Sri Lanka - Second Agricultural Extension Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No. 18758 IMPLEMENTATION COMPLETION REPORT SRI LANKA SECOND AGRICULTURAL EXTENSION PROJECT (CR.2380-CE) December 30, 1998 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: Sri Lanka Rupees (Rs) 1987: US$1.00 Rs 30.50 1993: US$1.00 Rs 48.25 1988: US$1.00 = Rs 31.80 1994: US$1.00 = Rs 49.41 1989: US$1.00 Rs 36.04 1995: US$1.00 Rs 51.25 1990: US$1.00 = Rs 40.06 1996: US$1.00 Rs 55.27 1991: US$1.00 = Rs 41.37 1997: US$1.00 Rs 58.50 1992: US$1.00 Rs 43.83 1998: US$1.00 Rs 65.11 FISCAL YEAR January 1 to December 31 WEIGHTS AND MEASURES (Metric system) PRINCIPAL ACRONYMS USED AEWG Agricultural Extension Working Group ASC Agrarian Services Center AERP Agricultural Extension and Adaptive Research Project ARP Agriculture Research Project CARP Council of Agriculture Research Policy CCB Coconut Cultivation Board DAPH Department of Animal Production and Health DOA Department of Agriculture EDU Extension Development Unit FET Field Extension Team FRG Farmer Representative Group GET Guide (District) Extension Team GOSL Government of Sri Lanka IDA International Development Association KVS Krushikarma Vyapthi Sevaka (Village Level Extension Agent) MAL Ministry of Agriculture and Lands PCC Project Coordination Committee PET Provincial Extension Team PC/PA/PS Problem Census/Problem Analysis/Problem Solving PMU Project Management Unit RTWG Regional Technical Working Group SAEP Second Agricultural Extension Project T&V Training and Visit Vice President: Mieko Nishimizu Sector Managers: Michael Baxter and Ridwan Ali Country Director: Roberto Bentjerodt Task Manager: Terrence Abeysekera  FOR OFFICIAL USE ONLY- SRI LANKA SECOND AGRICULTURAL EXTENSION PROJECT (CR. 2380-CE) TABLE OF CONTENTS PREFACE ................... i EVALUATION SUMMARY ................... . i PART I - PROJECT IMPLEMENTATION ASSESSMENT A. Introduction . B. Project Objectives . C. Achievement of Objectives .2 D. Major Factors Affecting the Project .6 E. Project Sustainability ............. . . , . 6 F. Bank Performance ..................7 G. Borrowqr Performance .7 H. Assessment of Outcome .8 I. Future Operations .8 3. Key Lessons Learned .8 PART II - STATISTICAL TABLES Table 1 Summary of Assessment .............................................. 10 Table 2 Related IDA Credits .............................................. 11 Table 3 Project Timetable .............................................. 11 Table 4 Credit Disbursements: Cumulative, Estimated and Actual ................... 11 Table 5 Key Indicators for Project Implementation ........................................... 12 Table 6 Studies Included in Project .............................................. 13 Table 7A Project Costs ............................................... 13 Table 7B Project Financing .............................................. 13 Table 8 Status of Legal Covenants .............................................. 14 Table 9 IDA Resources: Staff Inputs .............................................. 15 Table 10 Bank Resources: Missions ............................... 16 ANNEXES Annex 1 Mission's Aide Memoire ............................... 17 Annex 2 Government Contribution to the ICR ............................... 27 Annex 3 Changes in Crop Output, Yields and Land Use Intensity ...................... 41 MAPS IBRD Map No.23349R IBRD Map No. 14076R2 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.  IMPLEMENTATION COMPLETION REPORT SRI LANKA SECOND AGRICULTURAL EXTENSION PROJECT (Cr. 2380-CE) Preface This is the Implementation Completion Report (ICR) for the Second Agricultural Extension Project in Sri Lanka for which a credit (2380-CE) in the amount of SDR 10.5 million (US$14.34 million equivalent) was approved on June 15, 1992 and made effective on February 15, 1993. A sum of SDR 4.46 million (US$ 6.0 million equivalent) was canceled in May, 1998. The credit was closed on June 30, 1998. The last disbursement took place on September 30, 1998 at which time, the unutilized balance of SDR 0.39 million (about US$.54 million at current exchange rate) was canceled. The ICR was prepared by Terrence Abeysekera (Agricultural Economist), Nihal Fernando (Irrigation Engineer) of SASRD and Simon Hocombe (Agricultural Extension Specialist) of FAO-CP. Henry P. Gassner, SASRD Principal Operations Officer at headquarters, served as the peer Reviewer . Preparation of this ICR was initiated during the Bank's final supervision/completion mission in March 1998. It is based, among other documents, on a review of the Staff Appraisal Report and legal documents, supervision reports and project files as well as field investigations and discussions with the Bank staff and GOSL officials associated with the project. The borrower and implementing agencies contributed to the preparation of the ICR by providing data, arranging field visits and discussions with project staff, providing comments on the mission's Aide Memoire (Annex 1), preparing an evaluation of the project preparation (Annex 2), and commenting on the draft ICR. i  IMPLEMENTATION COMPLETION REPORT SRI LANKA SECOND AGRICULTURAL EXTENSION PROJECT (Cr. 2380-CE) Evaluation Summary Introduction 1. IDA has supported agricultural development in Sri Lanka since the 1970s, particularly through loans for rural development and irrigation. The Agricultural Extension and Adaptive Research Project (AEARP, Cr. 931 -CE) implemented from 1979 to 1986 aimed at introducing the training and visit (T&V) system of extension while the Agricultural Research Project (ARP, Cr. 1776-CE) implemented from 1988 to 1996 aimed at improving the research output and strengthening links with the extension system. Project Objectives 2. The major project objectives were to increase farmer income and agricultural production by improving farm productivity through, inter alia, promotion of innovative technology. To achieve this objective, the project aimed at: (a) strengthening agricultural extension services by integration of existing extension services; improved use of mass media; strengthened plant protection, quarantine and pest control activities, provision of training and establishing a private sector pilot extension program and (b) developing a national seed policy. The project objectives drew on past experiences and were consistent with sectoral needs at the time. The project was designed with the assistance of FAO-CP. Implementation Arrangements 3. The responsibility for implementing, monitoring and evaluating the project was vested in the Project Management Unit (PMU) while the Project Coordinating Committee (PCC) served as the apex body responsible for coordinating inter-ministerial activities and making critical decisions regarding project implementation. However, neither PCC nor PMU attained the necessary flexibility and autonomy to execute their functions effectively throughout the project life. Lack of effective coordination between the PMU and the Provincial Councils in particular was a major hindrance to project implementation. Implementation Experience and Results 4. Overall, the achievement of project objectives of increasing farm incomes and agricultural production was partial and unsatisfactory. The objectives of wider adoption of new production technology; integrating the services of extension line agencies at the field, district and provincial levels; introduction of a bottom-up approach for identifying and providing solutions through PC/PA/PS process and establishing group extension ii  methods were also partially successful. The pilot private extension program was not implemented. Although a few innovative technologies have been adopted by farmers in response to the project, their impact on farm incomes was minimal. The PMU performed its administrative role satisfactorily but failed to set up the monitoring and evaluation system. Following the mid-term review, the Bank proposed to restructure the project to restrict it to selected districts in selected provinces, until the problems observed in the functioning in the new systems could not be corrected. However, the Ministry of Agriculture insisted that the project continue to be implemented on a nationwide basis, leading to the eventual decision to cancel the project. The project was closed one year ahead of schedule. Establishing an Integrated Extension Service. 5. Moving away from the traditional system of top-down, individual line agency approach to extension with personal contacts with farmers was a major objective underlying the project. With this in view the project made some progress in establishing extension teams at the field, district and provincial levels. However, in most areas the effectiveness of the extension teams in interacting with farner groups and providing solutions to their problems was low. The activities of the extension teams in the provinces often were confined to compliance with procedures needed to receive funding and other support from the project. The level of integration of the activities of the four participating extension agencies at the national level and their commitment to implement the new extension system were less than required for project success. Establishing Farmer Groups and PC/PA/PS Process. 