Document of The World Bank FOR OMCAL USE ONLY Report No. 11300 PROJECT COMPLETION REPORT SOMALIA NORTHWEST REGION AGRICULTURAL DEVELOPMENT PROJECT (CREDIT 635-SO) NOVEMBER 2, 1992 Agriculture Operations Division Country Department II 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. CURRENCY EQUIVALENTS At appraisal: US$1.00 = So.Sh. 6.23 PCR: US$1.00 = So.Sh. 35.64 WEIGHTS AND MEASURES 1 hectare (ha) = 10,000 m = 2.47 acres (ac) I ton = 1,000 kg = 2,205 lb 1 kilogram (kg) = 2.2 pounds 1 kilometer (km) = 1,000 meters ABBREVIATIONS ARI Afgoi Research Institute DCA Development Credit Agreement ERR Economic Rate of Return FAO/CP Food and Agriculture Organization Cooperative Program GOS Government of Somalia PIU Project Implementation Unit PMC Project Ministerial Committee PPF Project Preparation Facility SCF Standard Conversion Factor WDA Water Development Agency GOVERNMENT OF SOMALIA FISCAL YEAR January 1 - December 31 FOR OFFIAL USE ONLY THE WORLD BANK Washington, D.C. 20433 U.S.A. Otfice of Diretor-Gonerad Operadon. Evaluation November 2, 1992 MEMORANDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT SUBJECT: Project Completion Report on Somalia - Northwest Region Agricultural Development Prolect (Credit 635-SO) Attached, for information, is a copy of a report entitled "Project Completion Report on Somalia - Northwest Region Agricultural Development Project (Cr. 635-SO)" prepared by the Africa Regional Office. No audit of this project has been made by the Operations Evaluation Department at this time. Attachment This document has a restricted distribution and may be used by recipients only In the pert omnce of their offical duties. its contents may not otherwise be disclosed without World Bank *uthorization. FOR OFFICIAL USE ONLY PROJECT COMPLETION REPORT SOMALIA NORTHWEST REGION AGRICULTURAL DEVELOPMENT PROJECT (CREDIT 635-SO) TABLE OF CONTENTS Page No. PREFACE ........................................ i BASIC DATA SHEET ........................... iii EVALUATION SUMMARY ......................................... iv I. Introduction ........................................ 1 Background ........................................ 1 Project Identification and Processing . .................................. 1 Project Design and Objectives ........................................ 3 H. Project Implementation and Operating Performance ............................ 4 Main Components .......................................... 4 Start-up ........................................ 5 Construction of Field Bunds and Gully Erosion Controls .................... 6 Development of Small Irrigated Farms .................................. 8 Building Construction . ............................................... 8 Construction of Water Points ........... .............................. 9 Crop Trials at Aburin Dry-Land Demonstration and Experimental Farm .... .... 9 Extension of Geed-Deeble Horticultural Station ........................... 11 Expansion of Extension Services and Training Facilities ..................... 12 Consultancy Services and Feasibility Studies .............................. 13 Staffing ......................................................... 14 III. Project Costs, Disbursements and Procurement ............................... 15 Disbursements . ................................................... 16 Accounts and Audits . ............................................... 16 Procurement ...................................................... 17 IV. Institutional Performance and Development .................................. 17 V. Project Impact . ........................................................ 19 Dryland Farming . .................................................. 19 Small Irrigated Farms ............. .................................. 20 Human Water Points ............. .................................. 20 Livestock Water Points .............. ................................ 20 Gully Erosion Control .............. ................................ 20 Project Design Problems ............. ............................... 21 Transfer of Technology ............. ................................ 21 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. VI. Economic Re-evaluation .................................... 21 VII. Bank Performance ..................................... 21 Lessons Learned ............................. 22 ANNEX I - TABLES Table 1 Bunded Areas and Beneficiaries .................................... 25 Table 2 Building Program .................................... 25 Table 3 Extension Staff .................................... 26 Table 4 Project Stafing .................................... 27 Table 5 Credit Disbursement .................................... 28 Table 6 Procurement Procedures and Amounts .................................. 28 Table 7 Sorghum Production in Bunded Area ................................... 29 Table 8 Small Irrigation Farms - Crop Production ................................ 