Uganda - Agricultural Development Project

Documnent of The World Bank FOR OMCLIAL USE ONLY Report No. 13812 PROJECT COMPLETION REPORT UGANDA AGRICULTURAL DEVELOPMENT PROJECT (IFAD LOAN 159-UG AND IDA CREDIT 1539-UG) DECEMIBER 16, 1994 Agriculture and Environment Division Eastern Africa Department Africa Regional Office This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  UGANDA AGRICULTURAL DEVELOPMENT PROJECT fIEA 2LQ-flI- & IDA-Crediti~LU 1 G1( CURRENCY EQUIVALENTS Currency Unit US$1.00 (at Appraisal) = USh3'5i 00 US$1.00 (July 1993) - US1i,200.( SDRI.00 (at Appraisal) US$ 1.00c SDRI.0O (July 1993) = TSS1.40 WEIGHTS AND MEASURES Nletric Ssstem ABBREVIATIONS ADP Agricultural Development Project AEP Agricultural Extension Project ARP Agricultural Rehabilitation Project ARTP Agricultural Research and Training Project DFI District Farm Institute EEC European Economic Community ICB International Competitive Bidding IDA International Development Assoclation IFAD International Fund for Agricultural Development MAAIF MinistrY of Agriculture, Animal Industry and Fisheries MAF NMinistry of Agriculture and Forestry MAIF Ministry of Animal Industry and Fisherices MN&E Monitoring and Evaluation NURP Northern Uganda Reconstruction Project PCUJ Project Coordination Unit PIU Project Implementation Unit PMU Project NManagemernt Unit T&V Training and Visit FISCAL YEAR Government of Uganda = July I June 30 IDA = July I - June 30 IFAD = January I - December 31  FOR OFFICIAL USE ONLY THE WORLD BANK Washington, D.C. 20433 U.S.A. Office of Director-General Operations Evaluation December 16, 1995 MEMORANDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT SUBJECT: Project Completion Report on Uganda Agricultural Development Project (IFAD Loan 159-UG and IDA Credit 1539-UG) Attached is the Project Completion Report on Uganda -Agricultural Development Project (IFAD Loan 159-UG and IDA Credit 1539-UG). Parts I and III were prepared by the Africa Regional Office. Part II was prepared by the Borrower. The project was designed to increase the production of food-crops, livestock and fish in seven economically disadvantaged districts in Eastern and Northern Uganda. The project also supported considerable institutional strengthening of agricultural support services and a project Monitoring and Evaluation Unit. The Loan and Credit became effective in May 1986 after a one year delay due to civil unrest in the project area, which continued throughout much of the project life. This risk was not adequately assessed at appraisal. This coupled with delays in procurement and the continuous inadequacy of counterpart funds, resulted in three Closing Date extensions totalling three and a half years. The project was restructured in 1991, adding three components. One was directed towards women's economic activities-which had been reviewed, but not supported, at appraisal. The component was successful. The project is rated as satisfactory by both the Borrower and the Bank although no ERR was calculated at completion. Substantial institutional development took place. The Monitoring and Evaluation Unit within the project was strengthened considerably and now serves as a core unit of the proposed M&E unit of the Ministry of Agriculture, Animal Industry and Fisheries. Training was also successful with much of the staff remaining in government service. The Borrower's report (Part II) is highly commendable for its thoroughness and quality. In general, the Borrower was satisfied with the level of assistance provided by the Bank. However, while admitting their own weakness with regards to local tendering procedures, the Borrower cited difficulties with IDA procurement procedures (notwithstanding five supervision missions including a procurement specialist), the lack of authority devolved to the Bank's Resident Mission which required costly reporting to Washington, and lack of continuity of Bank task managers. The Borrower is rightly proud of the institutional development components. In addition, the Borrower discusses the high level of participation involved, including the close interaction of staff with beneficiaries to assess their changing needs and circumstances so as to adjust project activities accordingly. This project may be audited. Robert Picciotto by Hans Eberhard Kopp 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 UGANDA AGRICULTURAL DEVELOPMENT PROJECT (IFAD LOAN 159-UG & IDA CREDIT 1539-UG) TABLE OF CONTENTS Currency Equivalents/Abbreviations PREFACE ...................... i EVALUATION SUMMIARY .................................... iii PART I: PROJECT REVIEW FROM THE LENDERS' PERSPECTIVE ... ... 1 1. Project Identity ......................................... 1 2. Background and Sector Context .............................. 1 Background . ............................... 1 Sector Development Context .............................. 2 3. Project Objectives and Description ............................. 2 Objectives ................................ 2 Description ........... ..................... 2 4. Project Design and Organization .............................. 3 Design ................................ 3 Organization and Management ............................. 4 5. Project Implementation ................................ 5 General ........ ........... . ........... 5 Input Supplies . ............................... 5 Research and Surveys ............................... 6 Livestock and Veterinary Services . .......................... 7 Institutional Building ................................ 7 Other Components ................................ 8 6. Project Results and Impact ............................... 9 Agricultural Development ............................... 10 Livestock Development .................... 10 Fisheries Development .................... 10 Capacity Building .................... 11 Formulation of Sector Plans ..................... 11 Economic Rate of Return .................... 11 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 Banlc authorization.  7. Sustainability .11 8. IDA/WFAD Performance ................................... 12 9. Borrower Performance ............ 12 10. Project Relationships ........... 12 11. Studies and Consultancies.. . ........ 13 Studies.. . ........ 13 Consultancies ............ 13 12. Project Documentation and Data .. ......... . 13 13. Lessons Learnt ............ 14 PART II: PROJECT REVIEW FROM THE BORROWER'S PERSPECTIVE .. 15 PART III: STATISTICAL INFORMATION .50 TABLES 1. Related Bank Loans/Credits 2. Project Timetable 3. Cumulative Credit Disburs'ments 4. Key Implementation Indicators 5. Project Costs and Financing 6. Project Results 7. Status of Major CovenenDs 8. Use of Bank Resources ANNEXES 1. Quantity of Inputs Procured 1986-93 2. Output of Food Crops In ADP Area rIRD Maps: 18036, 18035R1  PROJECT COMPLETION REPORT UGANDA AGRICULTURAL DEVELOPMENT PROJECT (IFAD Loan 1 59-UG & IDA Credit 1 539-UG) PREFACE This is the Project Completion Report (PCR) for the Agricultural Development Project in Uganda, for which IFAD Loan 159-UG, in the amount of SDR 14.6 million, and IDA Credit 1539-UG, in the amount of SDR 10.0 million, were approved by the respective Boards on 10 November, 1984 and 8 January, 1985. The Loan and Credit closed on 30 September, 1993, three and a half years behind schedule. The total undisbursed balance of over SDR 2.6 million (of which, IDA's share was SDR 858,534), was cancelled in February 1994. Parts I and III of this PCR were prepared by the staff of the FAO/World Bank Cooperative Program (CP) for the Agricultural Operation Division (Eastern Africa) of the World Bank. This report is based inter alia, on the Staff Appraisal Report; the Loan and Credit Agreements; supervision reports; correspondence between IFAD, IDA and the Borrower; internal Bank memoranda; and information gathered by the CP mission which visited Uganda in July 1993. The Borrower has provided its own evaluation of project performance, which is included as Part II of the PCR. i   - PROJECT COMPLETION REPORT UGANDA AGRICULTURAL DEVELOPMENT PROJECT (IFAD Loan 1 59-UG & IDA Credit 1 539-UG) EVALUATION SUMMARY Objectives The primary objectives of the project were: i) to increase the production of food- crops, livestock and fish and to contribute towards higher rural incomes; and ii) to increase the availability of data for the formulation of agricultural sector development plans. The restoration of small-holder agricultural production to the peak levels of the 1970s was viewed as a major means of ensuring food security, generating rural employment and boosting per capita income. The main constraints were lack of inputs and spare parts, as well as inefficiencies or weaknesses in the institutions supporting agriculture. The project strategy was therefore to achieve the above objectives through the provision of inputs, improvement of adaptive research, statistical surveys, and, institutional support. Seven economically disadvantaged districts in Eastern and Northern Uganda were targeted. Implementation Experience The Loan and Credit became effective in May 1986 after a delay of one year because of insecurity in the project area. This localised insecurity was to continue throughout much of the project life. It was not until 1991, five years into project implementation, that it was able to operate in all the areas intended at appraisal. Time lost due to these interruptions was never regained. This, coupled with the delays in procurement and the continuous inadequacy of counterpart funds, resulted in the closing date of the project being extended thrice involving a total of three and a half years. The project closed at the end of September 1993. Total project cost was some SDR 23.5 million, or about 25 percent below the appraisal estimate of SDR 31.1 million. The procurement of inputs was considerably delayed due to: project staff having inadequate knowledge of ICB procedures; lengthy and bureaucratic delays in approving the items to be procured, as well as belated approval and award of tenders and delays in opening letters of credit. In the early period of implementation there was sporadic looting of project stores in the insecure areas. Further losses were incurred as a result of cash losses, bad debts due to non-payment by the input buyers, and writing off of stock which had become obsolete or had expired. The project also incurred losses at the hands of foreign suppliers who, on some occasions, delivered goods differing from those specified both in quality, quantity and physical specifications. This was possible to a certain extent allegedly because an iii  international firm retained by GOU to carry out pre and post-inspection of imported goods did not render satisfactory service. The adaptive research activities were transferred to Tororo as it offered better security than Serere. Very little was done in the tsetse control till 1993. The fisheries stock survey of Lake Kyoga was completed in 1988 while the catch assessment survey was completed in 1991. Both started late because of insecurity and delays in procuring the required equipment. The livestock activities of the project were severely affected by the heavy loss of cattle through rustling which saw their population decline from 1.3 million heads in 1985 to about 0.3 million heads in 1990. The focus in the latter part of the project was therefore shifted to the improvement of small ruminants and poultry which are still relatively many in number, are kept by a majority of farmers, and take a shorter time to regenerate. Institutional capacity building was supported through extensive staff training overseas and in neighbouring countries, in disciplines such as, extension and research. The training was in form of short term courses and study tours or long-term courses leading to higher degrees. In all, 87 staff received training either overseas or in neighbouring countries. The project also assisted by providing transport, accommodation and equipment. At Tororo District Farm Institute (DFI), a media resource centre was established and produced a substantial amount of audio-visual materials which were useful in propagating extension messages. Many of these have been translated into vernacular languages to facilitate communication. The project was restructured in 1991 during which several components were added including women's activities, pilot credit scheme, ox-plough revival