Sri Lanka - National Irrigation Rehabilitation Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No: 19384 IMPLEMENTATION COMPLETION REPORT SRI LANKA NATIONAL IRRIGATION REHABILITATION PROJECT (Cr. 2260 - CE) May 28, 1999 Rural Development Sector Unit South Asia Region 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 Currency Unit = SLRupees (Rs.) Average Yearly US$ 1.00 Equivalent 1991 1992 1993 1994 1995 1996 1997 1998 41.9 45.4 48.0 49.5 53.5 55.0 59.9 68.6 FISCAL YEAR OF BORROWER January 1 - December 31 WEIGHTS AND MEASURES Metric System ABBREVIATIONS AND ACRONYMS AR&TI Agrarian Research & Training Institute IRMU Irrigation Research Management Unit CMC Central Management Cell MIP Ministry of Irrigation and Power D&F Distributory and Field MTR Mid-Term Review DAS Department of Agrarian Services NE North-East DCA Development Credit Agreement NGO NON-Government Organization E&A Engineering & Administration NIRP National Irrigation Rehabilitation Project ERR Economic Rate of Return O&M Operation and Maintenance EU European Union PCC Provincial Project Coordination Committee FAO Food and Agriculture Organization PD Project Director FAO/CP FAO/World Bank Cooperative Program PIAs Provincial Irrigation Agencies FOs Farmers' Organizations PMCs Project Management Committees FRs Farmer Representatives PPCC Provincial Project Coordination Committee GOSL Govermment of Sri Lanka QA Quality Assurance Ha Hectares Rs Sri Lanka Rupees lAs Implementing Agencies RST Regional Support Team ICR Implementation Completion Report SAR Staff Appraisal Report ICRC International Committee of Red Cross SASRD South Asia Rural Development ID Irrigation Department SCF Standard Conversion Factor IDA International Development Association UNHCR United Nations High Commission for Refugees IIMI International Irrigation Management Institute VIRP Village Irrigation Rehabilitation Project IMF International Monitory Fund WB World Bank lOs Institutional Organizers Vice President Mieko Nishimizu Country Director Mariana Todorova Sector Manager Ridwan Ali Task Leader Nihal Fernando  FOR OFFICIAL USE ONLY IMPLEMENTATION COMPLETION REPORT SRI LANKA NATIONAL IRRIGATION REHABILITATION P'ROJECT (Credit 2260-CE) Table of Contents Page No. Preface .........................................................i Evaluation Summary ........................................................ i PART I: PROJECT IMPLEMENTATION ASSESSMENT .........................................................I A. BACKGROUND AND STATEMENT/EVALUATION OF OBJECTIVES . .....................1 B. ACHIEVEMENT OF PROJECT OBJECTIVES .......................................................3 C. MAJOR FACTORS AFFECTING THE PROJECT ........................................................7 D. PROJECT SUSTAINABILITY .........................................................9 E. BANK PERFORMANCE .........................................................9 F. BORROWER PERFORMANCE ......................................................... 10 G. ASSESSMENT OF THE OUTCOME ........................................................ 11 H. FUTURE OPERATIONS .........................................................11 I. LESSONS LEARNED ........................................................ 11 PART II: STATISTICAL TABLES ........................................................ 13 Table 1: Summary of Assessments 14 Table 2: Related Bank Loans/Credits 16 Table 3: Project Timetable 18 Table 4: Loan/Credit Disbursements: Cumulative, Estimated and Actual 18 Table 5: Key Indicators for Project Implementation 19 Table 6: Key Indicators for Project Operation 20 Table 7: Studies Included in Project 21 Table 8A: Project Costs 23 8B: Project Financing 23 Table 9: Economic Costs and Benefits 24 Table 10: StatusofLegalCovenants 25 ThIs document has a restricted distribution and may be used ly recipients only in the performance of their official duties. Its contents may not othervwise be disclosed without World Bank authorization.  Table 11: Compliance with Operational Manual Statements 26 Table 12: Bank Resources: Staff Inputs 26 Table 13: Bank Resources: Missions 27 APPENDICES A. Mission's Aide-Memoire 28 B. Financial and Economic Re-evaluation 38 C. Cofinancier's (EU) Contribution to the ICR 55 D. Borrower's Comments and Observations on the Draft ICR 58 E. Borrower's Project Evaluation Summary 64 MAP No. - IBRD 22876  IMPLEMENTATION COMPLETION REPORTF SRI LANKA NATIONAL IRRIGATION REHABILITATION PRCOJECT (Credit 2260-CE) Preface This is the Implementation Completion Report (ICR) for the National Irrigation Rehabilitation Project (NIRP), for which Credit 2260-CE in the amount of SDR 21.9 million (US$ 29.6 million equivalent) was approved on June 6, 1991 and made effective on October 21,1991. The credit was closed on December 31, 1998 for the project activities in the Northeast Province. The credit for all other project activities closed on June 30, 1998. Final disbursement took place on May 11, 1999. Following the refund of unutilized balances in the special account, it is estimated that total disbursement will be about SDR 17.5 million (US$24.6 million equivalent). In 1997, the amount of SDR 3.65 million (US$5.0 million equivalent at that time) was cancelled from the IDA credit. An additional amount of approximately SDR 700,000 (US$945,000 equivalent), including SDR 442,500 (US$ 597,657 equivalent) of unused funds in project Special Account, is expected to be cancelled when the credit accounts are closed. Cofinancirng for the project, in the amount of US$4.5 million plus about US$9.2 million equivalent in food aid was provided as grants by the European Union. The ICR was prepared by Mr Nihal Femando (Irrigation Engineer, SASRD), the task team leader, and Mr Raffale Suppa (Economist) of (FAO/CP)Y. The ICR was reviewed by Mr Henry P. Gassner, Principle Operations Officer (SASRD). The Co-financier (EU) provided comments that are included in Appendix C to the ICR. Preparation of this ICR was begun during the FAO/CP mission to Sri Lanka from June 20 to July 3, 1998. The report is based on material gathered from the project files, supervision reports, various project related evaluation and research studies, findings from field investigations, and discussions with Bank staff, project implementing agency staff, project consultants and beneficiary farners. Borrower contributed to the preparation of the ICR by: (a) participating in the ICR mission; (b) providing data and substantive comments on the ICR mission's Aide-memoire (Appendix A); (c) contributing to the statistical information and reassessment of ERR (Appendix B); (d) commenting on the draft ICR. (Appendix D); and (e) preparing its own project completion report and evaluation summary (Appendix E). Some of the Borrower's comments on the draft ICR have been incorporated in the final ICR. Food and Agriculture Organization and World Bank Co-operative program. Mr Suppa was the leader of the ICR mission.   IMPLEMENTATION COMPLETION REPORT SRI LANKA NATIONAL IRRIGATION REHABILITATION PROJECT (Credit 2260-CE) Evaluation Summary Introduction 1. In 1980's the Government of Sri Lanka's irrigation subsector strategy emphasized preserving and improving the performance of existing irrigation schemes. ID)A supported this strategy by financing a series of irrigation rehabilitation projects, including the Village Irrigation Rehabilitation Project (VIRP) (Cr. 1160-CE). The NIRP was proposed in 1987 as a way to continue the rehabilitation of minor irrigation schemes started under the VIRP. The IDA supported the NIRP, particularly because of its strategy of improving the capacity and efficiency of public irrigation services and promoting increased participation of farners' organizations in irrigation system management. Project Objectives 2. The main objectives of the NIRP were to stabilize and increase agricultural production and incomes, and raise the standard of living of project beneficiaries through rehabilitation and improved O&M of existing irrigation schemes. Subsidiary objectives included: (a) upgrading the skills of farmers and staff of the implementing agencies (lAs); and (b) creating viable FOs for managing the rehabilitated schemes. The schemes for rehabilitation were selected after reaching prior agreement with the respective FOs that they would bear: (i) the full cost of O&M of minor schemes and Distributory and Field (D&F) canal system of major/medium schemes after rehabilitation, and (ii) the full cost of O&M of headworks and main canal systems of major/medium schemes beginning two years after rehabilitation. 3. Project objectives were in line with Govemment's irrigation management policies and Bank's sector assistance strategy for Sri Lanka, which inter-alia supported transferring the responsibility of operation and maintenance of irrigation schemes to beneficiaries. However, those objectives were too demanding for the implementing agencies in view of the absence of a coherent agriculture sector policy, effective land markets and water rights. Implementation Experience and Results 4. Overall the project achieved most of its physical objectives. IHowever, it has not yielded and is not expected to yield substantial development results, as evident from project's failure to establish promising conditions to ensure sustainability of project's outcome. 5. The project has reached physical rehabilitation targets fully. It is estimated that all civil works of 1,048 minor schemes and 34 major/medium schemes would be completed by the end of June 1999 (Part II, Table 5). It also established scheme level FOs in all minor schemes, D&F level FOs and Project Management Committees (PMCs) in all major/medium schemes, registered the FOs under Clause 56A of the Agrarian Service Act, and handed-over 64 % of the completed minor schemes and 32 % of D&F canal systems of major/medium schemes to registered FOs. Despite these quantitative achievements, the organizational and financial viability of the FOs remain uncertain. The handing- over process in many schemes is supply-driven, and is symbolized with a written agreement of the handing-over and taking-over, signed between the lAs and FOs respectively. FOs have accepted handing over through sheer necessity rather than by conviction. It is unlikely that those FOs could and .i  would bear the full cost of maintenance of those schemes/D&F canal systems in the foreseeable future, although only a few FOs could and would participate effectively in system operation. 6. Compared to previous projects, NIRP included a number of positive actions to improve the quality of civil works, including the fielding of technical consultants in the regions (Regional Support Teams) and providing training to farmers on basic construction aspects and O&M. The civil work quality assurance system could not be enforced at optimum levels mainly because of inadequate transport facilities, and shortage of qualified and trained Work Supervisors. 7. Overall, the achievement of the training programme for project staff and farmers is considered satisfactory. Overseas long-term and short-term training/study tours for project staff were successfully completed. In-country training was also successfully conducted and included courses for project staff, Institutional Organisers (1Os) and Farmer Representatives (FRs). Training of project staff recorded an implementation rate of 84%. However, training opportunities were not equitably distributed to all implementing agencies; it was largely the Irrigation Department (ID) staff who benefited from the overseas training program. Achievement of IOs and FRs training program is not satisfactory, mainly because of the delayed commencement of rehabilitation itself and the shortage of trainers. 8. The project included provisions for establishing a Dam safety Unit and an Irrigation Research Management Unit (IRMU), and for strengthening the Civil Works Quality Control Unit, all within the ID. While a Dam Safety Unit was established in 1994, the ID was not able to activate it as expected, and consequently, EU support was limited only to short-term training. Although dam safety monitoring is carried out as a routine activity by the ID, the need to upgrade dam safety arrangements and services prevails. An IRMU too was established and institutionalized within the ID. It was able to conduct a number of research studies and seminars, but its performance and effectiveness have declined. At the end, the research issues identified at appraisal yet remain to be addressed. The Civil Works Quality Control Unit has been satisfactorily strengthened. 9. Of the studies envisaged, only the Benchmark and Evaluation study was initiated during the project. The Benchmark was established but final evaluation is yet to be completed. Useful project related technical guidelines, manuals and studies have been concluded by project consultants and by IRMU. 10. Total project cost is estimated to reach US$ 44.7 million by Credit Closure, or about 89% of the original project cost of US$49.8 million. Final overall disbursement is estimated to include IDA's US$24.6 million, or about 83% of the total amount approved of US$29.6 million, EU's grant of US$4.5 million, EU's contribution of US$9.2 million as Food Aid and GOSL's contribution, including FOs' contribution, of US$6.4 million. The project's Economic Rate of IRetum (ERR) has been reassessed at 13.9 % compared to an estimate of 31% at appraisal (Part I, section 25). The main factors that have reduced the economic impact of the project are the less than expected increases in cropping intensity and yield, and the longer than expected implementation period.. The reduced production benefits are attributable to inadequate agricultural policies and farmer incentives (see Part I, section 28), while the delays in implementation were mainly due to delays caused by the late appointment of project consultants and mobilization of project staff and resources in the early years of the project, and weak cash-flow management in the later years. The financial analysis indicates increases of net farm income ranging from 28% to 50% (Appendix B). Although considered satisfactory at farm level, these results are below appraisal expectations. Also, the ERR was calculated assuming adequate post-rehabilitation maintenance of the completed works. However, this is considered unlikely. A sensitivity analysis was carried out to assess the ERR, assuming 20 % reduction in production phased over time due to poor maintenance. For this scenario, the ERR is 10 %. 11. Despite the limited impact, the project strengthened the participatory irrigation management culture amongst the irrigation agency staff and fartners. It planted participatory irrigation Iii  management as a principle and a mandatory role of their technical staff. The project created countrywide precedence for: (a) partial rehabilitation cost recovery from beneficiaries; (b) getting the FOs involved in construction activities to strengthenFOs financial status; amd (c) O&M turnover of rehabilitated schemes to FOs. Project Sustainability 12. Project sustainability is rated unlikely. The O&M of the rehabilitated irrigation schemes remains highly dependent on external financial support. While about two third of the completed minor schemes and about one third of the D&F canal systems of major/medium schemes are reported to have been handed over to FOs, the outlook for sustainability of the project is promising for only a few of the schemes. For the vast majority of the schemes, the sustainability will depend on further efforts, in forming strong FOs to adequately organise O&M operations. Without adequate O&M, the physical condition of the rehabilitated irrigation infrastructure and agricultural production under those schemes are likely to deteriorate soon. Key Factors Affecting the Project 13. Among many factors that affected project outcome and sustainability, the low productivity and poor profitability of irrigated farming in Sri Lanka are the main ones. These two main factors are consequences of the lack of coherent agriculture sector policy, absence of effective land markets and water rights. The overall project implementation has been behind schedule due to a number of reasons: (a) the late appointment of project consultants; (b) the IAs could not mobilise their resources fully; (c) delays in procurement of project vehicles and equipment; (d) the poor cash-flow management after 1996 which caused shortfalls in the availability of project funds to meet with project expenditures at crucial civil work construction periods; (e) the general strike of the Technical Officers of the ID for three months during July-September 1997; and (f) delayed but heavy Northeast monsoon rains during 1994-1997 which obliged the farners to opt for an early Yala ('dry season) cultivation in the following year reversing their consent to forego Yala cultivation to permit rehabilitation works. The project was benefited by periodic quantitative reviewing and approving of project work plans by the Project Coordination Committee (PCC). But the monitoring and evaluation of the project in respect of the institutional aspects and project sustainability issues was not effective. IDA Performance 14. The Bank was not directly involved in project identification/preparation but IDA promptly indicated support for the project as a follow-up of the previous irrigation rehabilitation project (VIRP). The WB Resident Mission assisted on a part-time basis in the preparatory work of the project during 1988-1990. The project was appraised by IDA in November/December 1990 and approved in June 1991. The project objectives were in line with GOSL national strategy for irrigation management and Bank's assistance strategy. Bank's performance in project identification and preparation assistance was satisfactory. 15. Many external factors, not foreseen at appraisal, had a negative effect on the implementation time table, project outcome and sustainability. The feasibility of the project objectives should have been carefully analyzed in the context of the agriculture sector policies, and land and water rights issues. The project overestimated the viability of FOs, and farmers willingness and capacity to finance O&M of rehabilitated schemes. The main lesson that resulted from the VIRP clearly indicated that "... the formation of FOs should come before physical implementation ..." was not given adequate attention at project design. The design of project imlplementation management arrangements and assessment of staff requirements to implement civil works were not adequate. In view of these shortcomings, Bank's performance in project design and appraisal was deficient. 16. Overall, eleven supervision missions visited the project. A comnbined IDA/EU Mid-Term Review (MTR) mission was carried out in May 1995. The early supervision missions assisted the .i.i  borrower to alleviate implementation constraints and accelerate the implementation. However, all supervision missions did not address the project sustainability issues adequately and allowed the project to be completed without accomplishing most of its institutional and social objectives. IDA management did not support borrower's request for an extension of the closing date beyond 30 June 1998 for the project as a whole. However, IDA responded positively to a six-month extension for the project activities in the Northeast in view of delays resulting from civil disturbances. Overall, IDA performance in project supervision is considered partially satisfactory. Borrower Performance 17. The project was prepared by the borrower through a Project Preparation Cell comprised of the ID and DAS staff assisted by consultants. The performance of the borrower is, at this stage, considered satisfactory. But there were a few shortcomings in the detail design as reflected under the section on 'Key Lessons Learned' (Section 20 below). The GOSL performnance in project implementation was mixed. The project implementation pace and progress were unsatisfactory from the inception, but improved substantially towards the end. Apart from completing civil works and forming FOs, the project did not accomplish most of its social and institutional objectives. The quality of the civil works in some schemes were below expected standards and requirements. The sustainability of the project outcomes is unlikely. However, the GOSL is to be particularly commended for its decision to allocate sufficient funds to enable the completion of the ongoing rehabilitation works of the schemes whose completion dates were beyond the credit closure on June 30, 1998. 18. The borrower's compliance with accounting, financial reporting, auditing, procurement, and progress reporting requirements was satisfactory. The legal covenant that envisaged the reorganization of the public irrigation sector was waived by mutual agreement with IDA and the borrower. The co-operation between the borrower and IDA remained firm through the implementation period in spite of the IDA decision to extend for six months the original credit closing date to December 31, 1998 only in respect of the works and studies in the Northeast. Assessment of Outcome 19. Overall, the project's outcome is rated unsatisfactory. Since July 1994, the project has been rated unsatisfactory due to the very slow pace of project implementation of the rehabilitation works and non-compliance with major project covenants. At the end, the project achieved its physical targets and major objectives of increasing agricultural production and incomes of farmer in the project areas, but failed to create viable FOs and to set up an efficient O&M system for the newly rehabilitated schemes. Consequently, the sustainability of the project is unlikely. The sustainability of the project outcomes still remains highly dependant upon external financial support. Future Operations 20. The Government provided finances to enable completion of ongoing rehabilitation of schemes whose completion dates are beyond credit closure. The government has secured EU assistance under 'aftercare' program to consolidate the FO development and institutional development programs initiated under the project. Despite unsatisfactory outcome of the project, the NIRP was a major national attempt to set in place a sustainable O&M system in the irrigation sector. It provides a wealth of lessons and experience for designing future projects. It is, therefore, recommended that the Bank together with the Government, carry-out a post-project evaluation of the status of the FOs, and the O&M arrangements of the schemes rehabilitated under the project. iv  Key Lessons Learned 21. The main lessons learned from the implementation of the project include: (i) without up- front commitments on agriculture sector policy and related institutional reforms, it is difficult to achieve successful irrigation projects; (ii) a situation of low productivity and profitability of irrigated farming systems inhibit the development of viable FOs, and establishment of beneficiary-driven, sustainable O&M arrangement in the irrigation sector; (iii) formings FOs and carrying out rehabilitation at the same time do not work well; before starting rehabilitation works, it is essential to maintain for at least a year, an effective institutional support; (iv) improvements to one scheme affects hydrology and water availability at other downstream schemes in tank cascade systems; this factor needs to be adequately addressed in scheme selection, planning and dlesign of future irrigation rehabilitation projects; (v) it is essential to develop and enforce pragmatic civil work specifications, and contract packaging, administration and quality assurance systems suitable for rehabilitation to ensure sustainability of the rehabilitated schemes; (vi) in line with point (v) above and to assure the achievement of good quality standards, consideration should be given for a pre-qualification system of contractors for larger civil works of similar projects in the future; (vii) the establishment of a maintenance fund to cover O&M costs with actual support from farmers has proved to be one of the possible solutions to the problem of project sustainability; (viii) UN agencies and international NGOs such as ICRC can make valuable contributions in implementing project activities in areas with civil security concerns; and (ix) for national projects involving provinces, project implementation authority and responsibility should be devolved to provinces to the extent possible to ensure efficiency and effectiveness. v   IMPLEMENTATION COMPLETION REPORT SRI LANKA NATIONAL IRRIGATION REHABILITATION PROJECT (Cr. 2260-CE) PART I: PROJECT IMPLEMENTATION ASSESSMVENT A. BACKGROUND AND STATEMENT/EVALUATION OIF OBJECTIVES Background 1. The project was a follow-up of the previous IDA-funded Village Irrigation Rehabilitation Project (VIRP) (Cr. 1160-CE) that completed the rehabilitation of about 1,675 minor scheme4 covering about 44,000 hectares (ha) of irrigated farmlands. Implementation of the VIRP had been adversely affected from 1985 by security problems in seven Northeastern districts. However, by the end of 1990 and as a result of four extensions of the IDA Credit, the government had rehabilitated about 1,675 schemes covering about 44,000 ha. In addition, some 90 schemes covering about 2,600 ha were at various stages of completion at the end of 1990. The National Irrigation Rehabilitation Project (NIRP) was first proposed in 1987 to continue rehabilitation and modernization of minor schemes started under the VIRP. All remaining VIRP works too were scheduled to be fully completed in 1991, using retroactive financing under the NIRP. The IDA support for the NIRP was justified as the latter was expected to: (a) improve the capacity and efficiency of public irrigation services at the center and in provinces; and (b) promote increased participation of farmer organizations (FOs) in irrigation system management. 