6. The objective of this component was to establish small groups of farmers with common interests and similar resource endowments as an entry point for the extension team working at the village level. The project goal of establishing about 20 Farmer Reference Groups (FRGs) per extension worker was rarely achieved and in most FRGs the number of farmers actively involved was fewer than expected. The extension coverage of the project has thus fallen well below expectations -- estimated at less than 10% of farmers in the country as compared to SAR target of 20%. The new extension methodology based on farmer group formation and the introduction of the problem Census/Problem/Analysis/ Problem Solving (PC/PA/PS) process raised initial enthusiasm among extension workers and some farmers. However, most issues raised through the process were associated with problems to which the field extension teams could not directly provide satisfactory solutions. This led to an eventual erosion of farner confidence in extension personnel, with serious consequences on project performance. Dissemination of Innovative Technology. 7. Promoting new agricultural technologies was a major goal of the project. However, except in some isolated instances, farmers, did not have the benefit of accessing new technologies through the project. In fact, the project design implicitly assumed that a large volume of research output would be generated by the IDA-funded iii  Agricultural Research Project (ARP). The ARP, however, failed to generate new technologies and with this failure, the extension system was starved of new technologies for dissemination among farmers. Research/Extension Linkages and Impact on Research. 8. The linkages between the national research system, operated by the central government, and the extension system operated by the Provincial Councils are extremely fragile. There is little evidence to suggest that specific farmer needs identified through the PC/PA/PS process were communicated to researchers to make any real impact on the research agenda. The provincialization of crop and livestock extension system further weakened links with the national research system, which remained a national responsibility. Neither the National Agricultural Extension Committees nor the Provincial Technical Working Groups (PTWGs) established under the project became effective in restoring research-extension links. Use of Mass Media Approach. 9. The mass media approach to extension planned under the project aimed at taking advantage of the generally high farmer literacy in Sri Lanka. At the national level, mass media communication approach received significant attention as reflected by the emphasis placed on using printed material, local radio broadcasts, television programs. However, at the provincial level, the mass media extension approach has received little or no priority. In fact, most of the printed material produced by the DOA has not reached the farmers mainly because Provincial Council funds were not available to purchase them or due to lack of enthusiasm. The overall impact of mass media program at the farm level could not be assessed. Human Resource Development. 10. This component was designed to provide training and related physical facilities for the extension staff and farmers and also to bring in international expertise to strengthen local extension capacity. Achievements are generally satisfactory and targets were mostly achieved. There were some delays in implementing the short term training, but these were later rectified. No forrnal evaluations of training of farmers and extension staff are available. Two major foreign consultancies were used to assist the project, to introduce the new extension methodology, and then to identify the problems and improve the PC/PA/PS process. The latter consultancy proved less effective and resulted in a collection of a mass of field information which could not be used productively. Improvements in Plant Protection and Pesticide Control. 11. The sub components for improving plant protection services and pesticide control activities achieved 85% of physical and 90% of financial targets and have strengthened government capability for regulatory activities concerning plant protection and pesticide use. Seed Policy. 12. A seed policy approved by the Cabinet of Ministers in March 1996 is now in iv  place. The policy provided extra impetus to the creation of a better environment to reduce government involvement in commercial seed multiplication, and to transfer these responsibilities to the private sector. Laws and quarantine regulations to put the seed policy into effect are now being framed. A National Seed Board with private sector representation for overseeing the seed industry activities has been set up. Project Management. 13. Despite some staffing problems, the PMU performed effectively in undertaking most of its mandated functions including financial control, contract administration, awarding fellowships and arranging training sessions. The M&E system, set up after a long delay, was designed only to track physical implementation progress and compliance with project procedures. Furthermore, major project management activities of the PMU were largely influenced by the MAL limiting PMU's role mostly to routine administrative matters. The project management tasks, in general, were complicated due to the multi- institutional nature of the project involving 3 Ministries, 4 Departments and 8 Provincial Councils. The ineffectiveness of the Project Coordinating Committee (PCC) also led to major problems in project management. Project Sustainability 14. Overall, prospects for sustaining the project activities in the long-haul are poor. The project failed in its efforts to convince farmers and the provincial officials of the advantages of the group approach to extension and the desirable aspects of the PC/PA/PS process. Most farmers have lost confidence in the new extension methodology underlying the PC/PA/PS process and as a result, farmer groups formed under the project are unlikely to continue without external assistance. A major problem for sustaining project activities is the lack of commitment and resources at the provincial Council level. Bank Performance 15. IDA was diligent in project design, appraisal and initial supervision, though with hindsight it appears that it took insufficient account of the potential vulnerability of the project. The Bank fielded missions at regular intervals initially and flagged potential problems. The Mid-Term review also identified a series of critical implementation problems affecting the project to which the borrower failed to give adequate attention. Contrary to the expectations of the Bank the project was viewed by most implementors as outside their normal tasks. Since SAEP provided an assured source of funding, the implementing agencies continued to comply with project requirements, only to the extent needed to maintain access to vehicles, training or operating expenses. Borrower performance 16. Overall, the government performance was weak. Three fundamental factors are responsible for this situation. Firstly, premature extension of the application of PC/PA/PS methodology from the pilot area to cover the whole country, (b) overlaying the project with "AMA program", a broad rural development approach introduced by the Ministry of Agriculture, to which extensionists were required to give priority, and (c) weak administrative linkages between the MAL and Provincial Councils. v  Key Lessons from the Project: 17. There are several important lessons to be drawn from this project. They are: i) There is an urgent need for a national extension policy which re-defines the development roles of public and private sectors in agricultural extension. The current, public sector operated extension system reflects a supply driven activity which shows a meager response to the actual needs of the farmers. There is a high potential for the private sector including NGOs to play a useful role in providing extension services to the farming community. Attempts must therefore be made to exploit this opportunity. (ii) An enabling policy environment is a pre-requisite for successful implementation of extension projects. Government must ensure a consistent, long-term agricultural policy to which it is fully committed before investing in major projects with radical changes to the extension system. (iii) An extension system, built on a single extension methodology to cover all farmers in all regions of the country is unlikely to deliver expectations. Sri Lanka has a substantial diversity in farming systems within which the extension needs are highly diverse. The extension needs of farmers vary substantially depending on a range of factors including climatic endowments, access to resources, educational level and commercial orientation. (iv) New extension methods should demonstrate their feasibility and acceptance by farmers prior to implementation in the field on a larger scale. Costly and high intensity extension systems that rely on individual contacts should be avoided, in particular, where real opportunities for technological change at farm level are few. vi  IMPLEMENTATION COMPLETION REPORT SRI LANKA SECOND AGRICULTURAL EXTENSION PROJECT (Cr. 2380-CE) PART I: PROJECT IMPLEMENTATION ASSESSMENT A. INTRODUCTION 1. The Second Agricultural Extension Project (Cr. 2380-CE) evaluated in this report was prepared by the Government of Sri Lanka (GOSL) and the World Bank with the assistance of FAO-CP between June 1989 and February 1991. The project was appraised in July 1991, negotiated in April 1992, and a credit of US$14.34 million approved by the Board in August 1992. The credit was effective in February 1993. The project was the first of its kind in Sri Lanka, showing a major departure from the previously adopted Training and Visit system of extension and relying on a mass media oriented, participatory approach. The project included four participating extension agencies; the Department of Agriculture (DOA), the Export Crops Department, the Coconut Cultivation Board and the Department of Animal Production and Health. The project commenced in 1993 and was scheduled for completion in 1999. However, due to poor performance and the failure to show satisfactory progress the project was closed in mid 1998. B. PROJECT OBJECTIVES 2. The main objectives of the project were to increase farmer's income and agricultural production in the non-plantation crop sector by improving farm productivity through, inter alia, promotion of innovative technology. The objectives were to be achieved through: (a) Strengthening the extension services by: i) creating an integrated extension system with a smaller cadre of better trained staff; ii) establishing a participatory, cost effective extension methodology, involving mass media techniques; iii) strengthening the Plant Protection and Pesticide Control Divisions of the DOA; iv) providing formal and in-service training to extension staff, training farmers, improving physical capabilities of the national extension system and mobility of its staff; and, v) implementing a pilot private sector extension program, to conduct specific extension programs with smallholders and; (b) Improving the production, import and distribution of seed and planting material by institutionalizing a national seed policy that would enhance private sector participation in seed supply and distribution. 