Sierra Leone - Second Northern Integrated Agricultural Development Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No. 9245 PROJECT COMPLETION REPORT SIERRA LEONE NORTHERN INTEGRATED AGRICULTURAL DEVELOPMENT PROJECT II (CREDIT 1128-SL) DECEMBER 28, 1990 Agriculture Operations Division Country Department IV Africa Regional Office This document has a restricted distribution and may be used by recipients only in the performarce of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  Currency Equivalents Name of Currency - Leone 1980 US$1.00 = 0.9528 1981 US$1.00 = 0.8639 1982 US$1.00 = 0.8078 1983 US$1.00 = 0.5958 1984 US$1.00 = 0.3984 1985 US$1.00 = 0.2114 1986 US$1.00 = 0.1191 Abbreviations ACRE - Adaptive Crop Research and Extension ADB - African Development Bank ADF - African Development Fund ASSP - Agriculture Sector Support Project BADEA- Banque Arabe pour le Develop-ent Economique en Afrique CARE - Cooperative American Relief Ererywhere EIADP- Eastern Integrated Agricultural Development Project FAO - Food and Agriculture Organization FFB - Fresh Fruit Bunch GOSL - Government of Sierra Leone IADP - Integrated Agricultural Development Project IDA - International Development Association IFAD - International Fund for Agricultural Development IPAM - Institute of Public Administration and Management MAF - Ministry of Agriculture and Forestry MANR - Ministry of Agriculture and Natural Resources MDEP - Ministry of Development and Economic Planning mOW - Ministry of Works MNR - Ministry of Natural Resources MPP - Model Production Programme NAPMU- Northern Area Project Management Unit NIADP- Northern Integrated Agricultural Development Project NUC - Njala University College PEMSU- Planning, Evaluation, and Monitoring Services Unit PESU - Project Evaluation and Services Unit PFP - Policy Framework Paper RMWA - Regional Mission in Western Africa (World Bank Office) RRRS - Rokupr Rice Research Station SAR - Staff Appraisal Report SLPMB- Sierra Leone Produce Marketing Board SMP - Seed Multiplication Project SMU - Seed Multiplication Unit T&V - Training and Visit UNDP - United Nazions Development Programme VCDU - Village Cattle Development Units FISCAL YEAR July 1 - June 30  FOR OmCAL USE ONLY THE WORLD SANK Washington. D.C. 20433 U.S.A. Ofrc of Dirctot.Cnral Opmatrw Evakiateon December 28, 1990 MEMORANDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT SUBJECT: Project Completion Report on Sierra Leone Northern Integrated Agricultural Development Proiect II (Credit 1128-SL) Attached, for information, is a copy of a report entitled "Project Completion Report on Sierra Leone - Northern Integrated Agricultural Development Project II (Credit 1128-SL)" prepared by the Africa Regional Office. No audit of this project has been made by the Operations Evaluation Department at this time. Attachment This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  FOR OFFICIAL USE ONLY PROJECT COMPLETION REPORT SIERRA LEONE NORTHERN INTEGRATED AGRICULTURAL DEVELOPMENT PROJECT II (Credit 1128-SL) TABLE OF CONTENTS Page No. Preface ..................................... i Evaluation Summary ...................................... PART I PROJECT REVIEW FROM BANK'S PERSPECTIVF. Project Identity ............................. 1 Background .1 Project Objectives and Description. 4 Project Design and Organization. 5 Project Implementation. 8 Project Results .11 Project Sustainability .13 Bank Performance .14 Borrower Performance ........... ............. 15 Project Relationship .16 Consulting Services .16 Project Documentation and Daa .17 PART II. PROJECT REVIEW FROM BORROWER'S PERSPECTIVE .18 PART III. STATISTICAL INFORMATION Related Bank Loans and/or Credits .19 Project Timetable .20 Disbursements .20 Project Implementation .21 Project Costs and Financing .22 Project Impact. Direct Benefits .23 Economic Impact .24 Financial Impact .25 Studies .26 Status of Covenants .26 Use of Bank Resources ........................ Staff Inputs .27 Missions .27 Map of the Project Area This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  i PROJECT COMPLETION REPORT SIERRA LEONE NORTHERN INTEGRATED AGRICULTURAL DEVELOPMENT PROJECT II (CREDIT-1128 SL) PREFACE This is a Project Completion deport (PCR) for the Northern Integrated Agricultural Development Project II (NIADP II) for Sierra Leone, for which Credit 1128-SL in the amount of USS 8.5 million was approved on April 7, 1981. The NIADP II Credit closed on June 30, 1987. Disbursements under the Credit were completed on May 31, 1988 with full recovery of the Special Account. The unused balance of USS 5.43 million was cancelled on July 7, 1988. Since the suspension of disbursements to Sierra Leone in early 1987, no further supervision work has been undertaken. No PCR for the project has been prepared by the Borrower. This PCR was prepared by the Agriculture Operations Division of the Africa Region. Parts I and III were sent to the Borrower for comment with a request to provide a Review of the project from the Borrower's point of view, but none was received. Preparation of the PCR, started in the third quarter of FY90, is based, inter alia, on the Staff Appraisal Report, Development Credit Agreement, available supervision reports, correspondence between the Bank and the Borrower, and internal Bank memoranda. Disbursements to Sierra Leone remain suspended.  ill PROJECT COMPLETION REPORT SIERRA LEONE NORTHERN INTEGRATED AGRICULTURAL DEVELOPMENT PROJECT II (Credit-1128 SL) EVALUATION SUMMARY 1. Obiectives 1.01 Following Sierra Leone's declining traditiolLal exports, a rapidly rising budget deficit and food import bill, and wide income disparities, the Government around the mid 1970s proclaimed agriculture to be a priority sector in the country's economic development. New overall objectives for the sector included an ambitious 5.4 percent real growth, self-sufficiency in rice, improved nutrition, and increased production of major export crops. The selected vehicle for the development and growth of the agriculture sector was the integrated agricultural development project (IADP) approach. In Sierra Leone this approach tended to be narrow and area specific with little emphasis given to reducing sector inefficiency. Consequently, the IADPs focussed more on production activity and essential infrastructure and less on the policy and incentive framework within agriculture. 1.02 Prolect Obiectives. The main objective of the Northern Integrated Agricu'.tural Development Project Phase II (NIADP II) was to increase food production and farm incomes by expanding assistance to all farmers that had been assisted under the Phase I project in the Northern Province, and to introduce fruit tree cultivation and promote the use of livestock. No new institutions were to be set up and no change in existing cropping patterns was envisaged. Increases in productivity were to be obtained through the use of high yielding varieties, higher input intensity as prescribed in the recommended technical packages, and considerably stronger extension surport. NIADP II was a multipurpose project that had the primary objectives of increasing, over a five year period, food production and farm incomes begun under the Phase I project of the Northern Province. The development of upland and swamp rice, groundnut, and maize was to be extended to reach farmers within an additional area of 8,500 km2, and fruit tree cultivation and livestock improvement introduced. The project also aimed at strengthening sector institutions, and constructing village wells, feeder roads and chiefdom tracks. 1/ 2. Implementation Experience 2.01 Project implementation overall was dominated by Sierra Leone's serious financial crisis. Inadequate counterpart funding, mismanagement of procurement, and suspension of disbursements had a profound bearing on the 1/ Chiefdom tracks were crude access roads constructed at the time that local governments were headed by village/community Chiefs.  iv inability of the project to fully meet its objectives and physical targets. Among project specific factors that hampered performance were: shortages of seedlings for oil palm. citrus, coffee and rice; weak credit and input supply systems; absence of a strong adaptive research component; generally weak technical packages, in some cases dle to insufficient testing at the farm level, and in others due to the state of technical development of the packages themselves; and shortag.s of fuel. 3. Results 3.01 The overall project outcome has been unsatisfactory with the project unable to meet the SAR's physical targets. There were large shortfalls in several areas of crop development, and in the construction programes for roads and wells. Progress in institutional development and training was generally poor. The contribution of agricultural research, particularly adaptive research, was negl'3ible. Agricultural extension in comparison showed significant improvemer. and this was reflected in the relatively large numbers of farmers who were made aware of extension messages. The translation of these extension messages into higher yields and incomes was inhibited by constraints on input supply, availability of credit, and poor mobility. 3.02 More specifically, except for swamp rice, all other food crop SAR targets, and those for tree crops were not met. While livestock health training was retained, the livestock development component had to be discontinued. The targets for village wells and roads were alsc not fulfilled and had to be scaled down. As was the case in the Ea.tern Integrated Agricultural Development Project (EIADP III), the implementation of which overlapped with that of NIADP II, a suggested and more costly new design of village wells in divergence with SAR specifications impeded progress of the village wells program. In the case of feeder roads, while new construction was justified, there was also the equally strong requirement that roads already constructed be adequately maintained. Because of the growing constrain- on financial resources and the poor state and neglect of roads already constructed, a trade-off had to be made between continued additional construction of feeder roads on the one hand and their proper maintenance on the other. In the case of agricultural services, the quality of extension staff was improved, and some progress made in the area of commercial accounting training. Research remained weak, and so did training for middle and higher level management staff. 