Malawi - Fisheries Development Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No.: 21560 IMPLEMENTATION COMPLETION REPORT ON AN IDA CREDIT IN THE AMOUNT OF US$8.80 MILLION TO THE REPUBLIC OF MALAWI FOR A FISHERIES DEVELOPMENT PROJECT (CREDIT NO. 22250-MAI) DECEMBER, 29, 2000 Rural Development Operations Eastern & Southern Africa This document has a restricted distribution and may be used by recipients only in the perfonnance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  CURRENCY EQUIVALENTS (Exchange Rate Effective October 2000) Currency Unit = Malawi Kwacha (MWK) At appraisal: US$1 = MWK2.8 At completion: US$1 = MWK55.0 WEIGHTS AND MEASURES Metric System FISCAL YEAR OF BORROWER July I - June 30 ABBREVLATIONS AND ACRONYMS BVC Beach Village Committees CP FAO/World Bank Cooperative Programme DCA Development Credit Agreement ERR Economic Rate of Return FD Fisheries Department FDP Fisheries Development Project FDPD Fisheries Development Project Document FRR Fishery Research Unit FTC Fisheries Training College GEF Global Environment Facility GOM Government of Malawi ICEIDA Icelandic International Development Agency ICR Implementation Completion Report IDA Malawi College of Fisheries MOF Ministry of Finance MONREA Ministry of Natural Resources and Environmental Affairs MOWS Ministry of Works and Supply MRFC Malawi Rural Finance Company MTR Mid-Term Review NDF Nordic Development Fund PCC Project Coordination Committee SACA Smallholder Agriculture Credit Administration QAG Quality Assurance Group Vice President: Callisto Madavo Country Manager/Director: Darius Mans Sector Manager/Director: Sushma Ganguly Task Manager: Satish Kumar  FOR OFFICIAL USE ONLY IMPLEMENTATION COMPLETION REPORT MALAWI FISHERIES DEVELOPMENT PROJECT (Credit No. 2225-MAI) TABLE OF CONTENTS Page No. 1. Project Data I 2. Principal Performance Ratings 2 3. Assessment of Development Objective and Design, and of Quality at Entry 4 4. Achievement of Objective and Outputs 7 5. Major Factors Affecting Implementation and Outcome 17 6. Sustainability 18 7. Bank and Borrower Performance 20 8. Lessons Learned 23 9. Partner Comments 25 10. Additional Information 26 ANNEXES Annex 1. Key Perforrnance Indicators/Log Frame Matrix 27 Annex 2. Project Costs and Financing 31 Annex 3. Economic Costs and Benefits 32 Annex 4. Bank Inputs 33 Annex 5. Ratings for Achievement of Objectives/Outputs of Components 34 Annex 6. Ratings of Bank and Borrower Performance 35 Annex 7. List of Supporting Documents 36 Annex 7a - ICR Mission Aide Memoire by FAO Annex 7b - Background of Fisheries Sub-sector & Project Implementation Experience Annex 7c - Summary of the Evaluation of Components by NDF & ICEIDA Annex 7d - Government Evaluation Report Annex 8. Beneficiary Survey Results Annex 9. Stakeholder Workshop Results This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not be otherwise disclosed without World Bank authorization.   1. Project Data Name: FISHERIES DEVEVELOPMENT L/C/TF IDA-22250 PROJECT Number: Country/Department: MALAWI Region: Africa Regional Office Sector/subsector: AF - Fisheries & Aquaculture; VM - Natural Resources Management KEY DATES Original Revised/Actual PCD: 08/16/89 Effective: 09/04/91 09/04/91 Appraisal: 05/30/90 MTR: 06/30/95 03/11/96 Approval: 04/02/91 Closing: 06/30/99 06/30/2000 BorrowerlImiplementing Govemment of Malawi (GOM)/Ministrv of Natural Resources & Agency: Environmental Affairs (MONREA): Fisheries Department (FD) Other Partners: Nordic Development Fund (NDF) & Icelandic Intemational Development Agency (ICEIDA) STAFF Current At Appraisal Vice President: Callisto Madavo E. Jaycox CountryManager: Darius Mans S. Denning Sector Manager: Sushma Ganguly C. Helman Team Leader at ICR: Satish Kumar Nwanze Okidegbe WB ICR Team Leader: Francisco Pichon ICR Primary Author: Pietros Kidane (Team Leader-FAO/CP) and A. Seymour (Consultant- Fisheries Specialist)  2. Principal Performance Ratings (HS=Highly Satisfactory, S=Satisfactory, U=Unsatisfactory, HL=Highly Likely, L=Likely. UN=Unlikely, HUN=Highly Unlikely, HU=Highly Unsatisfactory, H=High, SU=Substantial, M=Modest, N=Negligible) Outcome: U Sustainability: UN Institutional Development Impact: M Bank Performance: U Borrower Performance: U QAG (if available) ICR Ouality at Entry: U Project at Risk at Any Time: Yes Note 1: The rating for Outcome is based on the comparison between the final project outcome and the original development objective of assisting GOM to realise the potential contribution of the fisheries sub-sector to the national economy on a sustainable basis. As discussed in this ICR, the expected support to the artisanal and semi-commercial fisheries (from which thie bulk of the project production and development impact were to originate) was not provided by the proiect following concems of over-fishing in Lake Malawi and capacity constraints within the Fisheries Department (FD) which had been seriously underestimated at appraisal. Even though there is agreement that not all the original objectives of the project could have been maintained after restructuring, the total elimination of the production component (and with this also the objective of contributing to the national economy in terms of improving nutrition levels, employment generation and poverty reduction) was deeply resented by fishermen, especially by the semi- commercial fishermen, whose expectations of the projected investments were never met. While the borrower acknowledged its capacity limitations to implement some of the proposed production activities, it believed tlhat, rather than its total exclusion, restricturing of the production component through a more strategic capacity building programmne within the FD would have enabled gradual implementation of at least some of the key production-oriented investments. Since 1995 research had confirmed that the over-fishing noted in 1993 was localised in extent, and that with adequate management of the distribution of fishing effort the original production targets remained realistic and within the limits of sustainability. The overall incremental annual production by the project of 7,500 tons was therefore never achieved, and is really a missed opportunity by the country and unsatisfactory outcome of the project. Yet this goal remains a valid and important national target to this date. Accordingly, the ICR rating of Outcome reflects its assessment of the concerns and assumptions that led to the restructuring of the project in 1993 and 1996, and the modification of the project objectives and components that resulted from pro ject restructuring (see latest guidelines for preparing ICRs, Para 23). Note 2: Overall, project's sustainability can be rated as unlikely principally because of the removal of the essential parts of the production component as a result of project restrLcturing. If the assessment of project's sustainability is limited to the implemented institution building and research activities, sustainabilitv could be considered as likely. This is because most of the development activities carried out by the project appear to attract the necessary resources for their 2  continuation. Within the production support component, MALDECO' s activities are certainly sustainable as the company is in good financial standing. Its fishing and marketing activities will be able to stand competition from other firms. The FD/MRFC credit groups should also be able to continue on their own account. Neither of these activities are dependent on a great deal of back- up support from FD. Some concerns on sustainability remains with regard to the provision of the bulk of the FD's operational expenses which will have now to be covered by GOM. The level of sustainability of project activities is thus related to the amount of funds that GOM will be able to allocate to them. The Ministry of Finance (MOF) informed the ICR mission that it will increase the FD recurrent budget to incorporate the project's operating costs, but was not able to provide details on the level of increase. Note 3: It is difficult to give one rating for a nine-year project, in which the first five years were characterised by poor performance and the last three years by better perfonnance, albeit on a restructured project with significantly narrower objectives. If one rating should be made, however, the overall Bank and Borrower performance should be rated as less than satisfactory. This should be considered a fair rating even when taking into account the fact that project performance improved considerably following the 1996 restructuring and MTR, after it was significantly down-scaled. In the assessment of the ICR team, one of the main problems with the restructuring was IDA's decision to cancel the entire production component. instead of being open to discuss with GOM its possible reformulation. IDA also failed to address the impacts of the unprepared privatizations on fish production which could have been dealt with at the time of restructuring. In addition, the DCA was never revised to reflect the trimmed down number of components. On the other hand, the focus on strengthening the FD, which was largely accomplished during the last three years of the project, is to the credit of the project restructuring team. 3  3. Assessment of Development Objective and Design, and of Quality at Entry 3.1 Original Objective: As appraised, the project's main objective was to realize the potential contribution of fisheries to the economy on a sustainable basis through: (a) increasing fish production to improve nutrition; (b) generating additional off-farm employment to reduce poverty among the rural population, particularly women; (c) conserving the natural resources base of Malawi's water bodies and preventing environmental degradation through better management; and (d) improving institutional capacity in the fisheries sub-sector by strengthening the plannling, implementation and management capacity of the Fisheries Department (FD). While the objective of raising the contribution to the economy from fisheries was valid and desirable, the way the project was designed was complex in that it tended to cover too many activities that were beyond the limited capability of the main implementing agency, i.e. the FD. The Appraisal Mission had defined the FD as institutionally weak, yet it proposed numerous activities, some of a pilot nature (para. 3.3), without contemplating any persuasive project support aimed at capacity building of the Department. The project envisaged a technical assistance input only in financial management but not in support of other areas essential to project implementation in wlich the FD lacked experience. These included women's organisation, preparing and supervising major civil works and arranging for lines of credit for fishermen and women. Project management provided for an interagency Project Coordination Committee (PCC), but its role was expected to be of advisory nature, with little authority on decision making. As the major implementing agency, the head of FD was responsible for project implementation and was to be supported by two executives of FD, although it was clear that none of the FD staff had the experience to plan and implement a large project with many new activities and three donors (IDA, NDF and ICEIDA). 3.2 Revised Objective: The project became effective in September 1991, when start-up actions included the establishment of a Project Coordination Unit (PCU), the appointment of a Financial Controller, who established project accounts and initiated a procurement programme, and the start of an organisational review by the (then) Department of Personnel Management and Training (DPMT). A project launch workshop was convened in November 1991, after which very little activity took place for almost two years, mainly due to FD's management weakness and unicertainty on fish resource availability. Working papers of the FAO/UNDP Chambo Fisheries Research Project (FlI:DP/MLW/86/013) published in 1991-93 claimed that catch and effort data indicated severe over-fishing in the demersal trawl fishery of the South East Arm of Lake Malawi, and suggested that stocks had collapsed during the late 1980s. FD trawl surveys conducted under the Demersal Fisheries Reassessment Project (1989-93) also noted a decline in demersal stocks, although geographically unrelated to the distribution of the fishery. Subsequent studies also suggested that the Chambo project researchers had misinterpreted catch and effort data from fishing boats which were barely serviceable and confined to a very small area, while the FD survey had recorded a natural fluctuation in stock abundance which had more than recovered by 1993. Still, this apparent crisis in the trawl stocks led to a restructuring of the original project in August 1993. Basically, the restructuring responded to three areas of concem: (a) fears of over-fishing; (b) lack of FD capacity in both resource management planning and law enforcement: and (c) lack of progress in a complex project which appeared overloaded with pilot activities in addition to mainstream tasks. The restructuring proposed to shift project objectives away from production and towards improved resource management. Immediate objectives were: 4  (a) to improve the institutional and management capacity of the FD for policy formulation, planning (including donor coordination), research, extension, monitoring and regulation; and (b) to improve the enabling environment for private-sector enterprise in fishing; fish processing, distribution and marketing; boat building and the supply of gear and services Thie 1993 restructuring was accepted by the FD and the Ministry of Natural Resources (M`NR) but was never formalised by an amendment to the Development Credit Agreement (DCA), most likely because it was assumed that the restmicturing did not bring in new goals but rather shed some of the production-related objectives. In practice, however, the restructuring served to guide the project for the next three years. Specifically, the principal modifications proposed were the deletion of the aquaculture and beach access roads sub- components, on the grounds that the FD had already reached task saturation and that these items were the most expendable in terms of expected benefits. A modification to the production component was also proposed, removing a proposed expansion of the pair trawl fleet but retaining the planned replacement of existing vessels as well as investments in the artisanal and commercial sub-sectors. Boatyard privatisation was to be given heightened priority, and in the institution building component additional emphasis was to be placed on the development of management systems, extension services and law enforcement. A mid-term review (MTR) of the project was carried out belatedly in March 1996. It concluded that the project's perfornance was generally unsatisfactory, although the components supported by NDF and ICEIDA were implemented in a fairly effectively and timely manner. Poor performance in the IDA components was explained as the result of weaknesses in the central fisheries administration, including poor organisational structure, inappropriate staffing and lack of resources, compounded by extreme centralisation of financial control at FD Headquarters which hindered timely access to operational resources by field stations. The MTR mission insisted that the project should not continue unless important changes were made to improve its perfomiance, and recommended that IDA should not process withdrawal applications pending adjustments to some budget categories and completion of the following corrective actions by FD: (i) Preparation of a national policy statement for fisheries, and the review and approval of a resource management plan by the MFNR. (ii) Intemal organisation review of the FD to focus on essential research, extension, enforcement and training activities. (iii) Preparation of a fisheries development project document. (iv) Preparation of a new management and financial system. (v) Preparation of an annual work programme approved by the Director of Fisheries. The above actions were to form the basis for project submission to IDA for the resumption of processing withdrawal applications. A moratorium of 60 days was suggested as adequate for the FD to complete the actions. Following the MTR, FD prepared a Fisheries Development Project Document (FDPD) within 60 days, detailing the completion of actions in compliance with the requirements of the MTR mission. The FDPD, which became the basis for project restnmcturing in 1996, essentially confirmed the overall project goals and immediate objectives proposed in the 1993 restructuring, but the DCA was never revised to reflect the restructured project. These project objectives were designed to improve the FD's institutional capacity to enable it to more effectively implement its institutional functions and activities which included facilitation of the development of the private sector. While the focus of the original project objective was on increased production, employment and poverty reduction, together with improved institutional 5  capacity of the fisheries sub-sector, the revised objective de-emphasised fish production and emphasised fish resource management. Therefore, the credit programme, rural access roads and the aquaculture initiatives were dropped. The goal of the restructured projectwas to ensure that Malawi's fish resources were managed in a manner which maximises food production and employment within the confines of resource sustainability. 3.3 Original Components: The original project's components comprised: (i) institutional building aimed at strengthening the FD to enable it to concentrate on regulatory functions, staff training in policy promotion, extension, monitoring and evaluation and enforcement of fishery laws; (ii) strengthening FD's research capacity in fish stock assessment, environmental conservation and management programmes, and in the development of fish fanning models to integrate aquaculture into farming systems; (iii) rehabilitation and development of existing capture fisheries by supporting artisanal fishermen, semi-commercial and commercial fishiing activities, and establishing a pilot programme to increase the involvement of women in fish processing and marketing; and (iv) rehabilitation of infrastructure consisting of upgrading and building access roads, jetties for fish landing, and other shore-based facilities. Although the component mix was appropriate in relation to the set objectives, these were too demanding for FD as it lacked the necessary experience and know-how in handling other non-fisheries matters such as credit (para. 3.1). 