1  Evaluation of Project Objectives 3. Overall, the basic project objectives were relevant and consistent with the needs of Sri Lanka's non-plantation agriculture sector at that time. The objectives reflected the lessons learned from the previous T&V system of extension on key issues including the need to achieve cost containment via application of mass media extension approaches, group interactions, and to upgrade the quality of extension staff. They also reflected a move away from the prescriptive, input-oriented, top-down approach underlying the earlier T&V extension system. However, there was no clearly established national agricultural policy to which the project objectives could be anchored. As highlighted in "Sri Lanka: Nonplantation Crop Sector Policy Alternatives" (Report No. 14564-CE, 1996), a complex of policy related factors are responsible for most agricultural problems in Sri Lanka. These include a growing demand for non-agricultural labor, free irrigation, import protection for rice, low labor requirement of rice, weak credit discipline and restrictive land policies. Inadequate attention to analyze these issues and their impact on project objectives is a major weakness in project design. Project Cost and Financing 4. At appraisal, the total cost of implementing the project including physical and price contingencies, was estimated at US$18.32 million (Rs. 874.5 million). Of this, the IDA credit amounted to US$14.34 million (SDR 10.5 million), about 78% of the total cost. The remainder was to be financed by GOSL (US$3.79 million, 21%) and farmers (US$0.19 million, 1%). At project closure, one year before the due date, the actual project cost incurred was estimated at US$10.05 million, about 70% of the appraisal total. Of the actual cost, IDA contributed US$ 8.02 million (80% of the total) and GOSL US$2.03 million (20%) with no contribution from farmers. The IDA disbursement of US$ 8.02 million was about 56% of the appraisal estimate. 5. A sum of US$6.0 million from the loan was cancelled in May 1998. The undisbursed amount at project closure was about US$6.0 million. The expenditures for goods and civil works were reasonably in accord with expectations, 110% and 80% respectively of the appraisal estimates respectively. However, low disbursement rates were seen in pilot private sector extension system (0%), mass media (19%), operational costs (40%) and training/technical assistance (62%). The low disbursement of project funds was also a result of the depreciation of local currency from Rs. 42.20/US$ at appraisal to Rs. 65.1 1/US$ at closure. A sum of US$1.05 million, 10% of the total cost, set aside for activities in the Northern and Eastern provinces was not utilized due to the unfavorable security situation prevailing in these areas. C. ACHIEVEMENT OF PROJECT OBJECTIVES Overall Achievement 6. The project had negligible success in achieving three of its key objectives of: (a) integrating provincial and national extension line agencies; (b) institutionalizing the Problem Census/Problem Analysis/Problem Solving (PC/PA/PS) procedure as a means of identifying 2  farner problems and linking farmers to the extension and research systems; (c) promoting the adoption of innovative technology; and (d) introducing a privately run, pilot extension program. Evidence suggested that the project did not lead to significant increases in farm incomes and productivity. In fact, the crop output, yields, cultivated area and land use intensities in Sri Lanka show a major stagnation during 1990-97 (Annex 3). 7. The project did, however, provide the extension agencies with strengthened operational capacity in the form of buildings, equipment, vehicles, improved mass media support, training and consultancies. It also helped to re-build the field level agricultural extension cadre to some extent in two ways. Firstly, by bringing back 332 village level Administrative Officers (Grama Niladharis) with diplomas in agriculture who were transferred from the then Ministry of Public Administration and Home Affairs to the Provincial Council administration. Secondly, the project also brought back the Agricultural Instructors (AIs) and the Livestock Development Officers (LDOs) who had been placed under Divisional Secretaries, to the Provincial Councils. Establishment of an Improved Agricultural Extension Service 8. Establishing an Integrated Extension Service. A primary objective of the project was to move away from the traditional system of top-down, individual line agency approach to extension with one-to-one meetings with the farmers to a participatory approach using farmer groups and extension teams at the village level. Although the project made some progress in forming Field Extension Teams (FETs), the actual effectiveness of these teams in identifying farmers' problems and providing solutions was low. The effectiveness of the Guide Extension Teams (GETs) functioning at the district level and the Provincial Extension Teams (PETs) functioning at the provincial level coordinating and supervising the efforts of FETs, was even lower. The activities of these extension teams were largely confined to compliance with procedures needed to receive physical and training support from the project funds. 9. Formation of Farmer Groups and PC/PA/PS Process. This component aimed at establishing a large number of Farmer Reference Groups (FRGs) with common interests and similar resource endowments as an entry point for the extension team working at the village level. Although it was reported that about 11,000 farmer groups were formed under the project, the majority did not seem to be active. The SAR envisaged establishing about 15 FRGs per extension worker but this goal was seldom achieved and only in a few places was the extension worker able to set up more than 10 active FRGs. The number of farmers per group also varied significantly. Thus, the extension coverage of the project was well below expectations -- estimated at less than 10% of farmers in the country as compared to the SAR target of 20%. 10. The new extension methodology based on group formation and the PC/PA/PS process raised enthusiasm among field extension workers and some farmers initially. However, most of the problems which surfaced through the problem census process were associated with irrigation, land, credit, marketing, and the like to which the field extension teams could not respond. Although the project design anticipated that such problems would be referred to other relevant agencies for action, in practice this was not done in an effective manner. This led to a loss of 3  mutual confidence between farmers and extensionists. Thus, despite the large number of PC/PA/PS sessions conducted by the FETs, the project was not suiccessful in solving most farmer problems, promoting new technologies, establishing integrated farming systems or developing effective time bound farmer action plans for development and diversification, as envisaged in the SAR. Solutions offered to the farmers' problems through the PC/PA/PS sessions were more often based on conventional, input-based recommendations with emphasis on paddy, reflecting again the technical approach of the earlier T&V project. 11. Promoting Innovative Technology. Promotion of new agricultural technologies among farmers was a major project goal. Except in few isolated instances, the project did not contribute to wider dissemination of new agricultural technologies. The project design implicitly assumed that the IDA-funded Agricultural Research Project (ARP) would produce a stock of new innovative technologies that needed active dissemination. However, as noted in its ICR, the ARP did not generate many new technologies that are readily accepted by farmers. 12. Research/Extension Linkages and Impact on Research. The linkages between the national research system, operated by the central government, and the extension system operated by the Provincial Councils are extremely fragile. The complete lack of participation of the research scientists in the PC/PA/PS sessions has further weakened the linkages. Similarly, the Subject Matter Specialists (SMS), who were assumed in the project design to provide technical support to extension teams at the farm level and to play a part in the PC/PA/PS process, spent little time in the field. Specific farmer needs identified through the PC/PA/PS process have not been communicated effectively to the research system to make any real impact on the research agenda. Provincialization of crop and livestock extension services further weakened research- extension links. Neither the National Agricultural Extension Committees nor the Provincial Technical Working Groups (PTWGs) became effective in restoring research-extension links. 