4. Sustainability 4.01 The project closed almost three years ago. No PCR mission took place on completion and no PCR was prepared by the Borrower. Except for a recent visit by staff to selected project sites in connection with the Policy Framework Paper (PFP), and some communication with Government officials and donors, there has been no feedback on project developments per se. Based on these accounts, there are indications that the Bank- supported IADPs in Sierra Leone have been regarded as a failure with little impact on agricultural production. It is, therefore, unlikely that the project is sustainable. The integration of IADP staff into the regional MANR structures has proved difficult and, in spite of an encouraging start made toward this end as project completion approached, the anticipated  v integration has not been achieved. The high visibility of enclave projects such as the ADPs and the power that came to be associated with them hampered the planned integration. Resentment by existing MANR staff toward former IADP staff who had received relatively good training and enjoyed better remuneration than themselves prevented integration. Furthermore, the MANR is reported to be saddled with a budget that is used almost in its entirety to financing staff salaries which are about seven months in arrear. 4.02 Sierra Leone's financial and economic si.uation remains critical. The prospects for recovery of project activities in the near future, thus, remain bleak. Moreso, when the macro and sector framework continues to be burdened with distortions that would impinge on sector efficiency. The project could have contributed considerably to improved incentives within the sector through ensuring that pricing policies were determined in relation to border pricing. 5. Fitdings and Lessons Learned 5.01 The main finding of the NIADP II project experience is that the country's macro and financial crisis was a major obstacle to project success. It is unrealistic to expect projects, and certainly complex integrated agricultural development projects, to succeed in a macroeconomic environment that is unstable and saddled with distortions in macro prices, particularly the exchange rate. At the project level, the main lesson is to design projects that are simple to implement. Less ambitious projects embodying relatively simple designs in a country and sector characterized by serious macro and financial constraints, and aggravated by limited absorptive capacity and poor or non-existing infrastructure are likely to be far more successful in achieving their objectives and contributing to longer term development and growth. This is particularly true in instances where the record of macroeconomic management has been poor and continues to deteriorate. 5.02 Other lessons for relevant projects include: (i) the importance of maintaining a policy and incentive framework that is conducive to project success and sector efficiency; (ii) the need to take i:lto account not only project related risks, but also those that may have political implications, and which are, in nature, macro and sectoral; (iii) the benefit of maximizing the number of staff trained through courses and on- the-job training so that training continues to be consolidated and disseminated long after the project has closed; (iva progressively reduced dependence of indigenous staff on teaching and demon!tration by experts financed under the project; (v) the importance of preparing initial assessments for the demand and supply of essential inputs, and contingency plans should the situation alter, particularly with respect to shortfalls in supply; (vi) the merit in adapting proven design and implementation techniques to the country's socioeconomic conditions where the latter are insufficiently studied and understood. It should also be noted that in spite of the relatively large numbers of farmers that oere exposed to extension messages, these messages could not be translated into higher productivity and incomes. The primary reason for this was the difficulty in obtaining inputs and credit on the one hand, and poor mobility and infrastructure on the other. Efficient extension without the adequate and  vi timely supply of inputs is clearly insufficient to bring about sustained growth and higher incomes. A long-term commitment to higher agricultural productivity and sustainability also hinges on the benefit to the producer at farm-or mill-gate. In NIADP II, while farmers and women displayed keenness to participate in the project, the logistics of input supply and credit dampened the fruits of their effort. Future projects should ensure that farmers and women do not become disheartened, and continue to engage in the project activities designed to promote their economic well-being.  PROJECT COMPLETION REPORT SIERRA LEONE NORTHERN INTEGRATED AGRICULTURAL DEVELOPMENT PROJECT II CRLDIT 1128-SL PART I: PROJECT REVIEW FROM BANK'S PERSPECTIVE 1. Project Identity Name Northern Integrated Agricultural Development Project II Credit Number 1128-SL RVP Unit Africa Region Country Sierra Leone Sector Agriculture 2. Background 2.01 Sierra Leone with a total area of 72,000 km2 has an estimated population of 3.9 million, as of mid-1989, growing at an annual average rate of 2.3 percent. Its per capita income in 1988/89 was estimated at SDR 160, fifty percent lower in real terms compared to 1979/80. Th_ country has good soil and climatic conditions and is also rich in minerals. It is strongly dualistic with a small modern mining sector which is a major source of government revenue and export earnings, and a large traditional agriculture sector that accounts for one third of GDP and employs 75 percent of the labor force. Agricultural productivity and yields are low. Fertilizer is believed to be used by only three percent of the farmers, a fifth of whom are said to depend on primitive tools and implements for cultivation. The resource base, including fishing, if well developed and managed, can lessen Sierra Leone's dependence on diamond mining and export, promote economic diversification, enhance agricultural growth, and increase employment and rural incomes. 2.02 As of June 1982, the Government had no official development strategy, although agriculture, education and health were emphasized as priority sectors in economic policy statements. A clear focus on the agriculture sector in its overall strategy for development was for the first time explicitly outlined in the 1974675-78/79 National Development Plan. 2/ Its main objectives were accelerated economic growth, balanced regional development, and equitable income distribution. Given serious financial and institutional constraints the Plan proved to be overly ambitious. Specific sector development objectives had called for a 5.4 percent growth in agriculture, self sufficiency in rice, the staple crop, cultivated by 80 percent of the farmers, improved nutrition, and increased 2J Compared to the late 1960s, agriculture became a priority sector with its share in the development budget increasing from four percent in 1969 to 32 percent in 1978.  2 production of major export crops. 3/ The expansion of agricultural production through integrated agricultural development projects (IADPs) with focus on strengthened sector support services, improved research and planning, and a more effective Ministry of Agriculture and Natural Resources (MANR) was to be the hallmark of the sector strategy. 4/ It was expected that, by the mid-1980s, 30 percent .. the country would be covered by IADPs. 2.03 Past Performance. Agriculture's potential and performance over the years hap been curtailed by poor incentives, excessive taxation of export crops, low budgetary allocations, and weak institutions and riral infrastructure. Sierra Leone's serious and prolonged financial constraints were largely responsible for the sector's progressive deterioration during the last fifteen years. This contrasts sharply with the impressive economic growth during the first decade of the country's independence starting in 1962. In fact, no economic growth took place in Sierra Leone in the 1980s. The decade was also characterized by high inflation which peaked at 170 percent in 1986/87. With sharp oil price increases in the 1970s, an overvalued Leone, a rapidly rising import bill, steady depletion of exploitable gold, diamond, and iron reserves, Sierra Leone's terms of trade declined and were further aggravated by reduced prices of coffee and cocoa. 5/ But the more serious and underlying cause of the economic decline was weak economic management and lack of budgetary control that progressively worst:.ed with time. Government spending on rice and petroleum subsidies accounted for 10 percent of recurrent expenditures in 1973/74 and absorbed more than one-third of the revenue increase. For fertilizer, and chemicals and sprayers for tree crops, subsidies were of the order of 50 percent and 100 percent, respectively making them a further imposition on the already difficult budgetary situation. 2.04 Efforts at Economic Recovery. Attempts have been made, on several occasions, to resolve, together with the Bank and Fund, Sierra Leone's dire financial situation. A number of financial arrangements proposed with the Fund in the last ten years include: a stand-by arrangement in 1979, an extended facility in 1981, a drawing under the 3/ Having been self-sufficient in rice until the early 1950s, Sierra Leone became a net importer with rice imports averaging 50,000 tons annually for the next twenty years. 4/ A lack of coordination among sector institutions, for example, MAF, MNR, SLPMB, etc. had made the development and operation of consistent sector policies difficult. 