3.4 Revised Components: As indicated in Section 3.2, after the March 1996 MTR and the restructuring agreed during the September 1996 Supervision Mission, all the components related to production were excluded. Thus, the main components remained to be: (i) institutional strengtlhening of the FD; (ii) improving fisheries research; and (iii) improving regulatory enforcement and participatory extension. Although the DCA was never amended to reflect these changes, it should be noted that the project that was finally implemented following the 1993 and 1996 restructuring was significantly different from what had been appraised, particularly in relation to the production support activities, which were largely excluded, except for the support to commercial fisheries, through MALDECO. Although there was a sense at the time that tlhe Bank was over-reacting to the issue of overfishing, the FD recognized that the weak capacity of the Department at project restructuring would have made implementation of the original production component an insurmountable challenge. Nevertheless, the FD believed that reformulation of the component, rather than its total cancellation, should have been a more appropriate response to allow the restructured project to address capacity limitations within the FD to implement at least some of the key production investments while helping establish more clarity regarding the status of fish resources in Lake Malawi. Perhaps fearing the prospect of having the whole project closed in 1996, however, GOM agreed with this context despite its reservations about the elimination of all production-oriented investments. 3.5 Quality at Entry: No performance rating was made by the Quality Assurance Group (QAG). Although the original project objectives that aimed at increasing fish production, poverty alleviation, ' The project objectives are as explained in the Fisheries Development Project Document. 1996-1999, prepared by FD, and were the basis for the restructuring of the project. Tl ere are some nmodified versions of these objectives in some of the IDA supervision reports, but they have no substantive difference. 6  resource conservation and improving institutional capacity of the fisheries sub-sector were valid, the project was too complex for a weak department to execute. Flaws in project design at appraisal can in part explain the difficulties encountered by the project. Firstly, even though acknowledging the weak institutional capacity of the FD, a complex project was proposed, including areas in which the Department had little experience, such as major infrastructure construction programmes, including access roads, credit and a women's programme. Secondly, project appraisal failed to finalise the tendering arrangements for the civil works contracts, which were cause for long delays, and equally neglected designing a suitable system for the utilisation of the credit line and detailing the requirements for project monitoring, evaluation and reporting. Thirdly, the appraisal document failed to specify performance indicators and project targets, leaving the task of performance monitoring unnecessarily vague and subjective. Finally, the requirement that GOM divest itself of some commercial undertakings that were servicing the fishing community, without ensuring the continued provision of such services to the sub-sector by the private sector, has proven to be a serious misjudgement. Because of all these shortcomings, the ICR mission's assessment of quality at entry is unsatisfactory. 4. Achievement of Objective and Outputs 4.1 Outcomelachievemnent ofobjective: The project objective was modified due to inadequate project design and implementation difficulties, but not due to unforeseeable extemal factors. Hence, project outcome and achievement of objectives should be compared to those established at project appraisal. In this context, the project outcome and achievement is rated as unsatisfactory, since the main development objective, realising the economic potential of the fisheries sector, was not achieved. Its performance was affected by the over cautious approach adopted to fisheries production, following indicative reports in 1992 about over-fishing in Lake Malawi, and underestimated capacity constraints within the FD, which made the original project objectives impossible to achieve. As indicated earlier, the essential parts of the fish production componenit were dropped from the project, and with this also the objective of contributing to the national economy in terms of improving nutrition levels, employment generation and poverty alleviation. However, these goals remain important national targets to aim at even now, and are consistent with the current Bank's assistance strategy for the next phase of the program in Lake Malawi. The rating of project outcome and achievement of objectives would have been highly unsatisfactory had the restructuring of the project not led to enhancement of the capacity of FD in planning and implementing development programmes and improving the management of the fish resources in cooperation with the local communities. 4.2 Outputs by components: a) Institution building. The institution building component was designed to address perceived weaknesses in the FD's capacity to fulfil its legal obligations to the fishery sub-sector by eliminating all commercial functions in order to allow the FD to focus more on management and regulatory functions, supporting institutional and human resource development, including reform of FD structure in order to clarify roles and simplify the line of command, the creation of additional senior posts, and training, at all levels; and providing infrastructure and equipment, including the construction or rehabilitation of offices, research and training facilities and staff housing, and the acquisition of vehicles, motorcycles, boats and ancillarv equipment. The outputs from this component are discussed below: At project appraisal the FD was engaged in commercial activities of two kinds: commercial fishing, and the provision of services to the fishing industry, including boatbuilding, 7  the production and sale of ice, and marketing of fishing gear, motor spare parts, and fuel. With the exception of boatbuilding at Mpwepwe, the FD was required by the DCA to relinquish all commercial activities according to an agreed timetable extending to March 31, 1992. The objectives of this divestiture were twofold: to enable the FD to concentrate its limited resources on fisheries extension, management and research; and to open the field of service provision to private sector entry. The divestiture programme was followed in a timely maniner, and by 1993 all commercial activities, with the exception of Mpwepwe Boatyard, had ceased. Unfortunately the privatization of these vital services and activities did not guarantee the continued supplv of services to the fishing communities. One boatyard and an ice plant were dismantled. The second boatyard was eventually sold in 1998 after a prolonged period of very low activitv, and has since failed to regain momentum because of the new management's lack of financial resources. There is no other functioning boatyard in Malawi, neither has there been a commercial service agent for the engines of the semi-commercial pair trawl fleet since the FD discontinued this function at an early stage in project implementation. The supply of fishing gear also remains problematic in those areas remote from the Blantyre Netting Company Ltd. (BNC) in southem Malawi, and in the Northern Region the main source of nets and twines is alleged to be illegal imports from Tanzania. The project has resolved FD's constraints in shortage of office space, staff housing and other facilities through the contraction of civil works. The Fisheries Training Centre-- renamed the Malawi College of Fisheries (MCF)--and the Fisheries Research Unit (FRU) were expanded and renovated and a new FD headquarters established next to MONREA. Fisheries offices in the districts have been provided with new office space., housing facilities and vehicles, including motorcycles for the extension staff and boats for enforcement staff A total of 41 vehicles, 57 motorcycles and 25 boats of different types and sizes were supplied to FD under the project. The provision of all these facilities is expected to improve the operational efficiency of the restructured department. The restructuring of FD under the project also included delegation of authority to the regional and district offices, including the transfer of financial management directly to four field cost centres from FD. The district focus adopted by the FD, together with improved mobility and training, should make the activities of the field staff more efficient. Extension services to fishermen and enforcement of fishery regulations were also stepped Up. The extension staff intensified their meetings and dialogue with fishing communities and induced them to establish Beach Village Committees (BVCs) in order to participaie in the sustainable management of the fish resources. The enforcement officers also started to work closely with BVCs, the by-laws of which are upheld and supported by the district staff. Participation of local communities in fish resource management has been boosted by the promulgation. in 1997, of a new Fisheries Conservation and Management Act, in the finalisation of which the project was instrumental. This Act has given the local communities the necessary legal backing in undertaking the management of fishery resources through their own organisations. such as BVCs. A total of 267 BVCs have been formed under the project, but only 34 have been given the training required to make them effective (see discussion of the findings from Beneficiary Impact Assessment Study in Annexes 8 and 9). The GTZ financed National Aquatic Resources Management Programme and its predecessor, the Malawi-German Fisheries and Aquaculture Development Project, provided the leading impetus for BVC formation in the Mangochi, Zomba, and, to a lesser extent, Salima and Nkhotakota Districts, with logistic and institutional support from the FDP. In addition, the project has strengthened the FD in its capacity to carry out policy analysis and development of extension services and resource management through research. To this end, considerable input has been made by the project in training the professional anid senior staff of the FD. So far the project has provided 19 staff with degree and diploma course training, anid a considerable number were given short-term training on various topics. Different types of 8  technical level training were also provided to the extension staff, mainly at the MCF which was substantially strengthened under the project in terms of construction of classrooms, dormitories, offices and other facilities and the provision of vehicles and equipment. The college's curricula have been revised to suit the current and future needs of the fisheries sector, and its teaching staff upgraded through training. At project completion, the college hlas emerged as a modern institution capable of responding to the demand for training by fishermen and staff. The facilities provided appear to exceed the national capacity requirement, but FD explains that the MCF will also service other SADC countries. Overall, the performance under the institution-building component was considered satisfactory, but only marginally so, given the significant delay in the implementation of civil works programme which resulted in serious cost overruns. Progress in the civil works programme was extremely delayed, with no activity on the ground during the project's first five years. There were several reasons for this. At appraisal no agreement was made with the Ministry of Works and Supplies (MOWAS) either to implement the programme or to confer this responsibility upon another suitable agent. Neither were construction specifications or tendering procedures agreed with MOWAS. All of these activities were left to the FD, which had no competence in this field. The amount of work involved in acquiring land, seeking designs and supervising the tendering process was prodigious, and successful implementation required coordinated and timely actions on the part of several players. MOWAS and IDA failed to agree, for a period approaching two years, on a format for civil works bidding documentation, which should have been settled at appraisal. In addition, the GOM decided to alter the location and design of the FD Headquarters, greatly expanding the project, rendering the completed design redundant and delaying mobilisation by three years. Finally, unreliable contractors occasioned several delays, in some instances necessitating a repeat tendering process for completion of the work which resulted in two project extensions. The impact of these changes and delays on the project budget was also extreme. A programme budgeted at US$ 2.41 million finally costed US$ 7.59 million. b) Research. This component was the one that progressed with minimal difficulties. Under the project, the Fishery Research Unit (FRU) was strengthened through the provision of a research vessel, improving its base of operation by repairing jetties and offices, establishing new staff houses and provision of staff training and equipment. Starting from 1993, the Unit has been actively involved in research, particularly in the controversial area of stock assessment, the outputs of which have important implications for management and future development of the fish resource base. Since then, thirteen surveys have been conducted by FRU, anld some of these have been decisive in indicating the availability of a large quantity of unexploited demersal fish capable of sustaining an annual harvest of about 33,000 tons. Information on the location of these stocks has been published by FRU. Through the collection and analysis of catch and effort data, FRU provided in 1999 a comprehensive set of statistics on fish stocks in Malawi waters. Most of the studies carried out by FRU have been used for developing fishery resource management in the country and are key inputs in the planning of the Lake Malawi Environmental Management Project. In addition, a research fund has been established under the project which is expected to continue after completion. ICEIDA, which provided the bulk of the support to research, was so committed to the project that it spent double the originally allocated funds to fisheries research. The research component has performed highly satisfactorily. c) Production component. The initial production component comprised six themes, with considerable inter-dependency between some of them. They were: support to the artisanal, semi- commercial and commercial fisheries; pilot scale aquaculture development; a women's programme with focus on fish processing and trading; and a credit sub-component. As elaborated earlier in the document, due to underestimated capacity constraints within the FD and following speculative reports by a FAO Chambo Project in 1992 that indicated a problem of over-fishing in 9  Lake Malawi, most of the production component was suspended in 1993, as a precautionary measure, until further studies were carried out by FD. Yet, since 1995 research had confirmed that the over-fishing noted in 1993 was localised in extent, and that with adequate management of the distribution of fishing effort the original production targets remained realistic and within the limits of sustainability. At appraisal, the project was expected to support artisainal fishermen, semi-commercial and commercial fisheries by enabling them to replace and buy new boats, as well as fishing gear and other facilities so that they could produce an incremental fish output of about 7,500 tons per year at full development. However, with the exception of the commercial fishery, which was expected to account for about 20% of total output, the project dropped the bulk of the production component in 1996. Most of the funds were then used in the civil works part of the institutional strengthening component. The overall incremental annual production by the project of 7,500 tons was therefore never achieved, and is really a missed opportunity by the country and unsatisfactory performance of the project. Both a description of each individual production sub-component and ani explanation of why they failed to be implemented are presented below (additional background is provided in Annex 7b): Expansion of the Artisanal Fishery into Offshore Waters: The rationale for support to the artisanal fishery was that the inshore waters were already fully or over-exploited, and that helping fishermen to exchange dugout canoes for planked boats would redistribute effort over a wider range of resources. Although it was known that offshore resources were available they had not been quantified at the time of project appraisal, and the scale of this eniterprise was therefore restricted to 50 additional units and 150 replacement units. The planned support to the artisanal fisheries was not altered in the August 1993 restructuring, since the view was taken that any transfer of effort further offshore would serve the interests of resource conservation. The vehicle of support was to be a line of credit (discussed below), and the source of boats and motors was to be the privatised Mpwepwe and Salima boatyards. For the reasons that are discussed in this document, neither the line of credit nor the boatyards functioned in the intended manner during the life of the project. The artisanal production sub-component was therefore never implemenited. Fleet Renewal in the ("semi-commercial') Pair Trawl Fishery: At appraisal, 15 pair units were licensed, but only 10 were operational. The vessels were old, in a poor state of repair, and annual landings were substantially below the assessed sustainable production targets. Therefore, FDP was designed to finance the construction of eight new pair units. Contrary to current belief, it was never the intention of the project to provide a credit facility for the purchase of these vessels, as many of the existing fishing companies were sufficiently creditworthy to obtain commercial loans. An important part of the plan for this fleet was to ensure that effort was suitably distributed in accordance with the known disposition of demersal fish stocks, a management measure that had not been enforced for some years. The development of the pair trawl fleet did not go according to plan. During the early years of the project, this was due mainly to uncertainty on the status of the trawl fishery. Although by mid-1995, FRU conclusively indicated the availability of under-utilised demersal trawl stocks, the project did not reinstate the previously planned support to the pair trawlers. It should be recalled that support to the trawlers was synonymous with support to the Mpwepwe Boatyard--another of the project's sub-components that was, for no clear reason, stalled. Throughout the project life the boatyard was able to build only a very small number of new vessels, and no power units or engine spare parts were supplied after the original Treasury Fund stock was expended in the early 1990s. Import support via NDF was available for the supplv of engines, spare parts, trawl nets and ice plants, but the FD made no use of this facility. The semi- commercial fishery has now virtually collapsed, with only 4 units operating at minimum capacity. 