13. Use of Mass Media Approach. The mass media approach aimed to reach farmers through printed material, radio and TV programs. Between 1996 and 1998, DOA has produced a large number of useful technical bulletins, leaflets and posters. The project has helped the extension agencies to use Radio and TV programs more intensively. Despite the expectation in the SAR that the private sector would play a major role in the production of communication material, most of this was done by the DOA. At the provincial level, however, the mass media extension system has received little or no priority. Most of the printed material produced by the DOA has not reached the farmers, mainly because of the non-availability of funds in some of the Provincial Councils to purchase them. There has been no assessment of the overall impact of the mass media program at the farm level. 14. Training and Consultancy Support. Achievements in training are generally satisfactory and most targets were achieved. By February 1998, of a planned 1,258 man-months of long-term foreign training, 372 man-months of short-term foreign training and 4,000 man- months of short-term, local training, achievements were 1,070 (84% of the SAR estimates), 184 (566%) and 2,153 (49%) man-months respectively. There were some delays in implementing the short term training, but, these were later rectified. No formal evaluations of training are available. 4  Of the provision of 25 man-months of foreign consultancy, 23 man-months (92%) had been utilized to provide two consultancies to: (a) introduce the new extension methodology, and (b) to make the PC/PA/PS process more effective. However, the latter consultancy was of limited use as it was focused on collecting a large volume of data with no arrangements to analyze and provide feedback to the stakeholders. 15. Private Sector Pilot Extension Activity. Despite repeated attempts by IDA to get this component implemented, no results were achieved. The international consultancy provided by the project to initiate work on this program was not utilized. The absence of a firm and clear government policy on the role of the private sector in this activity was a major constraint in implementing this component. 16. Improvements in Plant Protection and Pesticide Control. This component achieved 85% of physical and 90% of financial targets and strengthened government capability for activities relating to plant protection, pesticide control and registration activities. 17. Civil Works. The implementation of the civil works component of the project, is satisfactory. This is mainly due to effective procurement management, contract awards and follow-up by the PMU. Of 47 new buildings planned, 45 (95% of target) had been completed by February 1998. Of 12 building renovations planned, 10 (83%) were completed and the remaining civil works were nearing completion. The overall construction quality of buildings appeared to be satisfactory. 18. Vehicles and Equipment. Procurement of equipment and vehicles was also satisfactory. Of the target of 740 items of equipment, 65 motor vehicles and 1514 motorcycles, the project had procured 738 items of equipment (100%), 65 motor vehicles (100%) and 1,514 motor cycles (100%). 19. Seed Policy. A seed policy approved by the Cabinet of Ministers in March 1996 is now in place. The policy provides extra impetus to provide a better environment to reduce government involvement in commercial seed multiplication, and to transfer the responsibilities to the private sector. Laws and quarantine regulations to put the seed policy into effect are now being framed. A National Seed Board with private sector representation for overseeing the seed production, imports and distribution activities has been set up. 20. Project Management. Despite staffing problems, the PMU played a key role in carrying out most of its functions including financial control, contract administration, awarding fellowships and arranging training programs. However, its failure to set up the monitoring and evaluation (M&E) system for the project was a major shortcoming. The M&E system was set up after a long delay, but designed only to track physical implementation progress. The project management activities to a large extent, were influenced by the MAL limiting PMU's role mostly to routine administrative matters. The PMU was not granted the level of autonomy necessary to make the strategic management decisions needed to implement the project. The Project Coordinating Committee (PCC), though expected to meet at regular intervals to solve major problems and to provide guidance to PMU, did not live up to expectations. In recognition 5  of the ineffectiveness of the PMU to undertake monitoring and evaluation functions, the government in 1997 established an Extension Development Unit (EDU) under the MAL. The EDU so far has not achieved much progress in monitoring and evaluation of the project. It consists of few officers seconded from line agencies and has no fimn funding. D. MAJOR FACTORS AFFECTING THE PROJECT 21. The main contributory factors for the implementation delays/shortfalls were: Factors Not Generally Subject to Government Control. The major factor was the unsatisfactory security situation continued in the North and East throughout the project implementation period. Factors Generally Subject to Government Control. These include: (a) lack of a clear and consistent national agricultural extension policy and the absence of a long-term vision for developing domestic agriculture; (b) complex and uncoordinated institutional environment within which the project was implemented; and (c) weak administrative linkages between the Ministry of Agriculture and Provincial Councils. Factors Generally Subject to Implementing Agencies' Control. These include: (a) ineffectiveness of the Project Coordinating Committee (PCC) to guide project activities and to solve major problems and its failure to develop an effective partnership with the Provincial Councils; (b) lack of an effective project monitoring and evaluation system at the Central Government well as at the Provincial Council levels; (c) decision to expand the pilot phase of the project to the whole country too rapidly without adequate assessment; and (d) overlaying the project with a new rural development strategy (AMA) in 1995 by the MAL which competed with financial and other project resources. E. PROJECT SUSTAINABILITY 22. Overall, the project outcome is not considered sustainable. Despite some improvements in human resource capabilities, infrastructure facilities and the mobility of the national extension system, the project has not been able to institutionalize the new participatory extension approach in the long-run. The efforts to form integrated teams of extension personnel operating at the field, district and provincial levels came mostly from the project and remained as a project driven exercise. The Provincial Councils have not displayed a strong sense of commitment to undertake and continue the project activities. In some provinces, the early closure of the project has already led to significant reductions in budgetary resources, resulting in cut-backs in field activities of extension staff. While farmers initial response was enthusiastic, they have been disappointed by the lack of follow-up to the PC/PAIPS process. Therefore, the viability of FRGs is uncertain. 6  F. BANK PERFORMANCE 23. Overall Status. The Bank was diligent in its efforts in designing, appraising and supervision. It handled each of these activities in a relatively satisfactory manner, taking account of views current at the time on how the role of the public sector in extension could be made more relevant and cost effective through wider beneficiary and private sector involvement. Nevertheless, the project failed almost totally to achieve its development objectives. With hindsight, it is apparent that the Bank took insufficient account of the combined influence of the following factors: (a) the challenges and risks posed by introducing a radically new extension approach, involving major changes in the roles and behavior of government staff and farmers; (b) low farmer demand for input intensive technical innovations and insufficiency of suitable technologies to choose from; (c) the negative consequences of conducting a census of all problems faced by farmers, without having under direct project control, the means to respond to their non-technical demands; and (d) the long time required for provincialized extension services to become established. 24. Project Concept and Design. The identification and preparation of the project was undertaken through the FAO/World Bank Cooperative Program (FAO-CP). Overall, the process of project preparation was guided by the experiences gained from the previous project. The project was initially prepared as a broad activity aimed at supporting the agricultural diversification and export promotion goals at a cost of US3 1.7 million. However, at appraisal its scope was reduced to focus only on improving the extension services, while cutting the cost down to US$18.3 million. In view of its innovative features and consequent risks associated with the proposed extension methodology, the project was initially implemented on a pilot basis to cover three Divisional Secretariat areas in three provinces. The expansion of the project to achieve a national coverage by including all 17 districts only a year later was too rapid and was undertaken without any evaluation of the results of the pilot exercise. The plan of expansion including the criteria to be adopted as well as the timing, were not specified in the SAR or in the Development Credit Agreement and this is a major design shortcoming. 