and cassava mosaic control. The women's component has been successfully implemented and has attracted considerable interest among the local community. The pilot credit is only in its second year of operation, but it too has attracted keen interest and more ADP districts are requesting to be included in the programme. The temptation of expanding too quickly should however be avoided given the limited project capacity to appraise and supervise the credits. Ox-plough revival activities were constrained by lack of adequate numbers of oxen and implements. A vigorous campaign was carried out against cassava mosaic virus and the project spearheaded the multiplication and distribution of tolerant and resistant varieties to farmers. Project Results and Impact Overall project performance may be deemed satisfactory, particularly when viewed in the context of the major structural, institutional and financial constraints undermining the economy, in addition to the overwhelming security problems. The aim of the project was to improve agricultural, livestock and fisheries development and to increase rural incomes; while it is difficult to determine precisely how far these goals have been achieved and to what extent the improvements are due to the project per se, as opposed to other factors, e.g., improved security situation and weather conditions, it is reasonable to assume that the project contributed positively to agricultural growth. The production of food and cash crops in the area has increased during the project period. The inputs supplied, the improved seed varieties made available, and the extension messages propagated by the project iv  contributed towards this improvement. The livestock population, particularly of the small ruminants is reportedly increasing, partly as a result of project interventions in animal health and breeding, but cattle populations are unlikely to recover quickly without a massive restocking operation. The fisheries and catch assessment surveys of Lake Kyoga have re- focused the autherities' attention to overfishing in the lake. Steps are being taken to correct the situation. In the course of project operations, water hyacinth was discovered on the lake and steps are now being taken to control it biologically. At grassroot level, about 400 women's groups which have been formed with project support are helping in sensitizing women about social and economic issues. It is too early to assess the impact of the pilot credit scheme because it has been in operation for a very short time. However, if it is managed well and loan delinquency is kept to a minimum, it could form a basis for a rural credit programme on a larger scale. In terms of capacity building, the institutions associated with the project have benefited considerably from the extensive training programme, transport facilities, accommodation and equipment provided by the project. The project also contributed towards its other stated objective of increasing the availability of data for the formulation of sector development plans. The M&E Unit collected data on agricultural prices, costs and extension activities in the project area. It also prepared an excellent study of how to integrate ADP's activities into MAAIF. The Unit has now become a core unit for the proposed MAAIF ministry-wide M&E Unit. Sustainability The main risk to the sustainability of activities supported by ADP is the inadequate availability of government budget to fund the full cost of most of the non-input supply operations of the project. This however may not be an immediate problem because a number of IDA and [FAD supported projects will continue or have started in some of the ADP districts and will continue with some of the activities initiated by ADP. The projects however have a finite life; on their completion the funding burden will revert to GOU, unless timely measures are introduced to initiate cost recovery mechanisms, linked to the rate of recovery and growth in agricultural production. Fundings and Lessons Learnt The main findings and lessons learnt from implementing this project are operational. The adverse impact on project operations arising from inadequacy or untimely availability of counterpart funding can be considerable and such risk should be carefully assessed at appraisal. Secondly, public sector distribution of inputs should be discouraged as the function is invariably better performed by private distributors and traders. Thirdly, key project documents like the SAR should contain specific key implementation indicators and targets towards which the project focuses. Without these, there is no benchmark against which to objectively measure project progress, success or failure. And finally, procurement contracts should contain adequate force majeure clauses to protect the parties to the contract and allow a quick and smooth winding down of operations in the event of prolonged periods of insecurity. v   PROJECT COMPLETION REPORT UGANDA AGRICULTURAL DEVELOPMENT PROJECT (IFAD LOAN 1 59-UG & IDA CREDIT 1 539-UG) PART I: PROJECT REVIEW FROM THE LENDERS' PERSPECTIVE 1. Project Identity Project Name Agricultural Development Project Credit No. IFAD Loan 159-UG & IDA Credit 1539-UG RVP Unit Eastern Africa DeDartment Country Uganda Sector . Agriculture Sub-sectors Food Crops, Livestock and Fisheries 2. Background and Sector Context Background 2.1 The Agricultural Development Project (ADP) was a sequel to the IFAD-funded and World Bank executed Agricultural Reconstruction Programme (ARP) which was implemented between 1982 and 1984. ARP included financial support for the preparation of a follow-up project to further agricultural production and to improve supporting services. With the assistance of consultants, Govemment of Uganda (GOU) produced a preparation document which formed the basis for a request to IFAD and IDA to support an agricultural development project. ADP was appraised in October 1983. It was approved by WFAD and IDA Boards in November 1984 and January 1985, respectively, and became operationally effective in May 1986 after delays largely attributable to civil strife.  2.2 The project was scheduled to be implemented over a 4 year period through 31 March, 1990. However, due to delays in starting up operations in most project areas, the project was extended thrice involving a total period of 3 1/2 years. It finally closed on 30 September, 1993. Sector Development Context 2.3 GOU's national policies at the time of project formulation were set out in the Recovery Program 1982-84. The central theme of the program was to encourage agricultural production as a means of increasing exports and improving rural incomes. The restoration of small-holder agricultural production to the peak levels of 1970s was viewed as a major means of ensuring food security, generating rural employment and boosting per capita income. The main constraints were lack of inputs and spare parts and inefficiencies or weaknesses in the institutions supporting agriculture. 2.4 The ARP provided a range of inputs for crop and livestock production in 7 districts. The operation was viewed as having been implemented without major problems. GOU was therefore keen to spread the benefits of an ARP-type operation to other parts of the country, which were economically depressed because of relatively low agricultural production, lack of employment opportunities, recurring droughts and other factors. Initially, ADP was to include only 4 eastern and northern districts of Apac, Kumi, Lira and Soroti. However, before appraisal, at the request of GOU, the number was increased to 7 by adding the districts of Gulu, Kitgum and Tororo. 3. Project Objectives and Description Objectives 3.1 The objectives of the project were (i) to increase agricultural production and the incomes of small-holder farmers in the project area; and, (ii) to improve the availability of data needed for the formulation of sector development plans. Description 3:2 As appraised, the project comprised of the following three components: (a) Input Supplies: the provision of agricultural implements and inputs including flour mills, wheelbarrows, workshop equipment, spare parts, agricultural chemicals and forestry nursery supplies; provision of livestock inputs including veterinary drugs and equipment, vaccines, acaricides, spare parts and a mobile laboratory; and, the provision of fisheries 2  supplies including outboard motors, spare parts, fishnets and boat building materials. (b) Research and Surveys: establishment and equipping of an adaptive research section at Serere Research Station; launching of a pilot tsetse fly control program; and carrying out a fish stock assessment of Lake Kyoga and a fisheries statistics and evaluation study to assess the current status of fishery also in Lake Kyoga. (c) Institution Building: strengthening of general agricultural extension services in the project area, training extension staff and rehabilitating several farm training institutes; introducing a pilot training and visitation (T&V) extension system in Soroti; provision of equipment, vehicles and housing for extension workers; and strengthening the Monitoring and Evaluation Unit in the then Ministry of Agriculture and Forestry (MAF). 3.3 Appraisal project Costs were estimated at some SDR 31.1 million, and remained unchanged after project restructuring. IFAD and IDA were to contribute SDR 14.6 million and SDR 10 million, respectively, while GOU was to contribute about SDR 6.5 million. An additional SDR 0.3 million was financed under the Bank's Second Technical Assistance Project (refer para. 4.2). Total project cost (at completion) of some SDR 23.5 million was nearly 25 percent below the appraisal es1timate. 3.4 The project was restructured in March 1991 following a request by GOU to extend it for a further 2 years. The restructured project kept the original components, as well as adding several new ones, viz., credit, women's activities, ox-cultivation, and cassava mosaic virus control. Total project costs were not altered. 4. Project Design and Organization Design 4.1 As a follow-up operation to ARP, the design of ADP was quite simnilar to ARP both in concept and target group. The project was oriented towards input supply (almost 75 % of total project costs) and was to be of a fast disbursing nature. The project provided a balance of components, some with immediate effect and others with more long-term effect. It initiated action to increase the effectiveness of extension and research services by methods which, once proved on a pilot scale, would be replicated in other parts of the country to improve institutions and services. However, the project organisation as designed was cumbersome as discussed in para 4.3 below. 4.2 Project risks were recognised and safeguards provided in project design. For example, in order to avoid delays in the conmmencement of start-up activities before the project effectiveness, US$ 0.3 million was provided upfront (financed under the second Technical 3  Assistance Project (Cr. 1434-UG)). In order to ensure that staff engaged in project supported activities were paid their allowances on a timely basis, a bank account was established wherein at the beginning of each quarter GOU would deposit the amount estimated to be required to cover expenditure for staff travel and transport allowances. The design however failed to adequately take into account the likely risk of shortfalls in counterpart funding which plagued the project till 1992 when GOU authorised the project to use funds generated from input sales. And as the country was in an unstable state during the latter stages of project processing, the project should have taken into account the risks arising from civil instabilities spreading into the project areas; this turmoil did undermine the project's performance. Organization and Management 4.3 At appraisal it was envisaged that overall project coordination would be through an Inter-ministerial Coordination Committee which had been set up under ARP. The day-to-day management of the project would be carried out by a small Project Management Unit (PMU) headed by an Executive Secretary in MAF. Two Project Implementation Units (PIUs) were set up, one in MAF and the other in the then Ministry of Animal Industry and Fisheries (MAIF) to implement project activities in the two respective ministries and liaise with the PMU. This arrangement was cumbersome and did not work satisfactorily largely because the designated officers were busy with their normal routine jobs. A Project Coordination Unit (PCU) was therefore set up in the place of the PMU and the PIUs were disbanded. The PCU was headed by a Project Coordinator (PC) assisted by 2 deputies, one for crops and the other for livestock. A separate Monitoring and Evaluation unit was created under ADP. The PCU generally functioned well. The supervision mission which visited the project in February/March 1991 agreed with GOU that in order to improve performance, it was necessary that the PCU be moved from Entebbe to Tororo and two Deputy Coordinators (DC) - one for the north based in Lira and one for the east based in Tororo - be appointed. The PCU's transfer to Tororo was however not undertaken, because it was found to be unnecessary given the requirements for the PC and the head of the M&E unit to frequently travel to the north. The parent ministry also felt that the PC should stay in Entebbe in order to attend to cordination matten in Kampala and Entebbe. It was also felt that the two DCs were providing the required improved field level supervision. This arrangement worked satisfactorily and was later enhanced by a ministerial re- organization which put all project activities under one ministry following the partial merger of MAF and MAIF to form the current Ministry of Agriculture, Animal Industry and Fisheries (MAAIF). 