10  The net loss of annual production, subtracting the 1996 catch (1,130 tonnes) from the mean catch over 1986-91. (3,130 tonnes), is 2,000 tonnes (1996 is the most recent year for which an estimate is available). Reinvestment in MALDECO Fisheries Ltd: MALDECO was founded as a private company in the 1950s to catch high value fish by means of trawling and ring netting. It was nationalised after Independence, but although it continued to perform well in public ownership, its profits were diverted to subsidise less viable enterprises, and necessary reinvestment in capital equipment was discontinued. In 1989 the company was sold to its present owner, the Press Corporation, now a public company listed on the Malawi Stock Exchange. At project appraisal MALDECO was in visible decline. Refrigeration, freezing, cold storage and ice-making machinery bought 30 years earlier had exceeded its useful life, and the condition of vessels anid vehicles was equally poor. By the time of project start-up MALDECO was no longer profitable. Under the project, NDF funds were reserved to provide import support to MALDECO, administered as a local-currency loan by the Reserve Bank of Malawi on behalf of the MOF. This would provide for the supply of a new 17 m multi-purpose fishing vessel as well as a range of ancillary equipment. In order to prevent an excessive build-up of fishing capacity in southern Lake Malawi (given the projected support to the pair trawlers) MALDECO was required to remove from service two small stem trawlers after commissioning the new vessel: this was done in 1995-96. The supply of equipment proceeded fairly closely to plan, and between 1993 and 1997 the project provided the fishing vessel Kandwindwi, cold rooms, a blast freezer, a holding freezer, an ice plant, a delivery truck, two fork-lift trucks. a new operational base at Chipoka (including cold storage and a wholesale/retail shop) and an usipa fishing rig. The Kandwindwi exploits deep- water demersal trawl stocks in the southern part of Lake Malawi and to a limited extent in the Central Region. In pursuing this fishery MALDECO is making direct use of research conducted under the ICEIDA supported FDP research component. Kandwindwi entered service in 1994, but after a successful start was damaged by fire in 1996. The period 1997-99 saw continuous service, with annual production averaging 1,110 tonnes (see Table 4 in Annex 7b). In addition, each of the experimental usipa rigs, one of which was provided by the FDP, took on average 10 tonnes/annum during the period 1995-99. The two vessels that were replaced according to the FDP prescription caught an average of about 200 tonnes/ annuLm during their final four years of service. The net increase in annual production attributable to the project is therefore around 900 tonnes. The company's financial performance improved markedly during the life of the project (see Figure 4 and Table 4 in Annex 7b). The gross margin on fishing operations rose from a low of about US$390,000 in 1994 to an annual average of about US$1 million in 1997-99. In 1995, pre-tax profit emerged into positive figures for the first time since the late 1980s, at around US$310,000, and has since grown slowly to slightly over US$400,000 in 1999. Recently, sustained growth in the face of a short-term decline in fishing margins indicates improved efficiency and expanded retailing operations. The contribution to profitability made by the refrigeration equipment is included in the cost of sales and is therefore not separable from the fishing operation. Credit: At project appraisal it was acknowledged that small-scale fishermen and those engaged in ancillary trades were effectively locked out of the formnal financial system because of their inability to provide loan collateral and because of the perception of fisheries as a high risk sector. Accordingly, a line of credit was to be provided through a pilot project for implementation by the GOM's Smallholder Agricultural Credit Administration (SACA). on commercial terms and conditions. Loan applications to SACA would require certification from II  FD that the planned activity complied with appropriate technical, environmental and legal standards. SACA was to lend to artisanal fishermen, for boats, engines and fishing gear, and to women fish processors and traders. The project budget included US$662, 100 for investment in the artisanal fishery and US$200,000 for the women's programme. While credit terms and conditions were outlined clearly in the SAR, the practical mechanisms for setting up and nianaging the credit component were again left to the FD, without guidance or technical assistance. By August 1993 SACA had prepared an action plan to implement the credit scheme, and was awaiting approval of the MFNR to proceed. The IDA restnicturing mission of 1993 proposed to expand its scope to include small-scale boatbuilders, since the critical shortage of fishing craft exacerbated by the closure of Salima boatyard was becoming increasingly apparent. However at this time Malawi was undergoing a profound political change--the advent of multi- party democracy. Democracy brought with it some unforeseen effects, one of which was widespread political interference in the seasonal agricultural lending programme in 1993, resulting in loan default on such a scale that it bankrupted SACA. By 1994 SACA had been succeeded by a quasi-commercial venture--the Malawi Rural Finance Company (MRFC)-- initiated under the World Bank and IFAD supported Rural Financial Services Project. The breakdown in credit discipline made for a difficult start for MRFC, and despite approaches by FD. fisheries loans were not accorded high priority. It was only wlhen the WB task management of both the FDP and MRFC fell to the same individual in 1996 that productive discussions on the credit component were resumed. An important policy change made then was the deletion of the line of credit but retention of the credit theme through MRFC, although it was clear that this company would not extend medium- to long-tern credit to fishennen. The credit programme started at the operational level in 1997 on the direct initiative of MRFC. Fisheries staff received training in the modalities of MRFC business lending, and were themselves involved in the establishment of credit groups (women's, men's and mixed), in training each group over a 10-week period, and in supporting those groups which had passed MRFC's screening process and had actually been provided with loans. By April 2000, 259 groups had been formed and almost half of these had received loans (See Table 5 in Annex 7b). Groups ranged in size from 10 to 25 members, and it is estimated that close to 4,000 people were members of fisheries credit groups. Loans averaged about MK5,000 (say US$ 100) and were given for a maximum period of one year, but usually repaid within a few months. Most of the borrowers were women and they used the credit to trade consumer goods, with little of this being for fish trade. The objective of this micro-credit was, therefore, completely different from the original aim of the project of promoting investment in the developlment of capture fisheries, fish processing and marketing. During the latter part of the project interest rates increased substantially. The base rate rose from 23% in the final quarter of 1997 to 43% in the same period in 1998, and 48% in 2000. MRFC's current business lending rate is 49%. This may be acceptable in rapid-turnover, high-margin enterprises, and it is not surprising that trade in various commodities (crops, clothes, sugar, salt, fish, oil) dominates the business activities of the fisheries credit groups. Short-term loans at high interest rates are not suitable for financing capital items such as fishing craft or engines, and no loans for boats or boatbuilding have been granted or are anticipated. Useful as it is, the MRFC programme should not, therefore, be considered as a replacement to the original FDP credit sub-component, which was to support fishery production and processing. The Women 's Programme: The women's programme was conceived as a pilot activity to be carried out in three districts in support of women's role in fish processing, 12  distribution and marketing. It was to incorporate training, in fish processing and business management, and credit, for working capital and the acquisition of necessary equipment. The success of the pilot programme was dependent on the operation of the credit programme, which for reasons elaborated above was both delayed and downsized, and on the supply of ice, whiclh had more or less been removed by the project. TA was to be provided in the form of a baseline socio-economic study, and female officers were to be recruited and trained to manage the programme. During 1992-93 many women's groups were established by fisheries extension agents in anticipation of a substantial credit programme--even in districts outside the three proposed for pilot operations. In 1993 a Senior Women's Programme Officer (SWPO) was appointed to FD Headquarters, and in 1994 three WPOs were recruited and posted in the field. The planned baseline survey and action plan were both successfully completed during 1994. The needs of women in the lakeshore communities were found not to match the project's initial emphasis on fish processing, and were more broadly spread over a range of small business activities. Implementation of the action plan was seriously impaired by the continued delay in the availability of credit, and by March 1996 only 23 of the original women's groups remained. The restructuring of the FDP in 1996 removed both the credit and the women's programme, and three of the original four female officers left the FD in subsequent years (one stayed on as a District Fisheries Officer). The only activity remaining from this programme is the modest short-term credit programme extended to women through MRFC. Aquaculture: Despite the availability of suitable land and appropriate teclnologies, aquaculture has not had a history of success in Malawi. Past efforts have been characterised by sporadic donor support, excessive concentration on research ratlher than on the dissemination and support of proven methodologies, and the dispersal of limited human and financial resources over a number of experimental projects. At project appraisal the aquaculture sector was producing less than 50 tonnes/annum. The FDP pilot project conformed to the previous pattern of development in this field. Another new experimental station was proposed, and two locally untried technologies were to be the subject of pilot initiatives. By 1993 no progress had been made in the aquaculture sub-component. Two other aquaculture projects had recently started in both the Northem and Southern Regions, placing considerable demands on an overstretched FD professional establishmenit, and again the sub- component comprised only a sketchy design and a budget, with no technical assistance to help it off the ground. It was judged that implementation of this sub-component would place unsustainable demands on the FD and would duplicate work undertaken by the EU financed Central and Northem Regions Fish Farming Project (1989-95). Accordingly, it was recommended during project restructuring that the aquaculture component be removed from the project entirely. d) Infrastructure. The original infrastructure component comprised three elements. The most important of these was the restoration of Mpwepwe boatyard to full working capacity, also an integral part of the planned expansion of fish production. In addition, the project planned to build 25 fishing beach access roads and two jetties at FD field stations. The Mpwepwe Boatyard was not improved as envisaged under the project because it was targeted for privatisation. Hence, only minor wood working machines were provided by the project. Contrary to initial project plans, the construction of access roads was also excluded in order to simplify the project and because of the reduced focus on increased production. In spite of this exclusion, however, the limited infrastructural rehabilitation that was implemented under the project (new jetties at FRU in Monkey Bay and at the District and Regional FD Office at Nkhata Bay) has been beneficial for the operations of research and for other FD vessels. Hence, the outcome of this component is considered satisfactory. 13  A description of the outputs produced under the infrastructure component is presented below (additional background is provided in Annex 7b): Mpwepwe and Salima Boatyards: At project start-up there were two commercial boatyards in Malavvi, both owned by the FD. One at Salima produced very popular V-bottom wooden boats for the artisanal fisheries. This yard was closed by the FD in 1993 in accordance with the DCA provisions, but no attempt was made to privatise it. The staff were dismissed, and some months later the tools and equipment were auctioned to local joinery finns, with FD retaining the site. Supervision missions in 1993 and 1994 emphasised the need to attempt to salvage this situation, but by this time the skilled workforce had dispersed, and no action was taken. The situation with Mpwepwe Boatyard, in Mangochi District, was rather different. Mpwepwe was a larger establishment whose main line of construction was a 7.5 m V-bottom boat, the basis of the pair trawl fleet. Maintenance of this fleet was estimated to require the construction of six new vessels annually. By 1991 output had fallen to a very low level--one or two boats per year--and the yard was in debt and in urgent need of rehabilitation. Thne construction of new vessels was key to the proposed expansion of artisanal and semi-commercial production, and therefore the redevelopment of Mpwepwe was seen as a priority activity. It was intended under the project to rehabilitate the yard and to run it on a commercial basis within the GOM management context. This option was selected because at appraisal the prospects for privatisation appeared remote. In 1993 a consulting firm was appointed to provide technical assistance for the rehabilitation, resulting in the preparation of new boat designs and outline specifications for reconstruction. The boatyard staff was reduced to a bare minimum while the rehabilitation was under way. By this time a number of entrepreneurs had expressed interest in the acquisition of the boatyard, and privatisation appeared for the first time to be a real possibility. The rehabilitation was therefore deferred on the recommendation of the 1993 restructuring mission, which considered it more appropriate to make available to the new owner, on a loan basis, those funds which had been reserved for the boatyard redevelopment. The NDF approved this course of action. The privatisation of Mpwepwe took almost six years, and was not completed until May 1999. In a management buy-out the yard was sold to the Mpwepwe Boatyard Company Ltd., comprising of staff previously employed by the FD. However, by this time the yard was almost defunct, and most of the equipment unserviceable. ICEIDA made a small grant available to buy minor tools, but there was insufficient time to arrange the NDF/GOM subsidiary loan prior to project closedown. Furthermore, boatyard debts exceeding MKW 2 million formed part of the privatisation package. The yard is now on the verge of collapse, and stands little prospect of survival unless it obtains rapid financial and management support. Access Roads and Jetties: A construction programme for beach access roads--at 25 sites and of an average length of about 2km--was included in the project as a stimulus to trade in locations where artisanal production was high and access poor. About 10% of the routes surveyed were also to serve trawler landings and the two boatyards. No progress on this programme had been made at the time of the project restructuring in 1993, and its deletion from the project was recommended at this time and later in 1996. The principal argument used to support its removal was that the FD was overloaded and did not have the capacity to implement this activity in addition to other tasks which had higher priority. It was also claimed, however, that given the shortfall in the national supply of fish, road access could no longer be considered a serious constraint to production or distribution. Two new jetties did get constnicted under the FDP, one at the FRU in Monkey Bay and one at the District and Regional Fisheries Office at Nkhata Bay. Their primary finction is to provide safe landing points for research and patrol vessels, i.e. they are part of the FD infrastructure rather than services to the fishing industry. 14  4.3 Net Present Value/Economic rate of return: At appraisal, the project was primarily justified on the basis of increased fish production, which was expected to reach 7,500 tons per year or 10% of the tlhen national production. Such a level of fish production was expected to yield an economic rate of return (ERR) of about 20% over 20 years. Furthermore, significant improvement in the people's diet, food security and off-farm incomes were to have resulted from the incremental fish outpult by the project. Unfortunately, the project excluded the main production activities expected under the artisanal and semi-commercial fisheries, keeping only the support to commercial fisheries, which was a minor element in termis of project's expected production. As a result, the incremental annual production at project completion was only about 900 tons by the commercial sector, i.e. MALDECO, which accounts for only about 12% of the output forecasted at appraisal. This would generate a negative ERR if it was to be recalculated based on the appraisal approach, because the project costs at completion are about the same as estimated at appraisal, but fish production was only a small fraction (12%) of the forecasted figure. The qualitative improvement in nutrition, food security and off-farm incomes also were not realised. The divestiture of FD's commercial activities have negatively affected the fishing community, because the new buyers did not ensure the continuity of services in the making and selling of ice and supplying boats, boat engines, spare parts and fishing gear. 4.4 Financial rate of return: At appraisal, financial analyses were carried out for the artisanal fisheries, semi- commercial fisheries, the Mpwepwe boatyard and MALDECO development proposals, and all yielded satisfactory financial results. With the exception of MALDECO, whose financial rate of return (FRR) was estimated at 20%, all other enterprises were expected to generate FRRs of 30% or more. Since the proposed artisanal and semi-commercial fisheries and the Mpwepwe boatyard development were not implemented under the project, only comparison with the MALDECO fisheries development is possible. At appraisal, MALDECO was expected to improve its financial performance and raise its pre-tax profit from about MKW 2.03 million (US$0.73 million) in 1990/91 to MKW 3.9 million (US$1.39 million) at full development, starting from 1992/93 onwards, in 1991 prices. At completion, the pre-tax profit of MALDECO accounted for about MKW 18.63 million (US$0.43 million) in current prices, i.e. a third of what had been projected at full development in US$ terms. This lower profit, compared to the appraisal estimate, is due to reduced lower fish catch in general, and the proportional decline of the high value fish, i.e. Chambo. Because of lack of information on the earlier years, it was not possible to carry out FRR calculations. 