29 Map No. IBRD1119OR PROJECT COMPLETION REPORT SOMALIA NORTHWEST REGION AGRICULTURAL DEVELOPMENT PROJECT (CREDIT 635-SO) PREFACE This is the Project Completion Report (PCR) for the Northwest Region Agricultural Development Project in Somalia, for which Credit 635-SO was approved on June 3, 1976 for the sum of US$10.0 million. The six-year credit was declared effective on February 8, 1977; the closing date was extended for two years from June 30, 1983 to June 30, 1985. Final disbursement was made on January 14, 1986, at which date US$9.95 million had been disbursed. US$50,000 was canceled. This report was prepared by the Agriculture Operations Division, Country Department II of the Africa Regional Office. Since the preparation of the report, the Government has been overthrown. As of the finalization of the report, the country remains in a state of civil war. A draft could not be sent to the Borrower for comment. This PCR is based on a review of the 1973 IDA/FAO preparation report, the Staff Appraisal Report, the Development Credit Agreement (DCA) between GOS and IDA and its amendments dated August 10, 1977 and August 1, 1979, supervision reports, project files, and the Government's progress reports, discussions with project Government officials during field visits as well as consultation with Bank staff associated with the project. The Borrower provided data for the document and assisted in assembling the technical and financial data. - 111 - PROJECT COMPLETION REPORT SOMALIA NORTHWEST REGION AGRICULTURAL DEVELOPMENT PROJECT (CREDIT 635-SO) BASIC DATA SHEET KEY PROJECT DATA (US$ million) Actual or Actual as % of Appraisal Current Appraisal Expectation Estimate Estimate Total Project Costs (USS million) 13. 14.1 101 Credit Amount (US$ million) 10.0 9.95 99 Canceled (USS million) -- 0.05 -- Economic Rate of Return 11% n.a. 1 Institutional Performance Good CUMULATIVE ESTIMATED AND ACTUAL DISBURSEMENTS (US$ million) FY 7S 79 80 81 82 S 3 84 S5 86 Appraisal Estimate 0.7 3.3 6.0 7.5 9.0 10.0 -- -- -- Actual 0.6 2.7 3.6 6.0 7.1 7.9 9.5 9.9 9.9 Actual as % of Estimate 86 82 60 80 79 79 95 99 99 Date of Final Disbursements: January 14, 1986 PROJECT DATES Other Project Data Original Revised Actual Identification (FAO/IDA) - 1111972 Pmrparation (FAO/IDA) - 10/1973 Appraisal - 5/1974 Negotiations - 1/14-16, 1976 Board Approval - 6/3/1976 Credit Agreement Date - 717/1976 Effectiveness Date 10/5t76 12/6176 2/8/1977 1/6n7 2/4177 Closing Date 6/30/83 6/30/R4 6/30/85 1/ Cannot be calculated with reasonable confidence due to the lack of reliable data (see Section VI paras. 6.1.) - Iv - STAFF INPUTS (Staffweeks) FY73 FY74 FY75 FY76 FY77 FY75 FY79 FY80 FY81 FY82 FY83 FY84 FY85 FY56 FY87 FY88 Total Identificationl 2.9 18.9 21.8 Preparation Appraisal 27.3 43.0 4.6 74.9 Negotiations 11.2 11.2 Supervision .3 13.5 7.9 13.0 14.2 12.7 14.6 7.3 7.0 2.2 92.7 PCR 0.4 8.1 5.4 13.9 Total 2.9 46.2 43.0 16.1 13.5 7.9 13.0 14.2 12.7 14.6 7.3 7.0 2.2 0.4 8.1 5.4 214.5 MISSION DATA Date Performance No. of SpecIaizalon Types of (mo./yr.) Rating a Persons Represented b/ Problems Preparation 9173 5 PAAE,LS,EC Appraisal 4/74 4 ECAE,FA Supervision 1 8/76 1 1 AE MT Supervision II 3f77 1 1 A Supervision 11 6/77 - 2 FA, LS MT Supervision IV 3/78 2 1 A OMT Supervision V 11/78 2 2 E, LS MO Supeivision VI 6/79 2 2 A, CS MO Supervision VlI 1/80 2 2 A, E MTO Supervision VIII 3/80 2 1 A MT Supervision IX 10/80 2 1 A FTO Supervision X 3/81 2 1 A TO Supervision XI 10/81 2 2 AE,A TO Supervision XII 4/82 1 2 AE, A TPF Supervision XIII 2V83 1 2 AE,A PO Supervision XIV 9/83 1 3 FA,AE,A 0 Supervision XV 3/84 1 1 A 0 Supervision XVl c/ 9/84 1 1 A OT Supervision XVII 6/85 1 FA OT a/ I -Problem free or minor problems; 2-Moderate problems b/ AE-Agricultural Economist; FA-Financial Analyst; LS-Livestock Specialist; A- Agriculturalist; CS=Conservation Specialist; E-Engineer, EC-Eoonomist, PA-Project Analyst ec From late 1984 to mid 1985, there was a serious cholera outbreak in the Northwestern part of Somalia including the project area which prevented supervision during that period. Concluding supervisions were combined with Phase 11 project. The above does not include project implementation reviews held in Somalia by Government and IDA on an annual basis. OTHER PROJECT DATA Borrower Government of Somala Implementing Agency Northwest Region Agricultural Development Project Implementation Unit under Ministry of Agriculture Fiscal year of Borrower January I - Dccember 31 Follow-On ProJects: Name Northwest Region Agricultural Development Project 11 Credit Number 1538-SO Amount (USSmillion) 10.6 Approval Date January 8, 1985 - v - PROJECT COMPLErION REPORT SOMALIA NORTHWEST REGION AGRICULTURAL DEVELOPMENT PROJECT (CREDIT 635-SO) EVALUATION SUMMARY Project Design and Objectives 1. The Government's Five-Year Plan 1974-78 emphasized self sufficiency in basic food and cash crops. As part of this overall plan, GOS decided to initiate the first phase development project in the Northwest Region which was considered to be among the least developed regions in Africa and the world. 2. The project was designed to increase farm incomes by addressing the crucial issues of severe soil erosion, inadequate extension service, poor agricultural practices, poor seed quality, and inadequate and uneven distribution of water facilities for both humans and livestock. The project was to establish some basic agricultural services for the first time and strengthen others which had been allowed to deteriorate during previous years. It was to assist the trend toward settled agriculture and support the establishment of a technical basis for increasing agricultural production, and training of Somali personnel. An estimated 2,500 farm families were to benefit directly with social benefits accruing to the population in a 200,000 ha area. The project was to be implemented by a Project Implementation Unit (PIU) attached to the Ministry of