4.4 Project organization and management was later complicated by the addition of several more components after the project was restructured. The PCU, however, generally coped well in the circumstances. 4  S. Project Implementation General 5.1 The project was implemented under very tryig conditions. It became effeetive on May 8, 1986 one year later than scheduled due to civil unrest in the project area. All project districts except Tororo, were, at one time or another, distwrbed by sporadic waves of insecurity and civil strife. This included damage or looting of property including project stores and equipment and cattle rustling. Therefore some project components could not be implemented in the designated areas and had to be shifted to safer ones like Tororo. The security situation improved in Lira and Apac districts from 1988; and in Kumi, Soroti and Kitgum districts from 1990 and in Gulu from 1991. 5.2 Time lost due to the above disturbances was never regained. This, coupled with the delays in procurement and the continuous inadequacy of counterpart funds, resulted in the closing date of the project being extended thrice involving a total of three and a half years. The implementation of individual components is discussed below at some length. Input SuppUes 5.3 This was by far the largest component, accounting for nearly 75% of the ised total project cost. The key implementation indicators relating to procurements are set out in Part m, Table 4. Procurements were delayed considerably due to: project staff having inadequate knowledge of ICB procedures; lengthy and bureaucratic procedures for approving the items to be procured, as well as delays in approving and awarding tenders, and in opening letters of credit. 5.4 The early implementation of this component was characterized by sporadic looting of project stores in the insecure areas. This was aggravated by pilferage by some staff. This has however been largely controlled through various measures including dismissals and police arrests as well as an improved stock control system. It is estimated that about USS 270,000 worth of supplies were lost in this manner. Other losses have been incurred as a result of cash losses, bad debts due to non-payment by the buyers, and writing off of stocks which had become obsolete or expired. The losses due to bad debts have now been minimized since sales are being made ex-main store. This system has also considerably helped in reducing the selling and distribution costs incurred by the project. 5.5 Further losses were incurred allegedly because of the poor service rendered by an international firm which GOU appointed to carry out pre and post-shipment inspections of goods being imported into the country. As a result, the project lost financially at the hands of foreign suppliers. Goods were delivered and certified as being satisfactory by the firm, but on delivery to ADP stores they were found to differ from agreed tender specifications in terms of quality, quantity and physical attributes. This was particulary the case with normal hoes, lango hoes, wheel barrows and knapsack sprayers. 5  5.6 Procurement was carried out in four "Buys". All goods ordered in the first three "Buys' have been received. The fourth 'Buy' proceeded smoothly, and unlike the earlier "Buys", it is based on a needs assessment which should minimize the risk of buying unnecessary items as happened in the earlier procurements. Tenders for some agricultural implements whica are now locally available have been cancelled from the fourth "Buy". 5.7 In the early years of the project, inputs were delivered by the project to debtor stores on credit. The debtor stores would then re-sell the inputs to sales centres and retailers at a mark-up and then repay the project. This arrangement was not satisfactory because the debtor stores defaulted on their commitments to pay the project for inputs suppLied. Secondly, the project sold the inputs at fixed discounted prices which were substantially lower than those of similar products in the open market. This resulted in the leakage of inputs to non-project areas. In order to minimize this problem, after the project was restructured in 1991, it was decided to harmonize the project input prices with market prices by selling them at prices which reflected the target retail prices. Inputs were no longer to be sold through debtor stores; the project began to sell them from its own stores and on a cash basis. Research and Surveys 5.8 Adaptive Research: The research activities (mainly foi cereals and tubers), wer intended to be carried out at Serere and Ngetta. However, as these areas were unsafe, the activities were transferred to Tororo District Farm Institute (DFI). In 1989, another adaptive research station was established in Lira. The strategy adopted was to select new technologies through on-station and on-farm trials and develop closer links between research and extension by involving extension staff in trial programs. Shortage of local counterpart funds seriously affected the implementation of the trials. As a result, a large number of trial plots could not be harvested and the participation of extension staff was seriously curtailed due to non-payment of staff allowances. 5.9 Tsetse Fly Control: This component had a delayed start because of the civil strife in the targeted areas during which departmental equipment and vehicles essential for the programme were destroyed or looted. Lack of adequate counterpart funding was also a constraint. Therefore, despite the shift of emphasis in 1990 from experimental studies to actual controlling of the tsetse fly, the component registered little progress. The consultant appointed as a tsetse control specialist declined the job. The project was however able to distribute about 500 locally manufactured traps, teach some livestock farmers the application of live bait, train field staff in bait systems technology, and trap tsetse flies (in areas affected by the prevalence of sleeping sickness). 5.10 FLsh Stock Assessment and Evaluation Study: Progress on almost all the activities on this component was delayed due to insecurity and delays in prociring boats, survey equipment and fish gear. The fishery statistics data collection and analysis were completed in June 1988 and the catch assessment survey in July 1991. Boat building was not carried out. The rehabilitation of landing sites at Namasagali and Lalle did not take place due to security reasons 6  and the subsequent re-location of activities. An improved technique of fish preservation through construction of a smoking house called 'chorkor kiln' was demonstrated. In May 1989 the fisheries staff discovered the presence of water hyacinth in Lake Kyoga. In 1992, when the weed had begun to cause concern to MAAIF, ADP funded the study-tour of two senior research officers to Benin-and Nigeria, to study biological control measures being utilized there. Steps are now being taken to breed weevils and introduce them into the lake. Livestock and Veterinary Services 5.11 Veterinary inputs arrived much later than agricultural inputs. Due to lack of transport and simple equipment like syringes, only a small portion of vaccines were utilized in the early years. Later, large stocks of the drugs were however destroyed because they expired before they were utilised. Procurement of large quantities of acaricide without adequate number of bucket spray pumps resulted in the low utilization of acaricide. However, the situation is reported to have improved later on and more drugs are being utilised; rinderpest control has been extended to a number of modern poultry centers in the project area. 5.12 Before the outbreak of the civil strife, the project area was one of the leading livestock producing regions in the country. But, according to MAAIF, during the disturbances which ensued, the cattle population fell from about 1.3 million in 1985 to only 0.3 million in 1990. With most of the livestock lost, the strategy of disease control alone, as envisaged at appraisal, was not considered adequate for the long term development of livestock in the project area. The emphasis towards the latter part of the project was therefore shifted to the improvement of small ruminants and poultry which are still relatively many in number, are kept by a majority of farmers, and take a shorter time to multiply. Institution Building 5.13 Strengthening of Extension Services: Most of the activities under this component took place at Tororo DFI which was upgraded. Since the Visual Aid Centre at the MAF headquarters in Entebbe had been burned down in 1986, a media resource centre was established at Tororo DFI and produced a substantial amount of audio-visual materials such as cassettes, radio programs, leaflets, posters, pamphlets, calendars, handouts and flip chart sets. Many of these have been translated into various vernacular languages. The project also organized pre-seasonal and quarterly workshops during which extension messages would be disseminated to- staff and production targets set. Due to insecurity, the planned rehabilitation of Arapai Agricultural College and Ngetta DFI was not started till 1991-92. Rehabilitation work is still in progress. 5.14 T & V Extension System: For security reasons this component introduced in Soroti, had to be abandoned and shifted to Tororo district. Implementation faced other problems such as inadequate staff, limited staff housing, and scarce funds for staff allowances and vehicle running. The project provided motorcycles and bicycles for extension staff which facilitated their mobility. However, the SAR-envisaged extension worker to farmer ratio of 1:240 7  was found to be inappropriate and expensive to the project. Therefore, on the basis of experience, a modified extension system was developed in 1990 incorporating some elements of the T & V system while maximizing on the scarce resources available. This was later to evolve into a unified extension approach whereby an individual field extension worker would be responsible to advise farmers on crop and animal production, agro-forestry and fishery. Field extension staff would operate through farmer groups which they would visit on a regular basis. This system was adopted in Tororo district from 1992 and later in other parts of the country. 5.15 Monitoring and Evaluation: The project established and equipped a M&E Unit within the PCU. The unit has been responsible for project level data collection, evaluating and reporting. A management information system has been set up and is working generally well. It has collected data on costs, prices and extension activities. The Unit also prepared an excellent study on how to integrate ADP's activities into MAAIF. The unit has gained recognition within MAAEF and is now serving as a core unit of the proposed M&E unit of MAAIF which will cover all projects being implemented by MAAIF. The unit is well equipped and staff are computer literate both in terms of data and word processing, The unit is well-placed to be a good base for the assembling of sectoral data for planning purposes by the ministry. Their main problem appears to be the difficulty they face in obtaining data which are reliable. 