5/ Production and productivity of coffee and cocoa, Sierra Leone's two major export crops had been adversely affected not only by an overvalued Leone, but also by pricing and marketing policies detrimental to their progress. First, the Sierra Leone Produce Marketing Board (SLPMB) held the monopoly on the marketing and export of both crops, and second, the producer price for both was set at unremunerative levels, not to exceed 65 percent of the FOB price for coffee and 59 percent for cocoa. Each crop was subject to an export tax of 30 percent, and export proceeds from both were used to finance the import and distribution of rice.  3 compensatory financing facility in 1983, a stand-by arrangement in 1984, and an arrangement under the structural adjustmernt facility together with a stand-by arrangement in 1986. By and large, most of these arrangements were soon rendered inoperative due to inadequate implementation. Since 1986, yet more attempts have been made to draw up a workable adjustment program, but the country has been unable to fulfill its financial obligations under these. At end 1980, the country was behind in its debt service payments on eight Bank loans and IDA credits in the amount of USS 1.03 million, and since early 1987 accumulated arrears to the Bank led to its being placed on a nonaccrual status. Starting early 1989, a series of adjustment measures was implemented. The exchange rate was further depreciated and private trade in coffee and cocoa, and in the import of rice, sanctioned. Efforts to raise revenue and collection were stepped up, and redundant civil service staff laid-off. In mid-December 1989, except for trade in gold and diamonds, import and export licenses were abolished, and fiscal and monetary policy strengthened. As of end May 1990, Sierra Leone's overdue financial obligations to the Fund were approximately SDR 73 million. Efforts towards a satisfactory resolution of the country's financial and macro predicament are still being actively pursued. 2.05 In retrospect, the National Development Plan and its policy context proved to be basically ineffective in correcting key macro distortions because of the weak monetary, fiscal and exchange rate policies. These policies had focussed on resource mobilization to meet short term needs of the public sector rather than on Sierra Leone's long term development and sustained growth. The net effect of the country's economic policies was to emphasize consumption to the detriment of the trade and current account balances which continued to be negative and increasingly became a cause for concern. The timing of the Plan coincided with a rapid erosion of foreign exchange earnings and deteriorating fiscal performance. With lack of policy direction and budgetary control, it, therefore, had little succ-;s in restructuring the macro policy framework in favor of development oriented activities. 2.06 In line with the adjustment programme starting in 1986, efforts were made by GOSL to establish effective linkages between project, sector and macro policy objectives. Export crop prices were increased, subsidies reduced, public finances and public payrolls better controlled, and a limited float of the Leone introduced. These measures also produced two disturbing results. First, with government salaries reduced to less than subsistence levels because of devaluation and higher prices, the existing low motivation and morale in the civil service became worse. Second, discontent among the freshly laid-off staff soared, wlo, in the absence of viable alternatives, increasingly turned to agriculture for employment. The impact of these measures on agricultural production was mixed. While more remunerative export crop prices and the currency depreciation led to improved incentives for tree crop development, th2 overall positive effect of these was dampened by low morale and poor rural infrastructure. The magnitude of the marketable surplus that might otherwise have emerged was accordingly reduced and so, too, were farmers' actual and potential incomes.  4 3. Project Objectives and Description 3.01 Project Objectives. The primary objectives of NIADP II were to increase, over a five year period, food production and farm incomes by expanding assistance to small farmers begun under a Phase I project in Sierra Leone's Northern Province (Credit 568/Loan 1138-SL). It was envisaged that approximately 13,500 farm families would directly benefit from the project, that is, 40 percent of rural families in the combined area covered by Phases I and II. Taking note also of the anticipated improvement in general agricultural services in the area, and progress in the roads and wells components that had occurred in the past, and was envisaged to improve further as a result of the project initiative, almost 50 percent of the rural families under the two phases taken together were expected to benefit. 3.02 Compared with the Eastern and Southern Provinces, climate and soil conditions in the Northern Province are generally less favorable to agricultural development. Whereas conditions in the former are conducive to tree crop farming and the development of a rural cash economy, the north, with longer dry seasons can support mainly subsistence farming and shifting cultivation. An occasional surplus of rice, groundnut and citrus, when it occurs, is usually sold for cash. Farm size averages between two to three hectares. By and large, livestock is not integrated into the farming systems. Rural incomes are far below the national average and, within agriculture, incomes in the north are among the lowest. The project area has a network of unclassified roads, mainly chiefdom tracks and minor roads, that has been neglected over the years. The Northern Phase II project was the Bank's fourth project for Sierra Leone's agricultural development. 3.03 Project Description. The project was to cover a total area of 13,400 km2 and 22 chiefdoms. While its main focus was on new project areas, it was also to assist in supporting farmers not reached in Phase I by making available to them credit and extension facilities. In all, about 34,000 farmers or 55 percent of area farmers were to benefit. Project components and sub-components were to consist of the following: (a) crop development: seasonal and medium-term loans would be extended to farmers at 20 percent and 15 percent annual interest rates respectively for financing the hiring of labor and essential agricultural inputs for: (i) developing 2,000 ha of swamps for rice; (ii) improving water control and crop husbandry of partially exploited swamp land, also primarily for rice production; (iii) enhancing yields on about 20,000 ha of uplands cultivated with rice, groundnut, cassava and maize and other crops; and (iv) establishing about 750 ha of fruit trees on uplands; (b) civil works: first, the construction of: (i) 80 km of all-weather penetration roads, and rehabilitation of about 300 km of chiefdom tracks; (ii) 400 village wells; (iii) five houses for project management staff; (iv) ten farm service centers and sub-centers; and, second, rehabilitation and expansion of: (i) the existing workshop and provision of related equipment; and (ii) livestock facilities;  5 (c) extension services for: (i) the crop development program, and (ii) livestock development with provision for the supply of drugs; (d) studies: (i) evaluation of the program for construction of village wells; (ii) evaluation of training programs and courses on management training given to the Northern Area Project Management Unit (NAPMU) staff by area training institutions; (iii) reports on field trials to determine the potential for the development of tree crops, including cashew, coffee and cocoa; (iv) a research program to determine the appropriate tree species for use as fuel wood; and (v) a baseline socio- economic survey and the subsequent planned adoption of a monitoring and evaluation program for the project; (e) support services: (i) establishing a monitoring and evaluation unit for supervising the project's activities with provision for vehicles and equipment; (ii) a stronger Institute of Public Administration and Management (IPAM) for providing effective management training for agricultural development; (iii) a health education program for the use and maintenance of the village wells to be constructed under the project; and (iv) preparation and adoption of a soil conservation and land use program. 4. Project Design and Organization 4.01 During the project period, Sierra Leone was engaged in implementing a number of projects aimed at rationalizing activities in agriculture. These projects were integrated projects with an area specific rather than sectoral approach. The emphasis was more on enhancing production operations and essential infrastructure and less on tackling economic efficiency issues at the sector level. The design of the NIADP II project was aimed at increasing, as did Phase I, the living standards, agricultural production and cash incomes of as many farmers as possible in the Northern Province. Phase I had been the first major agricultural development project in the northern part of Sierra Leone. Given the historical imbalance in regional development between the progressive Eastern region and the relatively less developed North, the project was of great significance. As between appraisal and preparation, there was no marked difference in project design. However, so as not to over-complicate or burden the management function, it had been decided at appraisal to exclude from the design certain components, in spite of their importance, such as farm storage, forestry, and health. 