Agriculture. To support a smooth project implementation, overall responsibility was vested in a Project Ministerial Committee (PMC). Implementation Experience 3. The initial project performance was delayed due to a conflict between the project manager and the consultant management team, and the outbreak of the Ogaden war. The consultants left after two bombings of Hargeisa. The implementation was further delayed by the lack of basic data. The credit agreement specified that the Government should make available topographical maps of the region. This covenant was never fully met because of the resistance of the Ministry of Defence to release photographs of the area for security reasons. The lack of this basic data constrained the planning and execution of the project with regard to selection of areas to be bunded, pilot areas for gully erosion control measures, proper sites for water points and development of small irrigated farms. Eventually, the National Range Agency conducted an aerial survey of the project area as part of its survey of the region. The results became available to the project in 1982, five years after credit effectiveness. Implementation of each project component is summarized below. 4. Construction of field bundinas and gullv erosion controls. Field bunding was carried out successfully. In the absence of maps, the location of priority bunding areas was identified from a farm survey prepared by an experienced specialist. The design and construction generally followed those outlined in the appraisal report. At completion, the bunding component accomplished most of the objectives and benefitted a larger number of farmers than estimated at appraisal. Gully erosion control started in 1979 in selected areas but was discontinued in early 1980, because the - vi - work was not designed well and unlikely to remove the cause of gullying. The main reasons were lack of technical knowledge and qualified personnel. 5. Development of small irrigated farms. Due to the Water Development Agency's (WDA) inability to carry out the test drilling for the water resources study (consultants were eventually employed to do the work), project works did not start until 1982, two years behind schedule. 20 farms were identified and surveyed, and 19 of them were constructed by the end of 1984. The operation was successful. Crops grown on the farms generated substantial incomes. However, the development of 50 additional farms targeted at appraisal was not possible because of the delays in test drilling, and a shortage of funds. At appraisal the cost for development of a one ha farm was estimated at US$2,000. The actual amount was US$7,000 due to the high costs of imported materials. 6. Building construction. Bid response after local competitive bidding was poor with prices double the original estimates. Following negotiations, the work was divided into two parts to be carried out by two qualified local construction firms. The implementation experienced considerable delay due to a lack of building materials, logistical problems and severe water constraints. The building program was reduced by constructing fewer houses for junior and none for unskilled staff. At completion the total cost was So.Sh.7 million against So.Sh.2.2 million estimated at appraisal. 7. Construction of water points. Implementation lagged from the beginning because of the shortage of cement and diesel. With the extension of the closing date, the project continued to build water points with improved designs. At completion, appraisal targets of 100 human and 50 livestock water points were achieved. 8. Crop trials at Aburin dry-land demonstration and experimental farm. The crop trial program was to be improved at the Aburin Farm through the introduction of improved varieties, crop husbandry techniques, pest control, rotations and alternative crops. However, the trials failed with no tangible results. In retrospect, the existing Aburin station should have not been chosen for the dryland crop trials. The remoteness of the farm combined with the lack of water severely limited its usefulness as a research/demonstration/training farm. In addition, housing and training facilities were inadequate. All these factors affected recruitment of qualified personnel. The trials programs suffered due to discontinuity of staff. 9. Extension nf Geed-Deeble horticultural station. This program intended to expand the area under irrigation at the citrus farm and nursery at Geed-Deeble from 10 ha to 30 ha. This objective, however, was not met since the water supply was inadequate and unreliable at the station. A new site for the research station was established under the Northwest II Project (Cr. 1538). 