Somalia - North-West Region Agricultural Development (Phase II) Project

Documnt of The World Bank FOR OFFICLAL USE ONLY Report No. 11919 PROJECT COMPLETION REPORT SOMALIA NORTHWEST REGION AGRICULTURAL DEVELOPMENT PROJECT - PHASE II (CREDIT 1538-SO) MAY 28, 1993 Agriculture Operations Division Country Department II Africa Regional Office This document has a restricted distribution and may be used bv recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  CURRENCY EQUIVALENTS -1985 US$ 1 = So.Sh. 39.5 1986 US$ 1 = So.Sh. 72 1987 US$ 1 = So.Sh. 105.2 1988 US$ 1 = So.Sh. 170.5 1989 US$ 1 = So.Sh. 490.7 1990 not available WEIGHTS AND MEASURES 1 hectare (ha) = 10,000m = 2.47 acres ABBREVIATIONS EEC European Economic Community GOS Government of Somalia IFAD International Fund for Agricultural Development M & E Monitoring and Evaluation MOA Ministry of Agriculture, Government of Somalia NEAP National Environment Action Plan NWRADP I Northwest Region Agricultural Development Project - Phase I NWRADP 1I Northwest Region Agricultural Development Project - Phase II PMU Project Management Unit PPF Project Preparation Facility GOVERNMENT FISCAL YEAR January 1 -- December 31  FOR OMCIAL USE ONLY THE WORLD BANK Washington, D.C. 20433 U.S.A. Office of Director-General Operations Evaluation May 28, 1993 MEMORANDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT SUBJECT: Project Completion Report on Somalia: Northwest Region Agricultural Development Project - Phase H (Credit 1538-SO) Attached is the Project Completion Report on Somalia - Northwest Region Agricultural Development Project - Phase HI (Credit 1538-SO) prepared by the Africa Regional Office. In current circumstances the report could not be sent to the Borrower for comments. The project continued activities assisted under the first phase approved in 1976 (Credit 635-SO), which were designed to improve small rainfed farms (PCR dated November 18, 1992). Physical implementation ceased after three years because of civil unrest, when about a third was completed. With the outbreak of civil war the project was suspended and the Credit was closed half disbursed. As conditions precluded field visits to the project area after 1988, the ability to prepare a PCR with up-to-date findings was limited. However the project records should have enabled a more in-depth review of the contentious issue of appropriateness of the basic technology of the project. The report was sent by the Africa Regional Office to cofinanciers (IFAD and EEC) for comments, but none were received. It is intended to audit the project. 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 SOMALIA NORTHWEST REGION AGRICULTURAL DEVELOPMENT PROJECT - PHASE II (CREDIT 1538-SO) TABLE OF CONTENTS Pagze M. Preface .......................................................... i Evaluation Summary .................................................. iii PART I: PROJECT REVIEW FROM BANK'S PERSPECTIVE .................. 1 Project Identity . ............................................... 1 Background ................................................. 1 Project Objectives and Description .................................. 2 Project Design and Organization .................................... 3 Project Implementation . ......................................... 3 Project Results ................................................ 6 Project Sustainability . ........................................... 6 Bank Performance . ............................................. 7 Borrower's Performance ......................................... 7 Project Relationship . ........................................... 7 Consulting Services . ........................................... 7 Project Documentation and Data .................................... 7 Lessons Learned . ............................................. 8 PART II: PROJECT REVIEW FROM BORROWER'S PERSPECTIVE .... ......... 9 PART III. STATISTICAL INFORMATION ................................ 10 Table 1. Related Bank Loans and/or Credit ............................. 10 Table 2. Project Time Table ...................................... 11 Table 3. Credit Disbursement ...................................... 11 Table 5. Project Costs and Financing ................................. 16 Table 6. Project Results ......................................... 18 Table 7. Status of Covenants ...................................... 20 Map Nos. 18256 18257 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.   - 1 - PROJECT COMPLETION REPORT SOMALIA NORTHWEST REGION AGRICULTURAL DEVELOPMENT PROJECT - PHASE II (CREDIT 1538-SO) PREFACE This is the Project Completion Report for the Northwest Agricultural Development Project Phase II in Somalia for which Credit 1538-SO in the amount of SDR 10.7 million (US$ 10.6 million equivalent) was approved on January 8, 1985. Due to the civil war in the country the project was suspended in March 1991, three months before the official closing date of June 30, 1991. SDR 5.7 million remains undisbursed. The last disbursement was on June 11, 1992. The credit account cannot be closed because the special account cannot be reconciled. In addition to the IDA credit, the International Fund for Agricultural Development (FAD) provided a loan of US$ 7.0 million, and the European Economic Community (EEC) a grant of US$ 5.4 million as cofinancing for the project. Parts I and III of the report were prepared by the World Bank, Agriculture and Environment Division, Country Department II, Africa Regional Office. A draft could not be sent to the Borrower for comment. Copies were sent to the cofinanciers but no comment has been received. Due to the security situation in Somalia, no field visits could take place. The report was based on the Staff Appraisal Report, Development Credit Agreement, supervision reports, correspondence between the Bank and the Borrower, and internal Bank Memoranda.   - iii - PROJECT COMPLETION REPORT SOMALIA NORTHWEST REGION AGRICULTURAL DEVELOPMENT PROJECT - PHASE II (CREDIT 1538-SO) EVALUATION SUMMARY Project Objectives 1. The project objectives were to: (i) increase the small rainfed farmer's income and subsistence through improved production of grains by soil and moisture conservation, adaptive research, extension, farmer training and the supply of inputs; (ii) increase fruit and vegetable production from small irrigated gardens through construction of new gardens, rehabilitation of existing gardens, technical support, research, extension and the supply of farm inputs and; (iii) strengthen local institutional capacity to implement projects and to improve services to farmers. Proiect Design and Preparation 2. The Bank's involvement in the development of the Northwest region started with the Northwest Region Agricultural Development Project (NWRADP I Cr.635-SO). NWRADP II was designed to further the development of the region. A fundamental flaw of the project was that it was based on the belief that NWRADP I was a success. However, when a completion report was issued in June, 1992, it concluded that project results were inconclusive. An economic rate of return could not be calculated because of conflicting crop data from project management and consultant surveys. The second phase project was aware of the lack of data on crop production and included support for monitoring and evaluation of crop yields. 4. Another lesson drawn from the first project was the finding of a 1982 consultant survey which showed that bunding was most effective on lowland areas but not in areas with less than 3 percent slope. Since lowland areas account for only 16 percent of the project area, the consultants expressed doubts over the widespread construction of bunds, and suggested other soil conservation and water harvesting measures in addition to bunding. NWRADP II took this into account by expanding soil and water conservation works to include terracing and contour hedging as well as bunding. A pilot watershed management program was also introduced to deal with the ecology of an entire area. 5. A Project Management Unit (PMU) established under the first project was to continue to implement the second phase. For longer-term institutional building, the project planned to integrate PMU services within existing national institutions after the mid-term review. However, with the escalation of armed conflict, this mechanism could not be tested out. 6. Security had been an ongoing problem for many years in the Northwest Region. The appraisal mission was fully aware of it. But the full scale outbreak of civil war in the area in June 1988 could not have been foreseen at the time of appraisal although security was recognized as an ongoing risk.  - iv - Project Implementation 7. The project had a good start. The overall implementation during the first year was on schedule and efficient. However the situation deteriorated during 1987 because the Govermment replaced the PMU general manager with a person who was not qualified. This was aggravated by the political and social unrest. In mid-1988, fighting erupted in the regional capital. PMU staff were evacuated to Mogadishu, and all field activities stopped. When the security in the northwest did not improve, the Ministry of Finance requested the Bank to terminate the project in December 1990, six months before the official closing date. However, the Government was overthrown shortly afterwards and IDA suspended all ongoing credits to Somalia in March, 1991. 8. About 65 percent of the planned bunding was completed. Due to the civil war and the weakness of PMU's monitoring and evaluation capacity, only one yield survey was carried out, which was insufficient to verify the economic returns of bunding. Little data was available on the expansion of dryland farming extension and research. The irrigated component developed 75 new one-hectare irrigated gardens and rehabilitated an equal number of them, exceeding appraisal targets of 50 in each case. Under the water component, a total of 23 water points for human and 15 for livestock were built. It was estimated that about 80 percent of the planned quantities of farm inputs and equipment supplies were procured, although many reportedly did not reach the project area. To support the PMU, the project provided vehicles and office furniture as planned. The use of technical assistance was about 50 to 60 percent of the appraisal provision. The project also trained 15 staff overseas. Project Results 9. The project was not completed because of the civil war. In the absence of production and other data, no conclusion can be made on the appropriateness of the project interventions in terms of technology and institutional-building. However, the project undoubtedly provided impetus to agricultural activity in the northwest of Somalia throughout the 1980s. Sustainabilitv 10. The question of project sustainability remains unanswered since most of the project actions were interrupted. Lessons Learned 11. The main lessons are as follows: (a) Before launching an expanded second phase project, there should be an objective evaluation of the results of the first phase. The Bank should resist pressures to rush prematurely into financing a follow on project; (b) Given the deficiency in monitoring and evaluation in the first project, a yield survey should have been one of the first priorities. However, this was not done until the third project year; (c) Once a project is seriously jeopardized, orderly resolution of issues resulting from the curtailment of contracts should proceed immediately, as occurred in this case. Other winding down activities, such as recovery of the initial deposit of the special account should also be initiated.  