4.02 In its consistency with Phase I under which upland and swamp rice had been successfully promoted, and the T&V system successfully introduced, the project design was appropriate. Efforts to incorporate lessons from the earlier phase were not entirely satisfactory. Lessons from Phase I that were most apparent were related to the ineffectiveness of credit operations, poor availability of seed, and weak capability of indigenous staff, which was especially pronounced at the middle and senior management levels. The project design had been unable to recognize, to the extent nccessary, the actual degree of weakness of the institutions concerned even at the end of Phase I. Thus, less than adequate arrangement in the project design for the basic and longer term development of the relevant institutions was made. A further shortcoming in design pertained to the inclusion of the tree crop component without prior study of the  6 potential for tree crop cultivation in an area with little tradition for long term cash crops. This taxed the project's scarce funds and placed a burden on the research and extension staff which lacked training in tree crop cultivation. The same argument holds for the project's promotion of groundnut in a geographical area not considered conducive to groundnut cultivation. Another aspect of project design that was noticeably weak was agricultural research, particularly adaptive research. The provision for field trials and analysis was almost totally confit *d to studying the potential for cultivation of cashew, cocoa and coffee, and appropriate species for fuelwood. There was no systematic provision for research into food crops, for which the project area was suited, for which the need was crucial, and for which the scope for productivity increases was large. 4.03 The above not withstanding, there were positive and innovative aspects of project design that deserve mention. Soil erosion in the uplands due to population pressure and decreased fallow and the likely adverse effects on crop yields nad become a major cause for concern. The new agronomy unit was to take steps to halt this trend through the introduction of conservation measures such as contour bunding, permanent cropping, the planting of fruit trees, and the setting aside of grazing areas. Far-reaching implications for resource depletion and soil erosion were also evident from the rapidly rising demand for fuelwood for cooking, construction, and tobacco curing. The project, accordingly, recommended investigation into appropriate tree crop species for these purposes. 4.04 To tackle problems experienced by cattle-owning crop farmers in stock management, within a predominantly cropping system, the project provided for the education of cattle-owners in livestock management through communally operated Village Cattle Development Units (VCDUs). A renewed effort was also to be made to interest farmers in the use of draft oxen through training in livestock, and field trials with mixed cropping. This was important for northern Sierra Leone, a region not only where labor was believed to be in relatively short supply, but, also, where farmers were predominantly risk-averse, and driven to a large extent, by subsistence needs rather than profit-maximization. It was recognized, then, that the farmer's choice of crop combinations and farm development programs would not always be consistent with the profit motive. This and other socio- economic considerations, including the fact of labor supply being a key limiting factor in the project area, were, to the extent possible, taken iato account in the design of the project's crop development programs. 6/ 4.05 When viewed in terms of Sierra Leone's budgetary constraint and limited absorptive capacity, the scope and scale of the project could have been guided by greater pragmatism. There was already an awareness on the part of many decision-makers of the inevitable burden on agriculture sector finances and institutions of implementing a number of IADPs at about the 6/ The paradox of labor being in short supply in conditions of apparently considerable open unemployment and underemployment continues. The problem is posed on the one hand of genuine seasonal and/or regional shortages, and, on the other, of farm wages being insufficient to bring forth the required supply over and above the reservation wage.  7 same time. 7/ Bearing this in mind, evcn though the project did not opt for an outright and ambitious expansion of area and activities, it could have further reinforced initiatives that had met with little success in the earlier phase. Fewer components with sharper focus on strengthening sector institutions, input and credit supply, accompanied by reforms in pricing and marketing policy, would have had greater impact. 4.06 The project was identified during the supervision of NIADP I in January 1979. It was a follow-up to the Second IADP which had included a Phase I project for the Northern Province. Encouraged by the progress made under Phase I, the project continued to pursue opportunities, believed to be numerous, for further development of the northern region. NIADP I was considered, then, to provide a good basis for the development of a second phase. Project preparation was carried out by FAO/World Bank Cooperative Program, together with the Government of Sierra Leone, between October 1979 and February 1980, and was presented to the World Bank Group for financing in March 1980. Cofinancing from the ADB, ADF and IFAD was supplemented by a grant from UNDP. Staff who had worked in the previous phase also contributed to the preparation of Phase II. In view of Sierra Leone's acute financial situation, the project was well-timed. Appraisal took place before the close of NIADP I on June 30, 1981 inclusive of a six-month extension of the closing date. 4.07 Except for the planned major reorganization of MAF and the integration of IADPs into the MAF, no major changes in organization and management were envisaged for Phase II. To the extent that the initial burden associated with the setting up and operation of a new organizational structure was reduced, implementation was enhanced. The respective responsibilities and roles of the agencies and staff were to remain basically intact. Day to day management was to be provided by the Project Management Unit headed by an internationally recruited Project Manager. Overall project responsibility was to rest with the MAF. The terms of reference were clearly defined for all project entities. 4.08 More careful attention to project design in certain specific areas could have contributed to greater success in meeting overall project objectives. First, the technical design for the layouts of selected swamps was weak and this impeded the full achievement of swamp rice targets. Second, the project's promotion of groundnut, in an area known to be climatically inhospitable to its cultivation, on the premise that improved seed would make it worth the effort, was questionable; similar questions can be raised on the inclusion of a tree crop component in northern Sierra Leone (not the study for assessing its potential which was essential under the circumstances). Third, the lack of attention to marketing arrangements for agricultural produce and livestock compromised the success of both these components. Fourth, the feeder roads component did not fully consider the economic cost implications of agreeing to higher than necessary standards for their construction. During Phase I, costs per mile 7/ In early 1980, the number of externally financed IADPs in Sierra Leone, whether in operation or planned totalled seven. At the start of 1983, it was estimated that 20 percent of the nation's farmers had access to agricultural services through the IADPs.  8 were in exc.ss of appraisal estimates by five times. Fifth, exclusive reliance on the Seed Multiplication Project (SMP) for project seed, an integral part of Phase II, was a major drawback. The crop physical targets became especially vulnerable to any setbacks in the performance of the SMP. Sixth, the decision to share technical expertise of a Credit Specialist with EIADP III for credit strengthening, an area critical to overall project objectives, was a risk that could have been avoided. It turned out that the selected Credit Specialist proved to be effective in improving cost recovery and streamlining procedures. However, this might not have been the case. In any event, a single Credit Specialist for two IADPs spread across the country placed an unduly large burden on one individual, thereby reducing his ability to fully tackle all issues that required to be urgently and more completely addressed. 5. Project Implementation 5.01 The project became effective October 15, 1981 six months after Board approval. Because of a growing shortage of funds between the close of Phase I and effectiveness, implementation got off to a poor start. Suspension of disbursements at various times by IDA, ADB, ADF and IFAD together with the country's rapidly deteriorating finances led to the progressive reduction in project effectiveness. The project rating fell from 2 to 3, and then to 4, where it remained in the later years. There was increasing pressure on project managers to lower their implementation targets and tighten control on spending to fit the Government's budgetary constraints. This led to some staff lay-offs which were followed by labor unrest. Managerial and staffing problems further slowed implementation of field activities, civil works and road construction. After detection of irregularities in the accounts section, all its staff were replaced by Ministry staff who then had to be trained in commercial accounting. The situation was compounded by the continued lack of communication between various project units. Finally, toward project completion, the project was brought almost to a complete standstill, by the further worsening of project finances, rapid depletion of the credit revolving fund, increased reliance on commercial bank overdrafts, and the suspension of IDA and ADB disbursements. An exception was the tree crops component, the funding for which was to continue under the Agri *Itural Sector Support Project (ASSP). With respect to the remaining projec ctivities, as completion and closing approached, a start was made to integrate project staff with the MANR's regional program. 5.02 Financial constraints impeded several activities central to the project. Difficulty in obtaining transport and equipment because of financial and procurement delays continued to hamp-r field activities including credit recovery. Three critical bottle. -ks, namely, funding, shortage of improved seed, and lack of transport persis#'d through most of the project. The civil works program also suffered on atcount of cost overruns from substantial increases in the cost of labor and materials. Except for livestock health training, management of the livestock component was so poor that it had to be retracted. As the project progressed, its benefits were believed to be far lower than expected. This placed doubt on project justification, especially in the event that the problems were to remain unresolved.  