10. Expansion of extension services and training facilities. The main objective was to rehabilitate, strengthen, and expand the existing extension services for dryland farming to a total of 23 extension agents mainly directed towards farmers on bunded lands. Because of staffing constraints there was little extension activity during the first two years. During most of the implementation period, the project had no more than ten extension agents, which was far fewer than planned. But the agents' performance was commendable. In 1981, with active participation of the expatriate horticulturist, extension work on a modified T & V system was carried out. 11. Consultancy services and feasibility studies. The original idea of having one firm provide all services did not materialize. Due to the harsh living conditions in Hargeisa, it was difficult to - vii - maintain a complete team all the time. The project had to recruit consultants on an individual basis. Discontinuity of the services of certain specialists delayed project implementation, but the overall performance of the consultants was considered satisfactory. They prepared the feasibility study of the Northwest Region Phase II Project, and provided good technical services. Project Results 12. The main project activity was bunding, which was to result in soil conservation and moisture retention and thus increase crop yields. While the physical targets were attained, the impact of bunding on crop production remains inconclusive because of conflicting and inadequate data. According to project management, bunding caused a yield increase of 75 percent, or 38 percent more than the appraisal estimate of the yield increase resulting from the project. Other consultant surveys, however, questioned the project management's methodology and gave a more guarded assessment. A consultant survey in 1982 showed that the effectiveness 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. Another consultant survey in 1987 showed that the impact of bunding on yields was minimal. However, it recognized that the survey was only based on one season's crop cut data. It was to conduct another survey the next year, but in 1988, the civil war in the north intensified and all project personnel were evacuated. The survey findings also indicated that there was a need to examine issues of bund design, bund maintenance and agronomic practices in order to realize the full potential of bunding. 13. The accomplishment of the project in developing successfully small irrigated farms for the production of fruits and vegetables for local consumption and potentially for export have to be acknowledged. The demonstration effects on other farmers in the areas have been considerable, attested to by the significant number of similar irrigated farms which developed in the area without project assistance. An estimated 5,300 farm families were assured water provision throughout the year. The production of alfalfa, groundnut and groundnut hay as projected did not materialize. The gully erosion control work was abandoned because it proved to be prohibitively expensive and could not be maintained. 14. Due to the conflict in production data, the economic rate of return could not be calculated with reasonable confidence. From a long term point of view, the project can be considered satisfactory since it has laid the groundwork, and provided an infrastructure and administrative framework on which to continue improvements of food production and water supplies and to lessen the reliance on food imports and/or relief donations. The experience gained under this project should help to encourage the settlement of nomads and improve dryland farming. Furthermore, the project has helped to fill many gaps in the knowledge about the region through the resource inventory undertaken by the National Range Agency and the test drilling for groundwater studies and quantification of such resources. Sustainabilitv 15. One main issue relates to the economics and financial sustainability of project benefits. The economic issue will be clarified only over time. Cost recovery was not undertaken under this project. The project unit, with the assistance of the consulting firm preparing the second phase, -viii - had submitted proposals for cost recovery. The implementation of these recommendations would have resulted in the collection of about So.Sh. 18.0 million. However, GOS requested to delay collection by one year as a result of the 1984 drought, and no cost recovery was carried out by project closing in 1985. GOS agreed to commence cost recovery in the second phase project but attempts to this effect were frustrated by the severe civil disturbances. Furthermore, given the uncertain impact of bunding on yield increases and the ongoing civil war that brought most developmental activities to a halt, the prospects for sustainability are not good. Lessons Leamed 16. The lessons leamed are summarized as follows: - The project's focus on an appropriate technology was the key to its success despite very difficult circumstances. Although the impact of bunding is still inconclusive, the project has yielded useful findings about the technology (para. 12). Building on this experience, the second phase project was to extend bunding to areas where it was appropriate and complement it with other water and soil conservation measures. - Adapting a technology to a region requires a long time. After an eight-year effort to develop rainfed farming in the Northwest Region of Somalia, the project had reached a stage where it should be possible to determine the effectiveness of the bunding technology. Unfortunately, in large measure because of inadequate monitoring and evaluation, it was not yet possible to determine the technical effectiveness and the economics of bunding. It was therefore important for the follow-up project to ensure that: (a) the experiences made under the project were not dissipated; and (b) the cost/benefit of bunding was properly monitored and the results suitably assimilated into the strategy for developing agriculture in the country. - Strong Govemment commitment and a dynamic project manager are essential to project success. Despite the difficulties