programs of the various agencies in the district. The Regional Development Division of the Ministry of Policy Planning and Implementation provided overall supervision and coordination of the operations of the project offices. 10. To some extent, preparation and implementation of certain project investments and activities suffered from a lack of clarity in defining the role and expected coordination among the agencies concerned, particularly where different agencies were called upon to work on related at)ects of the same type of investment or activity. An example of this was clearly observed in coconut development which involved the participation of three agencies - the Coconut Cultivation Board admiristered the subsidy programs for plantings of this crop, the Department of Agriculture provided the agricultural extension work, and the Department of Minor Export Crops was responsible for  4 carrying out the program for intercropping in coconut holdings. Other examples are the involvement of (a) the Irrigation Department and the Department of Agrarian Services in the rehabilitation and operation of irrigation schemes. and (b) the Department of Highways and Department of Local Govornment in the rehabilitation and maintenance of rural roads. 11. Although the use of a multiplicity of existing agencies would have contributed to a slower than anticipated project implementation, overall organizational experience has been positive and more cost effective than the establishment of yet another agency. Pro1ect Imp1elentation 12. A summary review of the implementation experience of the project compon*nts with emphasi- on the directly productive components and the critical vAriances between planned targets and the actual results achieved are preser:ted in the following paragraphs. 13. Implementation of the project was initially phased over a five-year period from mid-1981 to mid-1986. However, implementation did not progress as anticipated and the project had to be extended four times, each time by one year, to achieve the set targets and to maximize utilization of the IDA Credit. Actual disbursements amounted to SDR 22.86 million or about 891 of the total IDA Credit of SDR 25.6 million. Actual project cost in local currency was about 33Z higher than SAR estimate mainly due to domestic inflation; but in US dollar terms actual cost was lower (US$39.3 million compared to SAR eb 'mate of US$50 million) due to devaluation of the Sri Lankan Rtupee (Part III, Table e^) 14. The main reasons for the delay in project implementation were: (a) manpower saortages in the initial stages in some of the implementing agencies; (b) severe budgetary constraints resulting in shortage of local counterpart funds; (c) worsening of the security situation in later years of the project which caused project staff and contractors to stay away from project sites which in turn had led to disruptions of project works; (d) shortage of construction materials; and (e) delays in procurement of project goods. Despite these delays, implementation of the project as a whole had generally progressed satisfactorily in that the physical targets set during appraisal were essentially achieved by the end of 1987, only about a year and a half behind schedule. The latter two extensions of the Closing Date were granted essentially to enable fuller utilization of the IDA Credit, the local currency equivalent of which had increased substantially due to continuous and significant depreciation of the Sri Lanka Rupee. Such exten*ions were justified given the fact that the project provided a line of finance for rural development that was insufficient to complete essential development works in the districts. Fuller utilization of the IDA Credit had in turn enabled the actual physical targets achieved by many of the project components to exceed significantly appraisal estimates and reduce the works ieft for future development initiatives. In providing the required housing, office facilities, and transport vehicles, the project enabled the staff of the concerned agencies to carry out their tasks and to do them more efficiently. 15. Irritation and water management. Accounting for about one fourth of the total project cost, the rehabilitation of irrigation schemes and improved water management to raise paddy production and yields constitute the single most important component of the project. A total of 15 major and 240 minor irrigation schemes covering a total of about 29,500 ac were to be rehabilitated. The rehabilitation actually completed amounted to 21 major and  5 360 minor irrigation schemes covering a total area of about 39,100 ac. This significant increase in actual performance vis-a-vis the appraisal targets was made possible by the substantial increase in the local currency equivalent of the IDA Credit (para. 14). 16. Although the number of schemes and area rehabilitated significantly exceeded appraisal targets, the absence of an effective institutional arrangement, which would reflect the needs of farmers regarding the design and implementation of the rehabilitation works, was a major deficiency. As a result, some farmers did not feel sufficiently committed to repair and maintain these schemesl". Other important constraints include: (a) programs for improved agricultural and water management practices had not been implemented in some of the rehabilitated schemes; (b) reluctance by the Department of Agrarian Services (DAS) to take over O&M for some of the minor irrigation schemes rehabilitated by the Irrigation Department (ID); and (c) lack of downstream works at some of the minor irrigation schemes. 17. DAS is responsible for O&M of minor irrigation schemes, and ID for the rehabilitation of all irrigation schemes. Shortage of .