25. Supervision. IDA undertook regular supervision missions from October 1992 to March 1998 with a mid-term review in March 1996. While reviewing the implementation progress of the project, the missions repeatedly drew attention to the implementational deficiencies. Delays in setting up the M&E system, developing action plans for extension and the need to initiate the pilot private sector extension program were flagged from the outset. Supervision missions were ineffective in persuading Provincial Councils to implement the project as planned. In response to increasing confusion arising from the launch of AMA in 1995 and the negative signals following the Mid Term Review, the Bank identified SAEP as a "problem" project in March 1996. Subsequently, in September 1996, the Bank proposed some remedial measures including the introduction of the Extension Development Unit (EDU), formulation of an explicit national agricultural extension policy, and the long-overdue introduction of a monitoring system. The steps taken by the project to remedy the problems failed to show adequate progress. 7  G. BORROWER PERFORMANCE 26. The project was prepared by the borrower with the assistance of IDA and FAO-CP. The borrower provided adequate attention to the appraisal of the project by IDA staff. However, the borrower's implementation performance was unsatisfactory. A fundamental cause for this was the failure of the borrower to give overriding priority to the project activities, as assumed by IDA. Because of their pre-existing programs and commitments, most extension officials in the provinces considered project activities as tasks additional to their normal duties. Failure to set up an effective monitoring and evaluation system and the decision to overlay SAEP with the AMA program affected the project adversely. H. ASSESSMENT OF OUTCOME 27. Despite strengthening of the physical and human resources of the national extension system, the project failed to deliver its primary outcome of establishing a more streamlined, client oriented extension service which would raise farm productivity and incomes through the transfer of innovative technology. It also failed to introduce the concept of private sector extension activity. Neither the PC/PA/PS process of farmer participation nor the three tiered institutional hierarchy introduced by the project involving village, district and provincial levels extension teams is likely to survive without further assistance. Project performance in introducing new improved technologies to smallholders was disappointing. The project has not established a sustainable capacity to monitor and evaluate agricultural extension under MAL. With its weak implementational performance, the project has failed to achieve an effective, low cost outcome. The project did not have a significant impact on improving the environment. I. FUTURE OPERATIONS 28. The unplanned withdrawal of the project has left the national agricultural extension system without adequate resources or direction. Modifications adopted in different provinces to cope with the withdrawal of project funds show a significant disparity. The situation will only improve when GOSL develops a clear and consistent national extension policy which recognizes and responds to the farmers' needs while maximizing the potential contributions of private individuals and farmer groups. It should be based on a well conceived vision of emerging sectoral opportunities and constraints, and also compatible with the emerging local and global economic changes. J. KEY LESSONS LEARNED 29. The following lessons can be drawn from the project: i) Sri Lanka urgently needs a national extension policy which re-defines the development roles of public and private sectors in agricultural extension: Programs to develop agricultural extension activities have so far been solely driven by public sector initiatives and resources, reflecting a supply driven approach and catering little to meet the user needs. This approach has resulted in a sub-optimal use 8  of public resources, under-utilization of private sector initiatives. The absence of a well defined and a consistent long-term national extension policy is a fundamental cause that has led to these problems. Sri Lanka should therefore move towards a planned, progressive disengagement of the public sector from extension functions, wherever appropriate. Possible options for such a move would include: increased reliance on private advisory services and input suppliers; NGOs and farmer associations; and extensive use of mass media and farner groups for information transfer. Costly and high intensity extension systems that rely on individual contacts should be avoided, in particular, where real opportunities for technological change at farm level are few; ii) An Enabling Policy Environment is a Pre-requisite for Successful Implementation of Extension Projects:Evidence gathered in this study suggested that the demands made by farmers for new agricultural technology primarily depend on the profitability associated with the adoption such technologies. In a policy environment where the agriculture sector is stagnant with declining farm level retums and off-farm employment is more attractive, only limited success can be expected from any extension project. Therefore, as a development pre-requisite, it is necessary for Sri Lanka to prepare a well conceived and an outward looking national agricultural policy that would cater to the long-term needs of the country; iii) A single, centrally operated extension system cannot meet the needs of all farmers in all agro-ecological regions. Given the widely different ecological endowments and farming systems adopted in various parts of the country, a single extension system that is applicable to the entire country is unlikely to deliver expectations. Also a clear distinction should be made between the extension services demanded by commercial farmers, as against services demanded by subsistence farmers. Commercial farmers, in general, demand conventional intensification technologies and are often willing to pay for extension advice; and iv) Introducing New Extension Methods should demonstrate their feasibility and utility in pilot exercises before being adopted as the sole methodology for a major investment project: It is also important to recognize that implementation of extension systems involves a social-engineering process which requires significant transformations of behavior and attitudes of many stakeholders including farmers, farmer groups, extensionists, researchers and officials. The change process, even if designed and executed properly, would require many years of sustained commitment and nurturing by all stakeholders involved. Therefore, when radically new extension methodologies are introduced it would be useful to try out the proposed new systems on a small scale and to try to develop the necessary changes in stakeholder's attitudes before commitment to a national project. It is useful to provide sufficient safeguards in the DCA against the tendency to expand the pilot systems to a national scale before the Bank is satisfied that they had been adequately developed. 9  PART II: STATISTICAL TABLES Table 1: Summary of Assessments A. Achievement of Objectives Substantial Partial Negligible Not Applicable Macro policies O O O 0 Sector policies O O E 0 Financial objectives E 0 l O Institutional development O 0 El 0 Physical objectives 0 E E E Poverty reduction E 0 E E Gender issues E 0 El Other social objectives E Cl El 0 Enviromnental objectives E E 0 El Public sector management E E El E Private sector development E E 0 0 Other (human resource development) El 0 E E B. Project Sustainability Likely Unlikely Uncertain (/) (1 ) (1 ) E 0 E C. Bank Performance Highly Satisfactory Satisfactory Deficient (/) (/ ) V/) Identification El 0 E Preparation assistance E 0 E Appraisal E E 0 Supervision E E E D. Borrower Performance Highly Satisfactory Satisfactory Deficient Preparation E 0 E Implementation E E 0 Covenant compliance El E 0 Operation (if applicable) El 0 Highly Highly E. Assessment of Outcome Satisfactory Satisfactory Unsatisfactory Unsatisfactory (9') (9') (9') (9') El El 0 El 10  Table 2: Related IDA Credits Loan/ Purpose Year of Status Credit Title Approval Preceding Operations: 1. Agricultural Extension Develop smallholder agriculture sector by 1982 PCR and Adaptive research improving and strengthening physical and issued project (Cr 931 -CE) institutional capabilities the research and in Apr. extension system. 