PROJECT COMPLETION REPORT SOMALIA NORTHWEST REGION AGRICULTURAL DEVELOPMENT PROJECT - PHASE II (CREDIT 1538-SO) PART I: PROJECT REVIEW FROM BANK'S PERSPECTIVE Project Identity Project Name : Second Northwest Region Agricultural Development Project Credit No. 1538-SO RVP Unit Africa Region Country Somalia Sector : Agriculture Background 1. The Somali economy in the 1980s was dominated by the livestock sector, which provided a living to the large nomadic population and generated 35 percent of GDP. Exports of live animals represented about 80 percent of total exports. Crop production generated 8 percent of GDP and roughly the same share of exports. Rainfed crop production was based primarily upon sorghum. Crop production under controlled irrigation was centered on bananas, sugar cane, rice and maize; production under flood irrigation comprised mainly maize and sesame. 2. The Bank's assistance strategy was to assist the Government of Somalia (GOS) to devise and implement policies and programs to improve the efficiency of the public and private sectors, to improve the macroeconomic framework for domestic and external resource mobilization and allocation, improve pricing and administration and marketing system , and to increase production and exports. Its lending operations included both sectoral adjustment programs supporting policy reforms and productive investments and rehabilitation programs. Over 30 percent of total IDA lending was for agriculture, focusing on agricultural extension, drought rehabilitation, range management and rural development in specific regions. 3. The Northwest region of Somalia covers an area of about 3.5 million ha, but only about 100,000 ha of land (situated mainly along the Ethiopian border) was potentially cultivable. Agriculture was almost exclusively practiced under rainfed conditions. Pressure on land resources was high, and was exacerbated by restricted access of Somali nomads to traditional grazing in the neighboring Ogaden region and the presence of thousands of refugees from disputed border territories. Consequently, the GOS's strategy  - 2 - for development of the Northwest region placed emphasis on a balanced regional development through the sedentarization of nomads and the stabilization and economic exploitation of available natural resources, particularly by improving rainfed agriculture. 4. The Banks's involvement in the development of the Northwest region started with the Northwest Region Agricultural Development Project (NWRADP I, Cr. 635-SO). NWRADP I was implemented between 1976 and 1985 and at completion, it was considered a successful project, although no formal evaluation had been done. NWRADP II was designed to further the development of the region. Project Objectives and Description 5. Objectives: The main objectives of NWRADP II were to: (i) increase the small rainfed farmer income and subsistence through improved production of grains (sorghum and maize) by soil and moisture conservation, adaptive research, extension, farmer training and the supply of inputs; (ii) increase fruit and vegetable production from small irrigated gardens through construction of new gardens, rehabilitation of existing gardens, technical support, research, extension and the supply of farm inputs and; (iii) strengthen local institutional capacity to implement projects and to improve services to farmers. 6. The first two main objectives of NWRADP II were identical to those under NWRADP I, whereas the third objective emphasized the integration of the various support service operations into the national institutions, such as extension, input supply, credit and marketing. 7. Description. NWRADP II comprised the following activities: (a) expansion of the present soil and water conservation works onto 25,000 ha of accessible cultivable farmland in the region and a pilot watershed program based on selected catchments; (b) provision for, and expansion of, advisory services to all dry land farmers in the project area; (c) the improvement and operation of the adaptive research facilities of the Aburin rainfed crops experimental farm; (d) the development of an additional new 50 small irrigated one-hectare farms; (e) the provision for, and expansion of, advisory services to existing and newly developed small irrigated farms; (f) the establishment and operation of a new site for irrigated adaptive research; (g) the establishment of 100 human water points and 50 livestock water yards using boreholes equipped with hand pumps wherever possible; (h) the provision of annual farm inputs, ox- drawn farm implements and irrigation equipment in conjunction with the private sector and also where medium term credit was required with the Somali Development Bank; (i) the provision of technical services and training; (j) the continuation of support to the Project Management Unit (PMU), and; (k) the continuation of financial support to a pilot research program and preparation of the semi-mechanized rainfed agricultural development at Kurtun Waarey and Sablaale. 8. In 1990 it was decided by the Bank and GOS to add another component to support the preparation of a National Environment Action Plan (NEAP). 9. Total project costs over a five-year period, were estimated at US$ 25.6 million. The IDA- approved credit for US$ 10.6 million was to fund 42 percent of these costs, while IFAD, EC and the GOS were to contribute US$ 7.0 million (27 percent), US$ 5.4 million (21 percent), and US$2.6 million (10 percent) respectively. 