4.5 Institutional development impact: The impact of the project has been modest compared to appraisal expectations, principally because of the removal of the production component. Its major impact was on the FD's institutional development, in that investments in physical infrastructure, vehicles and operational support have shifted the FD from a state of immobility and dependency into one which commands a significant field presence. More importantly, the FD is a better structured, more professional and more capable institution than it was at project start-up. With project assistance the FRU has conducted timely and relevant research, the results of which have been made available through publication, and made use of by the industry. This has contributed to our understanding of the exploited fish stocks, and has identified new stocks previously unexploited. During the course of the project the FD has consulted widely in the development of a new 15  national fisheries sector policy, and in 1997 Parliament approved a new and progressive Fisheries Conservation and Management Act (see Annex 7b). A national programme of fisheries extension has been established, many of its staff trained and its activities monitored and periodically reviewed. Co-management of the artisanal fisheries, provided for in the new legislation, has been promoted through the extension programme, and although implementation remains at an early stage, its principles are well understood within the FD and are becoming known and accepted in the fishing communities. The relationship between these communities and the FD has improved dramatically as a result of the latter's role in support of credit groups, and linking them to MRFC lending. These are significant achievements, and are a tribute to the FD and to the guidance provided by IDA during the later years of the project. When attention is turned to the industry, the impact of the project has been minimal. This is because only the MALDECO investment proposal was implemented, but not the artisanal and semi-comrnercial investment proposals from which the bulk of the expected project production and development impact were to originate. There are other extensive areas of zero- impact in which a major fisheries development project might have been expected to make improvements. The range of fishing gear available to the industry remains unchanged after 30 years: no new materials have been provided, or technologies extended. The pair trawlers continue to fish with trawls designed in the 1970s. Boat designs are also unchanged, and access to fishing beaches no better, or worse, than before the project. There have also been negative impacts, principally associated with the FD's withdrawal from boat construction and the supply of other services. The closure of Salima boatyard scattered skilled boatbuilders and their equipment to local joinery shops in an unseemly end to FD's many years of workforce development. As explained earlier, the Mpwepwe yard is in a very precarious state. After project start-up it was rmn at an extremely low level of output, and many of its staff were laid off. When privatised 8 years later, the boatyard was little nmore than a shell, its new owners inexperienced in business and potentially crippled by inherited debts. The reduced availabilitv of seaworthy boats to the artisanal fishery has contributed to the concentration of fishing effort in the inshore zone, to increasing poverty in the fishing communities and ultimately to the resource management problems whichi the FD now has to face. The mechanised fishery has been more visibly affected. The pair trawl fleet was in a poor state when the project started, but after a decade without replacement engines or spare parts, this fishery is now almost defunct. FD's withdrawval from boat building and other services did have the intended impact of freeing up the time of field professionals for other areas of responsibility. But overall the impact on the FD itself was strongly negative, since field officers saw development of the industrv as a prime responsibility and the dismantling of the services they had worked to establish as incomprehensible. It confused field staff and soured working relationships between the FD and the industry. It would be wrong to give the impression that FD's withdrawal from service provision has left a total vacuum, and there are several examples of private sector development in this field that probably would not have taken place had the FD continued its operations. Two small-scale self-financed ice plants are known to be in operation, and several former Salima boatyard employees continue to make boats to the popular designs developed by the FD. However, the ice plants service mainly non-fishery businesses, and the boatbuilders operate as village enterprises with very limited capital, and their inability to access other than locally available construction materials means that their products have a very short working life. Fishing gear is sold by a number of retail distributors throughout Malawi, but this is a risky and unprofitable business, and annual sales by the Blantyre Netting Company have fallen from 80- 100 tonnes in the 1980s to only 50 tonnes in 1999--a remarkable statistic given that the number of fishing craft has increased by 60% over the same period. 16  5. Major Factors Affecting Implementation and Outcome 5.1 Factors outside the control of government or implementing agency: The project was not affected by any factors outside the control of government or implementing agency. The change in the political and economic system of the country had no particular effect on the project if not the general impact that this has made on all sectors as a result of depreciation of the local currency, inflation, and the effect of measures aimed at restructuring and stabilising the economy. 5.2 Factors generally subject to government control: The main factor subject to government control that negatively influenced project implementation is MOWS negligence in the execution of project's civil works, as it took five years to commence works on the ground. Other factors (discussed in Annex 7b) included: delay in the recruitment of senior staff agreed for under the project; shortage of budgetary funds in the early years ofthe project; delay in the release of funds; = uncertainty about the site for FD headquarters; and lack of continuity in project coordination by the PCC. On the positive side, GOM's agreement to the decentralisation of FD (including the establishment of regional cost centres); preparation of the National Policy Statement on Fisheries; and promulgation of the Fisheries Act, were important factors in turning around the project's performnance from poor to better. 5.3 Factors generally subject to implementing agency control: The most serious factor that affected project implementation, especially during the initial years, was lack of project ownership. Overall, FD was passively accepting decisions made by other bodies such as the MOWS or missions from the International Development Agency (IDA), as if it did not have the most important role to play. FD took little initiative to turn around negative situations to the way it believed were right for the fisheries sub-sector. Specific factors subject to FD's control that affected project implementation included (see Annex 7b for a discussion): lack of decisive action in maintaining the production component within the project when it became clear that there were many areas underfished in Lake Malawi; weak project management and initial centralisation of project implementation; weak project monitoring, evaluation and reporting; and inadequate filing and documentation to the extent that important documents are not found in project files. An important factor that enhanced project restructuring and implementation performance was FD's determination to prepare the Fisheries Development Project Document (FDPD) along the lines indicated by the MTR for a new beginning. 17  5.4 Costs and financing: Total project costs at appraisal were estimated at US$15.50 million. This was expected to be financed as follows: US$8.8 million IDA; US$3.5 million NDF; US$ 1.0 million ICEIDA; and the remaining US$ 1 .1 million by GOM. At completion, total costs amounted to US$15.10 million consisting of US$7.88 million IDA; US$2.81 million NDF; US$2.13 million ICEIDA; and US$2.3 million GOM2. In terms of financing, both ICEIDA and GOM doubled their financing levels compared to appraisal, while IDA and NDF contributed slightly less than had been planned. About half the costs were spent on civil works, which is a direct result of delayed implementation and cost overruns. Most of the funds initially allocated to support fisheries production by IDA were transferred to civil works. NDF's main financinig items were support to commercial fisheries development (MALDECO) and ICEIDA's main expenditures were on improving fishery research. 6. Sustainability 6.1 Rationale for sustainability rating: Overall, project's sustainability, compared to appraisal expectations, cani be rated as unlikely principally because of the removal of the essential investments of the production component from which the bulk of the expected project production and development impact were to originate. If the assessment of project's sustainability is limited to the implemenited institution building and research activities, sustainability could be considered as likely. The sustainability of development interventions supported by the restmictured FDP may be assessed on several levels. Ultimately, the intention is to put in place a variety of mechanisms to ensure that Malawi's fish resources are used and conserved to the maximum economic and social benefit to the nation. In the artisanal fisheries these mechanisms will depend to a large extent on the emerging local-level fisheries institutions--BVCs and Fishermen's Associations. These are yet at too early a stage in their development to permit a meaningful discussion of sustainability. The only interventions that have gone all the way to the end-user are the MALDECO loan and the FD/MRFC credit groups. Both of these initiatives are likely to be sustainable on their own account, and are not dependent on a great deal of back-up support. The remaining activities of the project concern the activities of the FD, and here sustainability may be considered at both financial and institutional levels. Financial sustainability is of immediate concem to the FD--even if support to the fisheries sub-sector were to enter a second phase under the Lake Malawi Environmental Management Project, it is now inevitable that a substantial funding gap would precede the new project. Financial sustainability means essentially the sustained funding of field operations, including provision for the maintenance and ultimate replacement of project investments in infrastructure and equipment. Although GOM support to the fisheries sub-sector has declined over the life of the project, MOF's position is that the project phase-out is understood and will be compensated by an increase in the recurrent budget. The extent of the increase was. however, not conveyed to the ICR mission. The managers of two costly recurrent activities--the Ndunduma research programme and the MCF--have taken proactive steps to assure future financial security by taking on 2 GOM's contribution increased because operating costs and expendituires on civil works, the only two categories for which GOM financing was contemplated, had sky rocketted. 18  commercial enterprises in addition to their development functions. It is intended to sustain the Ndunduma programme through commercial fishing at times when the vessel is not required for research purposes and thus replenish its research fund continuously. Current thinking for MCF points towards some degree of financial autonomy, with the college paying its way through cost recovery, the renting of conference facilities, commercial fishing, competition in the regional market for natural resources training, etc. These activities would certainly improve the sustainability of project activities under these two FD Units. This thinking is currently in line with that of GOM, where a number of ministerial departments are being encouraged to raise revenues to finance their services and operations. However, they would take the FD through a full cycle and back into a commercial role deliberately closed by FDP. This raises an important point of development philosophy--should the GOM enter into commercial activities in order to sustain necessary government functions? What is the justification for removing such functions into the quasi-governmental grey area, and how do the answers to these questions match the likelihood that GOM will in fact finance these activities from taxation revenues'? These are questions only the GOM can answer, but they should be confronted openly and not answered by default. 6.2 Transition arrangement to regular operations: During the last budget piesentation for the fiscal year 2000/2001, the MOF has proposed to include a portion of the project investment and operating costs under the recurrent budget. However, it was not clear what level of the operating and maintenance needs of FD would cover the MOF proposal. Another arrangement being made by GOM is the preparation of a fisheries resource management and sector development component as part of the proposed Lake Malawi Environmental Management Project. Discussions on the second-phase project are being carried out with donors, including the World Bank and the Global Environment Facility (GEF), and its final formulation is awaiting the finalisation of this ICR and that of the recently completed (June 2000) Lake Malawi/Nyasa Biodiversity Conservation Project, financed by the World Bank and the GEF. Quite correctly, the GOM intends to include lessons learned under the two projects in the next phase. Apart from possible shortage of funds, the FD's present structure and system of work should be able to undertake the necessary development work and provide services to the fisheries sub-sector. The question is will the fishing community continue to listen to the FD's technical and resource management advice, while their capacity to fish where fish is abundant is impaired by lack of adequate boats and fishing gear? If the fisheries sub-sector is to develop further, and the FD is to be heard and respected by the fishing community, there must be a concentrated effort aimed at resolving the problems of fishermen in tenis of provision of adequate technology and equipment for improved and sustainable fishing. As findings from the Beneficiary Impact Assessment Study suggest (see Annexes 8 and 9), meetings and discussions with the fishing communities alone will not help the fishermen to improve their incomes, and hence contribute to the national economy, if they lack suitable means for undertaking sustainable fisheries. Future fisheries development efforts should therefore, endeavour to enable fishermen to catch fish in underfished areas, which are mainly offshore. Encouraging offshore fishers would generate significant benefits: higher catches (increased production to the nation and incomes to fishermen) and reduced pressure on the biodiversity-rich nearshore areas, while minimising environmental damage and threats to the biodiversity of Lake Malawi. However, offshore fishing will call for major investment in enabling artisanal fishermen to catch fish through training and provision of the necessary equipment and material. Certainly, FD should continue to monitor the fisheries sub-sector to ensure that the resources are not over-exploited and that the use of fishing technology and equipment are compatible with sustainable fishing practices. It should also continue to practise co-management of the lake resources and to provide the extension activities 19  introduced under FDP. Support to such strategies and activities is being considered under the proposed Lake Malawi Environmental Management Project. 7. Bank and Borrower Performance Bank 7. 1 Lending: The Bank provided crucial assistance in the review of the fisheries sub-sector in which the main issues facing fisheries in Malawi were identified. Th-e Bank s subsequent decision to send an identification mission to assist GOM in the conceptualisation of a development project and to provide IDA's financial support, opened a golden opportunity for inodernising the fisheries sub-sector in Malawi. Investment in fisheries was long overdue and Banlk's assistance in the review of the sub-sector and in the project identification in the late 1980s ivas satisfactory. Unfortunately, this was not the case with regard to Bank's performance in the appraisal of the project, which was clearly unsatisfactory. The appraisal mission came up with a complex project for implementation by the FD, which earlier had been assessed as weak by the same mission. Although the appraisal mission proceeded along the proposals made in the preparation report, it eliminated important technical assistance inputs that were designed to boost the implementation capacity of FD. The design of the supply of boats during project fornulation was to strengthen the boat making in the country, and to procure the boats tnder the project, but to finance them under a credit scheme through three different financing institutions in the country. At appraisal, this was changed in that the support to be provided to boat building was excluded and financing arrangements were expected to be with SACA to cater for small-scale fishermen, leaving the semi-commercial fisheries to finance boats with their own funds or borrow from any lender of the banking system. In addition, a new activity to support women involved in fish marketing and processing was included at appraisal without sufficient study and without equipping the FD to undertake the task. Other weaknesses that can be attributed to project appraisal are failure to devise a system to ensure the continuity of services rendered by the FD commercial undertakings after their divesture; lack of a clearly defined civil works programme, and failure to emphasise the importance of completing a manpower development plan and recruitment of professionals by making this an issue for agreement during negotiations (such conditions were imposed on the restructuring of the Mpwepwe boatyard and SACA's adoption of suitable credit procedures, although these were less critical items to the whole project compared to the need for strengthening of the FD). 