-unds may have prompted DAS's reluctance and therefore often delayed takeover of the rehabilitated minor irrigation schemes. ID and DAS are also responsible for the implementation of improved water management programs in the major schemes and minor schemes respectively, the underlying hypothesis of which is that water saved during the main rainy season (Maha) can be used to cultivate a second crop during the season of lower rainfall (Yala). Sowing method for the first crop, rotational use of water during Maha, crop diversification during the second crop season, use of short maturity varieties, and mobilization of farmer group action through education and training are some of the key elements in efforts to improve the efficiency of water use. Implementation of water management programs have generally been satisfactory. 18. The rehabilitation of the irrigation schemes and improved water management, which had increased the amount and reliability of water, together with increased cropping intensities and use of fertilizers and other inputs significantly increased paddy yields in the project areas to an estimated average amount of about 65 bushels (1.36 metric ton) per ac compared with about 45 bushels (0.94 metric ton) at appraisal. Total annual paddy production has been estimated to increase from about 34,000 metric tons at appraisal to 71,200 metric tons at full development. 19. Agricultural credit. The project provided agricultural credit to facilitate farmers' access to inputs. In terms of project costs, this component is the second most important accounting for nearly one fifth of the total amount. Credit funds to borrowers were channelled through the Bank of Ceylon and the Peoples Bank. These two banks granted a total of 27,715 loans for Rs 182 million. The amount of credit disbursed to borrowers was double 1/ This shortcoming has been recently corrected. Under the recently approved National Irrigation Rehabilitation Project, a farmer organization must be established for each scheme proposed for rehabilitation, and farmers have to contribute at least 1OZ of rehabilitation costs and undertake to operate and maintain the scheme. It is expected that this requirement will apply to rehabilitation works undertaken under rural development projects.  6 the amount envisaged during appraisal. About 59% of the credit financed seasonal inputs (fertilizers, pesticides, seeds, etc.) mainly for paddy production, and the remaining 41% financed medium and long-term loans for procurement of farm equipment (mainly for irrigation and cultivation purposes, including hand sprayers), plant nurseries, storage facilities, and transport vehicles. The recovery rates of project loans averaged about 85Z. 20. The increased use of seasonal inputs, made possible by credit funds under the project, contributed significantly to the increase in paddy production as they were an essential complement to the effective use of the increased amount and reliability of water brought about by rehabilitation of the irrigation schemes. Increased use of agricultural inputs was also facilitated by the improved facilities for these inputs financed under the agricultural support services' component, although this component accounted for less than one per cent of the total project cost. 21. Forestry. Sizable parcels of land in the two project districts (some 402 in Matale and 20% in Puttalam) have rough/steep topography and shallow soils, most of which are best suited to be under forest. As it is, some of these lands (government owned) are being used for Chena (slash-and- burn) cultivation and have been degraded through over-exploitation. This over-exploitation has also been taking place on some tea plantations which are steep-sloped and in the catchment areas of the Mahaweli Ganga in Matale. 22. To reverse the situation and to arrest further soil erosion and its adverse consequences, GOSL has been carrying out a program of reforestation and afforestation of these lands. The trees planted include timber, pulp and fuelwood species. In support of this program, the project provided for the afforestation and reforestation of 16,500 ac. The actual area of afforestation and reforestation achieved, however, amounted to 13,038 ac. The shortfall, which was due to lack of planting materials, is being rectified by the additional plant nurseries established under the project. Four nurseries have been set up instead of three envisaged during appraisal. The tractors provided under the project have all been procured. 23. Coconut development. At appraisal, it was envisaged that the project would provide funds to assist privately-owned coconut holdings to: (a) improve the productivity of about 10,000 ac by improvements in drainage, flood protection and water conservation; (b) rehabilitate about 7,500 ac by replantirg and planting; and (c) intercrop 1,700 ac with minor export crops, mainly coffee, cocoa and pepper. Responsibility for the first two categories was assigned to the Coconut Cultivation Board (CCB), and the third category to the Department of Minor Export Crops (DMEC). The actual results achieved for the three categories are 21,271 ac, 12,193 ac, and 281 ac, respectively. The significantly higher tha.i envisaged acreages achieved for the first two categories were due to increased farmer demand encouraged by the provision of GOSL subsidies to defray part of the cost of these activities, and by effective extension work undertaken by CCB. On the other hand, the poor performance of intercropping was due to the reluctance of farmers to accept crops selected for this purpose because the crops were unfamiliar and in some cases had long maturity periods and high development costs. Another contributing factor was the shortage of plants which farmers had wanted to use for intercropping.  