with security and the project's remote geographic location, the project was satisfactorily carried out. - Demonstration effects are important. If appropriate technology is introduced, farmers can find ways to develop with their own resources even in the least developed areas. - Covenants aiming at ensuring the availability of certain essential project requirements (such as the responsibilities of WDA) should not have been deleted and merely referred to in the Agreed Minutes of Negotiations. - The difficulty of recruiting consultants to a geographically isolated area should be recognized in the project design. Provisions for housing and other support should be adequate. - The thorough review by the various Bank departments prior to finalizing a project proved extremely valuable in identifying important issues and resulted in adjustments in the project design. The regular supervision of the project assisted in maintaining the momentum of project implementation through a flexible application of Bank guidelines and procedures to overcome specific project related difficulties. PROJECT COMPLETION REPORT SOMALIA NORTHWEST REGION AGRICULTURAL DEVELOPMENT PROJECT (CREDIT 635-SO) I. Introduction Background 1.1 The Governmnent's Five-Year Plan 1974-78 emphasized, among other priorities, self sufficiency in basic food and cash crops. As part of this overall plan the Government of Somalia (GOS) decided to initiate a first phase development project in the N.W. Region of the country which is considered to be among the least developed regions in Africa and the world. Because of its geographical isolation, with its considerable distance from the more productive areas of the country, exacerbated through the poor infrastructure and communication links, the Government attached high priority to improve the standards of living for the population in northern Somalia. Particular emphasis was placed on regional food grains self-sufficiency by promoting increased production of sorghum and maize through soil and water conservation, irrigation and extension services to farmers. A gradual transition from nomadism to settled farming was also encouraged. The severe 1973/74 drought highlighted the urgency for a speedy realization of these goals through the Northwest Region Agricultural Development Project as well as other donor financed operations. This project was the Bank Group's third agricultural project in Somalia. The others were the Trans-Juba Livestock Project (Cr. 462-SO; $10 million), which was targetted at the nomadic population, and the Drought Rehabilitation Program in the Juba-Shebelli region (Cr. 623-SO; $8 million). Project Identification and Processing 1.2 In May 1972 a joint Bank-FAO/CP mission visited Somalia to review project possibilities in the Northern Region and specifically assessed the need for further preparatory work on an earlier identified project in the Northwest Region. This tentative project included provisions to control soil erosion and maximize rainfall use, improve and extend staff and farmer training and adaptive research facilities, develop seed multiplication facilities, improve extension services and input distribution facilities, construct social infrastructure facilities and support the development of sedentary agriculture, improve existing small-scale pumped and gravity-fed irrigation systems and study the potential of new areas for similar developments. In addition, the project was to provide Technical Assistance to study and prepare investment proposals for further phases of this program and for other dryland areas of Somalia. Direct project benefits would derive from increased production of sorghum and other food crops and from diversification into cash crops. The main issues were organization and management, project size and the supply of inputs and cost recovery. 1.3 In late 1973 FAO/CP assisted GOS in the preparation of a project with the following scope: - Construction of bunding works over 20,000 ha, repair of existing field bunds over 10,000 ha so as to increase the retention of rainfall and stop erosion, and increase crop production on about 40,000 ha. - Establishment of an effective extension service, improvement of training facilities at the Aburin dry-land research farm, and development of improved seed production and distribution. - 2 - Establishment of a model farm at Tug Wajale to investigate large-scale mechanized wheat production and rotation with fodder crops for livestock fattening. Development through credit of 50 small-scale irrigated farms. Construction of 100 water points for water for human consumption and 50 water points for livestock. - Construction of additional grain storage facilities and of a small wheat mill. - Feasibility studies for irrigation, date development and an integrated dryland farming project in the Upper Juba Region. 1.4 Costs were estimated at So.Sh 46.4 million (US$7.5 million) over a four-year period. Foreign exchange requirements were about 65 percent. At full development annual sorghum production was expected to increase by 50 percent to 5,500 tons, 1,500 tons of wheat and an additional 235 tons liveweight of cattle; increased production of citrus fruit of about 400 tons and 100 tons of vegetables; the annual net production was estimated at So.Sh. 3.1 million. The economic rate of return was calculated to be 8 percent. 