7.2 Supervision: During the first half of the project until 1995, supervision by IDA was of mixed nature, with some individual missions performing well, but overall functioning below expectations. First of all, the supervision missions were not carried out regularly over the years: only one mission each was undertaken in 1991, 1994 and 1995 and no supervision was made in 1992. In addition, the mid-term review scheduled to take place before mid-1995 was delayed. No valid reasons were given for such irregularities, but an important factor might have been the frequent changes of IDA Task Managers, with four different people up to the project restructuring of 1996, and another one since September 1996. In addition to these anomalies, the earlier supervision missions failed to assist the FD to come oult of its impasse. Thleir reports were more of a criticising nature with generic recommendations and threats of suspending the project, instead of suggesting a clear action plan for turning the project around and taking a firm stand to implement them - like the approach adopted during the MTR. The supervision missions also 20  panicked at unconfirmed news that indicated overfishing in Lake Malawi, in spite of the thorough studies and analysis carried out during project preparation, and advised for the suspension of the production component. While such a precautionary measure may be justified until the issue of resource availability is sorted out, it is not clear as to why this component was not reinstated at least in partial form after it was made clear in 1995 that there were many areas with underfished demersal fish in Lake Malawi. In addition, the requirement that GOM divest itself of some commercial undertakings that were servicing the fishing community, without ensuring the continued provision of such services to the sub-sector, proved to be a serious misjudgment resulting from a project action. In the assessment of the ICR teamn, there is no doubt that the project needed to be restructured because of its complexity and the underestimated capacity constrainits within the FD (i.e. even if there had been no concems about overfishing in Lake Malawi), which would have made all the original objectives impossible to maintain. The main problem with the restricturing was IDA's decision to cancel the entire production component, instead of being open to discuss with GOM its possible reformulation. IDA also failed to address the impacts of the unprepared privatizations on fish production which could have been dealt with at the time of restructuring. In addition, the DCA was never revised to reflect the trimmed down number of components. On the other hand, the focus on strengthening the FD, which was largely accomplished during the last three years of the project, is to the credit of the project restructuring team. IDA supervision improved markedly after the restructuring of the project in 1996. By developing a clear programme of work and monitoring system with FD, the supervision missions were able to improve significantly after the restructuring of the project. which was confirmed (though never fonnalised in the DCA) by the September 1996 mission. Because of the availability of the workplans agreed by FD and IDA, the supervision missions (two per year) based their review on what had been planned under the previous six-month period of work. Having reviewed such a programme, the supervision missions used to agree on another six-month plan which would be reviewed by the next mission. This system of work enabled supervision missions to follow project implementation systematically and to focus on the major issues that required their assistance. In contrast to previous missions, IDA maintained the same Task Manager from the project restructuring in 1996 until completion. This continuity greatly helped the project as the flow of thought was not interrupted, and the working relationship with the FD improved markedly. Therefore, implementation proceeded according to plans. 7.3 Overall Bank performance: It is difficult to give one rating for a nine-year project, in which the first five were characterised by poor performance and the last three years by better performance. albeit on a restructured project with significantly narrower objectives. The true picture of Bank-s performance would have been unsatisfactory for appraisal and for supervision before the restructuring of 1996, and satisfactory thereafter. If one rating should be made, however, the overall bank performance should be rated as less than satisfactorv. This should be considered a fair rating if account is taken of the fact that improved performance in the last years was on a project that was significantly down-scaled after the main production activities were excluded. Although it does not change the overall assessment of Bank perfom-ance and final project outcome, it is important to understand the context under which the Bank adopted the position to suspend the production component at project restructuring. The task team could have felt that the precautionarv measure of suspending all production-oriented investments was justified until the issue of resource availability was more fully sorted out. There was only a short period of time between the date the issue of overfishing was made known and the refutation of 21  this finding by research conducted by the FD and others. Further, the issue of overfishing was raised at a time of global focus on resource conservation issues following the 1992 UNCED. The Bank had some reason to be cautious about its support for increasing fish production in the biodiversity-rich Lake Malawi after it had come under heavy criticism for its past environmental record. Together with lack of capacity within the FD to implement the original production component, this may have had a bearing on the precautionary approach taken bv the Bank at the point of restructuring. Still, questions remain on whether the correct intervention should have been dropping the production component completely or modifying it through a change of strategy or more targeted focus. Since research in 1995 had confirned that the over-fishing was localized in extent, with adequate management of the distribution of fishing effort the original production targets could have remained realistic and within the limits of sustainabilitv. Borrower 7.4 Preparation: T'he design of the project was too complex for the capacity of FD. but it was believed that, with the support of technical assistance personnel, implementation would have been less problematic. Agreement with the appraisal mission on the exclusion of most of the technical assistance envisaged during preparation amounts to lack of recognition by FD of the principle accepted during project formulation that project complexity was to be mitigated by the help that would be provided by technical assistance personnel. 7.5 Government implementation performance: The Borrower's performance in the early years of the project was characterised by slow compliance with the agreements reached in implementing the project. Notably, GOM was not able to provide the necessary budgetary funds and recruitment and appointment of staff on time, and this held back progress in project implementation. Another example of poor performance was the excessive delay (about 5 years) in tendering the civil works contracts by the MOWS, leading to significant cost overruns. While the shortage of funds is explained by the overall government budgetary control policy, the undue delay in finalising civil work contracts can hardly be justified. In addition, the PCC did not function as anticipated, and did little to resolve the problems faced by the project with regard to the availability of funds, staff recriitment and the impasse in civil works. During this period,GOM's performiance was below expectations, but improved significantly after the MTR. Signs of these improvements were the preparation of a National Policy Statement for Fisheries in 1996, the promulgation of the Fisheries Conservation and Management Act in 1997, and the decentralisation of FD, including financial responsibilities to the districts. 7.6 Implementing Agency: The passive stance of FD during the early years of the project was one of the main causes of the implementation difficulties. The FD was only peripherally involved in the most crucial events and decisions made on the project. For instance, it was a FAO-sponsored report which indicated that Lake Malawi was overfished, and not FD's learned judgement, that created the initial difficulties to the production component of the project, leading to its suspension and later exclusion. Yet, as the institution responsible for the country's fisheries, the FD did not challenge this report until a study by FRU invalidated it 1995. But even after that time, perhaps fearing the prospect of having the whole project closed, FD failed to make a case for the production component or its reformulation, although it is clear, with hindsight, that FD senior officials were in favour all along of supporting increased production on a sustainable basis. The same could be said about their silence conceming the lost services in the supply of boats, ice, 22  motors, spare parts and fishing gear, after the divesture by FD. The fishing community was expressing its disappointment and protesting on the lack of these necessary supplies, but the FD failed to attempt to resolve the problems within the framework of the project or at project restructuring. It should be very well acknowledged, however, that after restructuring of the project in 1996, the FD significantly improved its performance. It introduced district based planning of work, devolved power to the districts, and established a good working relationship with fishing communities by initiating fish resource management committees through establishment of BVCs. 7.7 Overall Borrower performance: Considering the delay and inadequate provision of funding, slow provision of key staff and poor implementation that led to the elimination of important project objectives such as the increased supply of protein and poverty alleviation, overall borrower performance is rated by the ICR team as unsatisfactory. 8. Lessons Learned The main positive and negative lessons that could be leanied from the implementation experience of the project are presented below (see also Annexes 8 and 9 for a discussion of the findings from the Beneficiary Impact Assessment Study): - While it may seem obvious, the experience of the FDP makes it necessary to re-state that the design of a project must be compatible with the capacity of the implementing agency to carry out the planned activities. Even though the weak institutional capacity of the Fisheries Development (FD) had been acknowledged, a complex project was proposed, including areas in which the Department had little experience, such as major infrastructure construction programmes, including access roads, credit and a women's programme. Despite the weak capacity of the Department during project preparation, the appraisal mission omitted important technical assistance inputs that could have boosted the implementation capacity of FD. There was also failure to specify performance indicators and project targets, leaving the task of performance monitoring unnecessarily vague and subjective. - The experience of the project also demonstrates that complex projects present substantial and often unique management demands. Procurement and financial management are critical functions that demand a level of expertise that is generally not found in govemment departments and is difficult to attract and retain given govemment salary constraints. Poor performance in these areas typically results in serious delays in implementation, frequent and recurring ruptures in the flow of funds to implementing agencies, suppliers and contractors, and additional cost to govemment and to the project. Technical project management, especially the coordination, tracking, and support of a diverse array of programs and activities is particularly time-consuming. Government departments are generally deficient in experienced project managers and lack sufficient senior technical staff to be able to dedicate an officer for project management, without pulling the officer away from other critical duties in the department. As a consequence, public sector capacity is further weakened by the demands of project management and reporting. The lesson here is that serious consideration should be given to placing project management, including financial management and procurement functions, under professional managers. This can best be done through contracting out these functions to a firm on a competitive basis. Performance of the firm can then be reinforced through regular annual evaluations. 23  - One positive lesson of the restructured project was the fact that the FD did not need to create a separate project implementation unit. In recognition of the slow project implementation progress during the early years, the organisation and management system was decentralised in 1996 to the district offices. Four cost centres were established in the field through which funds necessary for the project were transferred, This gave the FD greater ownership of the project and played an important role in the institutional development of the department. In practice, the project was absorbed into the organizational stnmcture of the FD, with the real planning of works being carried out by the district offices in agreement with the relevant line Deputy Director- who was also responsible for supervision and implementation monitoring. In addition to decentralization to district offices, a more efficient system of workplan preparation and implementation was introduced and remains in effect after project completion. These new capacities of the FD represent a solid basis for planning of extension support services in the next phase of the Lake Malawi Environimental Management Project. - [In every project involving sustainable management natural resources, the balance between environmental and social objectives should always be negotiated with care. In the FDP experience, an over cautious approach adopted to fisheries production was allowed to take precedence over the development objectives of the project. In hindsight, there should have been a more reflective discussion with the FD about the options available for development interventions under the restnictured project. Although it is clear that the project needed to be restructured because of its complexity and the underestimated capacity constraints within the FD, IDA's decision to cancel the entire production component, instead of being open to discuss with GOM its possible refonnulation, closed the door for implementation of essential production activities intended to achieve development impacts within the artisanal and semi-commercial fisheries. Since research in 1995 had confinned that the over-fishing was localized in extent, with adequate management of the distribution of fishing effort the initial production targets could have remained realistic and within the limits of sustainability. * An important lesson stems from the project's insufficient attention given to the role of the private sector in delivering the services that the FD was asked to divest. The requirement that the Govemment of Malawi divested itself of some commercial undertakings that were servicing the fishing community, without ensuring the continued provision of such services by the private sector to the sub-sector, was a serious shortcoming of the project. FD's withdrawal from boatbuilding and other services did have the intended impact of freeing up the time of field professionals for other areas of responsibility. But the overall impact on the FD itself was strongly negative, since field officers saw development of the industry as a prime responsibility and the dismantling of the services they had worked to establish as incomprehensible. It confused field staff and soured working relationships between the FD and the industry. However, it would be wrong to give the impression that FD's withdrawal from service provision has left a total vacuum, as there are several examples of private sector development (though serving mainly non-fishery businesses) that probably would not have taken place had the FD continued its operations. - Another lesson stems from the lack of a clearly defined civil works programme which resulted in a great deal of time and effort devoted to the civil works component at the cost of the mainstream fishery production activities. It is thus advisable to complete the necessary agreement on tendering procedures and to finalise designs during appraisal or before effectiveness in order to avoid lengthy discussions of these items during implementation, as experienced under the project. Progress in the civil works programme was extremely delayed, with no activity on the ground during the projectfs first five years. 24  The amount of work involved in acquiring land, seeking designs and supervising the tendering process was prodigious, and successful implementation required coordinated and timely actions on the part of several players. All of these activities were left to the FD, which had no competence in this field. Finally, unreliable contractors occasioned several delays, in some instances necessitating a repeat tendering process and, at the end. two contentious project extensions for completion of the work. - Bank supervision of the project was negatively affected by the frequent changes of IDA Task Managers during the first half of the life of the project. Thlese changes of Task Managers did not provide IDA with the necessary focus on the project or establishing a solid relationship with the FD. In contrast, IDA maintained the same Task Manager from the project restructuring in 1996 until completion. This continuity greatly helped the project as the flow of thought was not interrupted, and the working relationship with the FD improved markedly. - A positive lesson stems from the use of bilateral donor support which can provide flexibility to the implementation of specific components or project activities during unexpected situations. Both NDF and ICEIDA contributed substantially to the implementation of the project, particularly in support of commercial fishing and fishery research, respectively. They participated in joint supervision missions and endeavoured to respond positively to the emerging needs of the project in order to ensure its continued implementation. - The decision to hold a joint review meeting of theICRs for the FDP and the GEF/SADC Lake Malawi/Nyasa Biodiversity Conservation Project, given the significant leaming potential of the two operations and the obvious linkages to the Lake Malawi Environmental Management Project (currently under preparation), should be regarded as a good practice. In addition, the preparation of an intensive leaming ICR (ILI) for the Fisheries Development Project and the required Beneficiary Impact Assessment Study enabled a more complete assessment of the impact of the project on groups of beneficiaries at the village level, including shortfalls that affected the project's impact as perceived by these beneficiaries, and beneficiaries' perceptions of the corrective measures that would have been undertaken to address these problems (see Annexes 8 and 9). The beneficiary study complements the assessment conducted by the FAO ICR mission team while providing baseline data and field-level perspectives from selected stakeholder communities to help evaluate FDP support to fisheries extension. This data set, together with the lessons leamt from the GEF/SADC Lake Malawi/Nyasa Biodiversity Conservation Project ICR, provide critical inputs and baseline information against which the intended development and environmental impact of the next project will be evaluated. 9. Partner Comments (a) Borrower/implementing agency: From 19 to 30 June 2000, the ICR mission team (led by FAO) met will all beneficiary institutions and key govemment agencies that were involved in the implementation of the project. The mission worked mainly with the Fisheries Department (FD) and met with the Principal Secretary of the Ministry of Natural Resources and Environmental Affairs. Discussions were also held with the Ministry of Finance, the Malawi Rural Finance Company (MRFC), the Icelandic International Development Agency (ICEIDA) and the GTZ assisted National Aquatic Resources Management Programme. Based on the mission discussions and the Government inputs into the ICR process, the FAO-led ICR 25  team incorporated the comments and inputs into the final Aide Memoire that is presented in Annex 7a. The Government also participated at the ICR Review Meeting in Washington on December 19, 2000, via a video-conferencing link to the Malawi Country Office in Lilongwe. The Government's Evaluation Report is enclosed as Annex 7d of this ICR. (b) Cofinanciers: An executive summary of the evaluation of the components financed by the Nordic Development Fund (NDF) and ICEIDA is provided in Annex 7c. (c) Other partners (NGOs/private sector,): No additional feedback on the ICR document had been received from other partners by December 29, 2000. 