7 24. In addition, a 5-ac coconut research substation and a fruit tree nursery were established as envisaged during appraisal. The research substation is conducting applied research and the nursery has already begun to supply fruit seedlings to farmers for intercropping in their coconut holdings. 25. Although the acreages achieved under productivity improvements and rehabilitation were significantly larger than anticipated, the quality of implementation of these activities had suffered from a number of deficiencies: (a) contour drains to prevent soil erosion were poorly constructed; (b) inappropriate time of planting, poor maintenance and subsequent death of plants which required their removal and replacement; (c) delays in subsidy payments; (d) individual subsidies which were paid in several installments did not coincide with the timing and costs of the different phases of development; and (d) lack of coordination among the agencies concerned, of which good examples are (i) the absence of proper timing of coconut planting and intercropping which comes under the responsibility of CCB and DMEC respectively, and (ii) overlapping jurisdiction for extension support by CCB, DMEC and the Department of Agriculture (DA). This led to implementation delays, higher costs, and resultant improvements/rehabilitation which were less effective than expected. Efforts have been made to have a unified extension service but this has yet to materialize. The advisory services of the CCB, DMEC, DA, and the Livestock Department continue to render technical advice in their respective fields. 26. Minor export crops. This component was included in the project to support GOSL's crop diversification efforts, improve the capability of DMEC staff and its operations, and generate additional employment and incomes to farmers, especially those owning small plots commonly referred to as home gardens. It was envisaged at appraisal that the project would provide funds to: (a) farmers to replant/new plant 7,890 ac of minor export crops consisting mainly of pepper, coffee, cocoa, cardamon, and other spices; and (b) DMEC to (i) establish a nursery to provide improved planting material to farmers, (ii) strengthen the capability of the DMEC research institute on cardamon research, pest and disease control, cocoa fermentation, and crop processing, (iii) establish a center to train DMEC staff and farmers on cultivation practices and crop processing, (iv) construct an office building and additional staff quarters to alleviate the shortage of office space and staff housing, and (v) procure vehicles to improve staff mobility so as to facilitate their work. 27. The total area of minor export crops actually planted and replanted amounted to 10,525 ac or about one-third higher than the appraisal estimate (para. 26). The significantly higher than anticipated demand for such plantings was due mainly to the profitability of the crops and the fact that an important share of the cost of their establishment had been financed by subsidies provided by GOSL. However, the quality of implementation of this activity suffered from inappropriate time of planting and poor maintenance of plants, delays in subsidy payments, and subsidy instalment payments not coinciding with the timing and costs of the different development phases. These deficiencies, which also afflicted plantings under the coconut development component (para. 25), resulted in delayed and added implementation costs. 28. As for the other appraisal targets (para. 26), they have all been achieved. Thus, the nursery, albeit smaller than envisaged, is producing and  8 providing improved planting material to farmers; the training center for DMEC staff and farmers has been operating as intended; and the work being conducted by the research station include carrying out germ plasm collection, yield trials and plant hybridization. 29. Fisheries. This component, accounting for only about one percent of total project cost, was expected to improve landing and shore facilities at major fishing centers in Puttalam District. The improvements included construction of jetties at three sites, establishment of a fishery service center to facilitate the operations of fishermen and fish traders, and installation of beacon lights at selected locations to ensure safer and quicker access by fishermen to fish landing centers. The component also provided transport vehicles and housing for fisheries staff to improve their operations. In terms of their physical targets set during appraisal, all the improvements and facilities were essentially achieved. Indications are that the improvements and facilities have increased the operational efficiency of fisheries activities in the district. 30. Agricultural support services. This component provided additional and improved facilities for: (a) the sale and storage of fertilizers, pesticides, and seeds; and (b) veterinary care of cattle, including their handling. Facilities under (a) were implemented by DAS and those under (b) by the Livestock Department (LD). In addition, the component provided additional office space, staff housing and transport vehicles for DAS and LD to improve their operations. The targets set for these facilities at appraisal have been achieved except for a few minor items (cattle handling and seed/pesticide improvements), the actual achievements of which had fallen short of the appraisal estimates due to requirements which turned out to be less than envisaged. The shortfalls were not significant. The facilities provided, although small (one percent) in terms of total project costs, have expanded and improved the quality of the services concerned, particularly the establishment of new veterinary clinics (4) and improvements to existing veterinary clinics (4) which have been providing farmers with better access to and quality treatment of their cattle. 