9 5.03 The above not withstanding, there were instances where effective measures were taken to facilitate implementation. Some of these involved amendments to the Credit Agreement. In the case of large cost overruns on buildings, funds, initially earmarked for replacing vehicles in project years 4 and 5, were reallocated to the civil works component to prevent further delay. Funds were also reallocated in the Agriculture Sector loan to enable the financing of essential agricultural inputs. The problem of the project's cash flow had been accentuated by the small initial deposit in the Special Account of NIADP II compared to those in Eastern Phase III and the Magbosi projects respectively. Through appropriate amendments to the Credit Agreement the amount was increased from SDR 80,000 to SDR 270,000. While this was helpful and eased the project's immediate cash flow situation, it did not do so for long with the funding constraint assuming even more serious proportions. There were other areas, too, where implementation appears to have fared better. The introduction of a new strategy for improved credit recovery along with new procedures for loan recording, collection and monitoring strengthened credit services, even though the question of the financial viability of credit operations remained. Project accounts, too, in terms of internal controls and book- keeping were reported to have been well maintained, and any irregularities detected promptly dealt with. 5.04 SAR targets could have been attained through: (i) closer attention to the design of the swamp rice component; (ii) better assessment of the viability of tree crop and groundnut cultivation in the northern region, an area not generally known to have been conducive to these crops; (iii) detailed planning and alternative arrangements for good quality seed; (iv) regular monitoring of fuel supplies; (v) more effective monitoring and evaluation during implementation; (vi) a free standing credit component; and (vii) a firm and unambiguous agreement during appraisal between IBRD and ADB on the terms of loans to farmers for seasonal and developmental loans. Such an agreement would have avoided sending two conflicting signals by donors to Government on the same project and issue. Viable alternatives for tackling seed shortages that had adversely affected implementation of the crop production program through most of the project's life would have enhanced the achievement of project targets, particularly with shortage of seed also having been a problem in the earlier phase. In spite of assurances at Negotiations that MAF would obtain from SMP the required quantities of seeds, there were serious shortfalls in supply with negative implications for the achievement of crop physical targets. Apart from the inability of the SMP to cope with the demand for seed from a number of IADPs, there was lack of adequate planning that would have enabled placing orders with the SMP, as required at least a year in advance. 5.05 All three project risks stated in the SAR were indicative of Sierra Leone's weak institutional capacity. These related to likely setbacks in expanding and strengthening the extension system, improving credit recovery, and maintaining an adequate and timely flow of farm inputs. While the project attempted to minimize these risks through the design of its components, it did not recognize, sufficiently, the extent of weakness, at the time, of the existing institutions and systems for credit, input supply and extension. Furthermore, neither the repercussions on  10 implementation from the country's serious budgetary situation, nor continued sector distortions were examined. 5.06 Not foreseen and having the most adverse impact on the project was deterioration in Sierra Leone'. already serious financial situation. This manifested itself in: (i) inadequate counterpart funding; and (ii) suspension of disbursements by IDA, ADB. ADF, and IFAD in response to outstanding payments arrears. Other factors not foreseen were, first, the shortages of fuel, transport, planting material and fertilizer; second, an absence of a rational method for pricing seedlings; and, third, non- fulfillment of the covenant requiring the recruitment and maintenance of a training expert in the IPAM resulting in substantial shortfalls in the training program. 5.07 Actions and/or decisions that enhanced project implementation were: (i) steps taken to transfer funds and activities from the Credit to enable proper maintenance of roads already constructed; 8/ (ii) the decision to close Mara ranch which had been poorly managed, as well as the Feed Mill, and to utilize the savings to strengthen livestock health training; (iii) in spite of serious funding and transport constraints, enumerators continued to conduct surveys by foot and public transport while awaiting airport clearance for motorbikes ordered earlier; (iv) prompt efforts by Bank to assist Government to utilize external loan funds for procuring essential fertilizer supplies; and (v) on noting increased interest in swamp rice cultivation, Bank missions urged project management to utilize the swamps and its environs to the fullest through the promotion of double cropping and the systematic development of adjacent slopes. 5.08 Actions and decisions not taken and which would have helped project implementation include: (i) a clear demarcation between Phase I and Phase II activities, particularly with respect to the utilization of donor funds; (ii) an initial strong agenda for participation by farmers and workers in schemes where their active involvement was vital such as the chiefdom tracks component where it was generally required that unclassified roads be maintained at the village level; (iii) agreement with GOSL on the standard of roads to be constructed by CARE, and an early decision and firm commitment from CARE on its agreeing to undertake the rehabilitation and/or reconstruction of chiefdom tracks; 91(iv) a thorough moni.oring of seed availability, including the performance of the SMU, and steps to ensure the recovery of adequate seed from contract farmers through good field supervision; (v) close monitoring of the fuel supply situation on which the project, especially all field activity, was heavily dependent; (vi) based on SAR findings, exclusion, at the outset, of the piggery and Feed Mill 8/ There remained the overall concern by Bank staff of pursuing additional road development without ensuring the financing for maintenance of roads already constructed. The idea of transferring funds from new construction in existing road components for the purpose of maintenance was one that was finally supported. 9/ Two years into implementation, CARE had serious reservations, and was reluctant to construct chiefdom tracks on the ground that these were seldom maintained and usually reverted back to bush.  11 from the project design; (vii) assessment of the logistics of fertilizer use and the adoption of improved packages at an early stage to ensure that only farmers who had received improved seed fertilizer were issued fertilizer; (viii) greater realism and caution in the promotion of selected crops, including tree crops, in areas where they were not traditionally cultivated and which were considered climatically sub-optimal for their cultivation; (ix) less frequent changes in technical recommendations made to farmers and more consultation on such recommendations with other IADPs; and (x) agreement between IBRD and ADB, prior to effectiveness, on the interest rates charged to farmers for seasonal and development loans. 6. Project Results 6.01 Although there were elements of success, the project according to supervirion reports failed to achieve its targets and objectives. The overall outcome has, accordingly, been unsatisfactory. Responsible for the overall unfavourable outcome was Sierra Leone's weak macroeconomic management that was unable to reverse the economic decline and strengthen fiscal management. Because of funding constraints and the suspension of disbursements, the project's benefits and impact were much below SAR expectations. However, as compared to Phase I, there was a notable increase in interest by farmers in project activities and improvements in the standard of cultivation. The number of farmers estimated to have participated in the upland and swamp rice, groundnut and tree crops programmes was placed at 13,727 compared to the SAR estimate of 13,500 farm families. This excludes farmers engaged in the cultivation of other crops such as cassava, maize, tobacco and other crops and crop mixes. As the project progressed, funding constraints, poor logistics for input supply, difficulty in obtaining credit, and lack of mobility dampened this positive effect. Farmers became discouraged, while the scope for substantial increases in yields remained large. The problem of good quality seed and lack of effective research and inadequate testing affected project success adversely. In the initial years, the project's management focussed more on administrative and organizational matters and less on field activities. This pattern was reversed with a change in project management during mid- project years. Improved project organization and better administration were a reflection of serious efforts toward effective implementation made by the project management team. In fact, there was a phase early in the project when the administrative standards of NIADP II were considered among the best of the three IADPs then being implemented. 6.02 As regards physical targets, by the start of the third project year, in spite of financial constraints and shortage of planting material (particularly for upland varieties), achievements in upland and swamp rice cultivation were in excess of SAR targeto. In contrast, the performance of other crops (accounting for about 5,000 ha of the total SAR targets) such as cassava, tree crops, maize, tobacco and groundnut was considerably below set expectations. This was because of the delayed start in the propagation of planting material. However, there was some progress in the development of disease resistant varieties for cassava. Extension and training services provided to tobacco farmers by the Commercial Tobacco Company were reported to be excellent.  