5.16 Staff Training: Almost all components had a training element in one form of another. In all, 87 staff members received training either overseas or in countries within the region. Eleven staff went on long-term trainiig leading to higher degrees or diploma while the remainder attended short-term courses or went on study tours. Masters degrees were obtained in areas such as crop protection, agricultural management, extension, and economics. Short-term courses 'and tours covered diverse areas including project management, information systems, rural development, agricultural economics, monitoring and evaluation, credit, livestock husbandry, extension and adaptive research. 5.17 Saary Incentives: In order to boost the morale of project staff whose remuneration as set out in GOU salary scales was low, the project after long negotiations with GOU, gave staff incentive salary payments largely based on individual performance and responsibilities. Other Component 5.18 Ox-plough Revival Program: The loss of cattle due to cattle rustling significantly reduced the practice of ox-cultivation in the project area. Demand for ox-cultivation facilities is however still strong among the local population. Recently, a project staff member visited Zimbabwe to identify ox-cultivation equipment such as planters and weeders, which would be appropriate to Ugandan conditions. The results from these efforts would complement the restocking and revival of ox-plough cultivation techniques, which are also being supported under the on-going [FAD Smallholders Cotton Development Project, and the proposed IDA Cotton Sub-sector Development Project. 8  5.19 Cassava Mosaic Virus Control Program: This component was included in 1991 following a serious outbreak of cassava mosaic disease in Soroti district which resulted in a serious famine. The ADP, using funds re-allocated under the project, mounted a vigorous campaign against the disease including destroying the diseased plants and undertaking the multiplication and distribution of resistant and tolerant varieties to farmers. The project-provided varieties accounted for about 2.5% of the total stems planted in the project areas. Currently, about 410 acres in Soroti district and 56 acres in Kumi district have been planted under clean, uninfected cassava. 5.20 Pilot Credit Scheme: The credit scheme envisaged to give credit for farming and other rural activities is in its second year of operation having been added late to the project and having experienced a delay of almost one year before being launched. The scheme was conceived as a pilot operation in the districts of Apac, Gulu and Pallisa but there has been widespread demand for it to be extended to the other districts covered by the project. The scheme aimed at delivering credit to a total of 3,000 beneficiaries in two years (including about 50% women beneficiaries). It is unlikely that the target of 3,000 beneficiaries would have been achieved by the time project operations ended. According to the latest progress report on the component, only 1,760 beneficiaries had received loans by the end of April 1993 totalling about Ush 180 million. Of these, 1,086 or 62% are women. Project staff are of the opinion that, women borrowers have proved to be a better risk, repaying their loans more readily than men. Virtually all the lending has been short term, with only 3 medium term loans for threshes. No analysis of loan purpose is readily available to indicate which activities have been fianced but the bulk of the money was reportedly lent for agricultural production. A market rate of interest (positive in real terms) has been applied to the credit extended under this component. While precise loan collection data are unavailable, ADP staff have reported satisfactory loan recovenes. The Cooperative Bank (which is not involved in the initial selection of beneficiaries), has been responsible for channelling of the credits. 5.21 Womens' Activities: In order to promote the economic advancement of women, a specific component targeted towards them was added to the project in 1991. The aim was to encourage women to get together in groups to undertake activities for social and economic gain. Over 400 groups which have been formed in the project areas are involved in crop production, seed multiplication, livestock production and marketing. As groups, they have also been able to access project credit funds as indicated in the preceding paragraph. They have also been found to be a convenient focal point and forum for propagating extension messages since most rural farmers are women. 6. Project Results and Impact 6.1 Overall project performance may be deemed satisfactory, particularly when viewed in the context of the major structural, institutional and financial constraints undermining the economy, in addition to the overwhelming security problems. The project aimed to support agricultural, livestock and fisheries development and to increase rural incomes. While it is difficult to determine precisely how far these goals have been achieved or to what extent the 9  improvements are due to the project as opposed to other important factors like improvements in security and weather conditions, etc., it is reasonable to assume that ADP contributed positively to agricultural growth in the project areas. The inputs supplied, the improved seeds made available and the extension efforts under ADP have all contributed to this improvement. The livestock population, particularly of small ruminants is increasing, partly as a result of project interventions in animal health and breeding; while the fisheries and catch assessments of Lake Kyoga have resulted in steps being undertaken to correct the situation. The M&E Unit is organizing a socio-economic survey of the project areas, which would shed further light on some of the achievements under the project. The noticeable results and impact in various areas are discussed below. Agricultural Development 6.2 Annex 2 shows the improvements in agricultural production in the project districts over the last 8 years. In the absence of fertilizers, production increases largely came from the additional area brought under cultivation, and in some cases, from improved seeds such as finger millet P 224, upland rice IRAT 112, groundnut RMP 12, soybean Namu 1, maize composite Longe 1 and bean MCM 5001 which were developed and or distributed by the project. These varieties have become very popular with the farmers but lack of adequate seed multiplication system is an obstacle to their rapid adoption. Cassava, which is the most dominant crop in the project area, benefited from the prompt action taken by the project in controlling mosaic disease and the supply of disease free planting material. Had it not been for the project intervention, cassava production in the area would have been seriously affected, thus leading to considerable food security problems in Soroti and Kumi districts. Livestock Development 6.3 Project impact was severely curtailed by insecurity and the resultant loss of a large percentage of the cattle population. After security improved, the project stepped up animal disease control and treatment activities in the project area and increased farmers' awareness of improved animal production methods. However, there are no statistics to show to what extent these interventions have led to increased animal populations or improved live/carcas weight, etc. The achievements from tsetse control activities were not very significant as component activities started up rather late. Fisheries Development 6.4 The overall impact of this component was reduced because the fisherie inputs and equipment arrived late and other planned activities were abandoned due to insecurity. Additionally, since most of the activities undertaken were surveys and research, their impact cannot be visible in the short run. However, there was a demonstrated benefit from fisheries inputs supplies. When inputs were received and distributed in 1990 and 1991, the fish catch 10  dramatically increased to 95 tons in both years, as compared to 64 tons in 1989. The catch decreased to 63 tons in 1992 after project financed inputs were no longer available. Capacity Building 6.5 Sinct institution building is a long term process, it is not possible at this time to determine the impact of the capacity building activities (such as training), undertaken by the project. Project financed vehicles, office equipment and some buildings did however provide immediate relief. Additionally, most staff who obtained the training have reportedly found the courses useful in improving their job performance. Most of those trained are still in government service. At the rural level, the women's groups formed with project assistance could support the rural institutions through which development activities could be implemented. Furthermore, if the loan appraisal and supervisory capabilities of the credit scheme continue to be enhanced and loan defaults and delinquency are kept to a minimum, the scheme could form the basis for a rural credit program on a larger scale. 6.6 The project helped improve the morale and work performance of staff by successfully negotiating with GOU to give incentive allowances to staff working on project operations. Thes allowances were performance and seniority based. Formulation of Sector Plans 6.7 The M&E unit conducted a fisheries survey and collected data on prices, costs and extension activities. It also prepared an excellent study of how to integrate ADP's activities into MAAIF. The unit has now become a core unit for the proposed MAAIF ministry-wide unit. The project also provided useful input for the preparation of subsequent development projects, like the Agricultural Extention Project (AEP) and Agricultural Research & Training Project (ARTP). Economic Rate Of Return 6.8 No economic rate of return (ERR) was estimated at appraisal because of difficulties to accurately quantify and match project benefits. No ERR calculation has therefore been carried out for this PCR. 7. Sustalnabilty 7.1 GOU is disengaging itself from input supplies operations leaving the role to the private sector. Input supplies activities started by ADP will therefore continue, but largely under private traders and distributors. However, the other activities of the project such as extension, research, surveys, training and monitoring and evaluation will largely depend on the continuing availability of government financing which, during implementation, has been inadequate most of the time. It is difficult to envisage the situation improving in the near future given the fiscal limitations GOU is operating under. These activities are therefore likely to be unsustainable at 11  present levels. In the immediate short run some of the ADP activities will be incorporated in the on-going IDA and EFAD-funded projects such as the Northern Uganda Reconstruction Project (NURP), AEP and ARTP. But all these projects have a finite life and the ultimate responsibility for continuing the programs will eventually revert to GOU, unless timely measures are introducted to initiate cost recovery mechanisms (with NGOs, farmer groups, etc.), for the delivery of services. Provision for testing various mechanisms for delivery of services have been included under the AEP, and the experiences under ADP re-inforce the importance of implementing these proposals. 8. IDA/IFAD Perfornance 8.1 IDA and IFAD were correct in deciding to support ADP as it targeted an economically disadvantaged part of the country. (It was therefore important that the two institutions help towards balanced national development). However, the project's performance was adversely affected by the poor security situation in the project areas. IDA, somedmes with IFAD participation, actively supervised the project. As shown in Table 8, ten supervision missions visited the project to monitor implementation. In most cases, the same mision members were retained for repeat missions which facilitated continuity in supervision. However the mix of the missions would have been enhanced if agriculturists and financial analysts had been included more frequently. Their absence is probably one of the reasons for the apparent scant attention paid to the agricultural aspects and the delays in seriously addressing the financial accounting and management problems of the project. 