31. Sericulture. It was envisaged at appraisal that this component would finance the establishment of a land settlement scheme for 125 settlers who would engage in sericulture. The component was not implemented because it was not technically and financially feasible to establish such a scheme, including the prospective settlers' inexperience with sericulture and non- availability of a suitable site for the scheme. 32. Groundwater exploration. This component was expected to: (a) strengthen field investigations and advisory capabilities of the Water Resources Board (WRB) in Puttalam by alleviating the shortage of office space, equipment and transport vehicles for the staff concerned; and (b) finance the drilling of about 60 investigation boreholes and 60 test wells to facilitate the siting of shallow and deep tubewells which farmers were expected to construct with loans under the agricultural credit component. It was subsequently determined that investigation boreholes were uneconomic and not as useful compared to test wells for the purpose intended so that a significantly higher number of test wells (115) were drilled in place of investigation boreholes of which only five were drilled. Ninety-one of the test wells were subsequently converted to tubewells to supply drinking water  9 (para. 39). The appraisal targets for all the items to strengthen WRB had been met. 33. Roads. This is the third largest single component, accounting for about 13Z of total project costs. Under the component, rural roads (essentially feeder roads) were to be constructed and rehabilitated, and smaller roads used by man-drawn or animal-drawn carts and pedestrians upgraded. The Department of Highways (DH) was to carry out these activities using labor intensive methods supplemented by some equipment, including vehicles, to be provided by the project. The actual results achieved amounted to about 170 miles of roads rehabilitated, 255 miles of smaller roads upgraded, and 7 miles of new roads constructed. The corresponding mileage numbers set at appraisal were generally lower at 144, 175, and 7. The significantly higher mileage actually rehabilitated was due mainly to the increased availability of project funds in local currency equivalent (para. 14) which had also enabled a bridge and causeway, not envisaged during appraisal, to be constructed. The equipment and vehicles were provided to the DH as envisaged. 34. Although the actual mileage achieved was satisfactory, implementation of this component suffered from a number of constraints. These included poor quality of construction/rehabilitation because of inadequate supervision of the works, shortage of road construction materials, and procurement delays which led to delayed construction and added costs. Deterioration of the security situation in recent years, by preventing access of contractors and supervisors to project sites, contributed to inadequate supervision and also disrupted and delayed the implementation of project works. Inadequate supervision was also due to shortage of qualified personnel during the initial period of the project. 35. In addition, many of the rehabilitated and new roads have suffered from inadequate maintenance. Although funds allocated for this activity had increased in recent years, they have continued to be inadequate. DLG which was responsible for this matter has repeatedly cited the shortage of funds for this purpose as the major contributory cause. 36. Rural electrification. Under this component, a number of selected villages with good potential for rapid economic growth were provided with electricity to stimulate such growth. The electrification process involved mainly constructing 33 kv and 11 kv lines, substations, and low tension lines. Eighty electrification schemes were implemented instead of 10 schemes envisaged during appraisal. This component accounted for about 6 percent of total project cost. 37. The electrification schemes had benefitted significantly the economies of the villages concerned. In particular, rural enterprises were expanded and the efficiency of their operations improved. The well-being of households with electricity also improved. 38. Health. To alleviate the shortage of basic medical services and to improve their quality, this component was expected to provide: (a) a new operating theatre and surgical ward at a district hospital in Puttalam, a new operating clinic and outpatient units in Matale, and basic surgical laboratory, and equipment to these new facilities as well as to other hospitals; (b) living quarters for medical officers and other hospital staff;  10 and (c) ambulances, transport vehicles and equipment (mainly electrical supply schemes) for outlyintg hospitals. The targets set at appraisal for all the above have been achieved. In addition, the project financed two surgical wards, five maternity wards, nine health centers, and improvements to five hospitals which were not envisaged at appraisal. The actual results achieved compared with the appraisal estimates are presented in Part III, Table 4. It was quite apparent that implementation of this component improved significantly the welfare and well-being of many people in the districts. For example, more people in outlying and remote areas now have access to the services of medical officers, whereas in the past, these officers were unwilling or unable to work in these areas because proper living quarters for them were not available. 