1.5 The issues centered on the low rate of return, project organization and management due to the shortage of experienced managers and technicians in Somalia, land tenure and the wheat farm included at the specific request of GOS. Queried were also the project concept of constructing bunds (the project's main investment), and related issues of maintenance of bunds and farmer participation as well as cost recovery. Although the proposed project contained many uncertain elements, it was concluded to proceed with full appraisal of the project. After appraisal in May 1974, protracted discussions took place within the Bank between the Regional and Central Project Staff on the viability of the project because of the cost-effectiveness of the proposed bunding methods and the high cost per beneficiary in terms of money and staff input. Serious doubts were raised about the reliance on heavy equipment rather than on labor intensive methods. Other issues were again the cost recovery and the lack of technical information in support of estimated benefits and the calculation of the economic analysis. Final conclusions however were to proceed with the project, which was the first major agricultural project, in the northern part of Somalia and which had no experience to draw on. 1.6 Draft Green Cover documents were ready in April 1975 and the six year project, as approved by the loan committee for a $4.0 million IDA Credit, was estimated to cost $13.9 million of which $10.2 million was in foreign exchange, with an estimated internal economic rate of return of about 11 percent over 25 years. However, further processing of the documents was postponed, pending completion of co-financing arrangements. A relatively firm understanding was reached with the Saudi Fund to participate with $7.0 million and GOS to provide the balance. On December 9, 1975 IDA management authorized negotiation. On December 31, 1975 the Saudi Fund declined participation. From January 14 - 16, 1976 negotiations were conducted in anticipation that the Arab Fund would participate as new co-financing partner. In early March 1976 however, the Arab Fund also withdrew because of the small size of the project and the tight processing schedule. All other attempts to obtain co-financing failed. 1.7 After careful deliberations, IDA, in March 1976, concluded that in view of slippage of the project which had been scheduled for Board presentation in April 1976 and in view of the (i) importance of the project for a balanced agricultural development in Somalia in general, (ii) its role as the nucleus of agricultural development for Northwest Somalia in particular, and (iii) that a further delay would result in a costly re-appraisal, it was decided additional IDA resources would be made available. Consequently, IDA increased its funding to $10.0 million and requested GOS to increase its contribution from $2.9 million to $3.9 million. 1.8 The project was approved by the IDA Board on June 3, 1976. A condition of Board Presentation was that aerial photographs taken in 1958 and 1973 be made available for the project. Due to misunderstandings this condition was considered met whereas in fact the project never received such maps which consequently hampered considerably project implementation. Project effectiveness was officially postponed three times, the first postponement was to allow for the appointment of a project manager, the other postponements to allow negotiations and signing of a contract for the appointment of management consultants. On February 8, 1977 the project was declared effective. 1.9 Several covenants in the legal documents strengthened the capability of project management in resolving issues during project implementation. Unfortunately, two originally proposed covenants were deleted during negotiations at the request of GOS and referred to in the Agreed Minutes of Negotiation. These were: (i) the provision of assistance by the Water Development Agency (WDA) in assigning manpower and equipment for test drilling and pumping for carrying out the water resources studies, and (ii) the proposal for GOS to recover, in accordance with procedures to be agreed upon between GOS and IDA, at least 15 percent of the farmers incremental income stemming from the project in order to mobilize resources for field bunding operations in the Northwest Region. Neither of these agreements were honored. Non-observance of the first issue seriously delayed project implementation, the second issue weakened the fiscal viability and replicability of the project. Project Design and Objectives 1.10 The project had been designed to increase farm incomes by addressing the crucial issues of severe soil erosion, inadequate extension service, poor agricultural practices, poor seed quality, and inadequate and uneven distribution of water facilities for both humans and livestock. The project was to establish some basic agricultural services for the first time and strengthen others which had been allowed to deteriorate during previous years. It was to assist the trend toward settled agriculture and support the establishment of a technical basis for increasing agricultural production, and training of Somali personnel. The project was to be implemented by a Project Implementation Unit (PIU) attached to the Ministry of Agriculture. To support a smooth project implementation overall responsibility was vested in a Project Ministerial Committee (PMC) comprising the Secretaries of State for Finance, Agriculture and the President of the Somalia Development Bank. 1.11 The project included broadly the same features as the 1973 preparation report had recommended (see para 1.03), except that new bunding was on about 25,000 ha with repairs for only 2,000 ha, and the Tug Wajale model farm and construction of grain storage facilities were excluded. Also, feasibility studies to be prepared were only for a second phase of this project and an integrated rural development project in the Bay Region of Somalia. 