1992 2. Agriculture Research Improve and strengthen the Agricultural 1987 PCR project (Cr. 1776-CE) Research system issued in June 1997 3. Following Operations: None Table 3. Project Timetable Steps in Project Cycle Date Planned Date Actual/Latest Estimate Identification Jun. 1989 Jan. 1989 Preparation Nov. 1990 Feb. 1991 Appraisal Jun. 1991 Jul. 1991 Negotiations Apr. 1992 Apr. 1992 Board presentation Jun. 1992 Aug. 992 Signing effectiveness Sep. 1992 Feb. 1993 Mid term review May 1995 Mar. 1996 Project completion Feb. 1999 Jun. 1998 Loan closing June 30, 1999 June 30, 1998 Table 4: Credit Disbursements: Cumulative Estimated and Actual (US$ thousands) FY93 FY94 FY95 FY96 FY97 FY98 FY99 FY00 Appraisal estimate 0.4 1.4 3.7 6.0 8.3 10.6 12.9 14.3 Revised estimate 1/ Actual 0.60 1.36 2.34 3.68 5.71 7.99 8.02 - Actual as % of appraisal 150 97 63 61 68 75 62 Actual as % of revised estimate - - - - - - Date of final disbursement September 30, 1998 1! US$6.0 million was canceled in May 1998 11  Table 5: Key Indicators for Project Implementation Key Implementation Indicators in SAR SAR Estimated Revised Actual 1. Extension Staff: Agricultural Instructors 332 332 332 2. Plant protection Staff: Research Officers 3 3 3 Agricultural Officers 9 3 3 Agric. Instructors 9 4 4 Support staff 18 2 2 3. Pesticide Staff: Research officer 3 3 3 Assistant Registrar 1 I 1 Computer Programmer 1 I 0 Support staff 12 12 5 4. .PMU Staff: Project manger I I I Others 10 10 10 5. Civil Works: Residences 41 41 40 Service Buildings 6 6 6 Renovations 12 12 10 6. Equipment/Vehicles: _______ Office Equipment (items) 424 424 446 Training Equipment (Items) 96 96 106 field Equipment (Items) 6 6 6 Media Equipment 70 70 47 Lab Equipment (items) 38 38 1 8 Misc. Household Equipment (Items) 106 106 Cars I I I Double cabs/ Jeeps/vans 63 63 63 Buses I _I__ Motorcycles 1514 1514 1514 7. Training: Long Term Training (Pers. Month) 1258 1258 1070 Short Term Training (Pers.Month) 372 372 184 Local Training (Sessions) 4000 4000 2153 Demonstrations (Numbers) 4200 4000 1773 8.. Technical assistance: International (person months) 18 18 92 Local (person months) 7 7 114 Training of computer operators 10 10 10 9. Quarterly Reports 26 26 24 12  Table 6: Studies Included in Project Study Purpose as Defined at Status Impact of Study Appraisal/Redefined 1. Private sector To study the existing Completed No clear cut private sector extension initiatives in private sector extension Sept. 1994. extension activities on whole Sri Lanka. initiatives. farm approach identified. 2. CCB Extension To prepare a proposal to Completed Support under SAEP extended strengthening support CCB end. 1995 to CCB. 3. Gender study To determine the Completed Need to increase participation participation of women March 1997 of women in extension in SAEP identified. 4. Research extension To provide guidelines for Completed Agencies convinced about the linkages the extension workers need to strengthen linkages. and rubber growers in DOA initiated formal meetings selecting clones with provinces. Table 7A: Project Costs Appraisal Estimate (US$ M) I/ Actual/Latest Estimate (US$M) ITEM Local Foreign Total Local Foreign Total Costs Costs Costs Costs Costs Costs Civil works 0.30 1.90 2.20 0.22 1.21 1.44 Mach. Vehicles and 1.70 2.30 4.00 1.21 2.89 4.04 equipment Training and consultancies 0.00 5.30 5.30 0.07 2.83 2.86 Contractual services 0.60 2.20 2.80 0.05 0.21 0.26 Incremental staff& 1.50 2.50 4.00 0.50 0.88 1.38 operational costs Total 3.90 14.20 18.30 2.03 8.02 10.05 1/ SAR-Annex 2, Table 1. Table 7B: Project Financing Source Appraisal Estimate (US$M) 1/ Actual/Latest Estimate (US$M) Local Foreign Total Local Foreign Total Cost Cost Cost Cost Cost Cost IDA 4.9 9.44 14.34 4.72 3.30 8.02 GOSL 3.8 0.0 3.8 2.03 0.0 2.03 Farmers 0.18 0.0 0.18 0.00 0.0 0.00 Total 8.8 9.44 18.32 6.75 3.30 10.05 1/ SAR - page v. 13  Table 8: Status of Legal Covenants Agreement Text reference Covenant Present Original Revised Description of covenant Comments type status fulfillment date fulfillment date .02b I C Borrower shall maintain in US $ special Account in In compliance Central Bank Schedule P4.1 5 C 12/31/92 2/3/93 Establish Committee on Agric. Extension. In compliance P4.2 (a) 5 C 12/31/92 2/3/93 Establish PMU In compliance P4.2(b) 5 C 12/31/92 3/14/93 Recruit PMU staff Employ architect firm In compliance P 4.2 5 C 12/31/88 Employ architect firm In compliance P 4.3 (a) (b) 5 C Employ Organizational and Extension specialists In compliance P4.4 5 C 12/31/92 2/3/93 Establish project coordinating Committee In compliance P4.5 5 C 12/31/90 Establish Agric. Extn. Working Group In compliance P4 .6 5 CD 3/12/93 6/30/93 HRD Plan/Fellowship training In compliance P 4.7 5 CD Plan of action for improving Agric. Extension Head not appointed P 4.8 5 C Mass media Contract Services In compliance P 4.9. (a) (b) 5 NC 6/30/93 Pilot extension program Not in compliance P 4.10 (a) 5 CD 6/30/893 Develop National seed Policy In compliance, P. 10(b) 5 CD 6/30/93 Implement seed Policy In compliance P 4.11 5 CD 6/30/93 Annual Program of activities In compliance P .4.12 5 CD 6/30/95 3/15/96 Conduct Mid-tern Review In compliance Status: Covenant Class C - Complied with I Accounts/audit CD - Complied after delay 2 Financial performance/generate revenue from beneficiaries NC - Not complied with 3 Flow and utilization of project funds SOON - Compliance expected 4 Counterpart funding in reasonably short time 5 Management aspects of the project or its executing agency CP - Complied partially 6 Environmental covenants NYD - Not yet due 7 Involuntary settlement 8 Indigenous people 9 Monitoring reviewing and reporting 10 Implementation 11 Sectoral or cross sectoral budgetary or other resource allocation 12 Sectoral or cross sectoral regulatory institutional action 13 Others 14  Table 9: IDA Resource: Staff Inputs Stage of Project Cycle Planned 1/ Revised 1/ Actual 2/ Weeks US$ Weeks US$ Weeks US$ (000) (000) (000) Preparation to N/A N/A N/A N/A 59.8 157.6 appraisal I Appraisal N/A N/A N/A N/A 67.8 207.6 Negotiations through N/A N/A N/A N/A 8.6 29.4 Board approval Supervision N/A N/A N/A N/A 94.2 241.1 ICR Completion3/ N/A N/A N/A N/A 14.0 3.9 TOTAL N/A N/A N/A N/A 244.4 639.6 1/ Estimates for original and revised staff weeks and dollar budgeting introduced only in FY95 and therefore total original and revised staff weeks and costs cannot be computed. 2/ Data from World Bank MIS 3/ Actual cost relates only to staff in Colombo office. Four staff weeks contributed by FAO/CP 15  Table 10: IDA Resources: Missions Stage of Month/Year Number Days Specialized Types of Project Cycle of in Staff Skills Performnance Rating by Problems Persons field Represented a/ c/ Implementati Developmen on Status t Objectives FAO Initial Preparation Jun/Jul. 1989 1 9 Ag. assistance Fact Finding Mission Sep 1989 2 12 Ag (2) Initial preparation Sep/Oct 1989 5 25 Ag (2); Tr (1) I ___________________ ______ M k(l), Se(I) _ _ preparation Sep/Oct 1990 6 30 Ag (5),Mk (1) Preappraisal Mar 1991 5 20 Ag (4) Ec(l) Appraisal Jun/July 1991 7 19 Ag.(3), Ec (1)., l ______________________ ________________ Se(2), Fe (1) _ Supervision 1 Oct/Nov. 1992 3 11 Ag. (2), Pr.(1), I I - Supervision 2 Mar/Apr 1993 1 17 Ag., 1 1 - Supervision 3 Oct/Nov 1993 2 _23 Ag (2). 2 1 - Supervision 4 Mar/Apr 1994 1 17 Ag. 2 1 - Supervision 5 Oct/Nov 2 14 Ag (2) S 5 - 1994 Supervision 6 Apr/May 1995. 1 16 Ag. S S - Supervision 7 Mar 1996 3 2 Ag.,Ec.,Gn. _ _ _ _ _ _ _ _ _ U U Supervision 8 Oct 1996 1 17 Ag.,. U U M ICR Mar 1998 3 8 Ag(2) En(l). U U M a/ Staff specialization: Ag. - Agriculturist; EC - Economist; FA - Financial analyst, En- Engineer, Gn- Gender specialist; Mk- Marketing; Pr-Procurement Specialist; Tr- Training Specialist. b/ Performance ratings based on IBRD and IDA - implementation summary Form 590 (before FY94) =1- Problem free; 2- Moderate; 3- Major problems, Implementation/development impact status (from FY94) = HS - Highly satisfactory; S- Satisfactory; U- Unsatisfactory; HU- Highly Unsatisfactory c/ Types of problems: T - Technical; M - Managerial; and F - Financial. d/ Project identification. e/ Information not found in project files. f/ mid-term review. 16  ANNEX 1 SRI LANKA: SECOND AGRICULTURAL EXTENSION PROJECT (Cr. 2380-CE) March 2-20, 1998 Final Supervision/ICR Preparation Mission Annex 1:Aide Memoire 1. INTRODUCTION 1. The mission, comprising Messrs. Terrence Abeysekera (Agricultural Economist), Nihal Fernando (Irrigation Engineer) and Simon Hocombe (Senior Adviser, Extension Specialist, FAO/CP), carried out the final supervision of the implementation of the Second Agricultural Extension Project (SAEP) and gathered material to initiate drafting of the project's Implementation Completion Report (ICR), during March 2 to 20, 1998. 2. After initial discussions in Colombo with the key staff of the concerned Ministries and Government Departments, the mission made field visits to Central (Kandy, Matale Districts), North Western (Kurunegala District), North Central (Anuradhapura District) and Western Province (Colombo District). The mission had meetings with senior project staff in the Ministry of Agriculture and Lands, Ministry of Finance and Planning, Coconut Cultivation Board, the Departments of Agriculture, Export Agriculture and Animal Production and Health, and with the Project Management Unit (PMU) and Extension Development unit (EDU). Contacts were also made with the Council for Agricultural Research Policy (CARP) and the Extension Department of the University of Peradeniya. In the provinces the mission had discussions with provincial, district and field level project staff and farmers. The mission would like to thank all the staff involved for arranging the field visits, for the support extended, and for the good working relationship established. 3. This Aide Memoire summarizes the mission's observations, main issues and recommendations discussed at the wrap-up meeting held on March 20, 1998, chaired by Mr. S. Jayaweera, Secretary, Ministry of Agriculture and Lands (MAL). Project Background 4. The project was a follow-up to the previous IDA-funded Agricultural Extension and Adaptive Research Project (AEARP), which introduced the Training and Visit (T&V) extension methodology to Sri Lanka. AEARP was implemented from 1979 to 1986 with moderate success; it helped to establish an institutional framework for extension on all crops except tea and rubber, and to strengthen research-extension linkages. It failed however in its aim to unify crop and animal husbandry extension and did not use a whole farm extension approach, nor seek to develop a village-level process of needs identification and basic planning. 