12 6.03 The results from the tree crop component were mixed. Depending on the manner in i'hich one views the final outcome it can be said to have either succeeded or failed. It failed because targets were not met. It succeeded because it created farmers' interest in tree crop cultivation as judged by the large shortfalls in planting material (with the exception of coffee) relative to farmer's demand. 10/ This is significant since the North was not an area with a .radition for tree crop cultivation. In addition, since the emphasis of NIADP had been on the production of food crops, staff were not adequately trained and key inputs nc. available for the inclusion of long term cash crops. The component, which finally took off in project year 4, was under severe strain by early 1986 from problems related to funding and fuel, poor management, and low morale. While the specific target for nurseries and seedlings was exceeded, nursery management had suffered from constraints on supervision, irrigation, and availability of essential inputs and planting materials. The most serious problem was that of poor accounting of nursery operations and the establishment of nurseries in vicinities considered suboptimal for cultivation. Since nursery expenditures were not recorded, the problem of pricing seedlings could not be resolved. There was no suitably trained counterpart to the Tree Crop Specialist, and no planning appears to have been in force either for the nurseries or for the supply of farm inputs. The component had thus become excessively target oriented without attention being paid to quality or the actual demand for seedlings or the ability of the nurseries to cater to this demand. 6.04 The key to successful tree crop cultivation was the tree crop nursery at Makump. Initially set up to supply fruit tree seedlings to large farmers for supply of fruit to the Mabole Fruit Processing Factory, which was closed down in 1980, its terms of reference were changed to providing seedlings for other tree crop species. However, because of poor logistics and lack of understanding of priorities, the irrigation system at Makump was rendered ineffective and recourse taken to less efficient plant watering methods. The final results were varied. Of the two year production targets, 93 percent was met for coffee, 75 percent for citrus (not budded) and 31 percent (budded), over 50 percent each for oil palm and coffee, 30 percent for mango, 23 percent for guava, and three percent for cashew. 11/ In February 1987, the Tree Crops Unit was integrated with the regional administration. 6.05 The wide variation between targets and actual crop achievements was due to funding and fuel shortages resulting in a lack of essential inputs, low morale, and poor extension and supervision of sub and village nurseries. On civil works, by project year 4, infrastructure was incomplete and the program had to be scaled down and limited to operational buildings. The village wells program was reduced from 400 to 48 wells, and feeder road construction from 50 miles to 45 miles. With the exception of the livestock component which was retracted, there were no major changes in project objectives. Because of poor management and absence of progress, a 10/ The shortage of planting material was most severe in the case of oil palm which was by far the most demanded crop. l/ Seedling citrus was not recommended because of low yield.  13 decision was made to limit the component's objectives to the livestock health training program. This turned out to be relatively successful in spite of the reported lack of cooperation from MANR's veterinarian staff. 6.06 The NIADP II was completed almost three years ago. In the assessment of the supervision missions, its development impact was rated a 3 with large shortfalls in most crop and civil works SAR targets. There were no noticeable breakthroughs on the technological front although some investigative work had been initiated on citrus cultivars and tree crop species for fuelwood. Project management rating also averaged 3. On the institutional side, it may be said that the project made a greater contribution to local rather than national institutions by improving standards of accounting and administration within the project and despite acute funding constraints. Even though the project was unable to make a mark on raising the quality of middle and higher level management staff, it enabled technical staff to gain good working knowledge in accounting and extension, much of it through on-the-job training. Based on these achievements it can be said that the project had some positive impact on human resource development. Given enhanced farmer participation in project activities, especially for the rice and fruit tree components, it is expected that some improvements in the social environment would also have taken place. The impact on macro performance and sectoral policies was however limited. Except for the positive and higher rates of interest to be charged on seasonal and development loans, there were no pricing and marketing reforms that were evident or of any significance, even though the need for sector policy reforms was urgent and long overdue. 7. Project Sustainability 7.01 The economic rate of return at appraisal was estimated at 21 percent. There was neither a PCR mission at project completion nor a comprehensive final supervision. An exact assessment of whether the project can maintain an acceptable level of net benefits through its economic life is difficult. However, based on reports by Bank staff who recently visited selected project areas in connection with the Policy Framework Paper (PFP), it is unlikely that the economic rate of return could have exceeded seven or eight percent. These accounts indicate that Bank-supported IADPs are not perceived by government officials and donors as having been successful. They are said to have had little impact on Agricultural production and rural incomes and employment. Being commonly viewed as 'powerful' enclave projects, the integration of the Bank-funded IADPs with regional MANR structures has been difficult and has not been achieved. MANR staff are reported to view their counterpart staff in the IADPs as having been privileged in terms of better remuneration, training, and working environment. The MANR is reported to be burdened with a large staff of 10,000 whose salaries take up almost the entire MANR budgetary allocation. 7.02 The prospects of project sustainability, under Sierra Leone's macro circumstances, would have depended first, on the extent of setback to the project since the prolonged economic crisis and suspension of disbursements, and second, on the project's resiliency in being able to recover from the setback and the requirements for doing so. Clearly, none of these conditions has been evaluated. The crucial risk factors, namely.  14 the continued unsatisfactory macro and payments arrears situations, and delays in carrying out necessary sector reforms, remain even now. At the project level too, the anticipated risk factors do not appear to have been addressed. The activities of the Ministry of Agriculture are reported to be almost at a stand-still. Accordingly, the problem of irregular and inadequate funding and insufficient maintenanc3 funds for feeder roads and chiefdom tracks continues to be serious. The effort tc strengthen agricultural services through more effective institutions and a long-term commitment to agricultural research and training, too, appears to have lost momentum. The key to achieving some measure of meaningful and sustained growth in the project areas would be the removal of these impediments and a policy and incentive framework conducive to sector efficiency. 8. Bank Performance 8.01 The project benefitted from effective Bank supervision. There was regular dialogue with the Government on project developments and progress and serious efforts made to resolve implementation difficulties. This strengthened the supervision process. However, in the early stages of the project cycle, the Bank could have devoted more time and effort to dealing with issues that are generally proven potential drawbacks to IADPs. Accordingly, greater emphasis on creating an enabling environment for effective agricultural research, training, and a good understanding of labor, land tenure and socio-economic issues in the project area would have been a worthy investment of staff resources. Specific areas that could have received closer attention, prior to effectiveness, relate to nursery development, and measures to ensure the availability of credit snd other essential inputs at the time and in the quantities required. 8.02 During appraisal, in addition to steps designed to strengthen institutions, the Bank could have focussed on the importance of adopting necessary policy measures aimed at bringing about economic efficiency at the sector and project levels in the interest of the long term growth and development of the sector. Supervision reports and follow-up measures and correspondence appear to have reported well on project execution and management, and financial issues. On reporting on the progress in achieving physical targets, however, project supervision and reporting was unsatisfactory. Toward project completion, more specifically in the final year of the project, emphasis was placed on the need to collect data for the preparation of a PCR. While it is never too late to insist on good and regular project monitoring, to have done so at this late stage in project implementation was next to futile. Both implementation and the preparation of a PCR would have greatly benefitted had regular project monitoring been insisted upon earlier. 