39. Water supply. Under this component, 250 shallow wells for drinking water were constructed as envisaged during appraisal. However, installation of a number of wells had been delayed and completed at higher costs because handpumps were corroded by the high iron content of the water, end had to be replaced. In addition, three water supply schemes and 235 tubewells, including 91 converted from test wells (para. 32), were constructed, although they were not included in the project at the time of appraisal. 40. Education and vocational training. This component was designed to improve science, agricultural and technical training in the public school system, to attract and keep qualified teachers in more remote areas, and to upgrade selected senior secondary schools. It provided for: (a) construction of science rooms for junior secondary schools and workshops (for teaching of carpentry, metalwork, weaving, and other technical subjects), and procurement of basic science equipment for these facilities and existing science tunits; (b) construction of living quarters for teachers serving in remote areas; (c) upgrading the facilities for the teaching of science subjects in senior secondary schools and primary schools; (d) installation of wells for drinking water at schools in remote areas; and (e) procurement of t-ansport vehicles for personnel engaged in supervising the construction of these facilities. The appraisal targets set for all the above items have been met and exceeded in the case of a number of items. The actual results achieved, presented in Part III, Table 4, also included the provision of additional facilities and improvements not envisaged at appraisal, e.g. woodwork training center and renovation of existing weaving centers. 41. Project coordination and management. To enable project staff to carry out their functions efficiently, funds under this component were expected to finance: (a) construction of a project office building in Matale, and renovation of an existing office building to house the project office in Puttalam; (b) additional staff, staff quarters, transport vehicles, and office equipment and supplies for project coordination and management in the Matale, Puttalam, and the Ministry of Policy Planning and Implementation: (c) technical assistance in the form of local supervisory consultants to assist project directors and the Ministry in supervising and monitoring progress and quality of works, preparing reports and processing of reimbursement applications; and (d) employment of a local research institute or consultants to evaluate the project. The appraisal targets for all these items have been met. 42. Coordination and management of the project had generally been satisfactory. The four main reasons which had contributed to this were:  11 (a) the project offices were established on time; (b) the project offices had the same project directors throughout virtually the entire project implementation period (except for the recent replacement of the project director in Matale), so that the continuity of management provided had exerted a positive impact on the implementation of the project; (c) suitably qualified local supervisory consultants were employed which provided the required technical expertise lacking among project staff; and (d) effective coordination and close supervision of the project provided by the Ministry concerned. 43. Local consultants were recruited to undertake evaluation studies on selected project components. These studies were generally not well carried out mainly because they were not properly focussed and some had drawn conclusions based on data which were inadequate and seemed unreliable. Use of vague and unsubstantiated statements in some of the studies was not helpful and could have been misleading. Also, data of no direct relevance for evaluation were included in some of the study reports which therefore tended to be unnecessarily lengthy. A main contributory factor for the generally poor quality of the studies as reflected in their reports was the use of inexperienced and not properly qualified people to carry out the studies. However, some f.ndings of the studies were useful, particularly on irrigation, which in identifying poor construction works (also confirmed by supervision missions) had prompted corrective measures to be carried out. This in turn had contributed to improved implementation of the project. Project Results 44. Project objectives and physical achievement. Overall, the project can be considered a success as it has achieved its main objectives, namely to improve the productivity, incomes and living standards of people in Puttalam and Matale districts. Indications in support of this are: (a) the actual results achieved equalled or exceeded the physical targets set at appraisal for essentially all the project components; and (b) significant increases in estimated yields and production from paddy areas in rehabilitated schemes and coconut holdings and new/improved plantings of minor export crops (para. 46), as well as other direct benefits received by project beneficiaries (Part III, Table 6). 45. Project impact. It would not be feasible to provide a full assessment of the impact of the project because some data would only be available in a few years time, some would be unavailable, some not quantifiable, and some of the available data would not be reliable. However, there are indications, based on available data, which clearly demonstrate that the overall impact of the project has been favorable. These are discussed in the following paragraphs. 