10. Additional Information In order to provide more detailed information on the implementation of the project than is possible in the main text of the ICR, a note on the Background of the Fisheries Subsector and Project Performance has been appended to this document as Annex 7b. 26  Annex 1. Key Performance Indicators/Log Frame Matrix Outcome/Impact Indicators: Indicators Projected in SAR Projected in Informal Actual/Latest Estimate Restructuring, August 1993 Increased fish production Production from No target specified Net annual increase of 900 to improve nutrition and artisanal, semi- ton from MALDECO protein supply commercial and commercial fisheries to increase by 7,500 tonnes annually Production from Deleted No action aquaculture to increase: no target specified As per SAR No action Boat builders and ancillary No action fisheries industries added as target groups, but no target outcomes specified Increased off-farm Women to be assisted in General economic development 259 credit groups (c. 4,000 employment and income fish processing and trade in fishing communities added. women and men) were to alleviate poverty in in three pilot districts but no target outcomes specified formed and linked to an lakeshore communities, through training and independent lending especially women credit: no quantified institution: 123 groups targets set received small loans, Conservation of the mostly for trade in crops, resource base of processed food, clothing, Malawi's water bodies fish and other commodities Improved institutional Interventions to achieve Interventions to achieve New fisheries policy capacity for fisheries management objectives management objectives formulated and Fisheries policy formulation, indicated but impacts / emphasised in more detail, but Conservation and research, planning, outcomes not specified impacts / outcomes not Management Act passed monitoring and control specified by Parliament. Annual resource management plans initiated in 1995. Annual workplans for management-orientated extension and research initiated 1996 27  Annex 1. Output Indicators: Institution Building Projected in SAR Projected in Informal Actual/Latest Component Restructuring, Estimate August 1993 Civil works programme Fisheries Headquarters As per SAR Civil works Building programme completed in full 4 new Fisheries Offices 2 Fisheries Offices rehabilitated Fisheries Training Centre rehabilitated and expanded Staff houses constructed or rehabilitated Provision of vehicles, boats Not detailed As per SAR 41 vehicles, 57 and equipment motorcycles, 25 boats, computer equipment and office fumiture procured Reorganization of the 6 additional senior posts As per SAR 10 new senior posts Fisheries Department created; FD reorganised created; number of according to an agreed professional / senior human resources technical posts development plan increased by 25; number of junior technical posts declined by 79. New structure had increased efficiency Withdrawal from all Withdrawal from Already completed All commercial commercial activities with commercial fishing and activities except for the exception of Mpwepwe the provision of Mpwepwe Boatyard Boatyard commercial support relinquished by mid services by March 1992 1993 Planning, Monitoring and Targets not specified As per SAR PMEU established Evaluation cell created 1993 but became visibly effective 1996 Staff training program Targets not specified As per SAR An extensive staff implemented training programme completed. Studies conducted Study on attitudes of As per SAR Not completed fishermen to lake conservation measures Study on the role of As per SAR Completed women in fish processing satisfactorily in 1995 and marketing 28  Annex 1. Research Component Infrastructure development FRU in Monkey Bay As per SAR Civil works programme rehabilitated and completed in full expanded Provision of research vessel 17 m. 300 hp research 17m. 380hip Vessel commissioned vessel research vessel 1993 Vessel captain and As per SAR Captain and engineer engineer 6 m/m each provided as planned, local officers trained Technical Assistance Fisheries Biologist 12 As per SAR Fisheries Biologist m/m provided for 48 monthls. Senior and intennediate staff training provided. Assessments in S., C. and As per SAR Assessments conducted N. Lake Malawi as planned 1993-2000. Demersal fish stock Development of demersal As per SAR Annual management assessments trawl management plans developed from recommendations 1995. Lake resource conservation Development of a Lake resource 3 Fishenres Research studies monitoring system to conservation Officers seconded to the enable rapid response to studies deferred in SADC/GEF Lake lake resource hazards view of probable Malawi Biodiversity duplication of the Conservation Project, ecological studies under which ecological proposed and studies were conducted conducted under as planned. the SADC/GEF Lake Malawi Biodiversity Conservation Project Aquaculture pilot Outputs related to the Aquaculture No action. programme integration of fish programme deleted farming into existing cropping systems; development of fish ponds between reservoirs and irrigated fields, and integrated fish-with-rice production Fish Production Component Artisanal fishery 50 additional chirimila As per SAR 1 No action development. via support to units boat building and a credit 200 existing dugout As per SAR No action scheme canoes replaced with small planked boats 5 additional units Additional units No action introduced deleted 5 units replaced As per SAR No action 29  ANNEX 1 Semi-commercial pair trawl Trawl nets, marine As per SAR No action fisherv development, via engines, spare parts and support to boat building ice plants provided under an import support scheme New 15m 240hp multipurpose fishing vessel provided Cold stores, ice plant, freezing plants provided Commercial fishery 2 refrigerated trucks As per SAR Commercial fishery development, via import provided investment programme support and commercial proceeded as planned, lending with minor changes. Women's program, via Training of female fish As per SAR No action extension, training and a processors credit scheme Provision of insulated containers Provision of smoking kilns Credit scheme to support the Credit to support the As per SAR but No action artisanal fishing industry, supply of boats, engines boat building added fish processing and trade and fishing gear to the as a specific artisanal fishery, and additional category equipment and working capital to fish processors/traders l Infrastructure Component ; Improvement of Fisheries Construction of jetties As per SAR Actions completed in Department infrastructure and rehabilitation of fill marine workshops at Monkey Bay and Nkhata Bay Support to boat building Rehabilitation and re- Privatisation of Boatvard privatised 1999 equipping of Mpwepwe Mpwepwe Boatyard Boatyard Minor grant for hand Rehabilitation and tools. No rehabilitation re-equipping of or other re-equipping. Mpwepwe Boatyard unable to Boatyard after service the semi- privatisation. via comnmercial fishery import support and a commercial loan Privatisation of No action Salima Boatvard Improvement of beach Construction of 46km of Sub-component No action access beach access roads deleted 30  Annex 2a. Project Costs and Financing Project Cost by Component (in US$ million equivalent) - Appraisal Actual/Latest Percentage of Project Cost By.Component LEstimate Estimate Appraisal US$ million US$ million: Institution Building 6.20 12.07 195 Infrastructure 0.80 0.04 5 Research _ 1.61 2.27 141 Production 4.64 0.72 16 Contingencies 2.25 Total Baseline Cost 15.50 15.10 Physical Contingencies 97 Total Project Costs 15.50 15.10 Total Financing Required 15.50 15.10 Annex 2b. Project Costs by Procurement Arrangements (in US$ million equivalent) Expenditure Appraisal Estimates Actual/latest Estimate Percentage of Appraisal Categories -_______ IC: NCB Prudwft Tot ICB NCB Prudent Total ICB NCB Prudent: Tot B ___ Shopping alt Shopping _ __ _ lopping al 1. Civil - 7.59 7.59 Works 2. Goods & 0.45 3.29 4.27 Equip. _ _ 0.53 3. Training & 1.81 1.81 Consult. 4. Operating 1.43 1.43 Costs Total _ _ __ 0.53 8.04 6.53 15.1 _ --The column concerning appraisal estimates is left blank because there were no project cost estimates by procurement arrangements at appraisal Anex 2c Pn4ect Finaiang by Cnanent (m US$ rrilion equvaIt) irnfrstn'tre 004 a c-ir rg t = - -nw-1d. Resdwch _c | | -| oa -| 013| 213 - 213 Pro - n 351 - -- 0721 -t I- 4- t- ToW 1 71| 11 1.11 7 88j 231 281, 213 nLa| 9| 2991 2B:4213| na| 11 BenelidariesiniLrded M&IDECO US$ 0.6nrillion andtargetgroLp US$0 5 rrillion AlpDetcoqed eticn, detils of M&LDEC(Ys cxihJlonAerLern*tavailable, ard X cmmrnetomerurngrS etalrvtg pvasrot inplenEYed. 3 1  Economic Costs and Benefits Cost Benefit Analysis Annex 3 (MK '000) B0,,:"-.B.. .. .. . ; og0 : : . .. .. . ............................ _'0300 '' ~ ~ ~ cjoicA1~ FX,,. e*aBysiI Xr~~~Aprl Lates ppr"^a Lat Best-a3: 148,247 Co rt~~ ~ 43112 105'135 IF~RJNP (%) ~ 20 ____ ModelSv4 trR ~30 70 40 F:~I~IR~{W) ._ -n 4 t.-40 1/ Model A = Replacement of dug-out canoe with one-plank canoe. Model B = Replacement of two dug-out canoes with two-plank canoes. Model C = New Chirimila units consisting of one-plank boat propelled with an engine. Model D = New pairtrawl units consisting of two-plank boats propelled with two engines. Model E = Rehabilitation of existing pair trawl units. 32  Annex 4. Bank Inputs (a) Missions: Stage of Project Cycle - :No. -of Persons:and Specialty Performance Rating- (e.g. 2 Economists, I FMS, etc.) Month/Year Count Specialty Implementation Developaient .:___-__-_._____:_____: Progress Objective Identification/Preparation Not Avail. Not Available Appraisal/Negotiation Not Avail. Not available Supervision 10/91 3 FA, FI, T S S 01-02/93 2 E, FI S S 07-08/93 4 C, E, Fl, Fl U S 06/94 5 E, Fl, N, R U S 02-03/95 5 E, Fl, Fl, N, R S S 03/96 2 E, Fl U U 09/96 3 A, FI, R U U 12/96 2 A,R U U 03/97 A, FI, 1, N, R S S 07/97 3 A, I, R S S 12/97 3 A, I, R S S 07-08/98 4 A, E. , R S S 03/99 6 A, E, FI, I, N, R S S 09-10/99 6 A, E, E, I,N,R S S 03-04/00 5 A, E, I, N, R S S ICR 06/2000 2 FI, Fl U U 1/ A=Agriculturist; C= Civil Engineer; E=Economist; FA=Financial Analyst: FI=Fisheries Specialist; 1= ICEIDA Co-Lender; M=Management; N =NDF Co-Lender; R=Research Specialist; T= Training Specialist Task Managers: 1. Okidegbe = Financial Analyst; 2. Rozell = Economist; 3. Pohland = Economist; 4. Fulton = Economist; 5. Kumar = Agriculturist Note: The disconnect between the later supervision missions (since 1997) IP and DO ratings and the final project outcome can be explained by the obvious difficulty of providing a single rating for a restructured project, in which the first five years were characterised by poor performance and the last three years by better performance, albeit on a restructured project with significantly narrower development objectives and impact. When one rating is made based on appraisal expectations, however, IP and DO ratings are unsatisfactory because of the removal of the artisanal and semi-commercial investment proposals from which the bulk of the expected project production and development impact were to originate. Only the projected support to the commercial fisheries was kept, which was a minor element in terms of project's expected production. Consequently, the expected incremental annual production by the project of 7,500 tons was never achieved. Furthermore, the expected improvements in the people's diet, food 33  security and off-farm incomes that were to have resulted from the incremenital fish output by the project were never achieved. Still, the above SPN IP and DO ratings accurately reflect the satisfactory progress and relative impact of the remaining institution building and research activities that were implemented following project restructuring (see Section 4.5 for a discuission of the proj ect's institutional development impact). (b) Staff: Actual/]Latest Estimate Stage of Project Cycle No. Staff weeks US$ (,000) Identification/Preparation 44.4 34 Appraisal/Negotiation 51.7 133.8 Supervision 185.92 466.4 ICR 10 20 Total 292.02 654.2 Annex 5. Ratings for Achievement of Objectives/Outputs of Components (H=High, SU=Substantial, M=Modest, N=Neglig ble, NA=Not Applicable) Ratings Macro policies NA Sector Policies M Physical SU Financial M Institutional Development M Environmental M Social Poverty Reduction N Gender N NA Private sec tor development M Public sector management SU 34  Annex 6. Ratings for Bank and Borrower Performance (HS=Highly Satisfactory, S=Satisfactory, U=Unsatisfactory, HU=Highly Unsatisfactory) 6.1 Bank perfoormance Rating Lending U Supervision U Overall U 6.2 Borrower performance Rating Preparation U Government implementation performance U Implementation agency performance U Overall U Note: It is difficult to give one rating for a nine-year project, in which the first five years were characterised by poor performance and the last three years by better performance, albeit on a restructured project with significantly narrower objectives. If one rating should be made, however, the overall Bank and Borrower performance should be rated as unsatisfactory. This should be considered a fair rating even after taking into account the fact that project performance improved considerably following the 1996 restructuring and MTR, after it was significantly down-scaled, with the production component being dropped. In the assessment of the ICR team, there is no doubt that the project needed to be restructured because of its complexity and the underestimated capacity constraints within the FD, which made the original objectives impossible to maintain. However, IDA's decision to cancel the entire production component, instead of being open to discuss with GOM its possible reformulation, while also failing to address the impacts of the unprepared privatizations on fish production, were major shortcomings at project restructuring. After 1996 the performance of both the Bank and the Borrower improved significantly. The Borrower's performance improved markedly, enabling the restructured project, with its much reduced scope, to be implemented smoothly as agreed in the latest project document (FDPD). IDA also maintained the same Task Manager since 1996, and this benefited the project immensely as implementation of the restructured project proceeded according to plans. But this improved Bank and Borrower performance could not be considered to compensate for the unsatisfactory performance of the first five years which largely determined the final project outcome. 35  Annex 7. List of Supporting Documents Annex 7a: The final ICR Mission Aide Memoire prepared by FAO and discussed with the Govemment in July, 2000, is presented in Annex 7a Annex 7b: More detailed information on the project is given in Anlex 7b (below) entitled "Background of the Fisheries Sub-sector and Project Implementation Experience." Figures and Tables are also enclosed below. Annex 7c: The Executive Summary of the Evaluation of the Components Financed by the Nordic Development Fund (NDF) and ICEIDA is presented in Annex 7c. Annex 7d: The Government's Evaluation report is included below as Annex 7d. 36  Annex 7A ICR AIDE-MEMOIRE Table of Contents A. INTRODUCTION ............................................2 B. BACKGROUND ............................................2 C. IMPLEMENTATION EXPERIENCE AND RESULTS ............................................ 3 Overview ............................................3 Implementation of Specific Components ............................................5 Institution Building ............................................5 Research ............................................6 Fish Production ............................................7 Infrastructure ............................................7 Organization and Management ............................................7 Project Impact and Sustainability ............................................8 Borrower Performance ............................................9 IDA Performance ............................................9 Co-financiers ........................................... 10 D. LESSONS LEARNED ............................................10  Annex 7A A. INTRODUCTION 1. A mission1 from the FAO/World Bank Cooperative programme (CP) visited Malawi from 19 to 30 June 2000 to prepare an Implementation Completion Report (ICR) for the Fisheries Development Project (FDP). The mission worked mainly with the Fisheries Department (FD) and met with the Principal Secretary of the Ministry of Natural Resources and Environmental Affairs (MNNREA). Discussions were also held with the Ministry of Finance, the Malawi Rural Finance Company (MNRFC), the Icelandic International Development Agency (ICEIDA) and the GTZ assisted National Aquatic Resources Management Programme. Field visits were undertaken in four of the seven project districts, i.e. Karonga, Nkhotakota, Salima and Mangochi. The mission met with Beach Village Committees (BVCs), artisanal and semi-commercial fisheries, women's and men's credit clubs and MALDECO Fisheries Ltd, which is involved in the commercial fishery of Lake Malawi. 2. The mission would like to express its appreciation to the govemment officials, particularly the staff of FD, for the support and assistance provided during the ICR preparation exercise. The mission's findings and conclusion are presented below. The mission would like to acknowledge the FD's contribution to this aide memoire. B. BACKGROUND 3. The project stemmed from the Fisheries Sub-Sector Study carried out in 1988 by the Govemment of Malawi (GOM) with the assistance of the International Development Agency (IDA). This study identified the main issues facing the fisheries sub-sector and offered ideas and recommendations on how to tackle the main problems. Based on these, a World Bank identification mission visited Malawi and suggested that IDA provide assistance to the fisheries sub-sector. Following IDA's agreement to support Malawi's fisheries, the GOM established a task force to take the lead in preparing an investment project. With the active assistance of FAO, a preparation report was finalised in November/December 1989, and was appraised by IDA in June 1990. 4. As appraised, the project's main objective was to realise the potential contribution of fisheries to the economy through: (a) increasing fish production to improve nutrition and protein supply; (b) generating additional off-farm employment and income to reduce poverty among the rural population, particularly women; (c) conserving the natural resources base of Malawi's water bodies and preventing environmental degradation; and (d) improving institutional capacity in the fisheries sub-sector. 