8.03 The most striking, obvious and predominant lesson from the NIADP II is the reduced likelihood of project success when the project is undertaken in an economy consumed with macro and sector distortions. Under these circumstances, improvement of the macro and sector policy and incentive framework should be a basic and minimal requirement for success at the project level. Additional prerequisites for success would be strengthened and effective sector institutions for extension, training, research and credit on the one hand, and adequate transport and marketing infrastructure on the other. Further lessons for relevant Bank financed  15 projects based on the experience of NIADP I! may be suggested. First, where clear lessons from an earlier phase of the project are available, the project design should, as a matter of priority, first address issues that arose out of these, particularly tackling institutional weaknesses before designing more complex components. Second, in instances where macro and sector policies are distorted, macro reforms are necessary but not sufficient to create conditions for sector efficiency; these conditions must be btought about through sector policy reforms related, especially, to pricing, marketing and subsidies. Third, where technical packages are weak, in the sense of insufficient knowledge and/or testing, productivity gains based on anticipated yields must be cautiously and conservatively projected. Fourth, in a region dominated by poor farmers, the project must be able to offer farmers the choice of simple and cost-effective technology at an affordable price. Fifth, success of the project should not be overly dependent on a handful of specialists recruited under the project, who, when they leave, also leave a floundering department, project and/or project staff. Sixth, innovative and well-designed training programs capable of reaching a large and maximum number of project staff are essential to project sustainability and institutional development. Seventh, regular supervision of the project in close cooperation with the cofinanciers and all project entities must be of high priority. Eighth, where project success hinges on the availability of essential inputs, credit and other services, the project design must either ensure their timely availability in the quantities required or have contingency plans for their supply, otherwise the project proposal itself is not very meaningful with credibility of the project initiative itself open to question. Furthermore, the expected and vital impact of any proposed reforms, especially, those relating to producer prices for cash crops, is lost, and the damage to farmer enthusiasm to adopt improved packages and practices difficult to remedy in the short-term. Ninth, in projects that place as much stress on the achievement of physical targets as on institutional development and strengthening, the project must allow for either reduced scope and scale of the project or a longer period of implementation. Tenth, in areas such as procurement and cofinancing where the Bank has expertise and experience, that is, where it has some control over the outcome, it must firmly pursue necessary solutions together with its counterparts and cofinanciers to prevent avoidable delaya in project implementation and progress. Eleven, project design must openly recognize the risk associated with the success of an important component hinging on that of another completely separately funded project as was the case of the seed project financed by donor technical aid. 9. Borrower Performance 9.01 The Borrower's performance was marred by its inability to arrest Sierra Leone's economic decline. Adjustment measures were not fully implemented. Continued weak economic management and fiscal indiscipline had an unfavorable impact on the project's perforwance and sustainability. This was the Borrower's main weakness. Its main strength was reflected in its willingness to openly discuss, enquire and/or explain project related issues that came to the forefront, and to cooperate with the Bank, as much as it could within the overall financial constraint, in resolving difficult problems. The critical financial circumstances would normally have preempted any project progress. However, due credit must be given to  16 project management and staff who tried hard to meet project targets. On the whole, the project itself was fairly well managed, even though this had a lot to do with the particular individuals appointed as project manager. 9.02 The experience in implementation and performance of the Northern Phase II project was similar to the Eastern Phase III project which was being implemented during approximately the same period. The lessons for the Borrower are also similar. :irst, it is important for the Borrower to take measures at the macro and central budget level that will ensure adequate and regular funding for the project being undertaken. Second, the project design and scale should be prudently drawn based on the Borrower's own realistic assessment of its overall budgetary and institutional capacity. Third, where the projects are externally funded, the Borrower must ensure the inclusion of a strong training component where a maximum number of staff would benefit and obtain specialized training relevant to the project and sector, so that training imparted after project closing is of equal quality, and continues with the same momentum. Third, whether or not the project is funded externally, it is essential for the Borrower to establish a solid framework for financial and commercial accounting, and monitoring and evaluation, through well selected project teams rather than through the temporary dependence on a few consultants. Fourth, it is in the Borrower's interest to bring to the forefront at the preparation stage, or even earlier, any social issues that might impede project progress and the project's ultimate success. It should also highlight possible additional constraints that could arise at decision-making levels which might delay or interfere with smooth project implementation. 10. Project Relationship 10.01 The Bank-Borrower relationship was marked by good communication at the working and decision-making levels. The Bank relationship with the many project entities, too, was satisfactory. However, the same cannot be said of the Borrower's relationship with the ADB which was characterized by strain on account of payments arrears and procurement difficulties. ADB had, on several occasions, been unable to join Bank supervision missions and actively participate and facilitate the process of problem solving. The importance of overall and effective communication on project implementation cannot be underestimated, and, particularly, on complex project related issues, if the project is to succeed. The Borrower's relationship with other relevant parties such as CARE was somewhat mixed. The underlying cause was the Borrower's financial constraints. The main issues that were required to be addressed were design and organizational issues, for example, grade specification of project roads and the apprcpriate organization that would maintain the roads post construction. There was also the question of whether CARE would agree to undertake the construction of chiefdom tracks given its strong view that these are not maintained after completion. The relationship between different project units was reported to suffer from poor communication between them. This led to difficulties in da~i. reconciliation. Shortages of vehicles which constrained the mobility of 1a&E staff is cited as the primary explanation for inadequate and ineffective communication.  17 11. Consulting Services 11.01 Consultants were able to make important contributions in selected areas. The Credit Specialist introduced new procedures and helped to streamline financial operations and, in the process, improve credit recovery. Yet the services of the Credit Specialist were shared with the EIADP III project whose implementation overlapped partly with NIADP II. This was a poor arrangement since credit recovery and delivery systems and procedures were weak and constituted major issues in both projects. One individual having to strengthen, signif.'cantly, credit operations in two IADPs stretched across two regions was not cost-effective from a long-run point of view. CARE continued to construct good roads in Sierra Leone as it had done in the past. Furthermore, it was very forthright in its view on the construction of chiefdom tracks and the problem of their subsequent maintenance, and at the same time flexible on the issue. It cooperated fully with the Bank on the decision to scale down the program for construction of new roads so that investments already made could be protected through adequate maintenance and attention. 12. Project Documentation and Data 12.01 The legal agreements for the projects were clearly spelled out and appeared to lend themselves to careful and close monitoring during implementation and supervision. An added strength of the Development Credit Agreements was in the built-in flexibility that enabled the Bank and Borrcwer to make amendments with relative ease in circumstances that necessitated them. 12.02 During implementation, the SAR was useful for the Bank and Borrower in terms of explaining the institutional and infrastructure constraints in the project area and the measures necessary to strengthen these. The SAR was weak, however, in analyzing the policy and incentive framework for the agriculture sector. This was inevitably to entail considerable economic cost with related serious implications for the projects were the prevailing policies to remain intact. This was a shortcoming that compromised technical and economic efficiency in the sector and projects. With relative optimism that macro problems would be solved in the short term, the SAR did not alert decision-makers, and those closely connected with the projects, on the risks for the project due to deterioration on the macro front and continuation of the distorted policy and incentive framework within the agriculture sector. 