10. Project funds were expected to be invested in rainfed agriculture components (47 percent of base costs), irrigated components (14 percent), farm input and equipment supply components (17 percent) and  - 3 - support to the PMU (12 percent). The balance of the base costs was to be utilized for water points (3 percent), training (4 percent) and semi-mechanized rainfed agricultural development (3 percent) components. Project Design and Organization 11. A fundamental flaw of the project was that it was based on the belief that NWRADP I was a success. Although a project completion report had not been done at the time, the preliminary judgement was that NWRADP I was successful because project data on yields indicated significant increases and the actual bunding achieved was much higher than appraisal estimates. However, when a completion report was finally issued in June, 1992, it concluded that project results were inconclusive. An economic rate of return could not be calculated because of conflicting crop data from project management and consultant surveys. The second phase project was aware of the lack of data on crop production and included support for monitoring and evaluation of crop yields. 12. Another lesson drawn from the first project was the finding of a 1982 consultant survey which showed that the effectiveness of bunding varied for different terrain. It found that bunding was most effective on lowland areas but not in areas with less than 3 percent slope. Since lowland areas account for only some 16 percent of the project area, the consultants expressed doubts over the widespread construction of bunds. In their view, other soil conservation and water harvesting measures would have an impact on crop production at least as significant as pure bund construction. NWRADP II took this into account by expanding soil and water conservation works to include terracing and contour hedging as well as bunding. A pilot watershed management program was also introduced to deal with the ecology of an entire area. 13. Project implementation arrangements were also similar to those of NWRADP I. A PMU established under the first project was to continue to implement the second one. It was a unit separate from the existing government structure, and was headed by a Somali General Manager supported by an expatriate technical assistance team. While this arrangement facilitated project implementation in the short-term as consultants were hired whenever there was a skills deficiency, it did not contribute to longer-term institution building. NWRADP II recognized the importance of a gradual transfer of the various services, such as extension, research and input supply, to the Ministry of Agriculture (MOA). The integration of PMU services within existing national institutions was to be considered at the mid-term review. However, with the escalation of armed conflict, this institution-building mechanism could not be tested out. 14. Security had been an ongoing problem for many years in the Northwest Region. The appraisal mission was fully aware of it, but since NWRADP was successfully implemented despite the security problems, it was considered an acceptable risk. The SAR discussion on project risks focused on the provision of salary incentives, housing and other benefits to reduce hardships to project staff. The full- scale outbreak of civil war in the area in June 1988 could not have been foreseen at the time of appraisal although security was recognized as an ongoing risk. Project Implementation 15. The IDA credit for the project was approved in January 1985 and became effective in July 1985, one month after the NWRADP I closed. The project had a good start. The overall implementation  - 4 - during the first year was on schedule and efficient. By September 1986, 4,159 ha had been bunded, 17 percent of the target. Four water points for human and 12 for livestock were completed, and 12 were under construction. Preliminary designs for civil works were submitted to and agreed by the donors. Most machinery and equipment were on order and some were ready for shipment. All aspects of the project were proceeding as foreseen at appraisal. 16. The situation deteriorated during 1987 because of a decline in the dedication and capacity of project management. Without notifying IDA and IFAD, GOS replaced the PMU general manager in May 1987 with a person who did not have the technical competence and experience required for his position. As a consequence, implementation during the remainder of 1987 was unsatisfactory. Actions agreed with the Bank mission were either ignored or delayed. Little progress was made on the Pilot Watershed Management component. Building and training programs were behind schedule. The standard of field supervision by PMU deteriorated. Moreover, seven vehicles purchased under the project were missing. Simultaneously, the working conditions in the project area began to deteriorate as a result of the aggravating political and social unrest, leading to a high turnover of project staff. Attempts to retain staff by offering salary and other incentives failed. 17. In mid-1988, fighting erupted between the opposition and government troops in Hargeisa, the regional capital. PMU staff were evacuated to Mogadishu, and all field activities stopped. GOS held a review of the project and submitted to the Bank an action plan for continuing some project activities in the Awdal region, where security was better. A Bank supervision mission found the action plan impracticable and recommended that project activities be confined to: (i) resolving issues resulting from the forced suspension of contracts; (ii) the preparation and implementation of the external training program as identified in the SAR; (iii) restarting some limited activities on a small scale, with local resources in the Awdal region which would also serve to maintain a presence of the project in the north and facilitate the assembly of scattered staff and equipment; (iv) providing financing to meet the remaining contractual financial commitments, financial obligations resulting from the cancellation of these contracts, the services of the financial controller, external training, and urgently needed repairs and spare parts for the limited operation in the Awdal Region; (v) the preparation by the financial controller of a detailed and specific list of items for IDA/IFAD financing which would follow the above principles. 