46. The project had rehabilitated about 39,100 ac of irrigation schemes planted mostly to irrigated paddy, improved 33,460 ac of coconut holdings (through drainage, flood control and water conservation measures, and replanting), and replanted/new planted 10,500 ac of minor export crops. Together with increased fertilizer use and improved cultivation practices, it has been estimated that annual per ac average yields fer paddy have increased from about 45 bushels to 65 bushels (2.32 metric ton to 3.35 metric ton/ha), and for coconut holdings from 2,200 nuts to 3,100 nuts, or annual increments of about 37,200 metric tons of paddy and 28 million coconuts (35 million at  12 full development). The annual incremental production of the main minor export crops at full development would amount to an estimated 1,F40 metric tons of pepper, 1,100 metric tons of coffee, and 600 metric tons of cocoa. 47. Project objectives of alleviating the overall living conditions of the rural people and their work skills by improving the roads network, hcalth services, rural electrification, drinking water supply, and educational facilities were substantially achieved. Although not quantified, expansion of the rural roads have clearly benefitted the rural communities by improving access to markets and non-farm employment opportunities, facilitating the movement of labor and farm inputs, and improving the delivery of other rural services. Bringing electricity to villages which had none before not only improved the well-being of households but also stimulated the expansion of small rural industries and improved their operations. Access to better quality education and medical services improved the health and skill performance of the people. Some diversification of the agricultural production base and income was also promoted by the minor export crops component and to a lesser extent by the fisheries and livestock components. 48. Economic rate of return. At appraisal, economic rates of return (ERR) for the irrigation, coconut, minor export crops, sericulture, and forestry components were estimated, but no ERRs were estimated for the education and health components and other social and economic infrastructure items since many of their benefits and impact could not be determined and quantified. Estimating an ERR for the project as a whole would therefore not be feasible. However, rough estimates (because of the deficiency of data on benefits and costs relevant to some of the components)indicate that ERR would be above the opportunity cost of capital in Sri Lanka. It should, however, be borne in mind that the recalculated ERRs are still rough estimates. The actual contributions to yields, production and income increases for certain crops would not be known until they reach maturity and attain full development production levels. Project Sustainability 49. As mentioned earlier, the project was extended by four years in order to: (a) achieve the physical targets set during appraisal, (b) accommodate the slower than anticipated pace of project implementation due to the worsening security situation and shortage of local counterpart funds resulting from government budgetary constraints, and (c) maximize utilization of the IDA Credit tparas. 13-1.4). The budgetary constraints, which show no sign of getting any better, can adversely affect the O&M of certain project irvestments and therefore their sustainability. Of particular concern are the significant shortfalls in O&M funds allocated in recent years for project roads and irrigation schemes. With respect to the latter, GOSL's recent decision to transfer the responsibility for O&M of irrigation schemes to farwer organizations should improve the situation and result in improved O&M. 50. Although detailed data are not available, it is reported that following cessation of subsidies under the minor export crops and coconut components for construction of drains and terraces (to prevent water-logging, conserve water and minimize soil erosion), farmers do not provide sufficient  13 resources for their proper maintenance and upkeep. On the positive side, however, they have increasingly resorted to use of improved inputs, planting materials and cultivation practices. 51. The average recovery rate of loans (about 85Z) provided under the agricultural credit component is satisf&ctory. This rate of recovery, together with improved procedures to raise the recovery rates, indicates that the component appears sustainable. Bank Performance 52. The Bank's involvement has been quite intensive throughout the project cycle. An IDA mission visited Sri Lanka in August 1979 to assist GOSL in preparing the project. GOSL (concerned line agencies) prepared reports on various components proposed for inclusion in the project which were presented to an IDA mission when it visited Sri Lanka in November 1979 to appraise the project. However, the reports did not provide an adequate basis for appraisal of the project as many of them lacked the required data. These data were subsequently provided to the mission following its return to Washington and the appraisal was completed in June 1980. In hindsight, it seems that the timing of the appraisal mission was premature and project design could have improved had more time been available for preparation. 53. A total of fifteen supervision missions were fielded over a nine year period. Although most of these missions were multipurpose, supervising also other projects, they were all considered beneficial to the project. The project officials, in particular, indicated that the missions helped them identify and correct deficiencies in design and layout of project facilities and in many aspects of project implementation. 54. Supervision missions concentrated on monitoring physical achievements against appraisal targets and identifying and recommending corrective measures especially for design, engineering, and civil works aspects of project facilities, including quality of construction and maintenance works. On the other hand, the absence of agriculturists in many of the missions resulted in less-than-desired emphasis on technical aspects (agronomic and cultivation practices) implemented under the agriculture and related components. This situation can also be applied to the social infrastructure components (health and education). 