17  5. The Second Agricultural Extension Project was prepared between 1989 and 1991 by GOSL with IDA assistance and was appraised by an IDA Mission in July 1991. It was designed to be implemented over six year period (1993-1999). In addition to improving extension services through a new, more participatory, approach described below, the project design provided financing for improved use of mass media, training, strengthened plant protection, quarantine and pest control services, safer and more effective pesticide use, and a pilot private sector extension program. It had a separate small component to prepare an improved national seed policy. These activities were backed by an extensive training program. Project Objectives 6. The objective of SAEP was to increase farner income and agricultural production by increasing farm productivity through, inter alia, promotion of innovative technology. In particular, the project sought to introduce a new extension methodology having a whole farm, bottom-up planning approach, with preparation of local, district and provincial-level extension plans on the basis of farmers' expressed needs. The methodology involved creation of an Integrated Agricultural Extension Service (IAES) based on a three-tier structure comprising Field Extension Teams (FETs), Guide Extension Teams (GETs) and Provincial Extension teams (PETs). FETs include field extension staff of the two provincial line agencies dealing with crop and livestock extension, and of the national agencies responsible for coconut and export crop extension. FETs were required to identify and prioritize the needs of Farmer Reference Groups (FRGs) composed of farmers with similar production interests, using an interactive problem census, problem analysis and problem solving (PC/PA/PS) procedure. In addition they were expected to develop individual farmer action plans and field extension plans at the Agrarian Service Center (ASC) level. The Guide Extension Teams and Provincial Extension Teams were responsible for the consolidation of ASC-level plans into district and provincial extension plans. 7. The project's objectives were considered important and relevant to the needs of Sri Lanka's non-plantation crops sector at the time. Following a period of relative stagnation, GOSL had been searching for nearly five years for a successor to the previous T&V extension project, which had led to initial productivity gains but had latterly shown a reduced impact and became financially unsustainable. The new project approach, based on problem solving by fanners and extensionists acting together, embodied progressive ideas on farmer participation and demand-led design of extension programs. Previous Mission Findings 8. Following implementation difficulties detected by a Mid Term Review (MTR) in March 1996 and a subsequent supervision mission in November 1996, SAEP was rated a problem project; it has remained in that category. Following MTR recommendations, the Ministry adopted some measures to address the problems identified. However, a follow- up IDA supervision in November 1996 observed that the status quo had remained unchanged. Despite such attempts, the present mission observed that the changes have still had limited impact in improving implementation effectiveness. Agreement was 18  reached between the borrower and IDA in November 1997 to close the project one year early, on June 30, 1998. II. MISSION FINDINGS Establishment of Extension Cadre 9. As specified in the SAR, the project was successful in transferring 332 village level administrative officers with diplomas in agriculture from the Ministry of Public Administration to Provincial Councils, to function as Agriculture Instructors (Als). The project also enabled the transfer of AIs and Livestock Development Instructors (LDIs) from the Divisional Secretariats to the provincial line agencies. These two initiatives resulted in the establishment of the planned extension cadre at the field level, having more direct lines of command. Human Resources Development and Consultancy Support 10. The progress of training is satisfactory. Of a planned 1,258 man-months of long- term training, 310 man-months of short-term training and 4,000 man-months of local training, achievements are respectively 1,052 (84% of the SAR estimates), 174 (56%) and 2,153 (54%) man-months. Although there were some delays in implementing the short term training, these have since been largely rectified. Short-term local training focused on the introduction of the new extension methodology and was supported by operational guidelines written by project staff. Longer-term and overseas training aimed at more general enhancement of extension skills. No formal evaluations of training are available. Of 4,200 planned farmer demonstrations, 1,773 (42%) have been implemented. Of the provision of 25 man-months of foreign consultancy, 23 man-months (92%) have been utilized, to provide technical assistance firstly with the introduction of the new extension approach, then to introduce modifications to make problem analysis more effective; the latter international consultancy also helped to formulate a monitoring and evaluation system. The 3 months of consultancy provided for formulation of the seed policy were not used. Mass Media Program 11. Taking advantage of the generally high level of farmer literacy in Sri Lanka, this component aimed to establish a cost-effective communication strategy to reach farmers using printed and electronic media. Progress has been made in producing printed material, radio and TV programs, but the overall impact of this activity cannot be readily assessed. Mission observations suggest that the printed materials are being used by fanners as well as school children, and some of the costs are also being recovered from users as intended. Despite the expectation that the private sector would play a major role in the production of mass media, the bulk of the mass media material was produced by the DOA itself. Greater use of private sector contracts for mass media production, as intended by the SAR, could have reduced delays. 19  Farmer Reference Groups (FRGs) 12. - The project design stipulated that each extension worker at the village level would establish 20 to 30 FRGs, with each FRG consisting of 20 to 30 farmers. This plan expected to cover about 20% of the total farm population. However, mission observations indicate that on average extension workers have seldom been able to establish more than 10 active FRGs, that the number of active members of FRGs are often only two or three, activities of FRGs have fallen drastically in the recent years, and relatively few active FRGs now remain. As a result the actual coverage of farm population by the project is far below expectations. Integration of Extension Line Agencies 13. Field extension officers of the four line agencies that comprise the FETs, which are based in Agricultural Service Centers (ASCs), have shown varying levels of cooperation in implementing the extension program. Whether instances of good individual cooperation at field level reflect the impact of the integration of extension services, or comradeship of a group of officers working in the same environment, is difficult to distinguish. At the national, PET and GET levels the coordination and interactions among members are highly variable, but generally weaker. The lack of integration at higher levels appears largely due to (a) the presence of complex organizational arrangements and linkages within the system, and (b) continuing priority which the line agencies are required to give to their original work programs such as subsidy disbursement and input distribution for coconut and export crops. These have left less time than was anticipated in the SAR for joint PC/PA/PS activities. Thus, contrary to the project objectives, SAEP has been no more successful than the previous T&V project in forging a fully integrated national extension system. PC/PA/PS Process 14. In its initial stages of implementation the project gave prominence to problem census at field level; problem analysis and solving were introduced much later. Many of the problems raised by farmers in the FRG problem census sessions concerned irrigation or non-technical matters which were outside the direct control of the FET members. The declining performance of the FRGs noted above is partly attributed to a loss of mutual confidence between fanners and FET members because of the inability of the latter to offer solutions to irrigation and non-technical problems raised by farmers. Furthermore, there has been a tendency for technical extension plans developed by FETs to be dominated by existing technologies aimed at single commodities, primarily paddy. In addition, farmers and extension staff alike have found the documentation required by the PC/PA/PS methodology cumbersome and time-consuming, especially the requirement to draw up detailed farm plans for each FRG member. Consequently, over time, farmers have lost interest both in participating in the PC/PC/PA process and in attending group meetings. Research-extension links 15. Responsibilities for generating new technologies through research are a national responsibility, while the responsibility for providing much extension is with the 20  Provincial Councils. Provincial Technical Working Groups were proposed during implementation to foster the flow of information from the research centers to the farmers and vice versa, but have not become effective. Currently the formal links between the provincial extension system and the research centers are limited to Regional Technical Working Group (RTWG) meetings held once in six months. Thus the project has done little to improve research-extension links. The mission saw little evidence that the project farmers' demands had influenced the research agenda. Extension Impact 16. Many extension recommendations framed by FETs were focused on conventional, packaged or input-based intensification technology, rather than being dominated