9. Borrower Performance 9.1 The Borrower complied with all the major covenants under the Loan and Credit Agreements except those relating to the timely submission of audited project accounts. The accounts were submitted late most of the time. The Borrower also failed to provide adequate counterpart funding which was a major factor contributing to the delay of project activities. Finally, the Borrower does not appear to have adequately managed the technical assistance staff on the project, as evidenced by the rather excessive extension of consultancy contracts even for skills which are readily available locally and could be obtained at a fraction of the cost. 9.2 The PCU was composed of relatively young but well educated Ugandans whose leadership took their work seriously. Besides the early problems related to procurement, and the persistent constraints associated with the accounting and financial management functions, other aspects of project management were generally well executed. 10. Project Relationships 10.1 IDA and 'FAD enjoyed good relationships with the Borrower throughout the implementation of the project. The close cooperation among them enabled changes in the project programme, to take place smoothly, e.g., when ADP was restructured, or when there was need 12  to re-allocate funds arnong categories, or when it was deemed necessary to extend the lifespan of the project. 11. Studies and Consultancies Studies 11.1 As shown in Part m, Table 6B, the project was supposed to carry out two studies, the Fisheries Survey of Lake Kyoga and a Catch Assessment Survey of the same lake. Both were completed satisfactorily, and made recommendations to reverse overfishing which was depleting the fish stocks in the lake. A number of these recommendations, e.g, encouraging farmers to grow fish in ponds, are being implemented. Consultancies 11.2 The project funded the services of 10 long term internationally recruited consultants to help in the implementation of input supply management, research and surveys, extension, monitoring and evaluation, and credit. The consultant hired as a Tsetse control specialist declined the job. The plan to hire a consultant as a Fish Biologist was shelved as capable expertise was available locally. The list of con'sultants and their field of specializations is set out in Part m, Table 6C. All consultants completed their assignments but after lengthy extensions to their contracts. Whereas some of these extensions were warranted, others were probably not justified as the relevant expertise, such as in accountancy, is readily available locally. Additionally, for the expertise which was not easily available, the consultants should have been given time limits within which to train local counterparts to take over from them at the end of their contracts. 12. Project Documentation and Data 12.1 The Credit and Loan agreements and subsequent amendments were well-understood by all parties concerned. The SAR was of a good quality except that it lacked implementation indicators and targets to gauge implementation progress. This reduced its value as a management tool. The M&E unit has a lot of data on project costs, prices, input purchases and food crop output in the project areas. There were however weaknesses in the accounting procedures and internal controls as evidenced by the unsatisfactory reports from the external auditors. The PCU has taken steps to improve the situation and the auditors' report for 1992/93 is expected to reflect those improvements. 13  13. LOWS Learnt 13.1 The major lessons learnt from implementing this project are: risks of insufficient counterpart funding should be assessed more carefully at the time of project preparation and at appraisal; public sector input distribution should be avoided unless the distribution mechanisms are efficient and the prices are competitive with similar inputs being. marketed elsewhere by private channels; as a benchmark for project progress, the key project document, such as the SAR should contain key indicators and expected targets at various stages of project implementation; both the project staff and supervision missions would then use these to measure project progress in crucial areas; and, another lesson learned is that procurement contracts should contain adequate force majeure clauses to protect the parties to the contract and allow a quick winding down of project operations. If a substantial part of the project area is undergoing prolonged periods of insecurity which make project operations impossible, adequately prepared contracts (as suggested above) would facilitate the closure of project activities in a smooth manner. 14  PART II PROJECT REVIEW FROM THE BORROWER'S PERSPECTIVE BANK'S PERFORMANCE DURING PROJECT IMPLEMENTATION 1.1 Bank Performance 1.1.1 The Agricultural Development Project was jointly funded by the World Bank and IFAD. IFAD Loan, 159-UG, amounted to SDR 14.6 million, and IDA Credit 1539-UG, amounted to SDR 10.0 million. IFAD Loan was approved by the Board on 1O' November, 1984, and the IDA Credit was approved by the Board on 8th January, 1985. The Loan and Credit became effective on 8t' May, 1986, after a delay of one year because of insecurity in the project area. 1.1.2 IDA and IFAD enjoyed good working relationships with the Government of Uganda, the Ministry of Agriculture, Animal Industry and Fisheries, and ADP Project Management throughout the implementation of this project. This close cooperation among the Donors and the Government enabled the Project to be extended three times, involving a total of three and a half years. The Project was also restructured in 1991 to reflect the needs of the beneflciaries; the changed circumstances related to agricultural input pricing and distribution arrangements, and the cassava mosaic disease emergency programme. The Donors also approved a new programme by ADP to combat the issue of the water hyacinth on Lake Kyoga. This flexibility and positive response by the 15  Donors to new problems identified by the Project on the ground during implementation is highly commendable. 1.1.3 IDA and IFAD sent ten (10) Supervision Missions to monitor Project implementation. These were rigorous missions which greatly enhanced the performance of the project. Various management and generic issues were discussed with Project managemcnt and other related agencies such as the Ministry of Finance, Central Tender Board etc. and measures to solve them adopted. However, ADP was a complex project. There was the need for IDA and IFAD to field specialised staff for in- depth analysis of the implementation, especially in the Fisheries sector, the Veterinary sector, and Procurement. Some of the problems and issues could not be resolved until after 4 years of implementation of these activities. 1.1.4 During the life span of ADP, there were a total of six (6) Task Managers from IDA, two based in Nairobi originally; three in Washington and finally one based in Kampala. IFAD had a total of three (3) Task Managers based in IFAD headquarters in Rome. The Task Managers assisted Project Staff tremendously, and their visits to monitor major project activities made the staff of ADP to work harder. However, in some instances, some Task Managers expected too much to be achieved by the Project without realistically gauging the very difflcult circuimstances under which staff members were working. 1.1.5 For IFAD, it would be useful to strengthen its capability at the country level. For instance they could have a Desk Offlcer at the UNDP offlces in Kampala to effectively monitor their programmes on the ground. 1.1.6 Before the advent of fax machines, communication between Kamrpal-i/Entebbe and Washington were very slow. The cost of sending faxes, telexes and letters by Speed Delivery 16  Systems are exorbitant. 1.2 The "No Objection" IDA procedures require that projects must flrst obtain "No Objection" from Washington on various Issues related to Project implementation. The only problem here is that whenever there are staff changes in Washington, the Project experiences long delays even on straight forward issues. The Task Manager in Kampala should be delegated sufficient powers to handle urgent problems and thereafter explain his actions to Washington. 1.3 Procurement Procedures The International Competitive Bidding procedure was the major method of procurement under ADP. The Project experienced excessive delays due to the prevailing local tendering procedures. Some actions have been taken by Government to improve on these bureaucracies, e.g. establishment of Ministry Tender Board. IDA should however standardize procurement documents, particularly bidding documents to avoid delays, and the requirement to obtain "the No Objection" for documents which projects have used before. IDA finally organized a Procurement Seminar for all Projects in Uganda in October 1993, which resolved many problems that had bedeviled ADP since 1986. 1.4 Disbursement The Project experienced no problem with disbursement, besides some minor incidence when Withdrawal Applications were returned by IDA to the Project due to insufflcient documentation. 17  1.5 Special Accounts Contrary to intentions of IDA, ADP originally found difficulties in accessing funds in the Special Account which had a deposit of only US $ 200,000 and was initially with B.O.U. ADP would not recelve documents directly from the correspondents or from IDA and therefore did not get timely advice on the on- going transactions. Bank statements were also in arrears yet these were required to seek the replenishment of the Special Account from IDA. Step by Step Procedure ADP followed to withdraw funds from the Special Account through Bank of Uganda 1. Fill Withdrawal Application Form. 2. Draft covering letter to the Commissioner/Treasury Officer of Accounts for the signature of the Permanent Secretary, MAAIF. 3. Forward Withdrawal Application Forms together with signed covering letter to the Commissioner/Treasury Officer of Accounts. 4. Collect Withdrawal Forms duly signed by the Treasury for onward transmission to External Debt Management Office (E.D.M.O.), Bank of Uganda (B.O.U.). * 5. Follow-up of payments with Foreign Exchange Operations, B.O.U; Forward documents to Exchange Control Department, B.O.U; following submissions by E.D.M.O for payment to be effected. 18  EXAMPLE 1. Submission of WA Forms for endorsement by Permanent Secretary, MAAIF, 12 January, 1992. 2. Endorsement of WA Forms by Permanent Secretary, MAAIF, 14 January, 1992. 3. Submission of Forms to Commissioner/Treasury Officer of Accounts by ADP, 14 January, 1992. 4. Submission of approved forms from Treasury to External Debt Management Officer, Bank of Uganda, 20th January, 1992. 5. Submission of endorsed approved forms by E.D.M.O. to Foreign Exchange Operations for verification, 20th January, 1992. 6. Submission of verified forms to E.D.M.O by Foreign Exchange Control (B.O.U.) for forwarding to Foreign Exchange Operations (B.O.U.) - 26th January, 1992. 7. Submission by E.D.M.O of forms to Foreign Exchange Operations of payment: 3rd February, 1992. S. Payment of funds by Foreign Exchange Operations (B.O.U.) to a Supplier. 7th February, 1992. 1.5.1 However, following the Country Implementation Reviews, World Bank recommended that all Projects operate Special Accounts from Commercial Banks. ADP moved its Special Account to Grindlays Bank (U) Ltd (now Stanbic Bank) and operations were smooth and quick thereafter. 1.5.2 The Project took long to grasp the manner In which Special Account funds could be used to meet IDA share of local eypendituires, ADP only started to utilize this system after 19  the Project was restructured in 1991. 1.5.3 Budgetary and Finance Aspects Since the beginning of the Project's implementation, ADP suffered from a chronic shortage of local counterpart funds. The funding problem was further aggravated by delays in releasing funds to ADP. To help overcome this problem, the World Bank and IFAD agreed with the Treasury and the Ministry of Agriculture, Animal Industry and Fisheries, that ADP can go ahead and use revenues from the sale of inputs after the quarterly funds released by the Treasury are exhausted. It was further agreed that ADP use revenues for funding essential Project activities not covered under the existing approved budget, based on clear work plans and budgets. This agreement between IDA/IFAD and the Government in October 1991 enabled the Project to be implemented at an accelerated pace and solved the local counterpart funding problems. 