2. The NIRP was prepared during 1988-1990, appraised in Novemaber/December 1990, and presented to IDA's board on June 6, 1991. Subsequently, an IDA Credit consisting of SDR 21.9 million (US$29.6 million) was signed on July 24, 1991 and the Credit became effective on October 21, 1991. The Credit, to be disbursed over a period of seven years, covered about 59 percent of the total project cost estimated at US$49.8 million. The financing plan included IDA financing of US$29.6 million, EU grant of US$4.0 million equivalent, and Government of Sri Lanka (GOSL) contribution of US$16.2 million. The GOSL contribution included US$9.1 million equivalent of EU 'Food-aid' counterpart funds, and US$2.7 million equivalent to be contributed in labor, or in any other acceptable form, by the beneficiaries of the rehabilitation and improvement works. The IDA financed the cost of rehabilitation and improvement works, formation of farmers organizations (FOs), and procurement of a part of vehicles and equipment. EU grant included funding for training, institutional support and studies, environmental protection, technical assistance, and part of vehicles and equipment. The GOSL financed the remaining project cost except for the folmation of FOs. Project Objectives 3. The main objectives of the NIRP were to stabilize and increase agricultural production and incomes, and raise the standard of living of project beneficiaries through rehabilitation and improved operation and maintenance (O&M) of existing irrigation schemes. Subsidiary objectives included (a) upgrading the skills of farmers and staff of the implementing agencies, and (b) creating viable FOs for managing the rehabilitated schemes. The schemes for rehabilitation were selected on A minor scheme, by definition, commands an area less than 80 hectares. A medium scheme commands an area between 80 hectares and 400 hectares, while a major scheme commands more than 400 hectares. 1  the basis of a selection criteria that required reaching prior agreement with the respective FOs that they would (a) in respect of minor schemes, bear the full cost of O&M of schemes after rehabilitation and improvements are completed, (b) in respect of major/medium schemes, bear the full cost of O&M of distributory and field channels after the rehabilitation is completed, and (c) start bearing the cost of O&M of headworks and main canals two years after rehabilitation is completed. Project Components 4. The project included the following components: (a) Rehabilitation and improvement of about 1,000 minor schemes and 60 major/medium schemes covering about 25,000 ha and 12,500 ha respectively. (b) The establishment of FOs and the introduction of improved O&M practices in all rehabilitated schemes. (c) Training of farmers and staff of the implementing agencies. (d) Environmental protection studies and works. (e) The establishment of three new support units (Dam Safety, Irrigation Management Research and Civil Works quality control units) in the Irrigation Department (ID) and the execution of socio-economic and hydrologic studies. (f) Consultancy services for project planning, implementation, and impact assessments. (g) Procurement of vehicles and equipment. Project Changes After Appraisal 5. The target number of major/medium schemes was reduced from 60 to 35 after an IDA review in July 1994 although the target area was increased to about 13,600 ha. After August 1997, one more medium scheme was dropped from the schedule (Mahawewa, 100 ha) and the rehabilitation of another scheme (Rampathwela, 118 ha) was limited to headworks. IDA did not agree to the Borrower's request to extend the credit closing date for the total project. However, IDA agreed in April 1998 to extend the credit closing date by six months from June 30, 1998 to December 31,1998 to enable completion of the rehabilitation work in the war affected Northern and Eastem Provinces (Northeast), that were commenced only in August 1997 and were ongoing at that time. The credit closing date for rest of the project remained as June 30, 1998. The final work plan included the rehabilitation of 1,048 minor schemes with a combined command area of about 25,000 ha, and 34 medium/major schemes covering about 13,380 ha. The program included 106 minor schemes (3,200 ha) and one major/medium scheme (1,640 ha) in the Northeast. The EU increased the grant component in 1995 by US$ 0.5 million equivalent and set apart US$ 1.0 million equivalent under 'food-aid' grant for an 'aftercare' program to consolidtate the FOs and institutional strengthening programs initiated under the project. GOSL provided sufficient funds under consolidated budget to complete all ongoing civil works after the credit was partially closed in June 1998. Evaluation of Project Objectives 6. At the appraisal, the project's objectives were considered important and relevant in pursuing the Irrigation Management Policy of the government, because the NIRP was designed to (a) increase the participation and responsibilities of the farmers in irrigation system management, and (b) reduce the role of the public sector. The government policy is to establish FOs in all irrigation 2  schemes and hand over O&M responsibility to the FOs by the year 2000. Minor schemes are to be fully managed by the farmers, and major/medium schemes are to be jointly managed by the farmers and the Irrigation Agencies (IAs). The role of the public sector would be gradually limited to providing technical and institutional advice to FOs, and O&M of major facilities such as headworks and main canals. It is in this overall context that the NIRP was considered an important intervention. 7. However, the project objectives to stabilize and increase agricultural production by irrigation rehabilitation and improved O&M, and to institutionalize a farmer-driven O&M arrangement through system turnover to FOs, were too demanding and unrealistic. Within the agriculture sector, over the last 10-15 years, there have been a stagnation in crop yield per unit of land, total annual production, intensity of use of irrigated lands, and fanr family income. These are consequences of the absence of a coherent agriculture sector policy environment, effectively functioning land markets and water rights. As explained in Section 28 below, the land tenure pattern and resulting inefficient land markets have led to under-utilization of land, non-viable land holding sizes, low investments on new agricultural/irrigation technologies, and less profitable subsistence level rice farming. Complete absence of water rights further compounds the problem. These factors explain the stagnation of the domestic food crop a;griculture sector, low productivity and profitability of irrigated farning, farmer reluctance to diversify from rice to high value crops, and farmers limited willingness and capacity to share rehabilitation and O&M costs. Therefore, without a suitable sector policy environment and an up-front commitment on policy and institutional reforms, project's social and institutional objectives were unrealistic and demanding. Also it was too optimistic to expect significant diversification of crops, increase of cropping intensity and yields in a sustainable manner through irrigation rehabilitation and improved O&M alone. B. ACHIEVEMENT OF PROJECT OBJECTIVES 8. General: Although credit was closed on December 31, 1998, the GOSL committed funds for completion of all civil works beyond credit closure. Therefore, the project's quantitative achievements have been evaluated in April 1999, making best judgements as to how much the GOSL was going to disburse, how much of civil works would be completed, and what would be the resulting increase in production. By evaluating the physical progress achieved by the end of April 1999, it was estimated that the rehabilitation of project schemes would be completed by June 30, 1999. The reassessment of Economic Rate of Return (ERR) included costs and benefits of all project civil works though the disbursements would occur after IDA's credit closing date. Based on this analysis, it was found that the project would achieve its objective of stabilizing and increasing agricultural production and increasing income of project beneficiaries immediately after the rehabilitation. But the project failed to create viable FOs and establish an effective and sustainable O&M system in the rehabilitated schemes. Without viable FOs and sustainable O&M system, it is too optim istic to expect that physical condition of the schemes would continue to remain satisfactory and that ,agricultural production and income levels would sustain. 9. Increase in Cropped Area, Cropping Intensity, and Yields: When project achievements are compared with the estimates at appraisal, the following are indicated: (a) as a result of rehabilitation, total cropped area is estimated to reach some 47,500 ha, which is slightly above the appraisal estimate of 46,000 ha; (b) average cropping intensity increased only by about 5 percent which is low, and (c) average unhusked rice yields increased by about 38 percent, or 10 percent below the appraisal estimate (Appendix B). 10. Increase in Agricultural Production: At appraisal, it was envisaged that the project would: (a) stabilize the production of about 100,000 metric tons (mt) of unhusked rice from all project schemes, which would otherwise be lost through further deterioration; (b) increase of production by about 20,000 mt of un-husked rice and about 5,700 mt of other field crops. The final economic 3  evaluation reveals that the production of rice has been stabilized and increased by about 25,000 mt of unhusked rice, in line with the appraisal estimate (Part II, Table 9). There is no evidence that the production of other field crops has increased. 11. Rehabilitation of Irrigation Schemes: At the originally envisaged credit closing date (June 30, 1998), 830 minor schemes and 13 major/medium schemes were completely rehabilitated, while the rehabilitation of the remaining 218 minor schemes and 21 major/medium schemes were ongoing. The ongoing schemes included the 106 minor schemes and one major/medium scheme in the Northeast Province for which the credit closing date was extended by six months until December 31, 1998. At credit closure (December 31, 1998), 971 minor schemes and 22 major/medium schemes were completed. By June 1999, the project would be able to reach the full target (1,048 minor schemes and 34 major/medium schemes with a total combined command area of about 49,000 ha). 12. The rehabilitation works required measures to minimize environmentally harmful activities and alleviate existing problems. To achieve this objective, the activities proposed under the project included three categories: (a) minimizing damage to the eco-system during construction; (b) reducing existing environmental problems; and (c) considering environmental problems at the watershed level. The project was able to reduce existing environmental problems of the irrigation schemes fairly satisfactorily, but overlooked most of the proposed interventions to minimize the damage to eco-system and to consider a 'watershed' or 'sub-basin' approach. Irrigation tanks in the country are generally arrayed in cascades. A tank cascade is a hydrologically interconnected series of irrigation tanks organized within sub watersheds of the landscape. The project did not pay adequate attention to the hydrological interdependence of cascades in the planning, selection and design of rehabilitation works. 13. Procurement and implementation of civil works: The standard specifications developed by the Institute of Construction Training and Development (ICTAD) of Sri Lanka for civil works and which were used for the project are not the best suitable for the nature of civil works typical to rehabilitation of small irrigation schemes. Because the project adopted those specifications without modifying to suit the actual project requirements, there was little flexibility in many instances to enforce the Quality Assurance (QA) system. While this QA system was set in place mid way after the project activities have begun, the necessary QA checks and balances could not be enforced at optimum levels due to a number of reasons resulting in poor construction quality of some schemes: (a) shortage of qualified and trained Work Supervisors (WSs) until mid 1997; (b) Majority of WSs recruited and posted to field sites after mid 1997 lacked prior experience in civil works supervision-- although they were trained on QA aspects by the CMC prior to posting to the project sites, their contribution to the quality of works did not prove to be entirely successful; (c) the transport facilities made available to the QA staff were not adequate; (d) although, the project consultants were expected to certify contractor's reimbursement claims against civil works contracts, in addition to the certifications by engineers of the implementing agencies, the project management failed to establish and enforce this requirement fully; instead, contract payments were made only on the certification of the engineers of the implementing agencies; and (e) When the scheduled civil works items/quantities were completed, rehabilitation of a scheme was declared 'completed' without a final certification of construction quality and structural safety of the completed works being issued by the consultants (since this was not a pre-requisite indicated in the project SAR). However, it would have been best for physical sustainability of the rehabilitated schemes, if the CMC enforced the requirements (d) and (e) above on its initiative. Furthermore, the implementation progress could have been improved if the CMC managed civil works contract packaging, administration, and monitoring more systematically. 14. Creation of Farmer Organizations: As of April 1999, FOs have been established in all project minor schemes. In respect of the major/medium schemes, 207 distributory and field (D&F) level FOs have been formed and Project Management Committees (PMCs) have been set up except in one project scheme. All these FOs have been registered under Clause 56 A of the 4  Agrarian Services Act implying that they are formally accepted FOs. Of these, seven FOs in minor schemes and seven in major/medium schemes have been registered under Clause 56 B of the Act, meaning that they have attained legal recognition. Although not considered a project target, FOs registered under Clause 56 B are considered to have a better chance for long-term sustainability. Despite these quantitative achievements the organizational and financial viability of the FOs and hence their sustainability remains weak. Independent evaluations by the Agrarian Research and Training Institute (ART&I), Irrigation Research Management Unit (IRMU), and International Irrigation Research Management Institute (IIMI) indicate that only a few of the project FOs are reported to be functioning satisfactorily and overall sustainability of the FOs is unlikely. 15. Handing Over Rehabilitated Schemes and Improved O&1!: As of April 1999, 675 completed minor schemes (64 % of the project target) and 67 D&F systems of major/medium schemes (32%) were reported formally handed over to the concerned FOs. Despite these quantitative performance, the organizational and financial viability of theFOs remain uncertain. According to ICR mission's observations, and findings of various project related research studies, handing-over process in many schemes is supply-driven, and is symbolized with a written agreement of the handing-over and taking-over, signed between the lAs and FOs respectively. FOs have accepted handing over through sheer necessity rather than by conviction. Farmers willingness and capacity to fully finance the O&M are poor. A special evaluation study financed under the prDject and carried out by independent consultants in three completed major/medium schemes reported in November 1998 that the O&M performance of the handed over D&F systems were poor. However, there is evidence that many FOs could and would participate in system operation in handed over systems to a fair degree, but it seems unlikely that they could and would bear the full cost of operation and maintenance in the foreseeable future. 16. A related issue was the farmers reluctance to contribute 10% of the civil cost of rehabilitation and improvement works in labor or in any other acceptable means as envisaged by the project. While in some major/medium schemes, the expected 10% contribution was beyond farmers financial capacity to bear, generally farners have shown reluctance to share rehabilitation cost. No accurate and complete records are available to quantify the actual total beneficiary contribution to civil works. 17. The Department of Agrarian Services (DAS) set up O&M funds in 275 minor schemes. Farmers were encouraged and guided by the DAS to undertake civil works contracts of the project to the extent possible. The DAS convinced each FO to contribute 5 % of profits from contracts and each farmer to contribute Rs 500 (US$7) per hectare in cash or in kind to create the O&M fund. For each scheme, the contribution was invested as a fixed bank deposit saving for 20 years; the FO can withdraw money only with the joint consent of the majority of the FO members and the concerned DAS officer. The concept is to use the six monthly interest of this saving deposit to supplement part of the seasonal O&M costs. This is an innovative and commendable effort. The adequacy of the interest money to supplement O&M requirements and long-term effectiveness of this arrangement need to be evaluated yet. If successful, this arrangement would ease off the burden of D&M costs at least partly. 18. Technical Assistance: Project consultants have performed most of their responsibilities satisfactorily. However, although the consultant was responsible for checking and certifying contractor's claims against civil works contracts, the project management failed to enforce that requirement. Instead, the claims were paid on the certification of senior supervisory engineers of the implementing agencies. The consultants produced a number of technical guidelines, design and O&M manuals and training manuals, and developed procedures for civil works quality assurance and monitoring. These are useful by-products of the project. In respect of institutional strengthening support, consultants have to depend heavily on training capacity of IA, that was constrained by lack of trainers. This resulted in a less than optimal performance of the consultants in institutional strengthening in the field. 5  19. Training: Overall, the achievement of the training targets is considered satisfactory. The overseas training included three Master of Science (M.Sc.) courses and short-term training/study tours for 243 staff persons. While the overseas long-term training (MSc) schedule was completed as foreseen, the latter could only achieve an implementation rate of 70 percent (172 staff) because either the suitable courses could not be identified (Women's Role in Farming Development) or there were no funds (Irrigation Management, Institutional and Manpower Development). In-country training was successfully conducted and included ten courses for project staff, four courses for Institutional Organizers (IOs) and six courses for Farmer Representatives (FRs). Training of project staff was successfully carried out recording an implementation rate of 84 percent. However, it was largely the ID staff who have benefited from overseas training opportunities. More project training opportunities should have been allocated to the DAS and Provincial Irrigation Agencies (PIAs) as well. Achievement for IOs and FRs training program was not satisfactory. This is mainly because of the delayed commencement of rehabilitation itself and the shortage of trainers in the field level. No forrnal evaluation of quality and impact of training is available. Assistance was adequately provided under the project for the expansion and upgrading of the four in-service training institutes in Galgamuwa, Pothuhena, Kandy, and Bandarawela, as envisaged. 20. Dam Safety Unit: A Dam Safety Unit was supposed to be established to assist the Irrigation Department (ID) in carrying out regular inspections, surveillance monitoring, and preventive maintenance of storage dams maintained by the ID. The project included provisions for training, vehicles, equipment, expert advice, and funds to cover operating costs of the unit during the project period. With initial support from the project consultants this unit was established in April 1994. However, the ID was not able to activate the unit as expected, and consequently, European Union (EU) support was limited only to short-term training in 1995. Although, dam safety monitoring is now carried out as a routine activity by the ID under the supervision of the O&M Unit, the need to upgrade dam safety arrangements and services still prevails. 