18. When the security in the northwest did not improve, the Ministry of Finance requested the Bank to terminate the project in December 1990, six months before the official closing date of June 30, 1991. Steps needed to recover the special account and pending expenditures from the IDA credit and IFAD loan were agreed with the Bank. However, shortly afterwards, the civil war spread to Mogadishu. With the overthrow of the Government in early 1991 and the country's inability to service its debt, IDA suspended all ongoing credits to Somalia on March 15, 1991, including NWRADP II. However, since the special accounts were not reconciled, the credit accounts could not be closed. 19. The final project costs cannot be estimated due to a lack of data on GOS financing. IDA disbursement was US$6.3 million, 59 percent of the credit amount. IFAD disbursement was US$4.6 million, 66 percent of the loan amount. EEC disbursement was about US$2.0 million, 37 percent of the grant allocation. Data in the Bank's files showed the physical achievement of each component as follows. 20. Rainfed component. The initial targets for aerial photography and detailed mapping, financed under the PPF, were respectively exceeded. Based on extensive discussions during supervision missions, the Bank and project management reached agreement on substantial issues of bunding, including their  - 5 - location and design. By the end of 1987, about 65 percent of the appraisal targets for bunding were met. 21. In 1987, a Bank supervision mission observed the increasing extent of bunded but uncultivated land. No agronomic causes could be found at the time to justify fallow on bunded land. The mission suggested the M & E consultant find out the scope of the land involved, from which GOS could decide whether to engage a social anthropologist to determine the real causes of fallow. 22. Equipment and vehicles to be provided under the project were fully procured, but little data was available on the expansion of dryland farming extension and research. Extension was handicapped by lack of staff mobility and housing problems. There were not many extension activities till mid-1987. 23. Irrigated component. Seventy-five new one hectare irrigated gardens were developed and an equal number of them rehabilitated, exceeding appraisal targets of 50 in each case. It is not known whether the horticultural extension service was expanded to five teams as indicated in the appraisal report. Eighty ha of land was developed at Baki for the new horticultural research farm as against an appraisal estimate of 30 ha only, but the construction of housing, storage facilities and other buildings was not completed. 24. Water Points. The construction cost of human water points was much higher than appraisal estimate mostly because of the expenses on waterproofing. The management therefore decided to use containers to reduce the cost of the waterproofing problem. Five sites were selected for using containers in 1988. Due to the war, the project could not reach the appraisal target. A total of 23 human water points were built at completion, 23 percent of the appraisal target. 25. For livestock water points, siltation control was the major problem at the beginning. Following the recommendation of the Hydraulics Research Institute, the siltation rate was reduced and so was the construction cost. The management planned to import vetiveria grass for silt filtration since it was used successfully in soil conservation projects in other countries. At completion, 15 livestock water points were finished, representing 30 percent of the target. 26. Farm inputs and equipment supplies. Input distribution at the beginning was confined to irrigated areas where management supplied inputs directly to farmers. At the end of 1987, project management appointed an input supply and marketing specialist to develop the distribution system and to assist in establishing a new credit system. It was estimated that about 80 percent of the planned quantities of farm inputs and equipment supplies were procured, but almost three-fourths of the procured quantities (with the exception of those financed by EC) reportedly did not reach the project area. An estimated 10 percent of the inputs financed by the EC were either damaged, lost or stolen. From the quantities procured and supplied to the project area as stated above, it was estimated that about 60 percent of the needs, as planned at appraisal, for the irrigated area were met. 27. Training. The project was to provide overseas training for 33 staff. 15 staff were sent for training. When IDA operations were suspended in 1991, one student financed by the project was still overseas. Since there was no government authority to submit withdrawal applications to the Bank, the credit could not be used to finance the completion of his studies. For humanitarian reasons, the Bank used its own discretionary funds to finance the remainder of his program. 28. Support to PMU. Vehicles for PMU were procured and house and office furnishings completed.  The use of technical assistance, including short-term consultancies under this as well as under other components was estimated to be about 50 to 60 percent of the appraisal provision. In early 1989, when it became clear that field activities could not be resumed, the contracts of all but two consultants were terminated. The services of the two retained consultants were respectively for the preparation of the contribution to the mid-term review and for the control of finances. 