12.03 Sierra Leone lacks a national agricultural data base. Bank supervision reports had made some data comparisons on 'actual' crop production compared to SAR targets, but much of the data on other targets and achievements was lacking. Even for crop production, area achievements were measured by annual inputs supplied rather than measurement of the actual area cultivated. Available data thus have to be interpreted with considerable caution. It would be useful for agricultural projects in future to ensure that the Borrower prepare in consultation with the Bank a format for a project data base prior to project effectiveness and preferably at appraisal. Regular updating of the data base and follow-up analysis could then provide a useful source of project data for  18 supervision, implementation, regular evaluation, and special stua,es including a PCR. 12.04 Collection of Baseline Survey data was completed within the first two years of the project, and a questionnaire for a labor allocation study in the area was being initiated. The data that were collected pertained basically to household information. No crop yield data were gathered initially because of constraints imposed by shortages of vehicles, fuel, and weighing scales. Subsequent data gathered turned out to be unreliable and was not corroborated by checks in the field. With closer supervision, the M&E unit could have provided useful data in several areas. First, it could have tested the results of applying fertilizer to local traditional seed material in upland cropping and in the process could have helped in reducing random and wasteful use of fertilizer. Second, it could have provided more precise estimates on average yields for swamp rice based on field cuttings. This would have assisted extension staff to counter farmer reluctance to adopting improved nursery practices such as line spacing at planting and better water control measures. PART II. PROJECT REVIEW FROM BORROWER'S PERSPECTIVE (The Borrower was requested to provide a Review, but nz;ne was received).  19 PtT liI STAI=STICA. miWTIN helum Ittgt1tgt1 lqrcattural uluegi et prettCt iCrti4t IIN2 tI 1. InlatStSt t tAO" Mtg/ C'edats Ltiay tsi Crtattt Thtet twoEw a t areo/ l Stit 0 r s Ss t latwtntt 4icr ltor4i rtke project matc to uaild X tu "Morigtegtc 4/24c Cloil so Of 05/11317 Thle WAt tgettc rats of Rtvo 4 Is olw t t. the Aewlget Fria ct aitinlt S23A5) tf t oal Uts ltlte atelaISM to 195. by Project Perfesce hat btWt cn ta ratc t4 the insttri 04 of scotVtvs Mi btnttl retta, at 2.empsnocet. this so% calcvaatee ma toe Ienwcin ("I ebach U tt aqo basis d4 too Pu's tntc alowstauns ade atutetts titslthliet oi t Ws stdtap a" tithe o ttt as the id eal ll Aa Uwe me to late wnruiaq arwe. It as aie to mcose, tAfon atcm so tne t1i11 cru ,o. the *tovnflocatot da teetlheidt frotetq ItVot e4ficast race, cos of Oat hold pdfhtCl. is t4raeltvt l tleul.eet XtBttitv l to4 toH n"tor dclb ow cotSA"to san te h socusai tand ousti Met savarte the eat gall tl td iel estate. A prejut unenut seat " the NW o to he ntablaeig , d rimage $ttode a c tosttool vceatc ts ad itetant of relevut croi cmrtgo mt. Se1i iVnCIlttrti 1`1` C1tIat Th1 pIu Et bM te asst a ffaUKag 04/29/75 Ctglei a 04 09d/tt VA b W stauted a Ktc_ rate o4 ratmre of O2 CrtUtt le. S*L da ltrmtetal mdstaun A" far Incoese WrCOtt. 1The Pnrojct PerwVtCe heta t_Ws Lm Se t" a" to "tallish wegt av sasst entleute "I pWCet. riso we prelat actasatoan ewence tMet * tid htll n walt Is tkt Gtnt ad (attrn. begos oath steantc ayanlteIat letie t Ut coetraints ta fitn t4 rettn at oranata Of 14 wcst ae 25 0 tIWttve letti 0tt4 tlat as rWei Woe. WECIt rteWctavei. The t overall rate of retIre ts a retttagett f otitrtfalls ay eOt c Wfur KCE Mi tie _ao Of eatrKt Ihraeaciaraes Iach on *vttacuwlf led Ie the (aotms I5qnt. (atwo tett ipacetite the ens if thte netjt on as eCta m *_a I/IS/SI Clash as of 12/31/dI The ecugmc rAte r.4 rotors gstasited by the VA at !4 i,steget Prett (irot 1094-SI ) five-w wsar atWarttwo" atatn or the smCtt no1 MtiIAl t cao s P e" gM t. th attest fite colttvetae of nea race. MW aIltM race 04 rgtme nI likelys to hm t fa rglo 4os to Mi caSSewet a" a" stanta Of cncfgs, tIvg rhrtOlls lat an to trw itat egoreew Me cm s" . as eal Oa&e the SA tarst fsfr rdso Mi village gtll. Pow oftt fttuC sa nthil loca Of the tbirnt s ISIaW fasucKat satnataM 014 OamsgUini 04 finsimoants by dmns En to the ctutri tssag ddt t to stte arroew IaVents.  20 2. Project Timetable Date Date Item Planned Actual - Identification 03/30/80 03/03/80 (Executive Project Summary) - Preparation 12/78 02/80 - Appraisal Mission 06/30/80 06/30/80 - Loan/Credit 02/09/81 02/09/81 Negotitations - Board Approval 04/07/81 04/07/81 - Loan/Credit 05/13/81 06/13/81 Signature - Loan/Credit 08/13/81 11/15/81 Effectiveness - Loan/Credit Closing 06/30/87 12/31/87 - Loan/Credit Complotion 12/31/88 June 86 3. Loan/Credit Disbursements Cumulative Estimated and Actual Disbursements (USS '000) FY82 FY83 FY84 FY85 FY86 FY87 a/ Appraisal Estimate 2.6 4.2 5.6 6.7 7.9 8.5 Actual 0.6 1.4 1.9 2.4 2.8 3.0 b/ Actual as % of Estimate 20 33 34 36 36 36 Date of Final Disburseme nt 12/89 a/ End of second quarter of FY87. b/ The exact figure is USS 3.07 million. Actual cumulative disbursoments were low compared to estimated disbursoemnts because of lack of counterpart funding and suspension of disbursements.  21 4. Project Implentation -----------------------------------------..-----------__---------------------__--- Appraisal Actual Indicators Estimate Estimate Crop Development (ha) 1/: Upland Rico 10,760 6,437 Swamp Rico 4,000 4,668 Groundnut 4,886 1,826 Tree Crops 304 348 Other Crops 2/ 5,096 no Village Wells (no.) 3/ 400 4/ na Roads (km) Class IV 80 6/ no Chiefdom Tracks 300 6/ no -----------------------------------------------------------------------__ 1/ Area based on sales of inputs each year. 2/ Includes cassava and sweet potatoes, maize, tobacco, and crop mix on upland and swamp cultivation. 3/ The target for village wells was revised downwards to 160 in 1983 because of increased cost per unit of construction and th- low construction capacity in the project to carry out the target construction. 4/ Standards of construction would follow those *pproved by UNICEF. Direct costs allowed for Le 1,750 per well. S/ New construction at a cost of Le 18,760 per km. 8/ Rehabilitation of unclassified chiefdom tracks at an estimated cost of Le 4,100 per km including the cost of repairs of bridges and culverts. The total maintenance cost of the chiefdom tracks vas estimated at Lo 46,760 per annum. Comments: Crop development was constrained by shortages of quality *eed, lack of mobility, and difficulty in obtaining credit and inputs. There was poor adoption of agricultural packages and therefore a constraint on any significant improvement of yields. Yields of var;ous crops as determined by crop cuttings by the MAE unit demonstrated little difference in favour of project fa-mers. No cash loans had been advanced to farmers in the first thre- project years reportedly due to financi.l constraints. In both food and cash crop, and tree crop development, there was need for strict supervision of field staff. Farmers had been reluctant to accept the project due to previous mishandling of credit and inputs and repayments, past neglect of farmer super.'sion, excessive emphasis on targets, poor farmer selection, and frequent suspension of disbursements by IFAD, IDA and AD8. A year before project completion, financial analyses of the recommended crop packages for upland and swamp rice packages revealed that they were financially unattractive to the farmer. The success of the upland rice sub component varied with the availability of seed. Technical problems dominated the development of swamps with the result that mid way through the project the SAR swamp development concept was replaced by cultivation" and provision made for technical advice and inputs to promote it. Late cash releases to farmers in violation of a previously established schedule also contributed to reduced project credibility. The trp* crop programi fared somewhat better with farmers responding positively to higher producer prices. As of February 1986, the demnd by farmers for tree crop planting material exceeded capacity. In terms of hectar-s to be planted, except for coffee where capacity relative to demand was high at 140 percent, all other crops faced large shortfalls: oil palm, the most demanded crop, 20 percent, citrus 27 percent, mango 10 percent, guava 37 percent, and cocoa 70 percent. Poor managent of nurseries resulted in high losses of planting materials, and Improper plant watering and application of fortilizers led to nutrient deficiencies and poor plant growth. An added problem was the quality of seedlings and their transportation over long distances which was prone to reduce quality further. Technical improvemnts, proper upkeep of nurseries and their close supervision were also important requirements that were not always met. The targets for the construction of roads were excessiv-ly optimistic in the SAR. In early 1985, even though CARE had agreed to shift, in the next construction season, to the construction of lower level of chiefdom tracks using palm oil log bridges to reduce costs and utilize local labour, this was not done. In fact, roads were being constructed to Class IV A standard using imported iron culverts. The wells program proved to be overly ambitious, and as the project progressed the type of wells to be constructed and their exact specification becam less clear and more costly to the project.  22 5. Project Costs and Financing Appraisal Estimate Actual Foreign Local Foreign Local Costs Exchoo Costs Total Costs Exchonqo Costs Total (USI millions) (US3 millions) Project Management 0.7 1.8 2.5 na ns no Extension, Training A Agronomy 1.6 2.4 4.0 no na na Commercial Services 0.6 1.2 1.8 no no no Production Inputs 2.0 2.8 4.8 no no no Livestock Sorvices 0.6 0.9 1.6 no na no Engineering Sorvices 0.9 1.0 1.9 no no na Road Construction 0.3 2.6 2.9 no no na Consultancy and Forestry Rosearch 0.2 0.4 0.6 no no na Monitoring and Evaluation 0.2 0.2 0.4 no no na Contingencies 2.7 4.6 7.3 no no na ---------------------------------------------____-___------------------__----__------------------------- Total 7.1 13.3 20.4 no no na B. Project Finoncing Planned Loan/Credit Source A(r- m-nt Final (US* wil1l1) IDA & IFAD 14.S Personnel 2.2 no Vehicles A Equipment 2.0 Road Construction - na Buildings 0.3 no Operations A Maintenance 2.6 no Production Inputs 6.9 na Consultancy and Research 0.6 no Co-Financing IFAD 6.0 na ADF */ 8.5 no Other External UNDP 2.0 no Domestic GOSL 2.7 no