21. Civil Works Quality Control Unit: Provisions were made to enhance the capacity of the Quality Control Unit of the ID, including the establishment of nine field laboratories, purchase of vehicles, equipment and increment of operating costs during the project period. Four new field laboratory buildings were constructed, four existing ones were renovated, and the unit was provided with the operational support, including vehicle hire. Although, it received the equipment as late as May 1998, it was able to put in place a comprehensive quality control system, and with the assistance of the project consultants, to provide support to lAs as envisaged. However, the adaptation of the system of quality assurance and quality of project's civil works in some schemes did not reach optimum levels due to reasons explained in Section 13 above. 22. Irrigation Research Management Unit (ERMU): As envisaged, the IRMU was set up in 1992 jointly by the ID with the support of the Intemational Irrigation Management Institute (IIMI). However, from the inception, IRMU did not receive much support from the ID, including the appointment of counterpart staff especially in the non-engineering disciplines. IIMI focussed its attention in research but was not successful in fully establishing the IRMU as envisaged. During the first four years of its life, the IRMU completed a few research studies, produced useful research publications, and conducted a number of seminars. After IIMI completed its assignment for IRMU in 1996 it offered a no-cost extension, but the co-financier (EU) did not agree due to poor performance of the IRMU and limited support of the ID to the IRMU. The IRMU maintains a low profile at present. At the end, the research issues identified at appraisal yet remain to be addressed. 23. Other Outputs and Achievements: The project included provisions to continue the hydrological and water management study and socio-economic study initiated and implemented under the VIRP. It also included a Benchmark and Evaluation study in 30 minor and 9 major/medium schemes. In view of the limited outcome of the VIRP studies, the first two studies were not implemented under the project. The Benchmark and Evaluation study, which 6  commenced in October 1995 with the scheduled completion date in December1998 has not been completed yet. * The project generated a number of useful operation manuals, guidelines, and templates on rehabilitation planning and design, construction supervision, O&M, farmer institutional strengthening and training. * Despite the limited impact, the project strengthened the participatory irrigation management culture amnongst the irrigation agency staff and farmers. It planted participatory irrigation management as a principle and a mandatory role of their technical staff. * Although not very successful, the project created countrywide precedence for: (a) partial rehabilitation cost recovery from beneficiaries; (b) getting the FOs involved in construction activities to strengthen FOs financial status; and (c) O&M tumover of rehabilitated schemes to FOs. 24. Estimated Project Cost, Financing, and Disbursements: The total project cost is estimated at US$44.7 million, which is about 89 percent of the original appraisal estimate. Final overall disbursement includes IDA's US$24.6 million (83% of the original Credit Amount), EU's grant of US$4.5 million, EU's contribution of US$9.2 million as "Food Aid" and the govermnent's contribution, including FO's share, of US$6.4 million. Of the IDA Credit (US$29.6 million equivalent), US$5.0 million was cancelled in October 1997 at the borrower's request. An additional amount of approximately US$0.95 million is expected to be cancelled when the credit accounts are closed. 25. Economic and Financial Objectives: The project's Economric Rate of Return (ERR) has been reassessed at 13.9 percent compared to an estimate of 31 percent at appraisal (Part II, Table 9 and Appendix B). The main factors that have reduced the economic impact of the project are the less than expected increases in cropping intensity and yields, and longer than expected implementation period. The reduced production benefits are attributable to inadequate agricultural policies and farmer incentives (Section 28), while the delays in implementation were mainly due to delays caused by the late appointment of project consultants and mobilization of project staff and resources in the early years of the project, and weak cash-flow management in the later years (Section 29). The ex-post financial analysis, details of which are given in Appendix B, indicates increases in net farm income ranging from 28 percent to 50 percent. Although considered satisfactory at farm level, these results are below appraisal estimates, which corresponded to a range from 109 percent to 308 percent (Table 5.2, SAR). The ERR was calculated assuming adequate post-rehabilitation maintenance of the completed works. However, this is considered unlikely. A sensitivity analysis was carried to assess the ERR assuming 20% reduction in production phased over time due to poor maintenance. For that scenario, the ERR is 10% which is considered marginal. C. MAJOR FACTORS AFFECTING THE PROJECT 26. Project Implementation Record: The project was expected to commence in January 1992, but could not take off until May 1993 due to delays in appointment of project consultants (Section 29). With this initial setback, the overall project implementation remained poor until the Mid-Term Review (MTR) in March 1995. The implementation pace accelerated after the MTR remedial measures. The overall project implementation pace was satisfactory by the credit closing date, but the implementation progress, particularly in respect of the physical rehabilitation program and handing over, remained behind the targets. However, it is estimated that all project civil works would be completed by June 30, 1999. Despite heavy project investments on civil works and 7  institutional strengthening of the FOs, the 'software' achievements of the project remain below expectations. The factors that affected the project are as following. 27. Factors Not Generally Subject to Government or Implementing Agency Control: Delayed heavy Northeast monsoon rains in four consecutive years (1994-1997) obliged the farmers to opt for an early yala (dry season) cultivation in the following year reversing their consent to forego one cultivation to permit rehabilitation works. The tense security situation prevented the commencement of project activities in the Northeast Province until mid-1997. Also, the general strike of the Technical Officers of the ID for 3 months during July-September 1997 affected the civil works progress of all major/medium schemes and about 300 minor schemes. 28. Factors Generally Subject to Government Control: Bank's agriculture sector studies in Sri Lanka conclude that lack of active land markets seriously constrain long-term agriculture sector performance. Government owns about 55% of non-plantation agriculture lands mainly in major/medium irrigation schemes. Although lands under minor schemes are mostly privately owned, most of those lands do not have clear legal titles. Population pressure and societal inheritance practices have caused substantial land fragmentation of both private and state lands under those schemes, leaving average farm size below economically viable threshold. While, 72 % of farms are below one hectare, widespread informal multiple co-ownership and further fragmentation in privately owned land have resulted in decrease in effective farm size. According to. the findings of the Bank's study on 'Non-plantation Crop Sector Policy Alternatives (March 1996), farmers' inability to buy/sell land and to expand farm size leads to a situation of small part-time farmers dependent for a substantial part of their farm family income from off-farm employment. In this situation, the combination of low productive subsistence rice farming and off-farm employment provides the greatest income-earning opportunity. These factors reduce interest in diversification of crops, specialize in commercial agriculture, and explain low productivity and profitability of irrigated farming. This situation, compounded by complete absence of water rights, provides limited incentives to farmers for sharing O&M costs or taking over full O&M responsibility of irrigation assets. Until and unless these core policy issues are adequately addressed by the Government, it is difficult to contemplate successful O&M turnover, substantial improvements in crop diversification, sustainable increases in agricultural production and farner income. Neither the Government nor the Bank adequately addressed the implications of these issues on the project development objectives at the project design, appraisal and during implementation. 29. Project start-up was delayed by 18 months due to the government holding up the approval of local partner of the duly selected project consultants, which seems to be politically motivated. This in turn delayed the reappointment of the selected consultants by 18 months. The Provincial Irrigation Agencies, which were in infancy at the time of scheduled commencement of project activities, could not mobilise their resources fully as expected to enable a quick take-off. Also, the government's weak cash flow management after 1996 caused shortfalls in the availability of project funds to meet project expenditures at crucial periods of the construction of civil works. 30. Factors Generally Subject to Implementing Agency Control: At project inception, the Project's Central Management Cell (CMC) failed to ensure timely project readiness conditions. The CMC was set up in one of the implementing agencies (ID) and ID engineer was appointed the Project Director. This project management design and choice of the PD resulted in the CMC being under bureaucratic control of the IDs top management; this, in turn, resulted in restricted independence and autonomy of the CMC to ensure timely project readiness conditions, manage project activities efficiently and effectively, and allocate project resources, facilities, and training opportunities equitably among the ID, DAS, and seven provincial irrigation agencies. Weak procurement management initially by the CMC resulted in the delivery of the first batch of project vehicles and equipment under IDA financing in 1993, two years behind schedule, leading to reduced mobility of project staff. The CMC, together with the Implementing Agencies (lAs) and the project consultants, should have been more vigilant and responsive to ensure high quality of civil works and strongFOs. The project implementation progress benefited by periodic quantitative reviewing and approving of 8  project work plans by the Project Co-ordination Committee (PCC), chaired by the Secretary to the Ministry of Irrigation. However, the PCC was not effective in qualitative monitoring and evaluation with respect to the FO development, project sustainability issues, and providing policy guidance and strategic directions to the CMC. The Provincial Project Co-ordination Committees (PPCCs) too were not effective in addressing policy and strategic issues of the project. D. PROJECT SUSTAINABILITY 31. The project has achieved its objectives of stabilising and increasing agricultural production, as demonstrated in ex-post financial and economic analysis (Appendix B). A large number of FOs have been formed, and O&M responsibility of many completed minor schemes and a few D&F systems of major/medium schemes have been handed to those FOs. Although the ERR reassessment shows acceptable returns assuming adequate maintenance of the completed works, the project has resulted in much smaller effects on increasing cropping intensity and yields, creating viableFOs, and institutionalising a sustainable O&M system for the rehabilitated schemes as; envisaged. Consequently, the sustainability of physical status of the rehabilitated schemes, agricultural production, and farmer income levels, is unlikely. As indicated in Section 25, the ERR is marginal (10%) in the event of reduction in agricultural production phased over time due to poor maintenance. 32. Related to the issue of sustainability are the farmers' willingness, capacity, and commitment to undertake the O&M responsibilities. The O&M of minor schemes are usually not very complex and therefore with further institutional and technical support, the FOs would be able to operate and maintain those schemes without much difficulty. But in respect of the major/medium schemes, where the O&M is relatively heavier, it is doubtful whether the farmers would be willing and capable of undertaking the full responsibilities even of the D&F canal systems. At most, the FOs appear to be willing to and capable of operating the D&F systems and participating in some labour-intensive maintenance tasks under the technical guidance of lAs, but not in providing inputs in cash or in kind to finance full maintenance costs even of D&F systems. The project's expectation that FOs would take over the full O&M responsibility of headwork's and main canals starting two years after the rehabilitation works are completed, seems to be highly unrealistic and would not be met in the foreseeable future. Among many others, reasons for this situation seem to be the low productivity and profitability of present irrigated farming, complex land tenure problems, and lack of a strong commitment for the system turnover and cost recovery by all concerned parties. As a result, the project sustainability is rated unlikely. E. BANK PERFORMANCE 33. The Bank was not directly involved in project identification/preparation. The government prepared the project with technical assistance from a team of local consultants, under the guidance of the World Bank Resident Mission. The Bank supported the government's project preparation efforts because it was consistent with the government's irrigation subsection strategy at that time which emphasised completing projects under construction, and preserving and improving performance of existing irrigation schemes to stabilise and increase agricultural production. The project was consistent with the IDA assistance strategy, which supported those government subsection policies and strategies. Therefore, the Bank's performance in project identification/ preparation assistance is rated satisfactory. 34. Many factors that were not adequately addressed at project appraisal had a negative effect on project's sustainability. The feasibility of achieving the project's development objective should have been carefully analysed in the context of the agriculture sector policy environment and land and water rights issues. At the project design, the Bank overestimated the organizational and financial viability of FOs, and farmers' willingness and capacity to take over the O&M responsibilities. Particularly, the expectation that FOs would take over O&M responsibility of headwork's and main 9  canals of major/medium schemes was not realistic. The project overlooked in the design, one of the main lessons learned from the VIRP that emphasised that " ....the formation of FOs should come before physical implementation....." (ICR June 1992). Instead, the project was designed to form FOs and rehabilitate schemes simultaneously. By design, the project implementation management was vested at the center with the ID. Given the national coverage, project implementation would have been more efficient and effective, if implementation management authority was delegated to provinces with the appointment of field-based, provincial-level, deputy project directors. Preparatory work on staff requirement to implement civil works was not adequate. The project design overlooked the need to engage a cadre of work supervisors at construction work sites to oversee civil works. This caused difficulties and delays in the civil works progress during the first four years. In view of this, Bank's performance in project design and appraisal is rated deficient. 35. The Bank carried out eleven project supervision missions including a combined IDA/EU MTR in May 1995. Early Bank supervision missions identified main physical implementation constraints and took early action to alleviate these. Considerable flexibility, especially after MTR from both IDA and the government, made it possible for the project to accelerate the project implementation pace and progress. However, all supervision missions did not address project's sustainability in the context of the prevailing sector policy environment, and consequently allowed the project to be completed without accomplishing most of its social and institutional objectives. The mission timing and reporting on the implementation problems, progress, likely development impact, and the performance ratings were satisfactory. The composition of the supervision missions showed a marked bias towards engineering inputs at the expense of other complementary disciplines such as agronomy and FO institutional development, which seems to be due to the heavy engineering component of the project. The project supervision responsibility was shifted from the headquarters to the field office in August 1997 following the South Asia Region's renewal policies. Toward the end of the project, the Bank could activate the dormant project activities in the areas affected by civil disturbances in the Northeast Province through back-up supervision and monitoring arrangement with the offices of the International Committee of Red Cross (ICRC) and the United Nations High Commission for Refugees (UNHCR). The ICRC collaborated by liaising with the field security authorities and facilitating the movement of project staff and construction materials to project sites where there were security restrictions. Both the ICRC and UNHCR provided periodic monitoring reports to Bank's field office on project implementation progress and issues. In view of this, the project supervision is rated partially satisfactory. F. BORROWER PERFORMANCE 36. As mentioned, the borrower prepared the project with the assistance from local consultants and the Bank's major project preparation requirements were satisfactorily addressed. But there were a few shortcomings in the detailed design, which have been reflected as lessons learned under chapter I of this report. The Borrower's performance in project implementation was mixed. The project implementation pace and progress were unsatisfactory from the inception, but improved substantially toward the end of the project. The main factors contributed to this situation have been described in Sections 27-30. Besides, a few other factors affected the Borrower's implementation performance: (a) Project management responsibility and authority were concentrated at the centre, although the project was of a national nature heavily involving the provincial councils. (b) Although the National Steering Committee reviewed and monitored physical project work plans and progress satisfactorily, it overlooked the effects of sector policy issues on project design, implementation, and sustainability. (c) Most Provincial Project Co-ordination Committees were not very effective. 10  (d) Project monitoring of physical progress was satisfactory but monitoring and evaluation (M&E) of the 'software' aspects of the project were deficient. (e) The quality of civil works is not up to expectations in some schemes, despite the heavy investments and technical assistance provided for quality assurance under the project. The Borrower's compliance with accounting, financial reporting, auditing, procurement, and progress reporting requirements was satisfactory. The legal covenant that envisaged the submission to IDA by the borrower of plans for organizational arrangements for the irrigation sector and timetable for implementation of recommendations for organizational arrangements (Paragraph 2 of Schedule 4 of DCA), was amended in April 1997. It was agreed by both parties that in view of the on-going works of National Water Council on water policy formulation, the time was not opportune for reorganization of the public irrigation sector. However, the amendment required the borrower to keep IDA informed on a quarterly basis, of developments in its organization of the public irrigation sector; IDA received only the first quarterly report. The co-operation between the government and the IDA remained firm through the implementation period in spite of GOSL disappointment at the IDA decision to extend the credit closing date only by six months and only in respect of the works in the Northeast. G. ASSESSMENT OF THE OUTCOME 37. The overall outcome of the project is rated unsatisfactory. The project succeeded in achieving its physical targets and raising agricultural production and incomes of project beneficiaries at least initially. However, it has not yielded and is not expected to establish an effective and sustainable O&M system for the rehabilitated project schemes. O&M of the rehabilitated schemes would remain highly dependent on external financial support. Given Government's poor fiscal strength to fully finance O&M, weak organisational viability of FOs, and lack of farmer's willingness to contribute to O&M, there is the expectation that the rehabilitated schemes would deteriorate and production levels would decline soon. H. FUTURE OPERATIONS 38. The Government has allocated funds to enable completion of all project schemes, whose completion dates were beyond the credit closure on June 30, 1998 for the schemes outside the Northeast Province. Also, the Government has secured financial assistance from the EU for an "aftercare" program until December 31, 1999 to consolidate the FO development and institutional development programs initiated by the project. 