5. The project, to be implemented over seven years, was estimated to cost US$15.5 million, and its components comprised: (i) institution building aimed at strengthening FD to enable it to concentrate on regulatory functions, staff training in policy promotion, extension, monitoring and evaluation and enforcement of fishery laws; (ii) strengthening FD's research capacity in fish stock assessment, conservation and management programmes and in the development of fish farming models to integrate aquaculture into farming systems; (iii) rehabilitation and development of existing capture fisheries by supporting artisanal fishermen, semi-commercial and commercial fishing Comprising Pietros Kidane (Mission Leader, Economist, CP), and Antony G. Seymour (Fisheries Specialist, Consultant). 2  Annex 7A activities, and establishing a pilot programme to increase the involvement of women in fish processing and marketing; and (iv) rehabilitation of infrastructure consisting of up-grading and building access roads, jetties for fish landing, and other shore-based facilities. 6. In addition to IDA and GOM, the project was to be co-financed by the Nordic Development Fund (NDF) and ICEIDA. Overall project responsibility was to be vested with the Ministry of Forestry and Natural Resources (MFNR), presently MNREA, and FD would implement the project in collaboration with other relevant institutions such as the Ministry of Agriculture, the Office of the President and Cabinet (OPC), the Ministry of Finance and the Reserve Bank of Malawi. A Project Coordination Committee (PCC) was to be established, and its members would be those institutions most involved in project implementation. The PCC would be chaired by the Principal Secretary of MFNR, the FD providing secretariat, and was expected to play an advisory role and ensure conformity to government policies, resolve inter-agency problems and facilitate support from other institutions. 7. Within FD, its Chief (now Director) was to have overall responsibility for project execution, and would be supported by an intemationally recruited Financial Management Advisor and two senior staff that would be assigned to work full time for the project. This team would form the Project Coordination Unit (PCU) to deal with the day-to-day project implementation and communication with other institutions and donors. Under the Regional Fisheries Officers, the District Fisheries Officers (DFOs) were to take charge of data collection, extension services, women's issues and market survey, including quality control. The heads of the Fisheries Research Unit in Monkey Bay and the Fisheries Training Centre in Mpwepwe would, respectively, be responsible for the implementation of the research component and training programme. 8. Following the fulfilment of all the conditions for the project to become operational, the project was declared effective on 4 September 1991. After two extensions of six months each, the project is scheduled to close on June 30, 2000. At completion, total project costs are estimated at US$ 15.10 million, and overall disbursement at US$ 12.80 million (US$ 7.87 million IDA'; US$ 2.80 million NDF, and US$ 2.13 million ICEIDA). C. IMPLEMENTATION EXPERIENCE AND RESULTS Overview 9. The project had a slow start up and poor performance during the first five years, and was considered a problem project. It was restructured twice, in 1993 and in 1996 following the mid-term review (MTR). It was only after the restructuring of 1996 and the preparation by the FD of a project plan for 1996-1999 that the project turned around and improved its implementation performance. This prompted IDA to extend the project closing date by one year, reversing the pre-1996 inclination of closing the project prematurely. The project that was finally implemented was significantly different from what had been appraised, particularly in relation to the production support activities, which were largely excluded, except the support to commercial fisheries, through MALDECO. 10. According to the supervision mission of January/February 1993, the project was judged as progressing satisfactorily given that the PCU had been established, an action plan for project Excluding the funds that are deposited in the special account. 3  Annex 7A execution prepared and positive steps taken towards the reorganisation of the FD. In addition, the research vessel financed under ICEIDA had been delivered, and plans for handing over the fishing vessel to MALDECO by mid 1993 finalised. However, the same supervision mission also reported some difficulties caused by shortage of operational funds, delay in the recruitment of essential staff, and a lack of progress in the implementation of the production component. Hence it recommended steps that needed to be taken to remove these constraints. 11. Rather suddenly, during the following supervision mission of July/August 1993, a totally different picture was depicted and calls for immediate restructuring of the project were made. This was triggered by recent studies and research works that indicated over fishing in the near shore areas and southem arms of Lake Malawi due to increasing population pressure and failure to regulate the semi-commercial fishery. Widespread increase in fishing effort and use of small meshed nets in both the artisanal and semi-commercial fisheries were reported. Therefore, a restructuring was recommended in order to shift the focus of the project and the FD away from increased production and towards fishery resource management. Accordingly, it was proposed to intensify the institutional building components, particularly in the areas of extension and enforcement of fishery laws and fish stock assessment; while at the same time reducing the productive components by restricting the supply of boats, and omitting activities related to infrastructural development and rural access roads. By mid-1994, the FD was operating on the basis of the restructuring proposals, although the amendment to the Development Credit Agreement (DCA) requested by GOM was sent to IDA only in December 1994. These amendments were never reflected on the DCA, most likely because it was assumed that the restructuring did not bring in new goals but rather shed some of the production- related objectives. 12. Although operating de facto under the 1993 proposed restructuring programme, the performance of FD did not improve and little progress was made in project implementation. The FD was not able to implement the plans it prepared, especially in extension and enforcement, mainly for want of experienced field staff. The MTR reported that excessive centralisation of financial control and other resources in FD headquarters had resulted in implementation difficulties in the field. These were further compounded by the failure of the Ministry of Works and Supplies and IDA to agree promptly on tendering procedures in civil works contracts. The MTR considered the overall performance as unsatisfactory and recommended, among other things, not to take action on credit withdrawal applications submitted by FD. The MTR also recommended a number of specific actions to be taken by the FD within a defined timeframe. These included ministerial review and approval of a national policy statement on fisheries and a resource management plan; intemal review of FD with a view to producing a dated six month action plan that would improve its performance; preparation of a Fisheries Development Project Document (FDPD) to outline FD's strategy for implementing the remainder of the project until the closing date, and preparation of a management and financial system for the proj ect. 13. The findings and recommendations of the MTR have made an impact on FD as it has come up with a clear strategy for addressing the major constraints and with detailed proposals for implementing the project. By May 1996 all the tasks indicated in the MTR were completed and were assembled in the FDPD for the period 1996-1999. This document, along with submission of a six- month action plan and annual workplan became the basis for the supervision mission of September 1996 to agree with GOM on steps to proceed with project implementation. Subsequently, the mission recommended IDA to remove its reservations on the project, and to give FD once more a chance to tum it around. This was accorded by IDA, and project execution, along the lines of the 4  Annex 7A FPDD, has continued to be carried out to date. At completion most of the implementation plans have been realised. However, as indicated earlier, the project that was finally implemented was significantly reduced in its scope from what had been appraised, particularly in relation to the production support activities, which were largely excluded. 14. Flaws in project design at appraisal can in part explain the difficulties encountered by the project. Firstly, even though acknowledging the weak institutional capacity of the FD, a complex project was proposed, including areas in which the Department had little experience, such as major infrastructure construction programmes, including access roads, credit and a women's programme. Secondly, project appraisal failed to finalise the tendering arrangements for the civil works contracts, which were cause for long delays, and equally neglected designing a suitable system for the utilisation of the credit line and detailing the requirements for project monitoring, evaluation and reporting. Thirdly, the appraisal document failed to specify performance indicators and project targets, leaving the task of performance monitoring unnecessarily vague and subjective. Finally, the requirement that GOM divest itself of some commercial undertakings that were servicing the fishing community, without ensuring the continued provision of such services to the sub-sector, has proven to be a serious misjudgement (para. 17). Implementation of Specific Components Institution Building 15. The project has resolved FD's constraints in shortage of office space, staff housing and other facilities through the contraction of civil works. The Fisheries Training Centre - renamed the Malawi College of Fisheries (MCF) - and the Fisheries Research Unit (FRU) were expanded and renovated and a new FD headquarters established next to MNREA. Fisheries offices in the districts have been provided with new office space, housing facilities and vehicles, including motorcycles for the extension staff and boats for enforcement staff. A total of 41 vehicles, 57 motorcycles and 25 boats of different types and sizes were supplied to FD under the project. The provision of all these facilities has improved the operational efficiency of the restructured department. The restructuring of FD under the project included delegation of authority to the regional and district offices, including the transfer of financial management directly to four field cost centres from FD. The district focus adopted by the FD, together with improved mobility and training, made the activities of the field staff more efficient. Extension services to fishermen and enforcement of fishery regulations were stepped up. The extension staff intensified their meetings and dialogue with fishing communities and induced them to establish BVCs in order to be involved in the sustainable management of the resource. The enforcement officers also started to work closely with BVCs, the by-laws of which are upheld and supported by the district staff. Participation of local communities in fish resource management has been boosted by the promulgation, in 1997, of a new Fisheries Conservation and Management Act, in the finalisation of which the project was instrumental. This Act has given the local communities the necessary legal backing in undertaking the management of fishery resources through their own organisations, such as BVCs. A total of 267 BVCs has been formed under the project', but only 34 have been given the training required to make them effective. 'The GTZ financed National Aquatic Resources Management Programme and its predecessor, the Malawvi-German Fisheries and Aquaculture Development Project, provided the leading impetus for BVC formation in the 5  Annex 7A 16. In addition, the project has strengthened the FD in its capacity to carry out policy analysis, the development of extension services and resource management through research. To this end, considerable input has been made by the project in training the professional and senior staff. So far the project has provided 19 staff with degree and diploma course training, and a considerable number were given short-term training on various topics. Different types of technical level training were also provided to the extension staff, mainly at the MCF which was substantially strengthened under the project in terms of construction of classrooms, dormitories, offices and other facilities and the provision of vehicles and equipment. The college's curricula have been revised to suit the current and future needs of the fisheries sector, and its teaching staff upgraded through training. At project completion, the college has emerged as a modem institution capable of responding to the demand for training by fishermen and staff. The facilities provided appear to exceed the national capacity requirement, but FD explains that the MCF will also service other SADC countries. 17. In order to enable the FD to concentrate on regulatory and other management-related functions, the project proposed the sale of or withdrawal from all its commercial activities. Therefore the FD sold its commercial establishments, i.e. two boatyards and two ice plants, and discontinued the repair of outboard motors and the retail supply of fuel, fishing gear and marine engine spare parts. Unfortunately the privatization of these establishments and activities did not guarantee the continued supply of services to the fishing communities. One boatyard and an ice plant were dismantled. The second boatyard was eventually sold in 1998 after a prolonged period of very low activity, and has since failed to regain momentum because of the new management's lack of financial resources. There is no other functioning boatyard in Malawi, neither has there been a commercial service agent for the engines of the semi-commercial pair trawl fleet since the FD discontinued this function at an early stage in project implementation. The supply of fishing gear remains problematic in those areas remote from the Blantyre Netting Company Ltd. (BNC) in southern Malawi, and in the Northem Region the main source of nets and twines is alleged to be illegal imports from Tanzania. Research 18. One of the project components that took off the ground early on was research, thanks to the expeditious supply of a research vessel by ICEIDA with the necessary support in technical assistance and training. By 1993, the FRU had initiated, with project support, a new programme of demersal fish stock estimation in Lake Malawi, the outputs of which were to have important implications for management and future development. Since then, thirteen surveys have been conducted by FRU, and some of these have been decisive in indicating the availability of a large and previously unexploited demersal stock capable of sustaining an annual harvest of around 33,000 tonnes. Information on the location of these stocks has been published by FRU. Through the collection and analysis of catch and effort data, FRU provided in 1999 a comprehensive set of statistics on fish stocks in Malawi waters. Most of the studies carried out by research have been used for developing fishery resource management in the country. 19. In addition to the research vessel Ndunduma, another vessel (Nkhatamaran) was rehabilitated by the project. The research centre in Monkey Bay was renovated, rehabilitated and extended. New offices, staff housing and a jetty were constructed, and a number of staff members were given degree course training. A Fisheries Research Fund was established under the project to Mangochi, Zomba, and, to a lesser extent, Salima and Nkliotakota Districts, vith logistic and institutional support from the FDP. 6  Annex 7A finance research activities. It is expected that this fund will be sustained largely by revenues from the sale of fish caught by the research vessels (para. 29). Fish Production 20. The expected production support to artisanal and semi-commercial fisheries was not provided by the project following the worries of over-fishing expressed in 1993, and was formally excluded from the project in 1996. The lack of the promised support was deeply resented by fishermen, especially by the semi-commercial fishermen who are now reduced to marginal fishers. Of the 20 trawlers that were operating during the start of the project, only four are now operational, the remainder being grounded for lack of serviceable boats, engines or spare parts. Yet, research has since confimied that the over-fishing noted in 1993 was localised in extent, and that with adequate management of the distribution of fishing effort the original production targets remained realistic and within the limits of sustainability. 21. The only direct support to increased production was given to MALDECO in the form of a fishing vessel, a blast freezer, cold rooms, an ice making machine, a cold truck and fishing gear. To some extent, the project has provided support to improved quality of fish through training in fish processing by district extension services, which also demonstrated the construction of fish smoking kilns. The extension service also facilitated the formation and training of credit groups to receive loans for income generating activities, including fish processing and marketing and the purchase of fishing nets. In dialogue with the MRFC, the extension service established 259 groups, and 117 of these have already received loans, although the amount of money is small - below MK10,000 - with a maximum term of one year. Infrastructure 22. The project rehabilitated the shore-based facilities at Monkey Bay and Nkhata Bay. At Monkey Bay, a jetty and a slipway were constructed and the research vessels are the main beneficiaries of these facilities. At Nkhata Bay, a pontoon has been constructed, again for use by FD vessels. The Mpwepwe Boatyard was not improved as envisaged since it was targeted for privatization. Hence, only minor wood working machines were provided by the project. Contrary to initial project plans, the construction of access roads was excluded, probably due to the reduced focus on increased production. Organization and Management 23. Initially, the project followed a centralised project organisation in which the PCU under the Chief of FD controlled the project. All planning, budgeting and financial management was carried out at FD headquarters. This top-down project management style led to the concentration of both professional staff and expenditures at headquarters, at the expense of field operations. Although the PCC had been established and had met on a few occasions during the first three years of the project, its effectiveness faded as time passed by, and it failed to play its expected role in resolving project difficulties. Realising the slow implementation progress, the organisation and management system was decentralised in 1996 to the district offices. Four cost centres were established in the field through which funds necessary for the proj ect were transferred. The PCU was downgraded to an administrative unit in which one coordinator, one staff from the planning and monitoring unit and 7  Annex 7A a senior accountant assisted the Director in overseeing project activities. The real planning of works was carried out by the district offices in agreement with the relevant line Deputy Director, who is also responsible for supervision and implementation monitoring. In practice, the project has been absorbed into the organisational structure of the FD. 24. Together with decentralisation, a new system of workplan preparation and implementation was introduced and remains in effect to the present. This entails the assessment of the type of support needed by the communities and the preparation, on the basis of the needs assessed, of a work programme with clear targets and a timeframe for their realisation. Implementation of the workplan is monitored by the district office by comparing output with planned targets. Although this system is still in its infancy, it has proven to be a useful instrument for staff to think on what needs to be done within their jurisdictions by sounding the fishing communities, and preparing realistic programmes that are easily monitorable. The field workplans and achievements are regularly reported to FD headquarters. 