55. The main lessons learned from the implementation of the project are: (a) supervision would have been easier if the project did not include such a wide and diverse range of subprojects; (b) in view of the involvement of a multitude of ministries and line agencies in project implementation, the project design should have focussed more clearly on the project's or ,dnization and management; (c) project period should have been longer than envisaged given the inevitable delays in implementation when such a large number of line agencies are involved and in view of the shortage of local counterpart funds; (d) greater attention should have been paid to ensure sustainable O&M of project facilities after completion of the project; (e) the need for greater beneficiaries' participation in rehabilitation of irrigation schemes; and (f) more emphasis should be given to the implementation of water management programs aimed at more efficient and effective water utilization, including the formation of water user groups for this purpose.  14 Borrower Performance and Project Relationship 56. The budgetary restrictions of the borrower (GOSL) which resulted in shortages of local counterpart funds contributed to the delays in project implementation. Although coordination of project activities among the line agencies proceeded without any major problem, the overlapping, and in some cases, not clearly defined responsibilities of some of the agencies also delayed and added to the cost of project implementation. Much of the data required for ex-post evaluation of the project, especially its benefits and impact, were not collected and assessed by the project offices. Another problem was the lack of a well-planned training program for project staff. Except for these setbacks, the borrower generally performed well throughout the entire project cycle. The two district governments gave the project their full commitment and support. Competent project directors were recruited and retained throughout virtually the entire project period. All the concerned ministries and line agencies as well as the project officers collaborated fully with the Bank's missions during the project preparation and appraisal stages, and were supportive to the supervision missions and receptive to suggestions for improvements. The borrower complied with all the agreed covenants (although audited accounts had been received by the Bank with some delay). 57. Bank relationship with the government agencies at all levels, including the two project offices which were responsible for the overall implementation of the project, has been good and effective. Consulting Services 58. Local supervisory consultants were employed and retained throughout the entire project period to assist the two project offices in their implementation of the project except rehabilitation of the major irrigation schemes (para. 59). The consultants were used mainly to: (a) prepare/review plars and designs for civil works to ensure that they were in compliance with agreed specifications and guidelines; (b) check all completed works to determine the quality and quantity of work; (c) inspect field works; (d) notify the project offices regarding compliance of agency-certified works with agreed criteria, specifications and guidelines; and (e) assist in preparing reimbursement applications, budgets, progress reports and work programs, and procurement of project goods. These services, which were generally performed well, could not have been carried out by project staff since they lacked the required technical expertise. The consuA.ant services were therefore an essential input which enabled the project offices to carry out their implementation responsibilities effectively and efficiently. 59. Other local consultants were employed to assist the ID in planning, designing, and costing of rehabilitation works for major irrigation schemes, including preparing tender documents, evaluating bids, and supervising construction works. The consultants were employed only for the first few years of the project period (up to 1986) when the ID lacked the required expertise. Their services, which were generally performed well and completed as scheduled, provided a good start to the rehabilitation of the major irrigation schemes. The consultants also included an agricultural extension  15 specialist and a rural sociologist to advise on the implementation of crop development programs following completion of the rehabilitation works. Proiect Documentation and Data 60. The Staff Appraisal Report (SAR) explained the overall concept and strategy of the project and, along with the project preparation reports, provided detailed guidelines for project implementation. The SAR also provided a useful framework for the Bank and the Borrower to monitor the progress of the project against the targets set. However, the four successive one-year extensions of the project period and the significant depreciation of the Sri Lanka Rupee reduced the usefulness of the SAR as a yardstick to measure the actual performance achieved in financial terms. 61. Despite the fact that the project offices maintained project implementation records in a satisfactorv manner, much data relevant to the preparation of the PCR were not readily available, especially regarding the benefits of the project. This was mainly due to the fact that data were not collected and processed for ex-post evaluation purposes, but rather for short- term monitoring of the project results against targets.  17 IPRWbCT CO4LET.1N A