1.5.4 Project Closure The Agricultural Development Project flnally closed on 30th September 1993. According to a Telex Message from the World Bank, dated 28th September 1993, "RE: AGRICULTURE DEVELOPMENT PROJECT (IDA CR:1539 - UG /IFAD LN 159- UG) THE CLOSING DATE FOR THIS CREDIT AND LOAN IS SEPTEMBER, 30th 1993. THIS IS TO ADVISE YOU THAT IN ACCORDANCE WITH CURRENT PRACTICE OF THE BANK, FURTHER DISBURSEMENTS FROM THE CREDIT AND LOAN WILL BE MADE FOR WITHDRAWAL APPLICATIONS RECEIVED AT THE BAINK'S HEADQUARTERS BY CLOSE OF BUSINESS ON J.ANUARY 31, 1994, IN RESPECT OF ELIGIBLE EXPENDITURES INCURRED BEFORE THE CLOSING DATE 20  (PAYMENTS MIADE OR PAYMENTS DUE FOR GOODS, WORKS AND SERVICES WHICH HAVE BEEN PROVIDED PRIOR TO THE CLOSING DATE). PROCEEDS OF THE CREDIT AND LOAN RErvLANING UNWITHDRAWN AFTER DISBURSEMENTS HAVE BEEN MADE IN RESPECT OF THESE WITHDRAWAL APPLICATIO1NS WILL BE CANCELLED." There is need for the World Bank to indicate to Projects, three months prior to the closing dates, what is actually involved in the closing of a Project. These guidelines should spell out the full methodology of recoveries by the World Bank, of the initial deposit into the Special Account; payment to suppliers after the closing date; reconciliation of disbursements; and channels for clearing various outstanding issues which may need the Bank's clearance. 2. Government's Performance During Project Implementation 2.1 Timeliness of the Project Start-up and Implementation ADP was a multi-component project designed to traverse mainly the agricultural, livestock, and flsheries sectors. In effect, the project took off in May 1986, a year behind schedule due to civil unrest and cattle rustling which led to insecurity in most of the area. This insecurity was to plague the region for most of the project's life, largely contributing to the need for its extension of up to three and a half years in total. The prolonged insecurity stagnated movement within the region and kept activity at a very low ebb. Apart from the loss of time which was never to be regained, this greatly negated the project's objectives which were to increase food production, raise farmer and fishermen's incomes and generate data for the formulation of sectoral ar.d nat npal developmental plans. This was in as far as it rendered thc very same project beneflciaries unproductive, reducing them to levels of p.verty, beyond that at appraisal. 21  The project was set-back by the looting of inputs then in store, and the partial destruction and looting too, of those institutions envisaged to greatly facilitate project activities. Meanwhile, there had to be a relocation of many of the activities/components to Tororo and Pallisa districts which were relatively safe. 2.2 Services and inputs Provided by Government 2.2.1 Financial Contributions Of the total project costs, SDR 31.37 million, counterpart funding at appraisal was stipulated to be SDR 6.54 million, 21% of the total costs. By the en(d of the project, government contribution stood at about SDR 2.58 million, about 40% of its planned contribution, and even then releases were characterised by delays, contributing to lost time for the project. Especially as the day to day activities were greatly hinged on the availability of such funds. Table 1. GOVERNMENT FINANCIAL CONTRIBUTIONS Exchange UShs US$ Rate July 1993 -Sept 1993 1,200 118,000,000 98,33 March 1993 - June 1993 1,200 165,000,000 137fiOQ00 Nov 1992 -Jan 1993 1,180 109,647,000 92,921.19 July 1991 - Sept 1992 984 364,000,000 3E,91&70 July 1990 - June 1991 558 151,733,000 271,~94 July 1989 - June 1990 332 70,900,000 213,554.2 July 1988 - June 1989 162 23,000,000 141,75I31 July 1987 - June 1988 60 28,043,000 467,333 July 1986 - June 1987 14 11,983,000 857 TOTAL 1,042,306,000 2,649,437.59 Source: AlP Accounts Section 22  2.2.2 Staffing and Office-space Staff were to be seconded to the project by the respective implementing ministries. This was to a larger extent fulfilled except in the case of thc T & V extension component in which the staff requirement was quite high, and could not therefore be completely met. Office space was a problem that could not be satisfactorily met until the project's prefabricated buildings and uniports hadbeen erected. 2.3 Project Management The organisational set-up envisaged at appraisal was that overall coordination would be done by an Inter-ministerial Coordinating Committee chaired by the Permanent Secretary of the then MAF, a PMU headed by an executive secretary would monitor the day to day activities while PIU's would be established in both implementing ministries. At the district level, project activities would be overseen by a District Steering Committee. However, because of the busy schedules of the officers charged with these responsibilities, this set-up did not perform satisfactorily, leaving most of the work to be handled by the PMU. On the basis of this, overall project management was transferred to a PCU, within which was the M & E component and with the coordinator deputised in both the implementing ministries of Agriculture and Animal Industries and Fisheries. Initially, partly because of the poor deflnition of roles, the PCU seemed to be bogged down by too much work. With the roles of the different components clarified, this situation was rectifled and management fared well. As part of the restructuring of the project, deputies to the coordinator in the north and east were appointed. Later under the Public Service Reforms, the implementing ministries were merged. Both these served to enhance the coordination of prcoject activities. 23  During project design there were areas where, because of lack of local expertise, it w-as realised there would be need to temporarily employ consultants, Indeed 11 consultants In various flelds were contracted by the project. Local understudies were expected to take over at the end of the contracts which in all the cases but one, were extended. :4.4 Staff Motivation Poor staff motivation was one of the risks identified during appraisal. At project inception for instance, the extension staff had lost credibility in the eyes of the farmer due to irregular visits and uncoordinated and often conflicting extension messages from the different departments. The staff were often undermined as they seemed less knowledgeable than the farmers who had the experience to draw from. This was partly because of lack of refresher and in- service training to keep the field staff abreast with new practices and lack of logistics to facilitate the operations of the staff. Therefore in-order for the extension system to regain credibility, these short comings had to be addressed. One of the strategies adopted to enhance staff performance was by training. Staff were given t-he opportunity to undertake the relevant training. Indeed the pool of knowledge gained through such undertakings did revigorate the participation of staff In project activities. When the project was restructured in 1991, additional effort was made to improve staff motivation by provision of performance allowances. However, at all times the PCU did aim to achieve the set objectives of the project, portraying their resourcefulness in activities that were designed to reflect the changing circumstances of the beneficiaries and under which the project was operating while webbing the different components into an interlinked system. 2.5 Fulfilment of dated covenants According to the SAR, audited project accounts were to be availed to the do. or agerc'es not later than six months after the end of a 24  financial year. However as portrayed in the table below, this was never to be fulfilled. Table 2. SUBMISSION OF AUDITED PROJECT ACCOUNTS Fiscal Year Date of Submission Months Overdue 1986/87 27th September 1989 21 1987/88 20th May 1991 29 1988/89 2 nd September 1991 21 1989/90 16th December 1991 11 1990/91 4 th May 1992 3 23rd June 1993 6 1992/93 23rd May 1993 5 Source: ADP Coordination Unit The mid-term evaluation report was set to be out by the 30th September 1988 but was submitted in October of the same year. Planning targets set in annual work plans were to be submitted by the 1st April of every year. However, because this relied heavily on inputs from the different components from the districts, delays in submission from even one component In a district, meant delays in the submission of these reports as was often the case. The project completion report is being submitted more than a month late mainly because of lack of a qualified data analyst to help with the analysis of the data gathered from the terminal socio-economic survey carried out by the M & E component. 2.6 Government Policy Given that the country's comparative advantage lies in the agricultural sector, Government policy interventions have been geared towards creating a favourable environment for especially the 25  small scale farmers, who form the backbone of the economy. The project therefore operated under such conditions except for a few instances here cited, where government policy interfered with project implementation. This was especially in the field of input procurement. The only recommended company to clear and forward government inputs, Transocean (U) Ltd, lacked the capacity to do so efficiently and timely. This was a major handicap to the project on many occasions. Likewise in an effort to protect local companies, Government instituted a policy dictating that locally manufactured inputs should not be imported. This policy led to the cancellation of a number of contracts e.g. for the supply of kidlima hoes and fishnets. The local companies have, however failed to meet the local demand but the policy still holds. Government with the aid of World Bgnk and the [MF has since 1987, been instituting economic reforms aimed at restructuring the economy rendering it more productive. Among such reforms has been the devaluation of the previously overvalued USh vis a vis the international currencies, and the liberalisation of the foreign exchange market. What did operating in such a macroeconomic environment hold for the project? For the largest component, input supply, constituting up to 75% of the total project costs and relying heavily on importation, the devaluations which render imports more expensive locally, meant that the farmers had to meet higher input prices. However in a bid to ensure that the farmers were still attracted to the inputs, the project availed them at lower than market prices. This led to leakages to non-project areas and eventually during the restructured ADP, prices were liberalised and inputs sold to farmers at market prices. 26  In conformity with the economic reforms, interest rates had to reflect the actual costs of lending capital by becoming real and positive (they rose up to 43.2 percent in 1989). The feasibility of the Credit Component, introduced in the restructured ADP, was therefore questionable as the tendency was to think that farmers should get credit at subsidised rates. In spite of complaints about the interest rates which have continued to fall as the economy grows stronger, the Credit Component was well received and is eagerly awvaited in the districts in which it is not yet operational. 2.7 Achievernents and lessons learnt: In spite of having operated in a hostile environment for most of the project life, ADP did indeed make a remarkable effort to fulfil the objectives stipulated at project appraisal. However, given that tangible indicators to achievement were not stipulated in the SAR, it has been rather difflcult to exactly state the extent to which project objectives have been met. It can safely be said though, that without the project, the situation in these parts of the country would have been worse than before appraisal because of the devastation the civil war effected on the infrastructure and productive ability of the population, on which agriculture, the main stay of the peasantry in the region, is heavily reliant. 