29. PMU's monitoring and evaluation capacity was weak. The Monitoring and Evaluation Unit was not functioning effectively during the first two years. Evaluation of results in dryland farming was critical in deciding the development impact of the project but had been neglected until mid-1987 when the Bank mission insisted that a yield survey should be carried out. A monitoring and evaluation expert was hired to devise tests. The survey revealed no significant difference in sorghum yields between bunded and unbunded fields in the 1987 Gu season. Project management planned to repeat the survey in 1988 using the identical methodology so as to obtain comparable data under different rainfall conditions. However, this was overtaken by the civil war. The economic justification of the investment for the bunds construction, a major part of the project, was therefore not verified. The belief that bunding improved yields was based on the sustained demand for bunding and project recordings. However, there was increasing evidence that a reason for the high demand for bunding was that farmers felt they were gaining the right to land tenure by participating in the project. 30. Other components. Funds allocated for semi-mechanized rainfed agricultural development were fully disbursed. The National Environment Action Plan (NEAP) was prepared. The first drafts of the task force report were scheduled to be submitted by the end of 1990. This did not happen due to the civil war, but a report was prepared by a group of about ten task force members in July 1991 in Nairobi. 31. Cost Recovery. Down payment made by farmers prior to the bunding of their fields was satisfactory. But subsequent collection by the Treasury Department of GOS was much delayed. Till the end of 1987, six months after the amount had been due and payable, the collection work had not started. The partial recovery was an improvement over the first project, which did not recover any costs. Project Results 32. The project was not completed because of the civil war. In the absence of production and other data, no conclusion can be made on the appropriateness of the project interventions in terms of technology and institution-building. However, the importance of the project to the development of the area throughout the 1980s was important in that it undoubtedly provided impetus to agricultural activity in the northwest of Somalia. 33. The study of government recurrent expenditures, geared to help the GOS overcome recurrent cost problems through appropriate reforms in the budgeting and planning process, was not carried out. Also, the terms of reference for the feasibility studies on watershed control and small scale irrigation development, both financed under PPF, were inadequately prepared to ensure the required quality of work. Project Sustainability 34. The question of project sustainability remains unanswered since most of the project actions were interrupted. Cost recovery was started but was incomplete. The extent of the damage to project works  -7- is not known. Bank Performance 35. The Bank's performance until the emergence of serious security problems was generally satisfactory. Since the evacuation from Hargeisa in mid-1988, it had concentrated on finding workable solutions to minimize project implementation problems and to wind down project activities. When the Government was overthrown in early 1991, most contractual issues had been resolved. The other IDA- financed Somali projects that did not go through this winding down period resulted in a number of contractor's claims. However, there was not sufficient time to reconcile the special account and the credit account could not be closed. Borrower's Performance 36. GOS replaced the general manager without notifying the Bank. From then on, implementation progress declined because the new PMU head was not qualified for the job. The PMU's performance was further weakened by the civil war. However, the dedication of the staff was remarkable in spite of the difficult conditions. Other deficiencies in the Borrower's performance included inadequate allocation of counterpart funding, failure to audit accounts on time (the 1985/86 account was not audited till the end of 1987) and the impressment of project vehicles for purposes unrelated to the project. Project Relationship 37. The Bank maintained good relationship with GOS and PMU staff through its resident mission. The Bank's and the Borrower's relationship with other relevant parties, such as IFAD, EC and consultants were generally satisfactory. Consulting Services 38. Under the PPF, consultants carried out: i) an aerial photography/topography mapping; and ii) a watershed management plan/irrigated garden development plan. The performance of the consultants on aerial photography/topography was satisfactory. The recommendations of the watershed/irrigated garden development plan, however, appeared unrealistically sophisticated in the Somali context and far too costly for implementation. 39. The performance of technical assistance consultants was generally satisfactory. Monitoring and evaluation had a good start but could not be completed. Architects satisfactorily assisted the PMU with building procedures until construction activities stopped. Services of a hydraulics research institute were successfully used in improving the design of water points. Project Documentation and Data 40. Project files at Bank Headquarters revealed that the information relating to the IDA credit and the IFAD co-financing was satisfactory but the files contained little to no evidence on coordination of project matters with the EC. Some of the monthly, quarterly and annual progress reports were missing from the files.  -8 - Lessons Learned 41. The main lessons are as follows: a. Before launching an expanded second phase project, there should be an objective evaluation of the results of the first phase. The Bank should resist pressures to rush prematurely into financing a follow on project; b. Given the deficiency in monitoring and evaluation in the first project, a yield survey should have been one of the first priorities. However, this was not done until the third project year. c. Once a project is seriously jeopardized, orderly resolution of issues resulting from the curtailment of contracts should proceed immediately, as occurred in this case. Other winding down activities, such as recovery of the initial deposit of the special account should also be initiated.  - 9 - PART II: PROJECT REVIEW FROM BORROWER'S PERSPECTIVE Due to the civil war in Somalia, Parts I and III cannot be sent to the Borrower for comment.  - 10 - PART III. STATISTICAL INFORMATION Table 1. Related Bank Loans and/or Credit Loan/Credit Title Purpos Year of Approval Northwest Region Increasing farm incomes by 1976 Completed in Agricultural Development addressing the crucial issues of 1985. Project-Phase I severe soil erosion, inadequate (Cr.635-So; US$ 10 extension service, poor million) agricultural practices, poor seed quality, and inadequate distribution of water facilities for both human and livestock. Central Rangelands Project Increasing rangelands and 1979 Completed in (Cr.906-SO; US$ 8 million) livestock productivity by 1988. strengthening management and training. Agricultural Extension and Strengthening agricultural 1979 Completed in Farm Management support services. 1987. Training Project (Cr.905- SO; US$ 10.5 million) Bay Region Agricultural Laying the ground work for 1979 Completed in Development Project increasing crop and livestock 1989. (Cr.972-SO; US$ 12 production in the Bay Region. million)  - 11 - Table 2. Project Time Table Itm Date Planned Date Revised Date Actual Identification May, 1983 May, 1983 Preparation Sep. 1983 Nov. 1983 Dec. 1983 Appraisal Dec. 1983 Mar. 1984 Nov. 1984 Credit Negotiation Oct. 1984 Nov. 1984 Jan. 1985 Board Approval Feb. 1985 Jan. 1985 Credit Signature Feb. 25, 1985 Credit Effectiveness April 1985 July 1985 Jul. 10, 1985 Credit Closing June 30, 1991 Jun. 30, 1991 Credit Completion Dec. 31, 1990 June 30, 1991 Table 3. Credit Disbursement (US$ million) IDA Fiscal Year Appraisal Estimate Actual Actual as % of Annual Cumulative Annual Cumulative Appraisal Estimate 1985 - - 0.3 0.3 3 1986 1.0 1.0 1.8 2.1 27 1987 3.4 4.4 1.9 4.0 38 1988 2.2 6.6 1.8 5.8 54 1989 1.7 8.3 0.2 6.0 56 1990 1.55 9.85 0.2 6.2 58 1991 0.75 10.6 0.1 6.3 59  Table 4 Page 1 of 4 - 12 - Table 4. Project Implementation Appraisal Estimate Actual ' Realisation as % Indicatonr of SAR Estimate Unit Quantity 1. Rainfed Components Under PPF Aerial Photography ha 300,000 580,000 193 Detailed Mapping ha 60,000 60,000 100 Soil and Water Conservation Works Bunding ha 25,000 8,700 35 Initial Ploughing ha 12,500 3,700 30 Civil Works No of basins 3 2 6.7 Equipment and Vehicles Motorgraders No 4 Bulldozers No 8 Tractors No 4 4-WD Vehicles No 9 Lorries No 3 Fuel Tanker Trailer No I Mobile Trailer Homes No 2 . TA (Internationnaly Recruited): Soil Conservation Officer m/m 48 Surveyor n/nm 36 Chief Mechanic mn/m 36 Dryland Fanning Extension . Tractors with assorted machinery for No of Units 5 preparation of demonstration plots 4-WD Vehicles No 6 Scooters No 25 Audiovisual Equipment No of Units I International recruited extension and rn/m 48 training officer I Data collected from project files in the Bank. Much of the information is missing.  Table 4 Page 2 of 4 - 13 - Research Facilities for Aburin experimental farm station: 4-W)D Vehicles No 3 Scooters No I Tractors with assisted machinery No I Prefabricated housing m2 300 Irrigated Components SmaU Irnigated Farm Development New one-ha irrigated gardens No 50 75 150 Rehabilitation of old one-ha gardens No 50 75 150 Hortiultural Extension Service Extension teams to be developed No 5 Support for above teams: 4-WD vehicles No 5 Scooters No 6 Housing m2 500 Storage m2 125 . TA (internationally recruited) Horticulture extn officer mi/m 48 Short-term consultancies m/m 9 New HortiCulCtre Researb Fams Site Development ha 30 80 267 Prefabricated housing m2 825 ) Estimated achievement 70% of Storage and other buildings M2 450 ) planned Vehicles and equipment: 4-WD vehicles No 2 Scooters No 2 7-ton truck No I Tractors with accessories No 2 3. Water Points Human No 100 23 23 Livestock yards No 50 15 30  - 14- Table 4 Page 3 of 4 4. Farm Iuput and Equipment Supply Rainfed Area . Equipment: Ox drawn equipment No of units 1,000 ) Available estimates Barrel churns No 45 ) show that about 80% of Sprayers No 1,000 ) the planned equipment Scraper boards No 1,000 ) and inputs were procured, although most (with the exception of Inputs: ) those financed by EC) Fungicides Kg 3,370 ) did not reach the project Insecticide Tons ~~~~~~~74 area. A smiall part of Insectiizer Tons 74 inputs and equipment Fertilizer TONs 150 ) financed by EC was ) reportedly damnaged, Irngated Ara ) lost or stolen. * Equipment: Pumps No of units 200 Water delivery pipes 200 Hand sprayers 500) Inputs: Urea Tons 480 DAP 480 ) Seeds: vegetables Kg 574 ) potato Tons 138 ) TA (ntenmationaily recruited) Input supply and marketing officer mi/m 48 4-WD vehicle for above No I 5. Training (overseas) Post-graduate training No of staff 3 ) In all 15 staff have been trained (achievement Two-year course in finance I ) 45%) Professional training of 2 to 10 months duration 29 )  - 15- Table 4 Page 4 of 4 6. Project Management Unit . TA Cinternationally recruited): Technical Advisor to GM, PMU mi/m 48 Financial Controller nm/m 36 Short-term consultancies m/m 19 4-WD vehicles No 9 House furnishings No of units 5 Office furnishinga 2 Note: Under the 'Semi-Mechanised Agricultural Development' component, the project provided a lump sum of USS 600,000 to enable the Ministry of Agriculture to continue with the program of testing semi-mechanised rainfed technology. In addition, by mutual agreement between WB and Government, it was decided in 1990 that the project should support the preparation of a 'National Environment Action Plan'.  