39. Despite the unsatisfactory outcome, the NIRP is a major country-wide attempt of the government to set in place a sustainable O&M system in the irrigation sector. It provides a wealth of lessons and experiences in relation to formation and strengthening of FOs, and O&M handing over process. The project should, therefore, be considered a learning experience for understanding the feasibility and desirability of the present irrigation management policies of the government. It is therefore recommended that the Bank, together with the government, carry out a post-project impact evaluation on the status of the FOs formed under the project, effectiveness of the 'O&M funds', status of the O&M of the handed over schemes, and the role of irrigation agencies in irrigation management with a view to revisit the present irrigation sector policies and strategies. I. LESSONS LEARNED 40. The NIRP is the first country-wide project aimed to ensure the sustainability of irrigation schemes through rehabilitation and improved O&M and creation ofFOs to managing the rehabilitated 11  schemes. Although the project is not likely to achieve its development objectives fully, the planning, design, implementation, and outcome of the project provide a useful leaming experience both for the Borrower and Bank. * The main lesson is that without up-front commitments on agriculture sector policy and related institutional reforms, it is difficult to achieve successful irrigation investment projects. * A situation of low productivity and profitability of irrigated farmning systems inhibits the development of viable FOs, and establishment of beneficiary-driven, sustainable O&M arrangement in the irrigation subsector. * Forming FOs and carrying out irrigation rehabilitation at the same time does not work well. Before starting rehabilitation works, it is essential to maintain, for at least a year, an effective institutional support aiming at establishing well-motivatedFOs and self- sustaining agricultural input services.. * Improvements to one scheme affect the hydrology and water availability at other downstream schemes in tank cascade systems. This factor needs to be adequately analysed in scheme selection, planning, and design for future irrigation rehabilitation projects. e It is essential to develop and enforce pragmatic civil works specifications, contract packaging, administration, and quality assurance systems suitable for rehabilitation to ensure effective implementation of irrigation rehabilitation. * To assure the achievement of good quality civil works standards, consideration should be given for a pre-qualification system of contractors for larger civil works of similar projects in the future. * The establishment of a maintenance fund to cover O&M costs with actual support from farmers has proved to be one of the possible solutions to solve the problem of project sustainability. This approach should be studied further. * UJN agencies such as UNHCR and International NGOs (ICRC) can make valuable contributions in implementing project activities in areas with civil security concerns. The Bank and the borrower should elicit their participation to the greatest degree possible in the design and implementation of lending operations covering areas affected by civil disturbances. * For national projects involving provinces, implementation management responsibility and authority should be devolved to provinces to the extent possible to ensure effectiveness and efficiency. 12  IMPLEMENTATION COMPLETION REPORT SRI LANKA NATIONAL IRRIGATION REHABILITATION PROJECT (Cr. 2260-CE) PART II - Statistical Tables  PART II: STATISTICAL TABLES Table 1: Summary of Assessments A. Achievement of objectives Substantial Partial Negligible Not Applicable (41) (W() (W) Macro policies Ei E [ [ Sector policies L L [3 L Financial objectives LiI [I [ Institutional development L [Z L1 Physical objectives [3i [ Poverty reduction L Li Li Gender issues C n Li Other social objectives Li L [3K Environmental objectives i [I [fl ED Public sector management l l Li Li Private sector development i E] E1 Other (specify) L 17] B. Project sustainability Likely Unlikely Uncertain (1) (1) (/) Highly C. Bank performance Satisfactory Satisfactory Deficient (V) 0() (I) Identification L e7 1Li Preparation assistance L EK Li Appraisal L L [K Supervision L Li 14  Table 1: Summary of Assessments (continued) D. Borrower performance Satisfactory Satisfactory Deficient ) (If() (se) Preparation L [7 [3 Implementation LI] E E3Z1 Covenant compliance E E37 Operation (if applicable) Ei L L Highy Highy E. Assessment of outcome Satisfactory Satisfactory Unsatis-factory unsatisfactory (5) () 1 5  Table 2: Related Bank Loans/Credits Loan/credit title Purpose Year of Status" approval Preceding operations 1. Mahaweli Ganga To increase agricultural production and standard January 1970 Completed Development Project, of living of about 53,000 families by providing (Dec. 1979) Stage I supplemental irrigation water to about 51,000 ha Cr. 174-CE/Ln. 653-CE of existing irrigated land and full irrigation of (US$ 29 million) about 42,000 ha of new land to be developed in subsequent stages. To be achieved by construction of head works, diversion dams and tunnels at Polgolla and Bowatenne, and of a 40 MW power plant at Polgolla. 2. Tank Irrigation To improve irrigation supply, reduce seepage Nov. 1976 Completed Modernization Project losses, provide inputs, strengthen agricultural (June 1983) Cr. 666-CE extension and water management in five major (US$ 5 million) tank schemes in the dry zone. 3. Second Mahaweli To increase paddy production and employment April 1977 Completed Ganga Development of about 27,000 farm families in the Mahaweli (June 1985) Project System area (28,000 ha), and provide productive Cr. 701-CE support and social services throughout the whole (US$ 19 million) system area. 4. Kurunegala To raise productivity, employment, incomes and 1979 Completed Integrated Rural living standards of the rural population through (Dec. 1985) Development Project an integrated package of investments and Cr. 891-CE institutional improvements. (US$20 million) 5. Mahaweli Ganga To support the Government's efforts to January 1980 Completed Technical Assistance accelerate implementation of the Mahaweli (Sept. 1985) Project development program by assisting with studies Cr. 979-CE to select the best plan for a future transbasin (US$ 3 million) diversion project and to provide detailed engineering design for downstream development. 6. Village Irrigation To support rehabilitation of 1,200 village June 1981 Completed Rehabilitation Project irrigation schemes, modernization of about 500 (Dec. 1990) Cr. 1160-CE other small schemes, strengthening of (US$ 30million) government institutions involved in village. irrigation, and initiation of a systematic water management program. 7. Third Mahaweli To increase agricultural production providing June 23, 1981 Completed 1994 Ganga Development irrigation infrastructure for about 28,000 ha, PCR dated Project clear jungle, settle farmers, assist on-farm Nov. 22, 1993 Cr. 1166-CE development, construct roads and social (IJS$90 million) infrastructure and establish fuelwood plantations. 8. Third Rural To increase productivity, employment, income 05/12/83 Development Project and living standards in the districts of Mannar Cr. 1363-CE (US$ 23 and Vavuniya. million) 16  Table 2: Related Bank Loans/Credits (Continued) Loan/credit title Purpose Year of Statusa/ approval 9. Major Irrigation To rehabilitate existing irrigation infrastructure Dec. 1984 Completed Rehabilitation and to address water management issues and (June 1993) Cr. 1537-CE irrigation practices in order to achieve a more (SDR 17 million) efficient use of water. Following operations 1. Mahaweli To shift the focus of the Mahaweli Authority of July 1997 Ongoing Restructuring and Sri Lanka from a construction agency to a river Rehabilitation Project basin management agency, and to improve Cr. 3058-CE agricultural productivity of Mahaweli system (SDR 41.7 million) "H" through rehabilitation and improved O&M. 2. Northeast Irrigated To help conflict-affected communities of the tJnder Agriculture Project Northeast province and adjoining areas to re- preparation establish normal life through assistance to jump start agricultural production, and build their capacity for sustainable social and economic development. al Figures in brackets indicate actual or expected closing date. 17  Table 3: Project Timetable Steps in project cycle Date planned Date actual/ latest estimate Identification (Executive Project Summary) Preparation September 1989 Dec. 1989/Jan. 1990 Appraisal February 1990 Nov./Dec. 1990 Negotiations August 1990 April 22-25, 1991 Letter of development policy (if applicable) Board presentation June 18, 1991 June 6, 1991 Signing July 24, 1991 Effectiveness October 21, 1991 First tranche release (if applicable) n.a. n.a. Mid-term review (if applicable) May 1995 Second (and third) tranche release (if applicable) n.a. n.a. Project completion December 31, 1997 June 30, 1999 Loan closing June 30, 1998 December 31, 1998b/ aH Held in Washington. A representative of EEC participated as observer in the negotiations. b/ Only in respect of the works in the NE. The original closing date of June 30, 1998 was maintained for the rest of the project. Table 4: Loan/Credit Disbursements: Cumulative Estimated and Actual (US$ million) FY92 FY93 FY94 FY95 FY96 FY97 FY98 FY99 Appraisal estimate 2.0 3.9 9.4 15.0 20.6 25.6 29.6 Actuala/ - 2.2 2.7 4.3 7.8 11.3 19.3 24.6 Actualas%ofestimate ... 56.4 28.7 28.7 37.9 44.1 65.2 83.1 Date of final disbursement: May 11, 1999 a/ As of 11 May 1999. 18  Table 5: Key Indicators for Project Implementation Implementation Unit Planned Planned Actual Comments Indicators in SAR (SAR) Restructuring Civil Works - Rehabilitation and Improv. Works (i) Minor Schemes Nb. 1,000 1,048 1,048 Of which 830 schemes completed by June 1998 and 971 by the end December 1998. All schemes would be completed by June 30, 1999. (ii) Medium/Major Schemes Nb. 60 35 341 22 schemes completed as of Feb. 28, 1999. The remaining 12 schemes would be completed by June 30, 1999. - Tanks (Reservoirs) Nb. 700 483 - Anicuts (Diversions) Nb. 300 565 - Command Area (Minor Schemes) ha 25,000 29,0002 25,000 - Command Area (Med./Major Sch.) ha. 12,500 13,608 13,390' - Training Centres Nb. 4 4 Expansion and upgrading of three existing centres and construction of a new one in Pothuhena Farmers Organizations - Registered under 56A Minor Schemes Nb. 1,000 1,048 Scheme level organizations. Medium/Major Schemes Nb. n.a. 207 Distributory and Field (D&F) canal ! ~~~~~~~~~~~~~~~~~~~~~~organizations - Registered under 56B Minor Schemes Nb. n.a. 7 Medium/Major Schemes Nb. n.a. 7 Training - Overseas (i) Masters Degree Courses Nb. 3 3 (ii) Short-Term Courses Moderniz. of Irrig. System Nb. of 2 2 Not appropriate for senior particip. staff Dam-Safety I - Lack of funds Inigation System Manag. 48 30 Satisfactory Rehabilitation Design 25 2.5 Good for junior staff, not approp. for senior staff Environmental Issues 2 - Considered not necessary . O&M " 30 24 Satisfactory Women's Role in Farming 4 - A suitable course could Development not be found Institutional Development 22 22 Satisfactory Inst. Dev./Farmer Organ. 32 32 Satisfactory Of the revised work plan one scheme (Mahawewa of 101 ha) was dropped in 1997; only headworks were included in the program for another scheme. (Rampathwela of 118 ha). 2 Including 106 schemes (3,200 ha) in the Northeast Province. 19  . Project Management I 1 Useful . Financial Management 3 3 Useful. . Training of Trainers 3 3 Useful h Institut. Development 30 30 Satisfactory . Irrigation Management 30 - Lack of funds Institut. & Manpower Dev. 10 - Lack of funds Total 243 172 - In-Country (i) Staff- Minor Schemes p/day 4,384 3,688 - Med./Maj. Schemes p/day 6,600 5,544 (ii) lAs - Minor Schemes p/day 11,231 7,204 - Med./Maj. Schemes p/day 3,027 826 (iii) Farm's Reps - Minor Schemes p/day 62,505 43,524 - Med./Maj. Schemes p/day 21,444 8,900 (iv) Farmers - Minor Schemes p/day 24,384 20,007 - Med./Maj. Schemes p/day 4,950 2,778 Technical Assistance - Intemational Recruited p/month 50 115.5 - Locally recruited Professional p/month 1,020 1,237.0 Staff -Locally recruited Mid Level Staff p/month 1,080 1,015.0 Table 6: Key Indicators for Project Operation I. Key operating indicators in SAR/Presidents Report Estimated Actual Not identified at appraisal. 20  Table 7: Studies Included in Project Purpose as defned Study at appraisal/redefmed Status Impact of study 1. Benchmark and To assess schemes Inception and As the benchmark study Evaluation. conditions prior to Benchmark studies results were available in rehabilitation and completed and final 1996, its impact on introduction of improved reports accepted in rehabilitation planning O&M practices. Four special August 1996. Final was limited. Final studies were initiated during impact evaluation is in evaluation study results implementation focusing on progress but not is not available yet. O&M, FOs and completed yet. Special evaluation study implementation delays reports generated information on status of O&M, FOs and implementation delays 2. Hydrological and To optimize the use of water Not Implemented. See item 8 below. Water Management resources through better Deemed unnecessary in Study. management view of the studies conducted by IRMU 3. Socio-economic In-depth monitoring and Not implemented. See item labove and Study. analysis of changes due to Deemed unnecessary in items 5 and 6 below the project: changes in farn view of the impact production, incomes, evaluation study and irrigation practices and the studies conducted by effectiveness of institutional IRMU arrangements for rehabilitation and O&M in the sample schemes 4. Assessment of Not identified at appraisal Completed in 1997 Useful study highlight- turnover process in and included during ing status and problems NIRP schemes. implementation . Conducted related to FOs, O&M by AR&TI performance and turnover process. Provided useful background information for the ICR 5. Monitoring farmers' Not identified at appraisal Completed in 1995 Provided information on involvement in and included during farmers' involvement in rehabilitation (Phase implementation. Initiated rehabilitation. However, I). Study of five and conducted by IRMU. corrective measures irrigation schemes were not taken by the under NIRP. project based on this information. 21  6. Monitoring farmers' Not identified at appraisal Completed in 1997 Provided information on involvement in and included during farmers' involvement in rehabilitation (Phase implementation. Initiated rehabilitation. II) and farmer and conducted by IRMU. Corrective measures participation in were not taken by the rehabilitation. project based on this information. 7. Rehabilitation of NIRP Not identified at appraisal Completed in 1995 Provided information on schemes: farmers' and included during farmers' involvement in perspectives. implementation. Initiated rehabilitation. and conducted by IRMU. Corrective measures were not taken by the project based on this information. 8. Hydrology of Small To optimise use of water Completed in 1998 Study results were Catchments. resources in catchments available in 1998. It was (watersheds) too late to incorporate recommendations into rehabilitation planning & design. 22  Table 8A: Project Costs Appraisal estimate (US$M) Actual/latest estimate (US$M) Local Foreign Total Local Foreign Total Costs costs costs Costs I.Rehabilitation and 27.6 3.7 31.3 29.1 Improvement Works 2.Farmer Organization 2.7 - 2.7 2.1 3.Training 0.9 1.4 2.3 1.6 4. Environmental 1.0 - 1.0 l Protection 5. Institutional Support 2.7 1.1 3.8 3.0 and Studies 6. Technical Assistance 4.5 0.6 5.1 6.4 7.Vehicles and Equip. 1.8 1.8 3.6 2.5 TOTAL 41.2 8.6 49.8 36.6a1 8.1a/ 44.7 a/ Mission's estimate. Table 8B: Project Financing Appraisal estimate (US$M) Actual/latest estimate (US$M) Local Foreign Total Local Foreign Total costs Costs Costs,. Costs Source IDA 25.0 4.6 29.6 21.0 3.6a' 24.6 EU (formerly EEC) - 4.0 4.0 - 4.5 4.5 EU Food Aid 9.2 - 9.2 9.2 - 9.2 GOSL 7.0 - 7.0 6.4 - 6.4a TOTAL 41.2 8.6 49.8 36.65 8.1 44.7 a/ Mission's estimate 23  Table 9: Economic Costs and Benefits Unit SAR Estimates ICR Estimates A. Project Beneficiaries No. 100,000 85,000 B. Command Area - Minor Schemes ha 25,000 25,000 - Medium/Major Schemes ha 12,500 13,390 C. Incremental Rice Paddy Production Tons 25,700 25,100 D. Increases in Cropping Intensity at Full Development - Kaltota % 25 2 - Mahagal Wewa % 48 2 - Dorakada Liyadde % 42 2 - Kobeigane % 51 7 - Mahakiri Ibbewa % 87 7 - Nittewa % 65 7 E. Economic Rate of Return' % 31.0 13.9 F. Increases of Net Farm Income2 - Kaltota % 307 36 - Mahagal Wewa % 180 36 - Dorakada Liyadde % 135 28 - Kobeigane % 120 47 - Mahakiri Ibbewa % 183 48 - Nittewa % 109 50 The economic analysis at appraisal was carried out for the project as a whole and for the six appraised schemes of Kaltota, Mahagal Wewa, Dorakada Liyadde, Kobeigane, Mahakiri Ibbewa and Nittewa. Because of lack of data, the economic analysis at project completion could only be carried for the project as a whole. However, as indicated above, an ex-post-fmancial analysis was carried out for the six selected schemes as envisaged at appraisal. Detailed description of the economic analysis is presented in Appendix B. 2 An ex-post financial analysis has been carried out to assess the achievement of the stated project objectives. In line with the approach used at appraisal, farm budgets were prepared for six sample schemes namely Kaltota and Mahagal Wewa for the medium/major schemes and Dorakada Liyadde, Kobeigane, Mahakiri Ibbewa and Nittewa for minor schemes. Corresponding basic data was provided by DAS and CMC and are based on a study carried out during the course of the project. As regard to minor irrigation scheme parameters, ten tank schemes and ten anicut schemes were considered under the study and selection was done starting from samples in 1992/1993 in the various districts where project activities were implemented. The selection was considered by DAS fairly representative as far as the NIRP is concerned. Detailed farm income analysis is given in Appendix B. 24  Table 10: Status of Legal Covenants Original Revised Covenant Present fulfillment fulfilment Description of Agreement Section class status date date Covenant Comments 1. Credit 2.02(b) I C - Maintain Project Special PSA has been opened. C 22600 Account (PSA) Became functional in Feb. 1993 4.01(b) (i) I C - - Have the records and accounts referred to in para (a) in accordanca with appropriate auditing principles I (ii) I CD Furnish the Association not later than 9 months after the end of each year a certified copy of audit reports 6.01 11 C 21/10/91 - EEC Grant Agreement Fulfilled signing and submission of 1992 work plan to be condition of effectiveness Schedule 4, 5 C 30/06/94 - Work programme, budget Fulfilled para I and staff requirements for the following field year Schedule 4, 12 SOON 30/06/94 - Submit plans for Amended in April 1997. pam 2(a) reorganization of irrigation Amended covenant not sector fulfilled. Schedule4, 12 NYD 31/12/92 - Submit timetab]le for Amended in April 1997. para 2(b) implementation of Amended covenant not reorganization fulfilled. Schedule 4, 5 CD 31/12/92 - Appoint consultants for the Project consultants were para 3 project in place since Oct 1992 Schedule 4, 10 C 31/12/92 30/04/93 Submit proposads for Submitted to both IDA para 4 inspection of dams, etc. and EEC in 1993 Schedule 4, 10 C - - Selection of schemes in Fulfilled para 5 accordance with specified criteria Status: C - Complied with CD - Compliance after Delay NC - Not Complied with SOON - Compliance Expected in Reasonably Short Time CP - Complied with Partially NYD - Not Yet Due Covenant Class 1. Accounts/audit 2. Financial performance/generate revenue from beneficiaries 3. Flow and utilization of Project funds 4. Counterpart funding 5. Management aspects of the Project or of its executing agency 6. Environmental covenants 7. Involuntary resettlement 8. Indigenous people 9. Monitoring, review and reporting 10. Implementation 11. Sectoral or cross-sectoral budgetary or other resource allocation 12. Sectoral or cross-sectoral regulatory/institutional action 13. Other 25  Table 11: Compliance with Operational Manual Statements Statement number and title I Describe and comment on lack of compliance No deviation from applicable operational manual statements observed. However, OD 10.6 (revised OP/BP 10.02 was only partially compiled with since a detailed review of the financial systems was not conducted at the time of appraisal. Table 12: Bank Resources: Staff Inputs Stage of Planned Revised Actual project cycle P Weeks US$ Weeks US$ Weeks US$ Preparation to 14.3 33.4 Appraisal Appraisal 51.1 77.7 Negotiations through 30.9 47.3 Board approval Supervision 148.2 286.3 Completion 10.0 45.0 TOTAL 274.8 636.8 274.8 636.8 158.2 331.3 26  Table 13: Bank Resources: Missions Performance r Stage of Number Specialized Implemen- Develop- project cycle Month/ of Days in staff skills tation ment Types of year persons Field representedt Status objectives problenie Identificatione Preparation" Dec. 1989/ n.a. n.a n.a. Jan.1990 Pre-Appraisal Appraisal Nov./Dec. 1990 6 n.a n.a. Supervision: I May 1991 4 10 IE,IE,AE,AE I 2 Oct. 1991 5 10 IE,IE,AE,AE,T I I TA 3 April/May 1992 2 16 IE,IE 2 1 P, T, TA 4 Sept. 1992 3 11 IE,IE,IE 2 1 P, T, TA 5 March/April 1993 3 10 IE, IE, AE 2 2 P, T, TA 6 Nov. 1993 2 15 IE, AE 2 2 P, T, TA, F 7 July 1994 4 10 IE,AE,SS,,IE U U P, T, TA, F 8 (MTR) May 1995 5 15 IE,AE,AEIE,IE U U P, T, TA, F 9 April/May 1996 3 11 IE,AE,AE S U P,F.L 10 Nov./Dec. 1996 2 14 IE,IE S U L ICRe1' June/July 1998 2 12 E, E S U a/ P: Procurement Progress; F: Financial Performance; T: Training Progress; E: Environmental Aspects; TA: Technical Assistance; M: Management; S: Satisfactory; U: Unsatisfactory; L: Legal Covenant. bl AE: Agricultural Economist; E- Economist, IE: Irrigation Engineer; SS- Social Scientist; T: Training Specialist; '- Carried out by a project preparation cell comprised of ID and DAS staff assisted by consultants. d/ The updating of the draft ICR prepared by FAO/CP in July 1997 was carried out by a joint IDA and FAO/CP mission in December 1998. 27  APPENDIX A SRI LANKA NATIONAL IRRIGATION REHABILITATION PROJECT (Credit 2260-CE) AIDE- MEMOIRE I. INTRODUCTION 1. The Mission comprising Messrs. Raffaele Suppa (Economist, FAO/CP) and Nihal Fernando (Irrigation Engineer) carried out the final supervision of the implementation of the National Irrigation Rehabilitation Project (NIRP) and gathered material to initiate drafting of the project's Implementation Completion Report (ICR), during June 22-July 3, 1998. 2. The mission had initial discussions with key staff including the Secretary, Ministry of Irrigation and Power (S/MIP), Project Director and Project Consultants and made field visits to Central, North-Central and North-Western Provinces. The mission had meetings in the provinces with the senior project staff of the Irrigation Department (ID), Provincial Irrigation Agencies (PIAs), Department of Agrarian Services (DAS), Project Consultants of the Regional Support Teams (RSTs) and farmers. Contacts were also made with the Director General of the External Resources Department (Ministry of Finance), District Secretaries of Mannar, Mulativu and Vavuniya of the Northeast province, International Irrigation Management Institute (IIMI), Agrarian Research and Training Institute (AR&TI), Heads of the Irrigation Research Management Unit (IRMU), Quality Control (QC), Irrigation Management and Operation and Maintenance (O&M) Divisions of the ID. The mission would like to thank all the staff, including the Project Director, project staff, consultants, and farmers for the support extended and for the good working relationships established. 3. This Aide-Memoire summarizes the mission's observations, main issues, and recommendations discussed at the wrap-up meeting held on July 2, 1998, chaired by Mr Jaliya Medagama, S/MIP. Project Background 4. The project was a follow-up to the previous IDA funded Village Irrigation Rehabilitation Project (VIRP) (Cr. 1160 CE), which covered the rehabilitation of about 1,675 minor schemes covering about 44,000 ha of irrigated farm lands. The NIRP was first proposed in 1987 as a way to continue rehabilitation and modernization of minor schemes started under the VIRP. The NIRP was built on the experiences gained through VIRP and similar recent projects. IDA support for the NIRP was justified as it was expected that the NIRP would: (a) improve the capacity and efficiency of public irrigation services at the center and in provinces, and; (b) promote increased participation of farner organizations (FOs) in irrigation system management. 