2.7.1 Beneficiarv coverage and response Awareness of the project and participation in one or more of its activities did gain ground especially during the restructured ADP and with the return of security. Very notable has been the increased participation of the women folk in project activities. The number of women groups have grown from an average 60 per district to 360 in Lira, 123 in Kitgum, 350 in Apac and 100 in Gulu district. The prcjf "t closely interacted with the beneficiaries and was therefo-e able to e ssess their changing needs and problematic areas, ;: t're a-c! reflect them in sequential activities. Some p:ib' c ancl p--ivate institutions were also planned to benefit 27  from ADP. These institutions were to be developed and strengthened. The project did achieve a lot with regard to the training institutions which had been ravaged and rendered non- functional. Ngetta DFI, Ngetta Experimental station, Arapal Agricultural college, Serere DFi, and Tororo DFI were rehabilitated to an extent where they are now carrying out their normal activities and have the capacity te. carry out training for both farmers and staff. Twenty-six courses/workshops/seminars for staff and/or farmers were held during the project's implementation. The plan for Gulu DFI was drawn and approved but its construction was time barred. Another notable achievement has been the establishment of a well equipped media resource centre in Tororo. It has greatly facilitated the education of farmers and staff throughout the project area by way of visual aids, handbooks, flip charts, pamphlets and taped radio programmes. Likewise, Laliya fry breeding centre, Gulu, had to be reestablished from a pool of water, bush, and silted ponds. In addition to these institutions, the problem of office and staff accommodation was greatly alleviated. Prefabricated buildings were put up in Entebbe, Tororo, and Jinja, while uniports were erected in Entebbe, Apac, Gulu and Kitgum. The DAO's offices In Lira and Gulu also had some repair work done on them. Office space so created, did not only serve the interests of the project, but was utilised by other depa'rtmental staff and projects as well. At appraisal, it had been envisaged that staff would live in their work areas in order to enable them visit their farmer groups regularly. The necessary accommodation would be constructed on a self-hflA basis, but with the break down of societies during the period cf iqsecurity, this was not feasible. 28  2.7.2 Institutior Building Monitoring and Evaluation Testimony to the achievements of this component which is part of the institution building strategy of the project, is the recognition it gained throughout the ministry, becoming the model around which the MAJIF is establishing its M & E section. Considering that resistance to M & E is the norm, ADP's M & E had very good working relationships with the project management at all times. It did register recognition within the whole ministry and was referred to by other projects . However because of such exposure and derrand, the staff turnover has been high, much as the said staff remain to serve the interests of government elsewhere. The component did also provide the necessary logistics for the start-up of the MAAIF M & E section. Procurement The procurement unit also gained recognition by the MAAIF, and staff were drawn from it to contribute towards the creation of the ministry's procurement unit. This however does not overlook the fact that at project inception, the staff were not well versed with procurement procedures and had a lot to learn while understudying the TA that was contracted. Extension Methodology The Worla bank's T & V method of extension was introduced on a pilot ba'.is in part of the project area. Having tried it out for a while, the srstai'iabill 4y of the system was questioned and consequently it was r-ol lfied to form the ADP extension programme. It is on this modifed programme that the unifled extension programme, current'y being promoted nationally under the Agricultural Extens-:n Programme (AEP), is modelled. 29  Project Management The ability of project management to ably run an ambitious programlme like ADP has also been given due credit and their skills have be.n utilised in the implementation and formulation of other projects, notably South West Region Agricultural Rehabiltation Project (SWVRARP) and Smallholder Cotton Rehabilitation Project (SCRP). Beneficiary Institutions Realising among others, the cost effectiveness of the group approach to farmers problems, the project throughout its life, encouraged farmers to form groups through which the project would reach them and the spirit of self-help rekindled. That group formation was appreciated by the targeted population has already been alluded to in the case of women. They were found to have a higher affinity for groups and at the same time to participate more actively In group activities than the men. The groups have therefore enlarged the outreach of the project than -if the farmers were to be approached individually. Among the skills imparted to the farmers has been book-keeping which was a necessity for the success of the credit components and sale of Inputs. 2.7.3 Achievements of Project Objectives One of ADP's objectives was to generate data which would be used for sectoral and national development plans. The baseline survey, (1986) was carried out in five of the project districts, Kitgum and Gulu being ormitted because of insecurity. The information generated was utilised in the day to day activities of the M & E component while monitoring the progress of the project, and also in the drawring up of the mid-term evaluation report. The formulation of the 'CR1' alsc dcrived a lot of information from the survey report. 30  The introduction and testing of a trapping technique for tsetse fly control did not take place because it was tested and proven effective in Kamuli district under a project financed by EEC. The same technique was then introduced into the project area as a means of control. The fisheries survey yielded information which has been utillsed in drawing up recommendations and management measures to mitigate the decline of the fisheries industry on Lake Kyoga. Has the project contributed to increased production and incomes? Given the circumstances under which the project operated,it remains very difficult to say to what extent the project did or did not increase production and Incomes. However, there are indicators that the project did contribute to improved well being. Figures 2 & 3 indicate a cumulative increase in aggregate production of some of the crops traditional to the project area. However, drought conditions, increasingly frequent in most of Uganda, leaves a marked reduction in the levels of production in almost all crops during such bad years, e.g. 1986 and 1990. Even then, most respondents to the socio-economic terminal survey attributed most of their improved economic status, if any, to increased agricultural production mainly achieved through Increased hectarage, figure 4. Figure 5 shows the proportion of respondents who acquired various assets in the last four years. From the figure, it is clear that in the districts which were more stable, i.e Pallisa and Tororo, the proportion of beneficiaries who acquired assets were much higher. Acquisitic.n of assets is a proxy indicator of improved incomes. 2.7.4 Sustainability of Project Activities A major fa :tor related to project effects concerns the issue of sustain..bi ity cf project activities. Sustainability in this context rElrites to the capacity of the local economy to 31  continue to generate the beneflts of the project to the beneficiaries on a long term basis following the expiry of the financial assistance period. In the course of implementing ADP a need has been identified to assess the project for sustainability. In undertaking this assessment the project's activities lhave been divided Into three (3) categories namely:- Input distribution, Institutional Development and Research and Surve-s. The establ'kshrcnt of the input distribution system underADPis not meant to be a permanent institutional development but a short run solution to solve the problem of basic beneflciary inputs in the project area. The project does not therefore have a plan for distributirg inputs to beneflciaries beyond the project's closing date. Following the expiry of the project, the local private sector has now taken over the provision of inputs to the farmers. With regard to Institutional Development, components of Extension, Agricultural Training, Adaptive Research and Monitoring and Evaluation, these have now been taken over by GOU either directly or through new projects being implemented in the ministry. 2.7.5 Lessons learnt In designing programmes like ADP the number of components should be reduced for ease of management, and as much as possible, be kept to one implementing institution for better ccordiination and working relationships, otherwise a lot more effort is required on the part of management to foster a desirable working environment, and In so doing, a lot of time is wasted. Al0tl-vugh project sustainability was not given specific cons-deration in the original design of ADP, it became an iTr:pi. rtanr aspect of the project in the course of its inrxp'=-"rntation. The experience of ADP indicates that it Is pc s! :ble to attain project sustainability if it is given specific 32  ccns deration in project design. It also indicates that appropriate macroeconomic policy framework needs to be put in r ace in order to support project sustainability efforts. CouPled with it, there maybe need to consider lengthening the dura;ion of such programmes as by the end of five years, mrnaz activities would not have been fully entrenched into the targc ted population and therefore the sustainability would be qt'es ionalule at the end of the project. Ten years maybe found to h- ve a more marked and longer lasting effect. At rroject design, quantifiable indicators of achievement should be clearly spelt out. Whe-i project management is detached from the bureaucracies, well facilitated and delegated with sufficient powvrs, more is achieved than when they are amidst the system. At the same time good working relationships do reap high'y. Given government's poor reoord for availing the necessary counterpart funding when and as it is required, it is necessary that the consequences be carefully looked into during project planning. The success of an agricultural programme is greatly determined by the timeliness of all activities, of which funds will always be paramount. Wh-erever possible, local consultancies, better versed with lo" I conditions, should be engaged In place of external TA wbfl.-h should be limited to areas where the country is totally lic-ing Even then the TOR's should clearly include that the slkil. be imparted to the local understudy within the St ;4v\ lated period. 33  3. Project Rclat2onshipsE 3.1 Project S3zkcholders: The design of ADP was fairly complex, involving several components and stakeholders. With this level of complexity, the issue and nature of relationships between the various stake holders played a key role in project execution and performance. The stakeholders which were involved in the Project can be categorized into seven as follows:- 1. Funding agencies - IFAD and IDA 2. Implementing Agencies - MOA and MAIF 3. Cooperating Agencies - Ministry of Finance and Economic Planniing (MFEP), BOU, Central Tender Board (CTB), Ministry Of Lands Housing and Urban Development (MOLHUD), Ministry Of Ju-stice (MOJ), Transocean-Uganda Ltd. 4. Other projects. 5. Political Establishments. 6. Private sector agencies. 7. Beneflciaries. 