1.12 The project was to result in an estimated annual incremental production of 4,000 tons of sorghum, 400 tons of fruit, 100 tons of vegetables and smaller amounts of alfalfa, groundnut and groundnut hay valued at about US$760,000 at full development in year 13. 50 percent of the fruits and a somewhat smaller percentage of the vegetables were to be exported to neighboring countries - 4 - and the remainder marketed locally. An estimated 2,500 farm families were to benefit directly with social benefits accruing to the population in a 200,000 ha area. In addition to seasonal employment opportunities, incremental permanent employment for about 500 persons was expected. 1.13 Furthermore, a sizeable number of nomadic people were expected to benefit from project facilities, particularly watering points. The bunding operations were to help arrest severe soil erosion. Gully erosion controls were designed to reduce the volume and destructive force of seasonal torrents and utilize such waters. 1.14 Recurrent Government expenditures were calculated to average about US$0.5 million (So.Sh 3.5 million) per annum. The strain on the national treasury was expected to be reduced through the implementation of a cost recovery scheme and through savings obtained by avoiding transportation of food grains to the Region. II. Project Implementation and Operating Performance Main Components 2.1 The project's main components specified in the Development Credit Agreement, Schedule 2 as revised on August 1, 1979 were: (a) The construction of field bunds on about 25,000 hectares, restoration of existing field bunds on about 2,000 hectares and establishment of experimental gully erosions control works. (b) The development of about 50 small irrigated farms. (c) Crop trials and improvement of training facilities at the Aburin dry-land demonstration and experimental farm. (d) The extension of the Geed-Deeble horticultural station to improve fruit crop trials, nursery production and sorghum seed multiplication. (e) The rehabilitation and expansion of extension services to dry-land farmers. (f) The construction of about 100 water points for human consumption and about 50 for livestock. (g) Consultants' services to provide technical assistance, training and preparation of two feasibility studies. (h) Construction of buildings, concrete water reservoirs and fencing. The project was expected to be completed by December 31, 1982. -5- Start-up 2.2 At the outset, a conflict between the project manager and the consultant management team, and the outbreak of the Ogaden war delayed initial project performance. Furthermore, the lack of basic data to be provided by Government as specified in the Credit covenants, and the inability of the WDA to undertake field investigation work as agreed during negotiations caused considerable delays in project implementation. Although most of the above problems were outside the project's control, delays could have been avoided if project design and preparation had been more advanced before the Credit reached the negotiation stage or was declared effective. The following are the main causes of delay. 2.3 Project Management Team. In accordance with DCA's Section 3.02(b), the PIU was to consist of a Project Manager and a team of experts collectively called the Management Team, to be provided by the consultants and comprising a Land Use Planning Engineer (Team leader and Deputy Project Manager), a Soil Conservation Officer, Chief Mechanic, an Extension Training Officer and a Horticultural Extension Officer. The team was to be appointed as a condition of Credit effectiveness. A contract with a consulting firm was signed on February 3, 1977 to this effect. The firm had been selected through a short-list method in accordance with IDA's guidelines for the engagement of consultants. The team leader took up his post in March 1977 and was followed later by the rest of the team. In July 1977 he was removed from the project because of his poor working relationship with the Project Manager and replaced. Furthermore, due to misinterpretations of their contract, the consultant staff resisted moves to be stationed in the field away from Hargeisa in the absence of adequate housing facilities. After the start of the Ogaden War in July/August 1977 security issues added to the problems. After two minor bombings of Hargeisa the consultants withdrew to Mogadishu at the end of December 1977. Following continuing disagreements between the Borrower and the consultants as to the severity of the security situation, the consultants terminated the contract and left the country on February 5, 1978 invoking Force Majeure. 2.4 The project was left with only the Somali Project Manager and two local senior staff, one horticultural and one soil conservation specialist. Project implementation activity was minimal. In early 1979, the project, with Bank approval, recruited expatriate experts on an individual basis. These included a new team leader who also filled in as Soil Conservation Specialist, a Financial Controller, a Chief Mechanic and an Extension and Training Officer. A Horticultural Extension officer joined later. After one year's delay and with a new complement of senior staff in post the project at last started to make good progress in its work planning and implementation. 