5. The NIRP was prepared in December 1989/January 1990, appraised in November/December 1990 and presented to IDA's board on June 6, 1991. Subsequently, an IDA Credit consisting of SDR 21.9 million (US$ 29.6 million) was signed on July 24,1991 and the Credit became effective on October 21, 1991. The Credit, to be disbursed over a period of seven years, covered about 59% of the total project cost estimated at US$ 49.8 million. The financing plan included a Government of Sri Lanka (GOSL) contribution of US$ 16.2 million and a grant equivalent to US$ 4.0 million from the European Union (EU).  Prc,ject Objectives 6. The main objective of the NIRP was to stabilize and increase agricultural production and incomes, and raise the standard of living through rehabilitation and impiroved O&M of existing irrigation schemes. Subsidiary objectives included: (a) upgrading the skills of farmers and staff of the implementing agencies, and (b) creating viable FOs for managing the rehabilitated schemes. The schemes for rehabilitation were selected after reaching prior agreement with the respective FOs that they would: (i) in respect of minor schemes (command area less than 80 ha), bear the full cost of O&M of schemes after rehabilitation and improvements are completed, and (ii) in respect of medium/major schemes, bear the full cost of O&M of Distributory and Field (D&F) canals after the rehabilitation is completed and bear the cost of O&M of headworks and main canals starting two years after rehabilitation is completed. 7. The project's objectives were considered important and relevant in pursing the Irrigation Management Policy of the GOSL which was designed to increase the participation and responsibilities of the farmers in irrigation system management and reduce the role of the public sector. The present policy is to handover all irrigation schemes to FOs for self-management by the year 2000. Minor schemes are to be managed fully by the farmers and major/medium schemes are to be managed jointly by the farmers and irrigation agencies. The role of the public sector woulct gradually be limited to providing technical and institutional advice to FOs, and O&M of major facilities such as headworks and main canals. It is in this overall context that the NIRP was considered as an important intervention. Project Components 8. The project included the following components: (a) rehabilitation and improvement of about 1,000 minor schemes and 60 major/medium schemes covering about 25,000 ha and 12,500 ha respectively; (b) establishment of FOs and introduction of improved O&M practices in all rehabilitated schemes; (c) training of farmers and staff of the implementing agencies; (d) environmental protection studies and works; (e) establishment of three new support units in ID and execution of socio-economic and hydrological studies, and; (f) consultancy services for project planning, implementation, and impact assessments. 9. The target of major/medium schemes were reduced from 60 to 35 schemes after IDA Review in July 1994 although increasing the area to some 13,600 ha. After August 1997, one additional medium scheme was dropped from the schedule (Mahawewa, 100 ha) and the rehabilitation of another scheme (Rampathwila, 118 ha) was limited to headworks. The revised work plan included the rehabilitation of 1,036 minor schemes covering about 29,000 ha, including 80 schemes (3,000 ha) in the Northeast province, and 34 medium/major schemes covering 13,380 ha. Project Implementation and Coordination 10. The project was implemented by three agencies under the guidance and supervision of a Colombo-based Central Management Cell (CMC) headed by a Project DirectEor (PD) and assisted by a team of consultants and eight field based Regional Support Teams (RSTs). The ID was responsible for the rehabilitation and FO institutional strengthening of all major/medium schemes, except for FO institutional strengthening of two schemes which was the responsibility of thne Irrigation Management Division (IMD) of the MIP. In minor schemes, the ID, PIAs and DAS sharecl the responsibility for the rehabilitation, while the DAS was responsible for the FO institutional strengthening. In addition, FOs were expected to implement a designated part of the rehabilitation works in their respective schemes as a voluntary contribution, a portion equivalent to 10% of the civil works cost. 11. The project activities were coordinated by a Project Coordination Committee (PCC) at national level and by Provincial Project Coordination Committees (PPCCs) at provincial levels. The PCC was chaired by the S/MIP and represented by the heads of relevant line departments and Chief 29  Secretaries of the provinces. The PD was the secretary to the PPC. The PPCCs were chaired by the respective Provincial Chief Secretaries and represented by the provincial level line agency heads. Partial Extension of Closing Date 12. On March 19, 1998, the GOSL requested IDA, a six-month partial extension of the Credit Closing Date from June 30 to December 31, 1998, to enable completion of the ongoing rehabilitation works in the Northeast. Government also requested IDA to finance the feasibility studies of a selected sample of irrigation schemes in the Northeast under the Credit. Considering that it was only in August 1997 that substantial work could be initiated in the NE and that subsequent disruptions due to civil unrest caused a delay in sustaining the implementation pace of those works, IDA agreed to accede to the Government request under 'force-majeure' conditions. On April 9,1998, the IDA communicated to the Government, its decision for a six months extension of the Credit Closing Date to December 31, 1998 in respect of the works and studies in the NE. The Credit Closing Date for rest of the project remains as June, 30,1998. II. OVERALL PROJECT STAiTUS Overall Project Implementation 13. The overall project implementation, expected to start in January 1992, has been behind schedule due to a number of reasons: (a) the appointment of project consultants scheduled on December 31, 1991, but accomplished only in May 1993, delayed the mobilization of project activities by one and a half years; (b) the PIAs, which were in infancy at the time of scheduled commencement of project activities, could not mobilize their resources fully as expected to enable a quick take-off; (c) the first batch of project vehicles and equipment under IDA financing was delivered only in late 1993, and as a result the mobility of project staff were affected; (d) the poor cashflow management after 1996 caused shortfalls in the availability of project funds to meet with project expenditures at crucial civil work construction periods; (e) the general strike of the Technical Officers of the ID for three months during July-September 1997 affected the civil works progress of all major/medium schemes and about 300 minor schemes; and (f) delayed but heavy Northeast monsoon rains in four consecutive years (1994 -1997) obliged the farmers to opt for an early Yala cultivation in the following year reversing their consent to forgo one cultivation to permit rehabilitation works. Except (e) and (f) above, all the other factors reflect the failure of the Borrower to ensure timely 'readiness' conditions required to mobilize the project activities as scheduled. Central Management Cell and Project Management 14. The CMC was able to carryout majority of administrative and financial management responsibilities assigned to it in the SAR. It successfully carried out feasibility studies, preparation of technical guidelines and manuals, monitoring and record keeping of physical work progress, procurements and training. It maintained satisfactory accounts, reviewed statements of expenditure, and prepared and furnished withdrawal applications to IDA. 15. However, initially the CMC failed to ensure project readiness conditions timely. This lapse seems to be largely attributable to the fact that CMC faced limitations to function as a fully independent and autonomous unit during the initial project period. By design, the CMC was expected to be a sub-unit within the ID. The CMC was set-up accordingly, an ID engineer was appointed as the PD, but it took a considerable time before it was fully staffed by the ID and other line agencies. The CMC, being under the bureaucratic control of the ID top management, could not exercise adequate financial and administrative authority to ensure timely project readiness conditions, manage project activities, and allocate project resources and facilities efficiently and effectively. Also, the CMC failed to monitor and evaluate effectively the FO institutional development program, which was a key project activity. Considering the project's scale, magnitude and policy importance, it would have been 30  more effective if the CMC functioned as an independent unit directly under the guidance of MIP linked with Provincial Management Cells each headed by a Co-Project Director at provincial level. Project Coordination 16. At initial stages of the project, the PCC, established as the apex body to deal with all aspects of project management and implementation met once in three months. However, after the project gathered momentum, the PCC met only once in six months or so. The PCC effectively carried out the reviewing and approving of work plans, monitoring physical progress and resolving implementation problems. However, the mission is of the view that PCC should have provided more policy guidance to the CMC with respect to the FO institutional development program and sustainability issues of the project. 17. The coordination of the project activities at provincial level by the PPCCs is not satisfactory. The PPCCs proceedings were focused on physical civil works progress monitoring at the expense of monitoring the institutional development activities and coordinating the other project activities among the three lAs. With respect of the implementation of FO institutional development program of minor schemes, lack of provincial level coordination between the DAS and the other two lAs affected the quality and outcome of the program to a considerable degree. This situation would have been eased off if the project activities were implemented under the overall supervision of co- Project Directors fielded at provincial level. Rehabilitation and Improvement Works 18. The pace of implementation of rehabilitation work has improved substantially after September 1997. Of the revised target of 1,036 minor schemes, including the 80 schemes undertaken after August 1997, only 76 were completed by June 1995. This figure increased to about 400 by June 1997, to 715 by May 1998 is currently estimated at 817 indicating an overall implementation rate of about 80%. The implementation progress of the 80 schemes in the NE is currently varying from 20% to 40%. It is however estimated that the remaining 219 schemes would be completed by the end of the year. With respect to major/medium schemes, out of the total 35, only six were fully completed by May 1998 and it is currently estimated at eleven. Considering that one scheme was deleted, the remaining 23 have so far recorded an implementation rate ranging from 60% to 90%. This is not satisfactory. This means that unless rehabilitation would be completed, there is a risk that in a considerable number of schemes improved water management would not mnaterialize leaving farmers in a worse situation than they had been in before the rehabilitation program. However, the CMC forecasts that all scheduled rehabilitation works of major/medium schemes would be completed by March 31, 1999. Construction Supervision and Quality 19. Compared to previous projects, the project design included a number of positive actions to improve the quality of civil works. Among many actions, the fielding of RSTs and providing training to farmers on basic construction aspects are the main ones. Since July 1996, the CMC with the assistance from project consultants, was able to enforce a comprehensive civil works quality assurance (QA) procedure and system. While this QA system was set in place mid way after the project activities have begun, the necessary QA checks and balances could not be enforced at optimum levels due to a number of reasons: (a) the transport facilities made available to QA staff were not adequate; (b) shortage of qualified and trained Work Supervisors (WSs) until mid 1997; (c) Majority of WSs recruited and posted to field sites lacked prior experience in civil works supervision-- although they were provided with quick short-term training by the CMC, their contribut:ion to quality assurance did not prove to be entirely successful; (d) although, the project consultants through their RSTs, were expected to check and certify reimbursement claims against all civil works contracts and monitor disbursements, the project management failed to establish and enforce this important QA requirement --However, IDA continued to disburse Credit funds in spite of the non-establishment of this QA 31  requirement: and (e) scheduled civil works are declared 'completed' without a final certification being obtained from the project consultants both in respect of the construction quality and structural safety of the 'completed' works, since this was not a pre-requisite indicated in the SAR. However, from quality assurance point of view, it would have been best for the works executed, if the steps (d) and (e) above were implemented by the Project on its own initiative. This is an important lesson learned from the project. 20. The mission is of the opinion that under the circumstances mentioned above, the CMC had difficulties in enforcing the QA system at optimum levels. While the quality and precision of some randomly selected civil works of four schemes inspected in the field with the project staff and consultants appeared good, it was not the case with two schemes. Given the small number of schemes inspected, it was difficult for the mission to make generalized conclusions on the quality aspects of the project. The mission was concerned whether the implementation pace was accelerated at the expense of quality. According to the CMC that it is not the case. It is an accomplishment by the Project if the good progress was achieved, amidst the shortcomings mentioned in paragraph 19 above, without sacrificing the quality. Procurements of Civil Works 21. The mission, was informed by the project consultants that the standard specifications developed by the Institute of Construction Training and Development (ICTAD) for civil works are not the best suitable for the nature of civil works typical to rehabilitation of small irrigation schemes. Because the project adopted those specifications without modifying those pragmatically, there was little flexibility in many instances to enforce the QA system. Also, the mission considers that both the implementation progress and quality of civil works could have been improved if the CMC managed civil works contract packaging, administration and monitoring more systematically. While the mission could not ascertain these aspects in detail, it suggests that these lessons be addressed constructively by the MIP for the benefit of ongoing and future works. These are tvo important lessons learned from the project. Formation of Farmer Organizations 22. FOs have been established in all minor schemes. In respect of major/medium schemes too, 207 D&F level FOs have been formed and Project Management Committees (PACs) have been set up except in one scheme (Ranpathwila). All these FOs have been registered under Clause 56 A of the Agrarian Services Act. Of those, seven FOs in minor schemes and seven in major/medium schemes have been registered under Clause 56 B of the Act, meaning that they have attained legal recognition implying they have a better chance for long term sustainability. Another 157 FOs are being considered for the registration under Clause 56 B. 23. The CMC reports that of the 956 minor schemes outside the NE, FOs of 811 (85%) minor schemes have contributed fully, the 10% voluntary contribution. In respect of the major/medium schemes it is reported that the 10% civil cost recovery has not been very successful. In some major/medium schemes, the value of the 10% civil cost is beyond the capacity of FOs to bear. However, the total monetary value of this contribution secured from FOs by the project was not known to the mission. The mission noted that the manner in which this contribution was to be recovered neither had been clearly described in the project Staff Appraisal Report, nor specified by the CMC. Eventually, lAs used uneven approaches in securing and certifying the 10% share, some of which were not acceptable. Handing Over and Improved O&M 24. As of May 31,1998, 483 completed minor schemes (67%) have been formally handed over to respective FOs. With respect of major/medium schemes only 20 D&F systems in four completed schemes have been taken over by FOs yet. The mission noted that although a large number 32  of minor schemes were reported to have been handed over, the preparation of O&M manuals for many of those schemes were still due. According to mission's observations and two NIRP research studies concluded by IIMI and AR&TI, it appears that handing-over in many schemes has been supply-driven, and is symbolized with a written agreement of handing-over and taking-over signed between the lAs and FOs respectively, and FOs have accepted handing over through sheer necessity than conviction. However, there is evidence that FOs could and would participate effectively in system operation in handed over systems -- but it seems unlikely that they could and would bear the full cost of maintenance in the foreseeable future. To this end, the initiative of the DA'; to set up O&M funds in 275 handed over schemes is commendable. Technical Assistance 25. Project consultants have performed their responsibilities satisfactorily except for the institutional development support at the RST level. The TA is coming to ea end with the scheduled Credit Closure on June 30,1998. However, in respect of the ongoing schemes in the Northeast, for which the Credit Closing Date is December 31, 1998, there is an acute need to provide technical assistance to the implementing agencies in the NE to enable them to complete the work program in time with due diligence and in accordance with the required technical standards. The mission strongly urges the MIP to convince the EU to continue with the consultancy support to those schemes through the two RSTs based in Anuradhapura and Ampara until December 31, 1998. The mission also urges that, if the proposed arrangement does not materialize, the CMC, using IDA funds, employs two ex- RST engineers for six months from July-December 1998 for the construction supervision of those works. Environmental Protection 26. The project included provisions for: (a) minimizing the environmental damage to the ecosystem during construction; (b) alleviating existing environmental problems, (c) studying environmental problems at the watershed level, and; (d) environmental awareness training for irrigation engineers, FO representatives and Institutional Organizers (1Os). The environmental problems of all major/medium schemes were identified and environmental impact assessments were conducted as a part of feasibility report preparation. Also, the environmental conditions of a sample of nine major/medium and 30 minor schemes were studied in qualitative terms under the Benchmark and Evaluation study of the project (paragraph 33). The rehabilitation may have automatically eased off a few environmental problems such as waterlogging and water stagnation in canals. However, except for tank siltation field research study conducted by the DAS in three schermes under the project, the implementation and monitoring of this component have been neglected and remained unsatisfactory. Training of Project Staff and Farmers 27. Overall, the achievement of this component is considered satisfactory. The overseas training included three Masters (MSc) Degree courses and short term training/study tours for 243 staff persons. This category of training aimed at enhancement of skills of senior and middle level technical staff. While the overseas long-term training (MSc's) schedule was completed as foreseen, the latter could only achieve an implementation rate of 70% (173 staff) either because suitable courses could not be identified (Women's role in Farming Development) or because of lack of funds (Irrigation Management, Institutional and Manpower Development). However, it was largely the ID staff who have benefited from the overseas training program. The mission feels that overseas training opportunities would have been allocated fairly to the PIAs and DAS as well. 28. In-country training was successfully conducted and included ten courses for project staff, four courses for institutional Organizers (1Os) and six courses for Farner Representatives (FRs). Training of Project staff was successfully carried out recording an implementation rate of 84%. Achievement for IOs and FRs Training Program not satisfactory. This is mainly because of the delayed commencement of rehabilitation itself and the shortage of trainers in the ID field level. No 33  formal evaluation of quality and impact of training is available. Four training courses were included for farmers in the NIRP. Since 1996, three out of these four are incorporated in the Aftercare Program implemented under EU funding. Assistance was adequately provided under the project for the expansion and upgrading of the four in-service training institutes in Galgamuwa, Pothuhena, Kandy, and Bandarawela. Irrigation Research Management Unit 29. It was felt necessary at appraisal to establish an Irrigation Research Management Unit (IRMU) to coordinate research on issues related to poor O&M, questionable hydrological planning parameters, and mono cropping of rice. The IRMU was set up in 1992 jointly by the ID and IIMI. It was headed by an internationally recruited agricultural engineer and supported by a full-time team of multi-disciplinary researchers from IIMI. It continued under IIMI's support until July 1996. However, from the inception it did not receive much support from the ID, including the appointment of counterpart staff, especially in the non-engineering disciplines. IRMU conducted a number of research studies and seminars. However after IIMI completed its assignment in 1996, the IRMU remained without necessary commitment, financial support, and staffing from ID to be able to continue research activities. At the end, the research issues identified at appraisal remain to be addressed. Dam Safety Unit 30. A Dam Safety Unit was supposed to be established to assist the ID in carrying out regular inspections, surveillance monitoring and preventive maintenance of storage dams. The