by the systems-based, holistic approach implied by the SAR. There was also an emphasis on paddy which contrasted with the stress in the SAR on promoting commercial agriculture, diversification and subsidiary food crops. Nonetheless, certain low-cost or no-cost system based recommendations were issued. Those for Integrated Pest Management and deep ploughing in paddy, alternative uses of paddy straw to boost soil productivity, and maize planting on paddy land in the Yala season, were better received by farmers than most input-intensive recommendations. Pilot Private Sector Extension Program 17. The project was to prepare a plan of action for the introduction of a pilot extension program that would operate on a fee-for-service basis. Following repeated reminders by supervision missions to mobilize this activity, some tenders for private extension contracts were sought by newspaper advertisement. Although some bids were received, no contracts were awarded. Lack of response by potential private suppliers of extension services may reflect the generally low profitability of investment in agriculture, low expected rewards, and the absence of a favorable extension policy environment. National Seed Policy 18. A seed policy satisfactory to IDA, involving disengagement of the government from commercial seed multiplication, and transfer of responsibilities and some facilities for this part of the seed chain to the private sector, is now in place and was approved by the Cabinet of Ministers in March 1996. While there was already a trend towards greater involvement of the private sector in seed production and distribution, the inclusion of a covenant under the project requiring the government to prepare a national seed policy provided extra impetus to formalize this transition. Dependent seed laws and quarantine regulations to put the policy into effect are now being framed by a National Seeds and Planting Materials Committee and its sub-committees. Plant Protection and Pesticide Control 19. The components on plant protection and pesticide control achieved 85% of physical and 90% of financial targets. Support for plant quarantine, anticipated at appraisal as possibly coming from the Government of Japan, was provided from that source. Project support has strengthened DOA capability for activities relating to plant protection and pesticide control. 21  The Project Coordinating Committee 20. Because of the devolution of crop and livestock extension responsibilities to the provinces and the involvement of separate national extension line agencies to handle coconuts and export crops, project progress from its inception suffered severely from coordination problems and weak lines of authority. The Project Coordination Committee (PCC), established as the apex body to deal with all aspects of implementation, appears to have had limited influence in improving project progress. The PCC included as much as forty representatives from provinces and the center, and although it was expected to meet and review project progress at least once in three months, meetings were few and far between. Extension Development Unit 21. Subsequent to the limited success of the project in implementing the training strategy and extension program, an Extension Development Unit (EDU) was created in March 1997, following the November 1996 IDA supervision mission. The EDU was also given monitoring responsibilities, taken over from the PMU following its lack of progress in this regard. The EDU includes seconded representatives on a full-time basis from each of the four participating extension agencies and the PMU. With the initial consultant assistance, progress has been made in defining a planning and monitoring system for integrated extension programs. However, the monitoring system has been criticized by the provincial and lower level staff who are required to operate it as excessively time- demanding and complicated. For this reason, and perhaps also because the provincial agencies anticipate the imminent closure of the project, their response to these EDU initiatives has been hesitant. But although the EDU has so far not achieved much progress in establishing and coordinating provincial level monitoring units, the mission believes that, if further strengthened, it could play an important future role in defining and backstopping the future development of agricultural extension in Sri Lanka (see para. 37). The Project Management Unit. 22. Overall, the Project Management Unit (PMU) satisfactorily carried out the majority of the administrative responsibilities allocated to it in the SAR, despite difficulties in recruiting or retaining staff in some categories. There were some implementation delays compared to SAR estimates, and the times required for bureaucratic procedures and consensus-forming among implementing agencies were at times under-estimated: but ultimately the PMU successfully handled project fund and budget allocation to the participating agencies, maintained satisfactory accounts, reviewed statements of expenditure, provided assurances of eligibility of project expenditures and claimed reimbursements from IDA. Project audit is up-to-date. The PMU also satisfactorily supervised the architect engaged to oversee the planning and bid specifications for the civil works program, handled bidding and procurement, and successfully assisted the award of project-funded fellowships. 22  23. A shortcoming of the PMU was its failure, despite repeated prompting from IDA supervision missions, to establish the project monitoring and evaluation system called for at the start of project implementation. The PMU also failed to secure from the Agricultural Extension Working Group (AEWG) the operational plans for the installation of the SAEP extension methodology, which that group was required under the SAR to generate. No benchmark surveys were made of the pilot areas, against which progress could be measured. 24. It was envisaged under the SAR that the PMU would be assisted in day-to-day project operations by the provincial line agencies. As indicated above, they did not commit the expected resources to the project and hence their assistance was less than anticipated. The PMU was not delegated with the authority under the project to overcome this type of inter-institutional implementation problem and was not backed by a Project Coordinating Committee which could exert such authority from a higher level. Civil Works, Procurement of Equipment and Vehicles 25. Progress of civil works is satisfactory; of 47 new buildings and 12 building renovations planned under the project, 35 buildings (76% of estimate) and all renovations have been completed. The remaining civil works are nearing completion. The progress of procurement of equipment and vehicles to support IAES operations is also satisfactory. Of the target of 738 items of equipment, 65 motor vehicles and 1,514 motor cycles, the project has procured 705 items of equipment (95%), 65 motor vehicles (100%) and all 1,514 motor cycles. Vehicles, motorcycles and equipment have greatly facilitated field operations, especially by increasing the mobility of extension staff. Project Disbursement 26. At project appraisal the estimated project cost was US$18.3 million, of which the IDA credit was US$14.3 million. About 10% of total project cost was set aside for project activities in the Northern and Eastern provinces in the event of a cessation of civil unrest: since the unfavorable security situation persisted, no project activities could be carried out. Of the IDA credit, a total of US$7.5 million had been disbursed by the end of February 1998. With the anticipated work program for the remaining project period, a total of US$6.0 million are estimated to remain undisbursed by project closure in June 1998, including the 10% earmarked for the North and East. II1. OVERALL ASSESSMENT 27. SAEP has shown limited success in bringing field staff of the extension line agencies to work together to introduce a team/participatory approach and establish an integrated extension system at the farm level. The mission assesses the overall implementation progress and achievement of these institutional development objectives of the project as unsatisfactory. Even the few organizational achievements are for the most part unsustainable. The commitment of resources by the higher levels of the extension line agencies to the SAEP extension approach remains weak, so that the 23  national agricultural extension system overall remains largely unintegrated and ineffective. The mission considers that the new methodology which was the core of the project design is unlikely to persist following project closure. This limited impact can be attributed to various factors which are summarized below and will be further analyzed by the mission in the Implementation Completion Report. Policy Issues 28. The project was not anchored within a sound sectoral policy framework, nor a clearly articulated national agricultural extension policy and implementation strategy to which all the potential implementing partners had committed themselves in advance. An agreed extension policy would have ensured that all parties gave an extension project their undivided commitment. An agreed implementation strategy based on that policy would have resolved in advance the question of how diverse extension line agencies were to merge their efforts: it would also have included expansion criteria to prevent over- rapid expansion of any untried approaches. Design Issues 29. The mission considers that there were probably five factors which affected project outcome - some but not all of which received sufficient individual consideration during design, but the potential aggregate effect of which was given insufficient weight.