3.2 Funding Agencies: The relationship between the funding agencies, IFAD and IDA, was governed by the DCA and other project documents. To the extent that these formal agreements were adhered to by the project, the Borrower- Donor relationship remained good. But instances of non-compliance resulted in sanctions such as suspension of disbursement by the Donors. However, in times of genuine difficulties the donors showed exceptional understanding particularly during the period of insecurity 34  in the project area and the approval of up to three extensions of the project closing date. Another aspect of Borrower-Donor relationship was in regard to Task Management. The project saw a very high turnover of Task Managers. Each of the Task Managers had a slightly different approach and personal peculiarities. Depending on the approach and personality of the Task Manager, the relationship was either good or cool. 3.3 Implementing Agencies: According to design, ADP was planned to be implemented by the then Ministry of Agriculture and Forestry, and Ministry of Animal Industry and Fisheries. Before the two Ministries were merged to form the Ministry of Agriculture, Animal Industry and Fisheries, relationship between the project and MOA was excellent except with MAIF. This was partly because the Project Coordination Unit (PCU) was located in MOA, which caused MAIF to feel that they were a junior partner in the project and therefore gave the project only limited support. However, when the project was restructured and a Liaison Officer from MAIF was included as a full time members of PCU, cooperation and support from MAIF improved significantly. The lesson that should be learnt from this experience is that projects which cut across several implementing agencies require additional effort in coordination. If this cannot be established it would be better to formulate 35  smaller projects which are implemented by the same agency. 3.4 CooperatingAgencies: Cooperating agencies are those organisations that were independent of the project but were involved in some aspects of implementation or facilitation of the project. Their roles and the nature of the relationship were as follows:- a. MFEP was involved in the project in four main areas namely; approval of annual budgets, release of counterpart funding, processing of withdrawal applications, and certifications of audited accounts by the Auditor General's office. Although there were often complaints of inadequate allocation of counterpart funding and delays in release of funds and carrying out other roles, generally the relationship with MFEP was good. b. BOU's role in the project cor'prised; operation of the Special Account, processing of payments through Drafts, LCs etc, processing of import documents and maintaining of the project collection account. The relationship between the project and BOU was marked by cooperation and understanding. c. CTB was concerned with award of contracts in regard to goods, services and works. At the beginning of the project when CTB meetings were being held monthly, approval of contracts would take unacceptably long time. This strained the relationship between the project and CTB. Later on, how ever, when meetings were convened weekly, the perfortrance of CTB improved and a more cordial relationship began to prevail. d. MO>',UD in it's interaction with the project was involved in the approval of building plans, evaluation of bids and 36  supervision of construction for prefabricated housing and evaluation of consultancy bids for Gulu DFI. The performance of MOLHUD in regard to the approval of plans and evaluation of construction tenders for prefabricated housing was marked by long delays. This had a negative impact on the relationship betw.ven the project and MOLHUD. However, the performance of the Ministry In regard to the Gulu DFI was relatively more proficient. e. MO.1 involvement in the project was in connection with the app-oval of the contracts to be entered into by the project. Throughout the Ufe of the project relationship with MOJ remained amicable. f. Transocean (U) Limited was a clearing and forwarding agent appointed by the Government to clear project goods. It's performance was far from satisfactory and this led to a cool relationship for almost the entire life of the project. During the final years of the project, through Country Implementation Reviews (CIR) Workshops, some of the shortcomings in these agencies were addressed. For instance special accounts were transferred to Commercial Banks thus removing th-e role of MFEP and BOU from operation of special accounts. The role of MOLHUD in approving plans and evaluating bids for building was curtailed and transferred to appointed private consulting firms. 3.5 Other Prcjects: There were no major projects in the project area with which ADP interacted. But outside the area South West Region Agricultural Rehabilitation Project (SWRARP), had a significant relationship with ADP. SWRARP learnt a lot from ADP experiences in the area of procurement, monitoring and evaluation, adaptive research and rural credit. This collaboration and relationship w-.s a contribution to the achievement of the 37  project strategy of institutional building. 3.6 Political l;Zstablishw.mnts: The project interacted with the political eEtablishment at the national, district and local level through the Resistance Committee (RC) system. The project enjoyed overwhelming stupport from the RC's at the local and district level especially in regard to mobilisation of beneficiaries for project activities, credit recovery, input sales etc. However, due to lack of information or misinformation the relationship with some members of Parliament was marked by unfair comments about the project. With time, this was overcome through increased awareness and information flow. The lesson that should be learnt as far as relationship with political entities is concerned is the importance of regular flow of information and exchange of ideas. Production of newsletters and having a regular forum for interaction would be a suitable way of doing this. 3.7 Private Sector: The private sector agencies with which the project interacted were mainly manufacturers and suppliers of inputs. The relationship with most of these local manufacturers was cordial as long as they were winning contracts but turned sour when they lost. In such instances they accused the project of favouring foreign suppliers. 3.8 Beneficiaries: The relationship between the project and its beneficiaries can be measured in terms of beneficiaries participation. According to the Terminal Evaluation survey the level of p,;rticipation by beneficiaries in two key activities of the project i.e. extension illustrated by figures 38  6 (extension group formation), 7 & 8 (adoption rates) and input purchases illustrated by figure 9. Almost throughoi:t the life of the project the beneflciaries were unhappy about the prices of inputs and the interest rates levied under the Credit Component. 39  4. Consultancies 4.1 The folloving table shows the consultancies which the project employed, the actual man-months provided and the targeted man- months at appraisal: Table 3. CONSULTANTS CONTRACTED UNDER ADP Component/Specialltv Target Actual Actual as Man-months Man-months % of Target 1. Input Management - Procurement Manager 24 48 200 - Project Accountant 24 96 400 - Warehouse Manager 48 60 125 2. Research Surveys - Agronomist 24 48 200 - lTsetse Fly Control Specialist 8 1 1 - Fisheries Stock Assessment Specialist 24 48 200 - Fisheries Statistician 24 24 100 - Fish Biologist 12 0 0 3. Extension - Training Specialist 30 48 160 - T & V Extension Specialist 36 45 125 4. Monitoring and Evaluation - M & E Specialist 30 41 136 5. Credit - Credit Specialist 0 18 Source: ADP Coordination Unit 40  4.2 As can be seen from the Table, 10 long term Consultants were engaged by the project to facilitate the implementation of input supply management, Research and Surveys, Extension, Monitoring and Evaluation and Credit. Because the initial long-term Tsetse Control Specialist cnvisaged to be engaged by the project declined to take the offer of employment by the Project, a short term Tsetse Control Consultant was employed to help review tsetse trapping programmes in the country. The plan to recruit a Fish Biologist was not implemented as local expertise was available. 4.3 All Consultants completed their assignments with the project. It is difficult, however, to specify the extent of the skills transfer to Ugandan staff by the Consultants as no detailed independent assessment has been undertaken on the performance of the TA's. However, with the exception of the Accountant, the management of all components of ADP reverted to Ugandans from March, 1991 up to the closure of the project in September, 1993. The reports by the visiting Supervision Missions during this period have indicated enhanced project performance during this period. This therefore indicates that some skills transfer did take place between the Consultants and Ugandan counterpart staff. 4.4 It is, however, also evident from the Table that with the exception of the Fisheries Statistician, all TA's were engaged for an average of 1.4 times more man-months than initially targeted for in the SAR. WYhile the extension for the period of the Accountant was justifiable on account of the need to maintain flnancial discipline within the project against a background of lowly paid local staff and an initial breakdown in civility due to insurgency, other factors contributed to the extension of other TA's. 4.5 - One of the factors wjhich contributed to this extension had been the non-corairiehensive nature of the institution building objectives. Although the TORs for the TA's included skills transfer, this had in many caw,s been limited to only some of the activities to be undertak E-i by thC. TA's, especially in respect of on-the-job training. In addi ir: even ir: such cases the training programmes were not 41  tied to systematic action plan tailored towards achieving skills transfer to Ugandaris. This lack of clarity therefore meant delayed and inadequate transfer of skills to local staff. 4.6 The contribution of TA's to the project was also affected by the sometimes incorrect perception by the Consultants of the relationship between technical assistance and the project. Most of the TA's scmchov. had the perception that they were employees of the World Bank ra titer than the Uganda Government. This, coupled with the very high renumerations of the TA's on the one hand and the very low remutnerations of the local staff on the other, tended to delay the forgirg of early close working relationship between the two groups. 4.7 ADP has been a multi sectoral project, covering crops, livestock and fisheries. It's effective implementation therefore required an interdisciplinary approach to work. The traditional apathy in the working relationships between particularly the livestock and crop sectors initially made it difficult to forge this approach to work and therefore contributed to the delay in effective utllisation of the TA's. 4.9 There was also a problem with the delivery of technical assistance. There was an over emphasis on the need to recruit expatriate TA's rather than local Specialists. Because of lack of exposure to the local environment, expatriate TA's took disproportionately longer time to adjust to and understand the project area as well as project staff. The process of effective problem identification and development of solutions as well as the treatment of counterpart staff as eq,ual par.icipants in project implementation was therefore unnecessarily delayed and perhaps even consciously stalled by the expatriaite TA's. 4.10 Experinr,'e has alsc shown that some of the TA's were not very knowle*.'e -table i,] their speciflc areas of assignment or lacked initiati-e . v a'i-jul - rogrammes In light of changing circumstances. Such T! ' `iaw t h r-fore to spend considerable time in learning on- 42  the-job tc acqluire the required skills and in so doing delayed the process o' skills tr..nsfer to Ugandans. 4.11 Civil disturbances which characterised the project area during the earlyyeai s of projcc t implemcntationalsoaffected theperformance of the TA's. Up to 1989 and with the exception of Input Supply in the Nortl,crn D.s.r-cts, all TA's had to relocate to Tororo due to Insecurity in other project districts. The process of effective skills transfer h ad therefire to be delayed until after 1989 when normalcy started tc return tt. the affected districts. 4.12 All in all therefore, the experience of ADP indicates that it is possible for Institution building to take place through technical assistanc: in a project environment. However, the lessons learnt indicate that this process can be enhanced through a clear and comprehensive specification of skills transfer process, timely and action plan-oriented delivery of on-the-job skills transfer, secure project environment as well as capable TA's who are familiar with, or can adapt easily to, the project environment. 43  Figure 1 ADP TERMINAL EVALUATION SURVEY LEVEL Of AWA^RNESS Of AO; A