project included provisions for training, vehicles, equipment, expert advise, and funds to cover operating cost of the unit during the project period. With initial support from the project consultants this unit was established in April 1994. The ID was not able to activate the unit as expected, and consequently, EU support was limited to short-term training in 1995. However, dam safety monitoring is now carried out as a routine activity by the ID, under the supervision of the O&M Unit. Quality Control Unit 31. Provisions were made to enhance the capacity of the Quality Control Unit of the ID, including the establishment of nine field laboratories, vehicles, equipment and operating cost during the project period. Four new field laboratory buildings were constructed, four existing ones were renovated, and the unit was provided with the operational support, including vehicle hire. Although, it received the equipment as late as in May 1998, it was able to put in place a comprehensive quality control system, and with the assistance with the RSTs, to provide support to lAs as envisaged. However, the adaptation of the system of checklists has not reached optimum levels. Studies 32. The project included provisions to continue the hydrological and water management study and socio-economic study implemented under the VIRP. It also included a Benchmark and evaluation study in 30 minor and nine major/medium schemes. In view of the limited outcome of the VIRP studies, the first two studies were not implemented under the project. The Benchmark and Evaluation study, which commenced in October 1995 with scheduled completion date in December 1998, is ongoing. The Final Report on the benchmark status was submitted to CMC by study consultants in August 1996. In view of its usefulness to the CMC to prepare its own project ICR, the mission urges the CMC to request the study consultants to expedite the study and submit the final report as early as possible. In addition, useful project related technical guidelines, manuals and studies have been concluded by project consultants as well as by IRMU. The mission suggests that those publications are disseminated among the lAs. 34  Agricultural Demonstration Plots 33. The project allocated a small provision (SLRs 4 million) to set up agricultural demonstration plots in about 10% of the rehabilitated schemes to promote crop diversification. Although, the DAS set up a few agricultural demonstration plots, the progress of this component remained unsatisfactory throughout the project period. Procurements of Vehicles and Equipment 34. The overall progress of procurements of vehicles and equipment to support project operations and lAs was not satisfactory. This is mainly because of the failure to procure vehicles and equipment due under EU funding, although procurements under IDA funding were performed satisfactorily despite initial delays. Of the target of 75 vehicles, 87 motorcycles, 300 bicycles and 55 vehicles, the project has procured 55 vehicles, 85 motor cycles and 325 bicycles under IDA financing. Estimated Project Cost, Financing and Disbursements 35. Total project cost is estimated to reach some US$ 39.3 million by Credit Closure, which is about 79% of the original appraisal estimate. Final overall disbursement (US$ 39.3million) is estimated to include IDA's US$ 22.1 million (75% of the original Credit amount), EU's grant of US$ 4.5 million and GOSL's contribution, including FOs' contribution, of US$ 12.7 million. Of the IDA Credit (US$ 29.6 million), US$ 5.0 million was cancelled in 1997 on GOSL's request. The total disbursement as of June 30,1998, is US$ 19.7 million (66% of original Credit amount). There is a backlog of uncleared project expenditure since December 1997 to date. III. PROJECT SUSTAINABILITY 36. Although the Project seems to have somewhat achieved its objective of increasing agricultural production and farmer incomes, in-depth analysis of the economic and financial viability will be carried out later by the mission as soon as complete data are made available by the CMC. However, the O&M of the rehabilitated irrigation schemes remains highly dependent from external financial support. While a considerable number of completed minor schemes and a small number of D&F canal systems of major/medium schemes are reported to have been handed over to FOs, the outlook for sustainability of the Project is promising for only a few of the schemes where FOs have been well established and O&M plans implemented. For the vast majority of the schemes, the sustainability will depend on further efforts, in forming strong FOs to adequately organize O&M operations. While Sri Lankan village irrigation schemes by tradition have some form of beneficiary association, evidently only a few of these groups in the project area have had the strength to keep the schemes in good operating conditions after completion of rehabilitation. It seems that this endeavor has weakened by a decline in farmers' sense of responsibility for their schemes in cases where they were not fully participating in the planning, design and implementation of the rehabilitation works and where rehabilitation has not been completed to the full satisfaction of FOs. A possible solution to improve this situation is to continue to support the IOs effort in setting up O&M funds, organizing deferred repair and maintenance works and further strengthening of water rnanagement. Without such follow-up programs, the sustainability of the project achievements appears to be at risk in the majority of the rehabilitated schemes. To this end, the mission is pleased that CMC has requested EC to continue to support for the 'Aftercare' program. 37. Related to the issue of sustainability are farmers' willingness, capacity and commitment to undertake the O&M responsibilities. O&M of minor schemes are usually not very complex and therefore with further institutional and technical support, FOs would be able to operate and maintain those schemes without much difficulty. But with respect of medium/major schemes, where O&M is relatively heavier, it is doubtful whether farmers would be willing and capable of undertaking the full responsibilities even of the D&F canal systems. At most, FOs appear to be willing to and capable of operating the D&F systems and participating in labor-intensive maintenance tasks under the technical 35  guidance of lAs, but not in providing inputs in cash or in kind to finance full maintenance costs even of D&F systems. The project's expectation that FOs would take over the full O&M responsibility of headworks and main canals starting two years after the rehabilitation works are completed appear to be highly unrealistic and would not be achievable in the foreseeable future. Among many reasons for this situation, the low productivity and profitability of present irrigated farming, complex land tenure problems, and lack of a strong commitment for the system turnover and cost recovery by all concerned parties, seem to be the main ones. IV. LESSONS LEARNED 38. The NIRP is the first country-wide project aimed to ensure the sustainability of irrigation schemes through rehabilitation and improved O&M and creation of FOs to managing the rehabilitated schemes. While, the project is not likely to achieve its development objectives fully, the planning, design, implementation and outcome of the project provide a useful learning experience both for the Borrower and Bank. (i) Irrigation rehabilitation alone is a necessary but not a sufficient condition to increasing agricultural production, incomes and standard of living of farmers. In the context of Sri Lanka's irrigated agricultural subsector, it is only a very marginal intervention. Providing innovative agricultural technologies and other services including rural credit and market facilities would be crucial to generate more benefits from rehabilitation. (ii) Forming FOs and carrying out irrigation rehabilitation at the same time do not work well. Before starting rehabilitation works, it is essential to maintain for at least a year, an effective institutional support aiming at establishing well-motivated FOs and self-sustaining agricultural input services. (iii) Land tenure problem is a factor that inhibits the development of viable FOs. Unit cost of rehabilitation is higher for schemes in areas of difficult land terrain. Also, value of ten percent of the civil cost of major/medium schemes seems to be beyond FO's capacity to bear. These factors should be considered in scheme selection for rehabilitation. (iv) Improvements to one scheme affects hydrology and water availability at other downstream schemes in tank cascade systems. This factor needs to be adequately analyzed in scheme selection, planning and design for future irrigation rehabilitation projects. (v) It is essential to develop and enforce pragmatic civil work specifications, and contract packaging, administration and quality assurance systems suitable for rehabilitation to ensure sustainability of the rehabilitated schemes. V. NEXT STEPS Project Closure Decisions and Actions 39. A sum of SLRs 300 million (US$ 4.6 million) has been allocated by the Government in the current financial year to enable completion of ongoing rehabilitation of schemes whose completion dates are beyond partial Credit Closure on June 30, 1998. It is expected that the CMC will clear the backlog of all unclaimed project expenditure, and send all withdrawal applications for project expenditure for components not extended to reach IDA before October 31, 1998. 36  40. The CMC will expedite the implementation of ongoing works in the NE and ensure completion of those schemes before December 31, 1998, as scheduled. Proposed Northeast Rehabilitation Project 41. It is also expected that CMC will expedite the ongoing feasibility studies, including socio-economic studies, of the sample schemes in the NE and ensure completion by July 31,1998, as scheduled. To enable the speedy preparation of the project proposals for consideration of IDA funding, the mission recommends that the MIP, in consultation with the relevant Government and Provincial authorities, take the following actions. (a) establishment of a separate Project Implementation Unit (PIU), with necessary staffing including a Project Director within the project area and (b) appointment of a Colombo-based Coordinator to provide guidance to the PIU and coorclinate preparation activities with the MIP, Bank and the PIU. Implementation Completion Report (ICR) 42. The CMC is currently assembling data on the project perfomiance essential to carry out the ex-post economic and financial analysis of NIRP. As agreed with the PI), these data will be sent by the CMC to Mr. Raffaele Suppa at FAO Headquarters in Rome with copy to Mr. Nihal Fernando, Task Leader at World Bank Resident Mission in Colombo, not later than 10 July 1998. The Government is required to prepare its own project evaluation report. As the works in the NE is less than ten percent of the total project, the CMC will prepare its evaluation report on the basis of the data and records available with the CMC at present. The Government will submit this report to ID no later than November 30, 1998. The ID plans to complete the final draft of the ICR before December 31, 1998. The Government's report will be attached to the final ICR. 37  Appendix B SRI LANKA NATIONAL IRRIGATION REHABILITATION PROJECT (Credit 2260-CE) Financial and Economic Re-evaluation An ex-post financial analysis has been carried out to assess the achievement of the stated objectives of the project. In line with the approach used at appraisal, farmn budgets were prepared for the same sample schemes used for the baseline, namely Kaltota, Mahagal Wewa (major/medium schemes) and Dorakada Liyadde, Kobeigane, Mahakiri Ibbewa and Nittewa (minor schemes). The other corresponding basic data required for the analysis were provided by the DAS, based on a study carried out during the course of the project, in respect of 20 minor schemes. This study covered ten completed minor irrigation tank schemes, and ten minor irrigation anicut schemes, representing various districts, and whose implementation was commenced in 1992/1993. These 20 schemes included 3 sample schemes (Mahakiri Ibbewa, Kobeigane tank schemes and Korakada Liyadde anicut scheme) analyzed at appraisal. The selection of schemes is considered fairly representative (Tables 1 and 2). The average farm size of the 20 selected schemes range from 0.2 to 1.1 ha (Table 8). The cropping intensity for tank schemes without and with project is estimated at 111% and 119% respectively. Similarly related cropping intensity for the without and with project situations for anicut schemes is estimated at 165% and 168% respectively (Tables 1 and 2). Gross incomes and cost of production were estimated for a single crop, i.e. paddy, using actual farrn gate 1999 prices. Only rice is considered because it is by far the most significant and representative crop for all project schemes, reliable information on production costs and farm gate prices for rice are readily available, and there is no significant evidence that cropped area under other field crops increased significantly under the project. However, these production costs and underlying assumptions are to be considered as average values and are equally applicable to both major/medium and minor schemes. Production costs include the O&M cost of schemes to be bone by farmers, at an average estimated rate of Rs. 500/ha for the without project situation and Rs. 900/ha for the with project situation (Tables 3 and 4). In estimating future project benefits, and in line with the analysis given at appraisal, it has been assumed that, in the absence of the project, the conveyance system and control of water deliveries within the irrigation schemes would have further deteriorated with the likely result of a reduction of farm output by about 20%. On the basis of this assumption and as indicated above, increases of net farm income over the without project situation would range from 28% (DorakadaLiyadde - Anicut) to 50% (Nittewa - Tank). Although considered satisfactory at farm income level, these results are well below appraisal expectations. The results of the farm income analysis is presented in detail in Table 9 and are summarized below: 38  Net Farm Income Incremental Farn Nel; Farm Income Size Present With Over Over Increase Project Project Present Without _________________________________________________________________ Project. (ha) . . .(Rs.) ............................ . (%) A. Medium/Major Schemes - Kaltota (Anicut) 0.40 13,881 11,105 15,144 1,263 4,039 36 -MahagalWewa(Tank) 1.10 38,199 30,559 41,644 3,445 11,085 36 B. Minor Schemes - Dorakada Liyadde 0.20 4,540 3,632 4,637 97 1,005 28 (Anicut) - Kobeigane (Tank) 0.20 5,660 4,528 6,632 972 2,104 47 - Mahakirilbbewa(Tank) 0.35 9,757 7,806 11,581 1,824 3,775 48 -Nittewa(Tank) 0.60 16,734 13,387 20,066 3,332 6,679 50 Economic Re-evaluation The economic analysis at appraisal was carried out for the project as a whole and for six sample schemes (Kaltota, Mahagal Wewa, Dorakada Liyadde, Kobeigane, Mahakiri Ibbewa and Nittewa). Because of lack of detailed data regarding the actual project performance relative to the six schemes analysed at appraisal, the economic re-evaluation at project completion could only be carried for the project as a whole. The overall Economic Rate of Return (ERR) fer the project is estimated at 14.1% (Table 14) which, although considered satisfactory, is well below tie ERR of 31% assessed at appraisal. Major factors that have negatively affected the economic impact of the project include less than expected increases in both the cropping intensity and paddy yields (Tables 1, 2 and 5). The main assumptions underlying the analysis and the methodology followed are described below: (a) Agricultural production: In the absence of complete data regarding actual land use in different project areas, estimates for the entire project are based on production parameters, cropping intensity and cost data from the sixrepresentative sample schemes. The value of the agricultural production has been estimated taking into account only one crop, i.e. rice paddy, for both Maha and Yala seasons. Details of production-related variables are given in Tables 1 to 9. As indicated in Tables 6 and 7, total project benefits, estimated on the basis of incrermental rice paddy production, have been derived from a total cultivated area of about 47,500 ha adding together 28,400 ha in the minor irrigation schemes and 19,100 ha in the medium/major irrigation schemes, and corresponding to 1,036 minor schemes and 33 medium/major schemes respectively. (b) Input - output prices: Considering Sri Lanka remains a net importer of rice, the economic price of paddy is based on projected 2010 import parity price derived from the January 1999 World Bank Commodity Price Projection and expressed in 1999 constant prices as shown in Appendix B, Table 12, where economic prices of fertilizer are also given. While labour costs were adjusted by a conversion factor of 0.9, other operating costs were converted into economic value by applying a Standard Conversion Factor (SCF) of 0.85 (Tables 2 and 3). (c) Project implementation The first batch of completed schemes were delivered as late as in 1995. For the analysis, all minor irrigation schemes (1,048) and 33 major/medium schemes (out of a SAR figure of 60 subsequently reduced to 35 at Mid-Term Review) were considered fully rehabilitated under the project. The 39  analysis has been based on a project investment life span of 25 years, in line with SAR approach. (d) Investment costs: The analysis was based on actual project costs as provided by the CMC and given in details in Table 10. The local costs were adjusted by applying a SCF of 0.85. (e) Operation and maintenance costs: O&M costs as estimated in Tables 3 and 4, are included in the crop budget analysis and were adjusted by a SCF of 0.85. The ERR was calculated on the assumption of adequate maintenance of the completed works. However, this is considered unlikely. A sensitivity analysis was carried out to assess the ERR assuming 20% reduction in agricultural production phased over time due to poor maintenance. For that scenario the ERR is 10%, which is marginal. 40  Table 1. Cropping Intensities in Selected Minor Schemes (Tanks)1/ No. Name of Tank Const. District Farm. Irrigable Cropping Intensity (%) Total Cultivated Area (Acs) Year Families Area (Acs) W/O Project With Project W/O Project | With Project 01 Nelugollekada Wewa 1992 Anuradhapura 40 140 83 90 116.2 126.0 02 Mahakiri IbbewaWewa 1992 Anuradhapura 50 95 100 120 95.0 114.0 03 PethiyannakadaWewa 1992 Anuradhapura 65 210 110 115 231.0 241.5 04 Kobeigane Wewa 1992 Kurunegala 68 87 145 150 126.2 130.5 05 Karawadeniya Wewa 1992 Kurunegala 60 30 114 125 34.2 37.5 06 Thittawela Wewa 1993 Kurunegala 80 95 88 100 83.6 95.0 07 DunupothaWewa 1993 Kurunegala 44 80 200 200 160.0 160.0 08 Aulegama Wewa 1993 Kurunegala 60 40 80 90 32.0 36.0 09 MatihakkaWewa 1993 Hambantota 10 35 100 100 35.0 35.0 .r.- 10 Omara Gonawetuna Ara Wewa 1993 Hambantota 40 48 85 100 40.8 48.0 Total 860 954.0 1,023.5 Weighted Cropping Intensity (VVithout Project) = 110.9% (954.0/860) Weighted Cropping Intensity (With Project) = 119.0% (1,023.5/860) Incremental Cropping Intensity = 8.1% 1/ The irrigable area and cultivated areas are shown in acres to retain the originality of data provided by DAS and CMC.  Table 2. Cropping Intensities in Selected Minor Schemes (Anicuts)1/ No. | Name of Tank Const. District Farm. Irrigable Cropping Intensity (%) Total Cultivated Area (Acs) Year Families Area (Acs) W/O Project With Project W/O Project With Project 01 Dorakada Liyadde Anicut 1992 Kalutara 25 40 187 194 74.8 77.6 02 Mahakumbura Anicut 1993 Badulla 30 70 182 192 127.4 134.4 03 Ekke Amuna 1993 Kegalle 15 16 200 200 32.0 32.0 04 Edanwala Amuna 1993 Kegalle 84 27 200 200 54.0 54.0 05 Punmulla Anicut 1993 Ratnapura 68 52 196 196 101.9 101.9 06 Halangoda Amuna 1994 Matale 65 100 146 146 146.0 146.0 07 Ittawala Anicut 1993 Matara 240 120 130 135 156.0 162.0 08 Pallewela Anicut 1993 Badulla 30 27 200 200 54.0 54.0 09 Devale LiyaddaAnicut 1993 Ratnapura 46 50 142 150 71.0 75.0 10 Mitiwalatenna Amuna 1993 Matale 36 27 200 200 54.0 54.0 Total 529 871.1 890.9 Weighted Cropping Intensity (Without Project) = 164.7% (871.1/529) Weighted Cropping Intensity (With Project) = 168.4% (890.91529) Incremental Cropping Intensity = 3.7% 1/ The irrigable area and cultivated areas are shown in acres to retain the originality of data provided by DAS and CMC.  Table 3. Paddy: Crop Budget - Production Cost Without Project in Both Financial and Economic Terms Activity Unit No. of Unit Unit Coist Total Costl/ I I | (Rs./unit) (Rs./acre) 1. Land Preparation - Tractor L.S ... .. 1,600 2. Seeding - Seeds kg 44 18 792 3. Fertilizer - Base Application (Vi) kg 50 16 800 4. Chemicals - 1st Weeding L.S . . 250 - 1 st Pest Control L.S .. ... 220 - Hire of Spraying Equipment L.S ... . 300 S. Harvesting - Hired Labor pd 4 150 600 - Tractor for Threshing L.S ... 600 - Other L.S 300 6. O&M Costs acre 1 500 500 Total Production Costs 5,962 in Financial Terms (Rs./acre) Total Production Costs (Rs./ha) 14,726 in Financial Terms (Rs./ha) Total Production Costs 12,869 in Economic Terms (Rs./ha) 21 1/ Production costs are given on per acre basis to retain the originality of data provided by DAS and CMC (1999 prices). 2/ To convert financial values into economic values (as indicated in the text of Appendix B: Economic Re-evaluation), a SCF of 0.85 has been applied to the financial costs of mechanization, seeding, chemicals and O&M. A conversion factor of 0.9 has been applied to hired labor. Conversion factors for fertilizers are given in Table 12. 43  Table 4. Paddy: Crop Budget - Production Cost With Project in Both Financial and Economic Terms Activity Unit No. of Unit Unit Cost Total CostlI (Rs./unit) I (Rs.iacre) 1. Land Preparation - Tractor L.S ... ... 1,600 2. Seeding - Seeds kg 44 18 792 3. Fertilizer - Base Application (Vi) kg 50 16 800 - First Top Dressing (Urea) kg 25 7 175 - Second Top Dressing (Urea) kg 25 7 175 4. Chemicals -1 st Weeding L.S ... 250 - 1st Pest Control L.S 220 - Hire of Spraying Equipment L.S 300 5. Harvesting - Hired Labor pd 5 150 750 - Tractor for Threshing L.S ... ... 600 - Other L.S ... ... 300 6. O&M Costs acre 1 900 900 Total Production Costs 6,862 in Financial Terms (Rs./acre) Total Production Costs 16,949 in Financial Terms (Rs./ha) Total Production Costs 15,702 in Economic Terms (Rs./ha) 21 1/ Production costs are given on per acre basis to retain originality of data provided by DAS and CMC (1999 prices). 2/ To convert financial values into economic values (as indicated in the text of Appendix B: Economic Re-evaluation), a SCF of 0.85 has been applied to the financial costs of mechanization, seeding, chemicals and O&M. A conversion factor of 0.9 has been applied to hired labor. Conversion factors for fertilizers are given in Table 12. 