25. The FD did not produce progress reports until the mid-term review, and it is difficult to make an in depth study in project implementation before that period. In addition, the filing system of FD was not up to the mark, conceming the project, and it is difficult sometimes to obtain important reports on the project or produced by the project. Fortunately, the project accounts have been kept well since the beginning and are readily available for any type of analysis. Since 1996, the FD has been preparing six-monthly reports on the status of the proj ect, and these have facilitated the understanding of the exact situation of project implementation, enabling the identification of problems and the planning of remedial and follow on actions. Project Impact and Sustainability 26. The project's main achievements have been in strengthening the fisheries service in the country and in instilling fish resource management with the participation of local communities. At completion, the FD has emerged as a more focused and streamlined institution. Its capability in policy analysis, planning and monitoring of fishing activities and fish stocks has been clearly improved. In particular, fish stock assessment and monitoring capabilities have been markedly improved and this has enhanced FD's planning in fisheries development and management. 27. Following the decentralisation of fishery service management in the field and the provision of staff training and improved facilities and working environment, the district staff were better motivated to work with fishermen by providing extension services to BVCs and training them on how to manage the resource on a sustainable basis. This message has been well received by fishermen and they appear to feel a sense of ownership of the adjacent waters and are keen to protect and use them judiciously and with care. Many BVCs have developed their own by-laws regarding the use of the resource. An important achievement in this regard has been the excellent relationship established between the fishing communities and the district staff. 28. Sustainability of the project activities will depend on the level of resources that GOM will be willing to provide to operate and maintain what has been put in place by the project. Presently, most of the operating costs that are necessary for paying incremental staff, running of field offices and vehicles as well as field allowances were financed through project funds. After the project closes these funds must come from GOM's recurrent budget, if project activities are to continue in servicing the fishing community at current levels, and are to facilitate the efficient and 8  Annex 7A sustainable utilisation of the fish resources. 29. Many FD staff are apprehensive that they may not be able to secure the funds required to operate project activities at full scale. If this is the case, part of the project undertakings would not be sustainable. According to the Ministry of Finance, the GOM will increase the FD budget at project closure, but it cannot guarantee that all the funds needed to sustain current project activities will be provided. In order to obviate budgetary difficulties, some FD units are thinking of generating their own incomes in order to cover part or all of their operational costs. To this end, FRU is proposing to increase its fishing activities and sell more fish to the public, and MCF intends to earn more income through increased fishing and renting its facilities for seminars, workshops and conferences. These initiatives, which are commendable, are encouraged by GOM and will certainly improve the sustainability of these units. However, a careful balance must be struck between undertaking income generating operations and the mainstream activities of supporting the fishery sub-sector by carrying out meaningful research and appropriate training to staff and fishermen. Borrower Performance 30. The Borrower's performance in the early years of the project was characterised by slow compliance with the agreements reached in implementing the proj ect. Notably, GOM was not able to provide the necessary budgetary funds and recruitment and appointment of staff on time, and this has held back progress in project implementation. Another example of poor performance has been the excessive delay (4 - 5 years) in tendering the civil works contracts by the Ministry of Works and Supplies, leading to significant increase of costs. While the shortage of funds is explained by the overall government budgetary control policy, the undue delay in finalising civil work contracts can hardly be justified. The PCC did not work out as anticipated and did little to resolve the problems faced by the project with regard to the availability of funds, staff recruitment and the impasse in civil works. During this period the Borrower's performance was less than satisfactory. Since 1996, the Borrower's performance improved markedly, enabling the restructured project, with its much reduced scope, to be implemented smoothly as agreed in the latest project document (FDPD). The relationship between the Borrower and IDA also became cordial, contrary to the pre-1996 period, where tensions were prevalent when the project was facing implementation difficulties and when suspension or closure of the project were frequently talked about. Perception of lack of project ownership by FD started to emerge within the staff, with some indications of waning interest in the project. The frequent change of IDA Task Managers during the earlier years of the project (para. 32) also exacerbated this situation. IDA Performance 31. The root cause of the project's implementation problems is to be found in the complex design of the project. A weak FD should have been given a simple and straightforward project to execute. Altematively, if it had to implement a complex one, provisions should have been made to enhance its capabilities through a well thought out technical assistance support, until it was sufficiently strengthened through staff training. Neither of these was done at appraisal. Additionally, project appraisal failed to detail some components and activities (para. 14), as well as to provide clear guidelines for the monitoring and reporting of project activities. All these deficiencies have contributed to slowing project implementation, rendering IDA's performance less than satisfactory. 9  Annex 7A 32. IDA carried out a total of 15 supervision missions, including the MTR. In the earlier years, supervision was not constant. For instance, after the first mission in 1991, no supervision was carried out in 1992, and only one each were undertaken in 1994 and 1995. Since 1996, however, two supervision missions per year were carried out consistently by IDA. The first five years were undertaken by five different Task Managers, and following a line of thought on improved project implementation was continuously interrupted. The perception of the level of project difficulties were not the same among the different Task Managers, and emphasis on how to proceed varied, creating tension between FD and Task Managers. Since 1996, a clear programme of work and monitoring system was agreed between FD and IDA, and the supervision missions based their review on what had been planned under the six-month programme. Before leaving the FD, the IDA supervision agreed on workplans for the following six months. Following its promise, IDA maintained the same Task Manager since 1996, and this has benefited the project immensely as implementation of the re- structured project proceeded according to plans. Co-financiers 33. Both NDF and ICEIDA have contributed substantially to the implementation of the project, particularly in support of commercial fishing and fishery research, respectively. They have participated in many supervision missions and have given their input while firming up follow-up plans. They have endeavoured to respond positively to the emerging needs of the project in order to ensure its success. For instance, ICEIDA has spent about US$ 2.13 million against its initial commitment of US$ 1.0 million. Recently NDF has agreed to finance an IDA cost category, training of staff, because this category was closed by IDA in December 1999. In general, the co-financing arrangement has worked out well. D. LESSONS LEARNED 34. The main lessons that could be leamed from the implementation experience of the project are shown below, and are subject to further elaboration during the finalisation of the ICR. - Once cognisance is made of the institutional weakness of the designated implementation agency, project design should seek to come up with simple proposals that aim at strengthening the institution concemed and implementing activities that pertain to its field of specialisation and experience. Otherwise underperformance will be inevitable as shown with FD during the first five years of the project. * One of the reasons why FD faced implementation difficulties initially was the lack of a clear implementation plan for each component, attaching specific staff responsibilities for their realisation. To the contrary, the main reason for its improved performance since 1996 was better organisation and the preparation ofworkplans that are workable and monitorable, with clear staff responsibilities for each task. These observations imply that commencing implementation in the absence of detailed plan is a waste of time. * Disposing of Govemment commercial establishments without ensuring that the services they provided to fishermen were guaranteed can result in a significant weakening of the small-scale fisheries and undoing FD's past efforts in putting these services in place. Many fishermen have suffered from the closure of FD services and establishments, since some of these were subsequently dismantled by their new owners. 10  Annex 7A A great deal of time and effort has been devoted to the civil works component under the project at the cost of the mainstream fishery components. It is advisable to complete the necessary agreement on tendering procedures and to finalise designs during appraisal in order to avoid lengthy discussions of these items during implementation as experienced under the project. * Reservation on increased fish production are likely to linger even after conclusive scientific results show the availability of fish stocks that can besustainably fished in specific areas. This is probably due to excessive prudence in resource conservation, combined with reticence to adopt complex fish resource management systems for the over-fished and under-fished areas. The project's continued neglect of the fish production component and removal of the line of credit in 1996, even though the issue of over-fishing had been clarified, can be explained in this light. * The performance of a weaker institution can be markedly improved if it is provided with an adequate working system and environment and if equipped with the necessary facilities, mobility and up-grading of staff skills. This is proven by the field staff of FD who have tumed into better- motivated and active group after the project's restructured approach starting in 1996. * Because of the flexibility they offer, co-financing with bilateral donors can resolve funding problems for unexpected situations or items. Some financing of IDA cost categories that were closed in 1999 have been taken up by NDF when these were judged necessary in 2000, indicating the benefit of such co-financing in responding to emerging needs expeditiously. * Frequent changes of IDA Task Managers do not provide IDA with the necessary focus on the proj ect or establishing a solid relationship with the Borrower agencies. On the contrary, if staff continuity is maintained, IDA's focus on the project becomes consistent and issues become clearer, ensuring a lasting relationship with the Borrower and co-financiers as has been proven by IDA's last Task Team Leader. 11   ANNEX 7B FISHERIES BACKGROUND AND PROJECT IMPLEMENTATION TABLE OF CONTENTS A. INTRODUCTION .............................................3 B. THE FISHERIES SUB-SECTOR .......3......................................3 C. PROJECT CONCEPT AND DESIGN .............................................4 Project Objectives and Components ............................................4 Project Start-up ..............................................4 Project Restructuring in 1993 ..............................................5 Mid-Term Review .............................................S5 Project Restructuring in 1996 ..............................................6 D. PROJECT IMPLEMENTATION .......6......................................6 The Institution Building Component ..............................................6 General ..............................................6 Elimination of Commercial Functions ..............................................6 Institutional and Human Resource Development .............................................. 7 Training .............................................8 Provision of Infrastructure and Equipment ..............................................8 GOM Support Through the Budget Process ......................9.......................9 The Research Component ..............................................9 The Production Component ............................................. 10 Expansion of the Artisanal Fishery into Offshore Waters .................................... 10 Fleet Renewal in the ("semi-commercial") Pair Trawl Fishery .......................... 11 Reinvestment in MALDECO Fisheries Ltd ........................................................ 11 Credit ........................................................ 12 The Women's Programme ........................................................ 13 Aquaculture ........................................................ 13 The Infrastructure Component ........................................................ 14 Mpwepwe and Salima Boatyards ........................................................ 14 Access Roads and Jetties ........................................................ 15 E. PROJECT ORGANISATION AND MANAGEMENT . .................................. 15 F. PROJECT IMPACT ........................................................ 16 G. SUSTAINABILlTY ........................................................ 17  ANNEX 7B FIGURES 1. Fish Production in Malawi, 1976-96 2. Artisanal Fish Production from Lake Malawi and All Other Waters, 1976-96 3. Fisheries Department Recurrent Budgets 1990/91 - 2000/01 4. MALDECO Fisheries Ltd.: financial performance 1993-99 TABLES 1. Fish Production in Malawi, 1976-98 2. Fisheries Department Proposed and Actual Staff Complement 1996-2000 3. FDP Civil Works Programme, 1996-2000 4. Fishing and financial performance of MALDECO Fisheries Ltd., 1993-99 5. Status of FD/MRFC Mudzi Credit Groups, April 2000 2  ANNEX 7B FISHERIES BACKGROUND AND PROJECT IMPLEMENTATION A. INTRODUCTION 1. This Annex is intended to provide more detailed information on the implementation of the Fisheries Development Project (FDP) and on its bio-economic and institutional context than is possible in the main text of the report. It draws on information from many sources, but mainly from the Fisheries Department's published reports and Fisheries Bulletins, and from the series of IDA supervision reports which chart the progress of the project. Liberal use was also made of the wider literature on Malawi's fisheries, and as many people were consulted as possible during the recent Implementation Completion Report (ICR) mission. The FDP spanned a period in which Malawi underwent profound political change and adopted new policy principles for natural resources management. To some extent the project was caught up in the uncertainties of change, which slowed progress, and to some extent it has benefited from the more progressive policies and new institutional opportunities of the democratic era. Other opportunities have been missed though, and every attempt has been made to point these out for the benefit of future IDA investments in this important sub-sector. B. THE FISHERIES SUB-SECITOR 2. The fishery sub-sector in Malawi is large and extremely diverse, contributing 2-3% to GDP and employing in 1999 some 49,000 persons directly and an estimated 200,000 indirectly in fish processing, distribution and associated trades. Fish is an important element in the national diet, probably accounting for 60-70% of animal protein intake. The annual catch has varied widely between 30,000 and 80,000 tonnes (para. 3), with landings in most years between 50,000 and 60,000 tonnes. Most of the catch (more than 90% during the 1990s) is landed by artisanal fishermen deploying a range of simple gears from dugout canoes and small planked boats. Mechanised fishing is currently confined to the southem part of Lake Malawi, and is carried out by one medium-sized company, which owns steel fishing vessels up to 17 m and 380 hp, and a number of smaller enterprises including pair-trawl units based on wooden boats of 7.5 m and 30hp. 3. The catch is made up of a great variety of fish species, dominated by the endemic Lake Malawi cichlids (of which 1,000 species may exist) but also including a typical Zambezi fauna. Most of the commercially important species are cichlids and cyprinids of small size and subject to considerable fluctuations in abundance, although the largerbagrid and clariid catfishes also make up 10-15% of annual landings. Figure 1 and Table I provide an overview of fishery production from 1976 to 1996. The principal features of this period are: * National fish production increased irregularly to a peak of 77,000 tonnes in 1990. After this it declined steeply to 32,000 tonnes in 1995, recovering somewhat in 1996; * Most of this variability is attributable to the artisanal fisheries of waters other than L. Malawi, which declined massively from a peak of 46,000 tonnes in 1990 to a trough in 1995 at only 7,000 tonnes (Figure 2). L. Chilwa alone lost 23,000 tonnes when it dried up in 1995-96 in part of a long-term cycle of changing evaporation! inflow balance. Other 3  ANNEX 7B changes include the collapse of the L. Malombe fisheries as a result of over-fishing, and a downturn in production from the Lower Shire floodplains resulting from reduced effort and the effects of water hyacinth and changes in the flooding regime.