Table 5 Page 1 of 2 - 16 - Table 5. Project Costs and Financing A. Project Costs Appraisal Estimate Actual Iten L ocal Foreign Total Local Foreign Total .............................. US$ million .............................. 1. Rainfed components 2.8 6.6 9.4 2. Irrigation components 0.5 2.2 2.7 3. Water points 0.3 0.3 0.6 4. Farm Input and Equipment 0.2 3.2 3.4 supply 5. Training - 0.7 0.7 6. Project Management Unit 1.0 1.4 2.4 7. Semi-mechanised Rainfed 0.2 0.4 0.6 Agricultural Development Sub-total 5.0 14.8 19.8 Physical contingencies 0.5 0.7 1.2 Price Contingencies 1.1 3.5 4.6 TOTAL 6.6 19.0 25.6 1/ Relevant data not available. However, in 1990, a sum of US$ 420,000 out of undisbursed IDA credit was allocated for the preparation of the National Environment Action Plan.  Table 5 Page 2 of 2 - 17 - B. Project Financing (US$ million) Sources Planned Actual IDA Credit 1. Civil Works 0.4 0.2 2. Machinery, Equipment, 1.6 2.0 Vehicles and Prefabricated Buildings 3. Input Supplies 2.3 0.3 4. Technical Asst.& Training 1.7 1.5 under Part D 5. Operating Costs under Part 2.4 1.0 A&D 6. Technical Asst.& Operating 0.8 1.0 Costs under Part E 7. Special Accounts 0.1 0.1 Authorized Allocation 8. Refunding of Project 0.5 0.1 Preparation Advance 9. Unallocated 0.8 NEAP 0.1 IDA Credit Total 10.6 6.3 IFAD Loan 7.0 4.6 EEC Grant 5.4 2.0 a/ GOS 2.6 n.a. Total 25.6 n.a. a/ Estimated US dollar equivalent of ECU 2.775 million.  Table 6 Page 1 of 2 - 18 - Table 6. Project Results A. Direct Benefits Appraisal Estimate Estimated at Indicators Closing Without With Project Date 1/ Project Full Devipt Rainfed Agriculture . Sorghum: Area (ha) 9,000 13,500 Yield (tons/ha) 0.8 1.3 Production (tons) 7.20 17,550 . Maize: Area (ha) 3,000 4,500 Yield (tons/ha) 0.7 1.1 Production (tons) 2,100 4,950 . Straw: Area (ha) 12,000 18,000 Yield (tons/ha) 2.0 3.5 Production (tons) 24,000 63,000 No of beneficiaries (including - 42,000 NWRADP I area) Irrigated Agriculture Fruit production (tons) - 350 Vegetable Production (tons) - 225 No of beneficiaries (including - 3,000 NWRADP I area) 1/ No data available.  Table 6 -19- Page 2 of 2 B. Economic Impact Relevant data is not available. C. Financial Impact Relevant data is not available. D. Studies Purpose Defined at Appraisal Status xnpact of Study 1. Aerial Photographic cover and Study financed under PPF and Satisfactory mapping study. completed 2. Feasibility studies on - do - Not satisfactory, but accepted watershed control and small- due to weak Terms of irrigated garden design. Reference. 3. Study of government recurrent Not initiated. None. expenditure to explore the types of isues that the government would have to consider in taking acount of the recurrent cost problems, and examine the reforms that are needed in the budgeting and planning process.  - 20 - Table 7. Status of Covenants Covenants Status 4.02(a) The PMU shall be headed by a general The Government did not consult with the Bank manager whose qualifications and experience shall when it changed general manager in 1987. be satisfactory to the Donors. 4.03 Employ consultants in accordance with the In compliance. Bank's "Guidelines". 4.04(a) Insure the imported goods. In compliance. 4.04(b) Exclusive project use of goods. Seven vehicles procured by the Project was not used for the Project. 4.05(a) Furnish to the Donors promptly the Project In compliance. plans, reports, and contracts. 4.05(b) Establish Monitoring and Evaluation Unit to In compliance. maintain project records. 4.05(d) Project completion report contribution. Not in compliance due to the war. 4.06 Open and maintain Project Account in the In compliance. Central Bank of Somalia. 4.07 Transfer extension services. Not in compliance due to the war. 4.08 By 12/31/1985 open and maintain the Credit In compliance. Revolving Fund with the SDB. 4.09 PMU to (a) acquire and sell to contracted Not in compliance. wholesales and retailers annual farm inputs for onwards sale to farmers; (b) deposit sales proceed into Credit Revolving Fund; and (c) pay sale proceeds to Borrower. 4.10 By 6/30/1988 carry out with the Donors a Not in compliance due to the war. mid-term review. 5.01 Maintain separate accounts and expenditure Submitted late. 1985/1986 account was not audited records. Furnish annual audit report. till the end of 1987. 5.02 Cost recovery. Down payment by farmers prior to the bunding of their fields was satisfactory. But subsequent collection was delayed. 5.03 Furnish to the Donors detailed annual work Submitted late. programs By Nov. I each year.  - 21 - Table 8. Use of Bank Resources A. Staff Inputs 1983 1984 1985 1986 1987 1988 1989 1990 1991 Preappraisal 9.2 35.9 l Appraisal 20.3 10.4 l Negotiation 3.7 Supervision 2.8 19.5 15.3 11.9 4.4 3.1 0.5 PCR 7 Total 9.2 56.2 16.9 19.5 15.3 11.9 4.4 3.1 7.5 B. Missions Stage of Month/Year No. of Specialization Performance Types of Project Cycle Persons Represented a/ Rating Status b/ Problems c/ Preparation 10/1983 1 FA Appraisal 3/1984 4 FA, LS, Ag, LS Supervision 5/1985 1 FA Supervision 2/1986 2 FA, C, Supervision 3/1986 4 FA, F, ES, C 1 F, M, I Supervision 10/1986 3 FA, F, C 1 F, M, I Supervision 1/1987 1 FA Supervision 6/1987 1 FA 2 M, I Supervision 12/1987 1 Eco 2 M, I Supervision 12/1988 1 Eco Supervision 12/1990 1 FA 3 I, L, M a/ Ag=Agronomist, C=Consultant, Eco=Economist, ES=Extension Specialist, F=Forester, FA=Financial Analyst, LS=Livestock Specialist. b/ 1 =Problem free or minor problems, 2=Moderate problems, 3=Major problems. c/ F=Financial, I=Impact, L=legal, M=Management. * No Form 590.     IBRD I