44  Table 5. Minor and Medium/Major Irrigation Schemes: Yields of Paddy Without and With Project Without Project With Projeci= Name of Scheme/Type Maha Yala Maha I Yala ------------- (tonslacre) A. Minor Schemes I/ 1. Dorakada Liyadde-Anicut 1.2 1.1 1.3 1.2 2. Kobeigane -Tank 1.8 1.7 2.0 1.9 3. Mahakiri Ibbewa - Tank 1.8 1.7 2.0 1.9 4.Nitftewa - Tank 1.4 1.3 1.6 1.5 B. MedlumlfMajor Schemes 2/ 1. Kaltota 1.5 1.4 1.6 1.5 2. Mahagal Wewa 1.5 1.4 1.7 1.6 1/ Data provided by DAS based on a study covering ten Tank schemes and ten Anicut schemes. For the without project situation, average yields have been estimated on the basis of data over a period of ten years. The possible negative effect of three bad crop years has been considered in estimating yields for the with project situation. 2/ CMC's estimates. 45  Table 6. Minor Schemes: Total Project Area and Estimated Rice Paddy Production Command Irrigable Cropping Cultivated Yields 3/ Production Area Area 1/ Intensity 2/ Area (ha) (ha) (%) (ha) (tons/ha) (tons) I - Without Project -Tanks 17,500 14,875 111 16,511 3.99 65,880 -Anicuts 7,500 6,375 165 10,519 2.84 29,873 Total 25,000 21,250 127 27,030 3.54 95,753 11 - With Project CYN --Tanks 17,500 14,875 119 17,701 4.49 79,479 -Anicuts 7,500 6,375 168 10,710 3.09 33,094 Total 25,000 21,250 134 28,411 3.96 112,573 1/ Estimated at 85% of the Command area. 2/ As given in detail in Tables 1 and 2. 3/ Average rice paddy yields Maha and Yala seasons; based on data given in Table 5. Totals do not add up precisely because of rounding.  Table 7. Medium/Major Schemes: Total Project Area and Estimated Rice Paddy Production Command Irrigable Cropping Cultivated Yields 4/ Production Area I/ Area 2/ Intensity 3/ Area (ha) (ha) (%) (ha) (tons/ha) (tons) I- Without Project 13,390 11,382 165 18,779 3.58 67,231 11 - Future With Project 13,390 11,382 168 19,121 3.95 75,528 1/ Corresponding to 33 schemes actually rehabilitated by the end of June 1999. 2/ Estimated at 85% of the Command area. 3/ Average cropping intensity based on parameters as used for anicuts (Table 2). 4/ Average rice paddy yields Maha and Yala seasons; based on data given in Table 5.  Table 8. Rice Paddy Production for Selected Farm Models Average Cropping Cultivated Area Rice Paddy Yields 1/ Production Intensity 1/ Farm Size W/O Project With Project WIO With W/O Project With W/O Project With Project I Project I Project n Project (ha) (%) - (ha) ---------- ---------- (tons/ha) ---------- ---------- (tons) ---------- A. Medium/Major Schemes - Kaltota 0.4 165 168 0.66 0.67 3.58 3.95 2.36 2.65 - Mahagal Wewa 1.1 165 168 1.82 1.85 3.58 3.95 6.50 7.30 B. Minor Schemes - Dorakada Liyadde 0.2 165 168 0.33 0.34 2.84 3.09 0.94 1.04 (Anicut) - Kobeigane (Tank) 0.2 111 119 0.22 0.24 3.99 4.49 0.89 1.07 Z, - Mahakiri Ibbewa (Tank) 0.35 111 119 0.39 0.42 3.99 4.49 1.55 1.87 co - Nittewa (Tank) 0.6 111 119 0.67 0.71 3.99 4.49 2.66 3.21 1/ As indicated in Tables 6 and 7 for minor and medium/major schemes respectively.  Table 9. Summary of Farm Incomes Rice Paddy I/ Farm Gross Value of Product. Farm Product. Costs 3/ Net Farm Income Incremental 2/ Farm Gate W/O Project | With Project W/O With W/O With Net Farm Price Project I Project Project I Project Income (Rs.1kg) - - .---------- (Rs.) ----- A. Medium/Major Schemes - Kaltota 10.0 23,600 26,500 9,719 11,356 13,881 15,144 1,263 - Mahagal Wewa 10.0 65,000 73,000 26,801 31,356 38,199 41,644 3,445 4S B. Minor Schemes - Dorakada Liyadde 10.0 9,400 10,400 4,860 5,763 4,540 4,637 97 (Anicut) - Kobeigane (Tank) 10.0 8,900 10,700 3,240 4,068 5,660 6,632 972 - Mahakiri Ibbewa (Tank) 10.0 15,500 18,700 5,743 7,119 9,757 11,581 1,824 - Nittewa (Tank) 10.0 26,600 32,100 9,866 12,034 16,734 20,066 3,332 1/ Increased by 10% to take into-account the value of by-products. 2i Based on farm production data given in Table 8. 3/ Based on crop budget per ha given in Tables 3 and 4 and taking into account the respective total cultivated area indicated in Table 8.  Table 10. Total Project Costs in Current Terms 1991 1992 1993 1994 1995 1996 1997 1998 Sub- 1998 1999 Grand total Total upto May from June to to End of Project Dec. Rehabilitation & Improvement Surveys 0.2 3.2 5.9 8.8 9.2 15.8 0 0 43.1 - - 43.1 Civil 54.3 12.6 34.6 68.9 145.2 319.5 296.8 97.2 1,029.1 120.0 253.9 1,478.5 E & A 0 2.6 9.2 12.0 34.2 55.8 40.2 11.9 165.9 24.0 - 189.9 Sub-total 54.5 18.4 49.7 89.7 188.6 391.1 337.0 109.1 1,238.1 144.0 253.9 1,711.5 Farmer Organization 0 1.4 6.2 10.2 15.1 21.3 23.8 4.8 82.8 37.4 - 120.2 o-n Training 0 3.2 4.2 4.2 28.9 26.3 21.3 1.4 89.5 0.6 - 90.1 Environmental 0 0 0 0 0 0 0 0 0 0 0 0.0 Protection Institutional Support & 0 6.8 3.3 7.5 8.6 10.5 33.5 41.5 111.7 58.8 - 170.5 Studies Technical Assistance 0 19.4 24.0 33.7 32.3 41.8 93.6 88.0 332.8 34.6 - 367.4 Vehicle & Equipment 0 0.7 14.6 8.1 31.4 47.8 0 0 102.6 40.3 - 142.9 TOTAL COST 54.5 49.9 102.0 153.4 304.9 538.8 509.2 244.8 1,957.5 315.7 253.9 2,602.6 (Rs. Million) TOTAL COST 1.3 1.1 2.1 3.1 5.7 9.8 8.5 3.7 35.3 4.6 4.8 44.7 (US$ million) Source: Data provided by the CMC.  Table II. Total Project Cost in Constant 1999 Terms for Economic Analysis Consumer Deflators Investment in Investment Costs Price Index 1/ Current Terms 2/ in Constant 1998 Terms (1990 = 100) (Rs. Million) (Rs. Million) 1991 112.2 2.05 54.5 111.7 1992 125.0 1.84 49.9 91.8 1993 139.6 1.65 102.0 168.3 1994 151.4 1.52 153.4 233.2 1995 163.1 1.41 304.9 429.9 1996 189.0 1.22 538.8 657.3 1997 207.1 1.11 509.2 565.2 1998 226.5 1.02 560.5 571.7 1999 3/ 230.0 1.00 253.9 253.9 1/ IMF: International Financial Statistics - March 1999. 2/ See details in Table 10. 3/ Consumer price index for 1999 estimated by the mission.  Table 12. Derivation of Import Parity Prices for Rice and Fertilizer Parity' Units | [I] [I] RI Projected World Price (2010) Rice Urea DAP In 1990 constant dollars2 US$/MT 267.0 118.3 151.4 Adj. Factor to 1999 dollars3 1.054 1.054 1.054 In constant 1999 dollars US$/MT 281.4 124.7 159.6 Quality Adjustment % 80 100 100 World Market Equivalent US$/MT 225.1 124.7 159.6 International shipping US$/MT 30.04 50.0 50.0 CIF Value/Colombo US$/MT 255.1 174.7 209.6 Exchange Rate5 Rs./US$ 68.0 68.0 68.0 CIF Value/Colombo Rs./MT 17,347 11,880 14,253 Port charges6 Rs./MT 540 540 540 Internal transport6 Rs./MT 540 540 540 Internal handling6 Rs./MT 215 215 215 Processed Value Rs./MT 18,642 13,175 15,548 Processing ratio % 65 - - Processing costs6 Rs./MT (412) - - Value of by-products6 Rs./MT 463 - - Wholesale Value Rs./MT 12,168 13,175 15,548 Local marketing/transport6 Rs./MT (230) 230 230 Economic Farm-gate Price RS/MT 11,938 13,405 15,778 Financial Farm-gate Price Rs./MT 10,0007 7,000 16,000 Conversion Factor 1.19 1.92 0.99 Parity: [I] Import. Sri Lanka is a net importer of rice. 2 World Bank price forecasts January 1999. 3 MUV index. 4 Between Colombo and Bangkok. 5 Official exchange rate (February 1999). 6 Adjusted by SCF of 0.85. 7 Including value of by-products. 52  Table 13. Net Incremental Value of Production at Full Development for Economic Re-evaluation Present Ful:ure With Project WIO Project A. Production 1/ (Tons) - Minor Schemes 95,753 112,573 76,400 - Medium/Major Schemes 67,222 75,518 53,780 Total 162,975 188,091 130,180 B. Economic Price of Rice Paddy (Rs./kg) 2/ 12.18 12.18 12.18 C. Gross Value of Production (Rs. million) 1,985.0 2,290.9 1,585.6 D. Production Costs/ha "Economic" (Rs./ha) 3/ 12,869 15,702 12,869 E. Total Area Cultivated (ha) 4/ 46,807 47,529 37,445 F. Total Production Costs "Economic" (Rs. 602.4 746.3 481.9 million) G. Net Value of Production (Rs. million) 1,382.7 1,544.6 1,103.7 H. Net Incremental Value of Production (Rs. 440.9 million) 1 / Tables 6 and 7. 2/ Table 12. 3/ Tables 3 and 4. 4/ Tables 6 and 7. 53  Table 14. Economic Rate of Return Investment Net Incremental Balance Costs11 Value of Production2/ .................... (Rs. million) . 1991 98.3 0 -98.3 1992 80.8 0 -80.8 1993 148.1 0 -148.1 1994 205.2 0 -205.2 1995 378.3 88.2 -290.1 1996 578.4 176.4 -402.0 1997 497.4 264.5 -232.9 1998 505.1 308.6 -196.5 1999 223.4 352.7 129.3 2000 0 396.8 396.8 2001 -2015 0 440.9 440.9 Economic Rate of Return: 14.1% 1/ Based on data presented in Table 11; a SCF of 0.85 has been applied to the local cost component estimated at 80% of total investment costs. 2/ Presented in Table 13; values during project years are estimated as follows: 1995: 20%; 1996: 40%; 1997: 60%; 1998: 70%; 1999: 80%; 2000: 90%; 2001: 100%. 54  Appendix C SRI LANKA NATIONAL IRRIGATION REHABILITATION PItOJECT (Credit 2260-CE) Co-financier's Comments EUROPEAN UNION Delegation of the European Commission to Sri Lanka 81, Barnes Place, Colombo 7, Sri Lanka Tel: 699745 / 698819 Fax: 698820 Date: 09 March 1999 Ref: JHG/psa-99/141 Mr Roberto Bentjerodt Resident Representative World Bank DFCC Building (Ist Floor) 73/5, Galle Road Colombo 3 Subject: NIRP - Draft ICR Ref: Mr. Fernando's letter of 4 March Dear Mr Bentjerodt, The delegation has examined the draft report. We can broadly agree with the observations and would support the conclusions and recommendations. I would like to clarify the issue of the EC contribution, however, and ask that the relevant parts of the draft (e.g. Table 8B) be modified accordingly. Basically we do not agree that our Food Aid Counterpart funds should be classified within the GOSL contribution and would like it to be shown, correctly, as part of the EC funding. In our financing agreement with the GOSL that amount of 7.58 million euro (ecu) is clearly distinguished as a Commission contribution to the project. There was also an amendment to the F.A, made in 1995, in which we increased the grant component by 0.45 million euro. I enclose a copy of the Appendix I to Annex A (attachment to Appendix C of this ICR) of that Rider to illustrate that. It also shows that the anticipated GOSL component was less than half that indicated in your table 8B. The summary statement of expenditure enclosed, which was prepared by the CMC, also shows that the actual GOSL contribution was about 4.56 million euro, at the current exchange rate. 55  I am sure you want your report to be accurate in all respects. As the EC undertook to finance more than 25 percent of the estimated project cost that should not be misunderstood by the readers of the final version so I look forward to the necessary corrections being made. Yours sincerely, Ilkka Uusitalo Ambassador Head of Delegation Encl. as above 56  Attachment to Appendix C APPENDIX 1 TO ANNEX OF EU GRANT AGREEMENT FINANCING AGREEMENT: ALA/91/07 - National Irrigation Rehabilitation Project FINANCIAL BREAKDOWN Budget Item Initial Amount of EC Initial Amount of EC Modification Modification New Financial New Financial WB GOSL Contribution to Contribution to to initial to initial Breakdown Breakdown (IDA) Financial Agreement Financial Agreement amount in amount in ECU in ECU in in in M.ECU MECU M.ECU MECU M.ECU MECU Art. 73000 ECFoodAid Art. 73000 EC Food aid Art. 73000 ECFoodAid Counterpart Funds Counterpart Funds Counterpart Funds Surveys and -- 0.42 -- Investigations f_____ Civil Works 19.58 3.25 Engineering and 1.42 1.42 Administration __ _ Farmer Organizations -- -- 2.25 Training 1.070 0.850 (0.200) (0.070) 0.780 0.780 -- Environmental Protection -- 0.830 (0.830) -- -- -- Institutional Support and 1.010 2.160 (1.010) (0.603) 1.557 0.08 Studies __ _ _ Aftercare Program -- -- -- 0.900 -- 0.900 -- Technical Assistance 0.580 3.670 1.670 0.200 2.250 3.870 -- Transport and Equipment 0.680 0.070 (0.079) 0.120 0.601 0.190 0.91 1.25 Contingencies -- -- 0.159 0.283 0.159 0.283 -- -- TOTAL 3.340 7.580 0.450 0.000 3.790 7.580 24.66 5.92 Rider N* 01 to Financing Agreement ALA/91/07  Appendix D SRI LANKA NATIONAL IRRIGATION REHABILITATION PROJECT (Credit 2260-CE) Borrower's Comments and Observations on draft ICR (Note: the paragraph numbers indicated below refer to the paragraph numbers of the Part I of draft ICR. These paragraph numbers are differenet from that of the final ICR included in this report. Readers are therefore requested to read the comments in relation to the subheadings of the final ICR) Part I: Project Implementation Assessment A. Background and Statement/Evaluation of Objectives 1 to 4 Paragraph Nos 1 - 4 : Agreed. No comments Paragraph No. 5 : Project Changes after Appraisal The Project objective and purpose as stated in the logical framework prepared for European Commission were: Overall objective : Increase and more secure agricultural production in rehabilitated irrigation schemes. Project Purpose: Sustainably improved irrigation supplies in rehabilitated schemes. It may be appropriate to mention about this in this paragraph. Paragraph No 6 : Evaluation of Project Objectives. It is mentioned that the project was well conceived with it's objectives being consistent with IDA strategy. However, the main project objective to stabilise and increase agricultural production was too demanding and unrealistic for the following reason: Rehabilitation alone does not increase crop yields in irrigation schemes without an adequate agricultural component. In fact, it is the latter component (agricultural extension, improved agricultural practices etc) that contributes to a significant increase of production. In the project, only a very small fund provision was available (about 0.2% of project cost) for agricultural demonstrations. With such a low agricultural component, it is highly un-realistic to expect an increase of production as project outcome. In fact, in the Logical Framework Approach the project purpose is taken as the outcome of the project and the overall objective is considered as the ultimate outcome with additional inputs external to the project. 58  B. Achievement of Project Objectives Paragraph No. 7: It would have been more appropriate to state that the project objectives were achieved only partially through a partial project period. Further, at the time of SAR preparation, it was a very well known that the PCs were being yet established and the fact that they would require some time to organize themselves to undertake a task of this magnitude has not been duly recognized. Further, it is more appropriate to compare the achievements with the project purpose in the Logical Framework as mentioned under comments on paragraph No.6 Paragraph Nos: 8 & 9: Conclusions have been made based on a survey carried out on a very small sample of rehabilitated schemes. A better indication could be had after the Benchmark and E,valuation Study, undertaken by a firm of Local Consultants, is completed. Paragraph No. 10: The statement that only 6 major/medium schemes out of a target of 35 schemes were rahabiliated is incorrect and the total number of major/medium schemes completed by credit closing date was 13. It should also be mentioned that the bulk of the balance schemes had achieved a physical progress exceeding 75%. Paragraph No.11: The final sentence "Overall, only a few of the project FOs are reported to be functioning satisfactorily" does not reflect the correct situation. A Performance Assessment Study carried out by the Department of Agrarian Services on 953 schemes, have revealed that in 40% of the schemes FO development has been highly successful and in about 42% of the schemes there is a potential for success. Already considerable work has been done in rehabilitated minor schemes to ensure a sustainable O&M system. Steps have been taken to set up an O&M fund by the farmers. Even in major/medium schemes, several FOs are functioning satisfactorily. The World Bank seems to have used the number of FOs registered under clause 56b (of the Agrarian Service Act) as the indicator to arrive at this conclusion. The issue of FOs attaining a status under Clause 56b has been discussed at length during an earlier supervision mission. Also the Bank agreed by then that this parameter should not be used as a performance indicator of the NIRP. The Mission should amend its observations regarding this subject accordingly and throughout the document Paragraph No. 12: The number of completed major/medium schemes mentioned should read as 13 and not as 11 Paragraph No. 13: To state that the consultants have performed satisfactorily except for the institutional development support at field level is misleading. It should be noted that a total of 7 Institutional Development Officers (IDOs) were attached to the Regional Support Teams, and fiurther supported by a Senior Institutional Advisor and a Farmer Training Specialist. With a permanent representation in the field by these IDOs and through very frequent and intensive guidance by two senior specialists, consultants have done the maximum to contribute to the further development and strengthening of 59  FOs and the Implementing Agencies (IAs). However, to expect, from this team a full coverage of all 1000 minor schemes and 35 major/medium schemes is unrealistic. They had to rely (heavily) on the training and coaching capacity of the IAs who were constrained by the lack of trainers. This has - indeed - resulted in a less than optimal institutional strengthening ofFOs (and the IAs for that matter) but to "blame" the consultants for that is incorrect. At mid term review concern was expressed about this poor performance after which provisions were made available (by the EU) to include a specifically designed component to the project (the Aftercare Programme) to address the issue more directly. The development and initiation of this programme has taken more time than anticipated, mainly due to the pressure to complete as much as possible of the physical implementation. It seems that the programme is now firmly rooted, but it is too early too assess - in earnest - its impact. In addition it should be noted that the overall impact of the Aftercare Programme is expected to be limited due to constraints in available resources. It would be appropriate if this is mentioned in the completion report. Paragraph Nos: 14 & 16: Agreed. No Comments Paragraph No. 17 It should be noted that the IIMI offered a no cost extension of their services. However this was not agreed to, by the EC due to the general poor performance of the IRMU. IIMI is partly to be blamed for this poor performance by focussing very much attention to indulging in research instead of actively pursuing the establishment and strengthening of this unit (which was their mandate) Paragraph No.18 - 19: Agreed. No Comments. Paragraph No. 20: The project's Economic Rate of Return needs to be re-checked after completion of the Benchmark and Evaluation Study, when more data would be available. C. MajorFactorsAffecting the Project Paragraph No. 21 It is mentioned that the quality of Civil Works was below expectations. A more appropriate statement would be that the quality of Civil Works did not reach optimum levels in some of the schemes. Paragraph Nos: 22 - 23 Agreed. No Comments Paragraph No. 24 The statement that the PCC was in-effective in providing policy guidance and strategic directions to CMC is incorrect. The PCC was the apex body that directed the CMC in all policy matters connected with the project. Paragraph Nos: 25 - 26 Agreed. No Comments 60  D. BankPerformance Paragraph No. 27 The statement "Therefore the Bank's performance in project identification/ preparation is rated satisfactory" is not entirely correct (See comments Para.28) Paragraph No. 28 : It is mentioned that the important lesson learned from the implementation of VIRP, Viz. The formation of FOs should come before the physical implementation, was not given adequate attention at project design. This lesson was well known to the GOSI, as well as to the appraisal mission of the Bank. Two members of the appraisal mission had been members of VIRP review missions as well. Yet, at the time of appraisal, the World Bank team stressed for a programme which included physical works during the first two years of the project. The implementation of the NIRP further emphasized this lesson. Even at present, the World Bank seems to be not giving head to this important lesson in the proposed North East Emergency Irrigated Agriculture Project where steps are being taken to call for tenders to execute some physical works during the first few months of the project. The NIRP project criteria stipulated the handing over of head works and main canals of major/medium schemes to FOs for O&M. The fallacy of this stipulalion was emphasised by the GOSL delegation at negotiations. Yet, the Bank insisted on this stipulation. Other than for these two aspects, it is agreed that the Bank's performance in project design and appraisal was satisfactory. Paragraph No. 29 The role of the ICRC and the UNHCR as part of an 'innovative project back-up and supervision arrangement appears to be slightly overstated. E. Borrower Performance Paragraph No. 30: (a) It is agreed that if Deputy Project Directors with adequate authority and support staff were appointed in the provinces, the implementation could have been better coordinated and managed in an island-wide project of the nature of NIRP (b) The statement that the National Steering Committee failed to addlress the sector policy issues and sustainability aspects is not correct. It must be mentioned that the sector policy issues could not be addressed in isolation with respect to NIRP alone but should be addressed as a national issue. (c) Some Provincial Project Coordinating Committees were not very effective may be a more appropriate statement. (d) Agreed. (e) This statement should be re-worded to mention that the quality cf civil works were not up to optimum standards in some schemes. 61  F. Assessment of the outcome, Paragraph No 31: The project's overall outcome has been rated as unsatisfactory, as the project failed to establish an effective and sustainable )&M system for the rehabilitated schemes. This is very puzzling considering the steps the project has taken (and is taking) to address the issue of sustainable O&M. As pointed out under comments on paragraph No. 11, considerable work has been done to establish an effective and sustainable O&M System in the rehabilitated minor schemes. This aspect is being further pursued in the after-care programme. Even in major/medium schemes, considerable work is being done under the aftercare programme to strengthen FOs to execute O&M. Further, the outcome has to be compared with the project purpose of the Logical Framework. If this approach is adopted, the project outcome could be assessed as satisfactory. G. Future Operations: Paragraph Nos 32 : Agreed. No Comments. Paragraph No. 33 A very good suggestion has been made to conduct a joint (Bank and GOSL) post-project impact evaluation on the status of FOs, status of O&M etc of rehabilitated schemes. This study should be best done two to three years after rehabilitation, when the post project- production situation also could be assessed better. L LessonsLearned: Paragraph No 34 (a) Irrigation rehabilitation . ... Agreed (b) Forming FOs and ........ This is the most important lesson learnt from NIRP. Even then, the proposed NEEAP (to be funded by a IDA Loan) does not even currently recognize the necessity to develop and strengthen the FOs (established or to be established) prior to rehabilitation. (c) The land tenure ......... Agreed. Again at appraisal, the Sri Lanka delegation pointed out that the 10% contribution is beyond the farmers capacity to bear. (d) Improvements to One ...... Agreed. (e) It is essential ......... Agreed (f) To assure the ......... Agreed (g) From the point ........ Agreed. However, the Bank also had a responsibility to ensure that this system was adopted. (h) The establishment . .... Agreed. (e) International ....... This is a policy matter to be decided at National level. (F) Implementation responsibility: In a national project of this nature, the mainauthority (Project Director) should be at center and should be supported by Deputy Project Directors at provincial level. 62  J. General Comments (I) Paragraph on Key Factors Affecting the Project - of the Evaluatioln Summary. Notwithstanding the other factors, listed in this paragraph, thz delay in appointment of consultants was 17 months. The project therefore may be deemed to have commenced in June 1993 and closed in June 1998 when project performance was at its peak. The project period was therefore six years and one month. The project closure vwas at a time when project performance was at its peak. It is regretted that this has only been mentioned and not taken into account in the evaluation of the project performance. (II). It must be mentioned that the European Union, the Co-financier -- provided funds to the time of around 25% of the project cost. This included crucial components of the project such as training, technical assistance, quality control, Aftercare programme and the provision of training and laboratory equipment. The completion report does not reflect very much on this aspect. (III) The Project Completion mission of the World Bank had discussions with the European Commission. However, their views are not reflected in the report. (IV) It is erroneously mentioned in the evaluation summary that the progress of major/medium schemes was 17% at credit closure. This should be corrected to read as 75% 63  Appendix E Borrower's Project Implementation Completion Evaluation Summary DEPARTMENT OF EXTERNAL RESOURCES Ministry of Finance and Planning The Secretariat P.O.Box 277, Colombo 1, SRI LANKA 27t May, 1999 Ms. Mariana Todorova Country Director World Bank Office Colombo Dear Ms. Todorova NATIONAL IRRIGATION REHABILITATION PROJECT BORROWER'S PROJECT COMPLETION REPORT I forward herewith the Borrower's Project Completion Report of the National Irrigation Rehabilitation Project. While the contents of the report are endorsed the following need to be further highlighted: 1. Where beneficiary contribution is anticipated an institutional development phase must precede construction. 2. Bank funding cycles would not normally coincide with rehabilitation cycles, especially where major schemes are invited and cultivation is the only source of income of farmers which requires staggered rehabilitation programmes in line with the cultivation seasons. In developed major schemes rehabilitation project implementation has taken between 7-10 years to complete including in earlier Bank funded projects. 3. With the banks experience in similar projects the decision not to extend the loan closing date notwithstanding that the projects initial delays were beyond its control and progress and improved significantly in the last two years is inexplicable. This created a disharmony between the institutional and construction activities the latter having to be telescoped and 64  accelerated as a result. This has in some cases affected the turn over programme to FO and possibly sustainability as anticipated under the Project. 4. Notwithstanding great emphasis to and mention of the need for institutional development the composition of the many Bank Missions have been engineering biased and ultimate decision making highlight preoccupation with infrastructure. The EC focus on institutional aspects is heartening. 5. The need for after care maintenance of institutional programmes have been highlighted and are necessary in future projects if FO sustainability and financial viability needed to reinforce efficient O&M is anticipated. It is noted that at least a lead time of two years is required for initial institutional start up where FO do not exist and at least one year after care subsequent to completion of construction activities for institutional consideration. Issues relating to land tenure, quality control, selection of schemes etc. as highlighted in the Implementation Completion Report of the Bank have been noted for action. Yours sincerely, Nalini Madanayake Director/World Bank Division For DG/ERD cc: 1. Secretary/Ministry of Irrigation & Power 2. Project Director (NIRP) 65  SUMMARY OF BORROWER'S EVALUATION OF THE NATIONAL IRRIGATION REHABILITATION PROJECT (CR.2260-CE) 1. Background 1.1 Project Development Objectives: The project objectives at appraisal were; (a) Stabilize and increase agricultural production and incomes; and (b) Raise the standard of living of farmers, through rehabilitation and improved operation and maintenance (O&M) of existing irrigation schemes. Subsidiary Objectives were (I) Upgrading skills of farmers and staff of the implementing agencies; and (ii) Creating viable farmer organizations (FOs) for managing the rehabilitated schemes. At the time of preparation of Consultants Inception Report for European Commission (EC), the project objectives were reviewed and amended. The main reason for the review was that the project activities primarily focused on improving irrigation supplies, while only a very limited project activity, namely conducting agricultural demonstrations (amounting to 0.2% of the project cost) concerned the agricultural development. According to the EC, the overall objective was increased and more secure agricultural production in rehabilitated irrigation schemes, and project purpose was sustainable improved irrigation supplies in rehabilitated schemes. 1.2 Project Components: The main project components were: (i) Rehabilitation and Improvement of 1000 minor and 35 medium/major schemes covering a total irrigable area of 37,500 ha; (ii) Establishment of Farmer Organizations and Introduction of improved O&M practices in all rehabilitated schemes; (iii) Training of farmers and staff of the implementing agencies; (iv) Environmental protection studies and works; (v) Establishment of three new support units in Irrigation Department (Research Management Unit, Dam Safety Unit, Quality Control Unit); (vi) Aftercare Program (included after 1995); (vii) Consultancy services for project planning, implementation, and impact assessment; and (viii) Procurement of vehicles and equipment. 1.3 Project Cost and Financing The project was estimated to cost US$ 49.8 million at appraisal. The IDA provision was US$ 29.6 million while EC contribution was a direct grant of US$ 4.0 million and a food aid grant of US$ 9.0 million. The balance US$ 7.2 million was to be provided by the Government of Sri Lanka (GOSL). The last disbursement took place on May 1I, 1999. An amount equivalent to US$ 5.0 million was cancelled in 1997. With that the total IDA contribution to the project would be US$ 24.6 million. There is an unutilized balance of US$ 597,596.66 in the Project's Special Account maintained at the Central Bank of Sri Lanka. This amount will be returned to IDA soon. 2. Borrower's Assessment of project objectives and achievements 2.1 Appropriateness of Project Objectives The project objectives mentioned in para 1.1 were in conformity with the Government of Sri Lanka's policy on agricultural development and irrigation management transfer and were appropriate at the time of project design in 1991. 66  2.2 Size of the Project and Scope of Investment: Even though the project objective was to stabilize and increase agricultural production, only a very limited project activity, namely conducting agricultural demonstrations (fund provision of about 0.2% of the project cost) was directly related to increasing agricultural production. As rehabilitation and improved operation and maintenance alone would not ensure an increase in. agricultural production, an agricultural extension component with adequate financial provision should have been included. Other than this aspect, the other project components and financial provision were adequate. 2.3 Achievement of Project Objectives. Quantitative Progress: The project completed the rehabilitation of 830 minor schemes and 13 major/medium schemes on the original credit closing date on 30 June 1998. However, a total of 1,048 minor schemes and 34 major/medium schemes would be completed by June 30, 1999 using GOSL funds. Farmer organizations (FOs) were formed in all minor schemes and in 207 Distributary and field (D&F) canal systems by June 30, 1998. All these FOs were registered under clause 56 A of the Agrarian Services. Act, implying that they are formally accepted FOs. As of June, 30, 1998 550 numbers of completed minor schemes and 20 numbers of D&F systems were handed over to the FOs. As of April 1999, 675 completed minor schemes and 67 completed D&F systems were handed over. O&M funds have been set up in 275 minor schemes. As regards training, the project successfully financed the completion of three Masters Degrees and study tours/short-term training for 243 staff persons. Most of the short-term training courses and study tours have been rated very useful. The establishment of the Dam Safety Unit in the Irrigation Department (ID) could not be achieved as the EC funding was withdrawn. The project successfully supported the strengthening of the Civil Works Quality Control Unit of the ID, but could not build up the Irrigation Research Management Unit (ID) as envisaged. Consultancy services for project planning and implementation was successfully utilized and project procurement targets other than procuring some vehicles under EC funding were achieved during the proj ect period. Qualitative Progress: The project purpose (as per logical framework prepared for European Commission) was sustainable improved irrigation supplies in rehabilitated schemes and this was satisfactorily achieved by the project. However, the project requirement of 10% contribution by farmers was not fully achieved by the credit closing date, particularly in major/medium schemes. In some districts, similar rehabilitation projects being implemented concurrently with NIRP did not have a farmer contribution component, which caused difficulties in convincing the farmers for the necessity of this contribution. The establishment of farmer organisations was successfully executed. A performance assessment survey of FO status in minor schemes revealed that the FO development has been highly successful in 40% of the rehabilitated schemes while in another 42% of the schemes, a potential exists for success. The ongoing after-care programme included as a project component after mid-term review in 1995, is expected to improve the status further, especially in the schemes where there is a potential for success. With the setting up of O&M fund, the FOs in minor schemes should reach a sustainable status and execute O&M works as envisaged in the project. In major/medium schemes, even though the FOs have been established, motivation ofFOs to undertake O&M is difficult than in minor schemes. However, with the ongoing after-care programme, it is expected that the FOs would reach a status to execute the O&M of the distributory and field canals. However, the project's expectation of tuming over headworks and main canals of major/medium schemes to FOs for O&M would not be realized. 67  3. Borrower's Assessment of Implementation Experience 3.1 Implementation Problems The Central Management Cell (CMC) headed by the Project Director was responsible for the management of the implementation of all project activities. An integral part of the CMC was a team Expatriate and Local Consultants. The Project Director was appointed in 1991, just prior to project negotiations, but there was a long delay in appointing the other staff for the CMC. The Consultants were mobilized only in mid-1993 and the Project Accountant was appointed even later, in November, 1993. The non-engineering government staff for the CMC were never appointed. These delays caused immense difficulties in the management of the project upto 1994 and was one of the major causes for the poor progress in initial years. At the time of project appraised, the Provincial Councils (PCs) were at it's infancy and were lacking in resources (especially in experienced staff) to execute the project works. It took a considerable time, nearly two years, for the PCs to gear themselves by recnmitment of staff etc, to implement the project, thus causing a shortfall in progress. Another major problem was the irregular cash flow during the implementation. The rehabilitation activities generally peak during the period May-September each year and during this period difficulties arose in getting imprest released from the Treasury. This problem was an annual occurrence. During project design, sufficient provision has not been provided for the procurement of vehicle. Even some of the provided vehicles (under EU funding) could not be procured due to procedural delays. This caused difficulties in project supervision. Other problems that affected the implementation was a prolonged strike of the Technical Assistants of the Irrigation Department and excessive rainfall during three consecutive years. 3.2 Assessment of Borrower's Performance The feasibility studies for the project was based on detailed studies done on seventeen sample schemes. The number of schemes studied was too small for a project proposing to rehabilitate over 1000 schemes. The selection of sample schemes was not representative of different agro-climate conditions. Other than this aspect, the borrower's performance during the project preparation and design was satisfactory. The progress of implementation upto 1994, was not satisfactory for reasons already mentioned under paragraph 3.1. However, progress improved tremendously after 1995 and the overall performance of the borrower during the implementation phase was satisfactory. 3.3 Assessment of Bank's Performance The performance of the Bank during the project preparation, design and appraisal was satisfactory except for the following two aspects. (a) At project preparation, the Borrower with experience gained from the implementation of the Village Irrigation Rehabilitation Project (VIRP), fore-runner to NIRP, emphasized the necessity to 68  provide sufficient lead time in the project to establish FOs prior to commencement of rehabilitation and that there should not be any rehabilitation work during this lead period. However, the Bank disagreed and urged for an implementation programme providing for some rehabilitation works to be undertaken during the first two years. (b) The Borrower, at negotiations, emphasized the impracticability of the project criteria stipulating the handing over of head works and main canals of rehabilitated major/medium schemes to FOs for O&M. However, the Bank insisted on this stipulation. (c) The supervision missions of the Bank were timely and were very useful as valuable suggestions were given to improve perfornance. However, most of the mission members were of an engineering background and there were no members with adequate experience in farmer organization development. This was a drawback as the project experienced difficulties in executing the FO development component effectively. Eventhough, the project was financed jointly by IDA and EC, there were only two joint supervision mission. This was another drawback. During the letter stages of the project, the office of the Resident Representative of the World Bank in Colombo was upgraded to a Country Department. Subsequently, the supervision of the project was transferred to the Country Department and resolution of implementation problems requiring Bank intervention became easier. The refusal of the Bank to grant one-year extension was disappointing in view of the tremendous improvement of progress of implementation after mid term review 3.4 Assessment of Performance of European Commission (EC) The EC was not involved in the project planning and came to the picture only at negotiation stage. Most of the officers in the Central Managemeni Cell including the Project Director were not exposed to rigorous procedures of the EC earlier and experienced initial difficulties in the implementation of the respective project components. The absence of an EC Delegation in Sri Lanka at that time compounded the difficulties. However, later with the setting up of a Delegation in Colombo, adequate support was made available to the project to resolve implementation problems expeditiously. Invaluable help was rendered by the staff of the Delegation in the preparation of annual work plans. The Supervision missions of the EC also provided valuable guidance for project implementation and the mission members had the necessary background experience. The only instance where the EC performance was unsatis:factory was their failure to approve the tender dossier for procurement of vehicles without giving a valid reason. As such, the vehicles provided under EC funding for quality control purposes were not purchased. 4. Key Lessons Learned 4.1 Project Formulation (a) Sample Schemes The sample of 17 schemes studied in detail, for project feasibility, was not very representative of the agro-climate conditions and also was too small considering the total number of schemes to be 69  rehabilitated under the project. In future projects, a representative sample of appropriate size should be selected. (b) Implementing Agencies At project preparation stage, the capabilities of implementing agencies to execute the project activities should to be properly assessed. (c) Resources for Implementing Agencies The requirement of staff, vehicles and other resources should be assessed thoroughly at project preparation stage and once assessed, provision should be made without arbitrary reductions. 4.2 Project Planning, Design and Implementation (a) Project Logical Framework The Logical Frame Work Approach (LFA) is a very effective tool for project planning and implementation. Unfortunately, LFA for NIRP was prepared only at implementation stage. In future projects, this approach should be adopted from the project preparation stage. (b) Selection of schemes for rehabilitation. In NIRP, there was no hydrological criteria for selection of schemes. In future projects, hydrological criteria should be stipulated, especially for the selection of schemes in a cascade. (c) Farmer Contribution In future projects, the farmer contribution should be confined to earthwork in minor canals. During execution of NIRP, farmers experienced difficulties in fulfilling the 10% contribution, particularly in schemes where the quantum of minor works (which farmers could execute) was low. (d) Implementation Programme and Lead Time In future rehabilitation projects, an initial two year period (lead time) devoted entirely to FO development should be allowed. The rehabilitation should not commence until the FOs have reached a status to discharge their commitments. The recruitment of staff, procurement of vehicles etc could be carried out during this lead time period. This is the most important lesson learnt from NIRP. (e) Quality Control (i) Inexperienced Contractors The works in minor schemes under NIRP in most instances, were executed by inexperienced contractors which put an additional burden on quality control. In future projects, it is advantageous to pre-quality contractors. 70  (ii) Supervisory Consultants Employing independent consultants in future projects with authority to certify payments would enhance the quality of works. In NIRP, the Regional Support Teams were not empowered to suspend payments of poor quality works. 4.3 Project Sustainability (a) Income Generating Activities The sustainability of schemes rehabilitated under NIRP depends on proper execution of O&M by the FOs. In order to achieve this, FOs should be sustainable. A sound financial status is very essential for FO sustainability and income generating activities (other than farming alone) are needed for this purpose. The NIRP did not provide support for any income generating activities. In future projects, such support is a must to ensure FO sustainability. These activities could commence during i;he lead time period of the project. (a) Operation and Maintenance Fund In future projects, creation of an O&M fund by collection f-rom FO members during the lead time period should be stipulated as a criteria for selection of a scheme for rehabilitation. 4.4 After-care Programme This is another very important lesson learnt from the NIRP. The after-care programme was introduced into NIRP after mid-term review with the objective to strengthen FOs to eventually execute proper O&M of rehabilitated schemes to ensure sustainability. The initial results of this programme are very promising and this type of programme should be included as a component in future rehabilitation projects. 4.5 Project Management The Central Management Cell (CMC) headed by the Project Director (PD) and supported by the expatriate Co-Project Director was responsible for the implementation of all the project components. The CMC was set up within the Irrigation Department (ID) and the PD was at the third level of department hierarchy. The donors expected the PD to take many important decisions regarding project implementation, for which the PD was not delegated with sufficient authority. A vast project of this nature involving many implementing agencies could have been better managed if the CMC and the PD were outside the bureaucral:ic control of ID. The ideal situation would have been to set up the CMC under the Secretary of the Ministry as in the case of North-Western Resources Project financed by the ADB, This is also a very important lesson for future projects of this nature. 71  4.6 Project Account Time and again during project implementation, cash flow difficulties were faced by the implementing Agencies. This was mainly due to difficulties, the PD encountered in getting timely releases of imprest from the Treasury. To avoid this problem in future projects, the PD should be authorised to maintain a separate account and at the beginning of every financial year, the Treasury should advance an imprest to the PD (say about 75% of the expenditure forecast for the year as per approved work plan). The balance 25% could be released once the money allocated is used up. This procedure is already in operation for European Commission funded projects. 5.0 Future Operations NIRP was the first nation wide attempt to set up FOs and turnover of O&M to beneficiaries and a wealth of experience was gained during the implementation. A joint post-project evaluation by GOSL/Bank to assess the positive and negative aspects to would provide a valuable insight to the current irrigation management policies and turnover action plan of GOSL. 72  MAP SECTION   IBRD 22876 SRI LANKA NATIONAL IRRIGATION tJ ) REHABILITATION PROJECT LC)CATION OF SCHEMES OLOMETES 0 10 20 30 40 \0 Appraised Schemes: CM Medium and Major Schemes 0 Minor Schemes Prepared Schemes t \ 'L<Medium and Major Schemes 0 Minor Schemes 4 < _ < - Ntornobe - - - Agro-Climafic Zones ./ > 3X /> ( .' - - Major Roads f i Anuraclhapuradhap 12 DistrictCapitals 0 Province Capitals @ _ , , li Notioncl Capitol 0;b;>>/ - - - \ X ,.<ri' } Mft - International Boundary / I~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~l Batticloa ts vr0 | -X D =/ - I of .- X r o / * / / Kandy&