2.5 Although the above delays were partly attributable to war disturbances, most of them could have been alleviated if: (a) the field location of each member of the consulting team would have been more clearly specified in the contract, and (b) adequate housing facilities had been available for the consultants upon their arrival. Since the consultants were appointed as a condition of Credit effectiveness, the necessary housing and working facilities were not ready by the time they arrived. 2.6 Aerial Photographs and Topographical Maps. The Credit covenant (DCA Section 4.05) specified that the Government should make available topographical maps to be prepared from the aerial photographs of its Northwest Region taken in 1973. This covenant was never fully met because of the resistance of the Ministry of Defence to release such photographs for security reasons. The lack of this basic data constrained the planning and execution of the project with regard to selection of areas to be bunded, pilot areas for gully erosion control measures, proper sites for water points and - 6 - development of small irrigated farms. The 1952 aerial photographs were the only ones available to the project but considered inadequate for all but topographical purposes as they were out of date. Efforts to obtain new aerial maps and photos from other sources failed. In April 1981 the National Range Agency arranged for an aerial survey of the project area as part of their program of an inventory survey of the Northwest Region. The survey was completed in November 1981 and the results became available to the project in early 1982, five years after the Credit effectiveness. 2.7 Availability of the 1973 photographs to the project was made a condition of Board Presentation. Prior to Board presentation, a Bank mission, upon their return from Somalia, reported that the photographs were already in possession of the Project Ministerial Committee which was accepted as fulfillment of the Board presentation condition. That mission's reporting was later proven to be incorrect. 2.8 Test Drilling and Pumping for Water Resource Studies. The Credit provided funds for technical assistance for: (a) the design and construction of an irrigation system at the Geed Deeble Horticultural Station; (b) the development of 50 small irrigated farms; and (c) feasibility studieor a Phase II project. The Water Development Agency (WDA) was required to assist in the test drilling and pumping for the water resources studies but delayed the test drilling by more than two years until late 1979 before notifying the project formally that it could not undertake the work due to shortage of equipment and manpower. 2.9 In this context it became apparent that Credit funds for the above activities were not sufficient. This problem was eventually resolved through the provision of funds from the Technical Assistance Project (Cr. 821-SO) and the USAID - PL480 Fund which covered foreign and local currency expenditures of the drilling. 2.10 Finally, with more than two years delay, the test drilling work was put out to contract through ICB procedures. The contract was signed with a foreign contractor in January 1981 for USS 0.966 million. Problems started with delays in advance payment to the contractor compounded with the contractor's inability to procure materials required for testing (screens and casing) and disputes regarding tax liabilities. Drilling was delayed and in November 1981 the contractor declared bankruptcy. Another foreign contractor was selected through limited international bidding but by then a year had elapsed. A new contract was signed in April 1982 for a total amount of USS0.785 million. The work was completed successfully in March 1983 and a final report submitted in April 1983. The two years of delays caused considerable slippage to the project implementation. In retrospect, Government's commitment for WDA to undertake the test drilling should have remained clearly a covenant in the DCA. Construction of Field Bunds and Gully Erosion Controls 2.11 Field bunding was carried out successfully and eventually benefitted a greater number of farms than had been planned at appraisal. Due to the timely procurement of heavy construction machinery which arrived in October 1978, construction began in January 1979, one month ahead of appraisal schedule. In the absence of topographical maps, locations of priority areas of bunding in the Hargeisa, Gebile and Boroma districts were identified from a farm survey prepared by an experienced Somali Soil Conservation Specialist. The design and method of construction generally followed those outlined in the appraisal report except that only about 50 percent of the area (average farm size: 6 ha), being the lower part of each farm, was actually bunded as against 90 percent which - 7 - had been proposed in the appraisal report. The upper half was generally left fallow for grazing and acted as a catchment for scarce runoff to supply the lower bunded area. Surplus runoff, when it occurred, discharged over farm spillways at the end of each bund, and was channeled to the bunded area of the farm below. Farmers were given advance notice of bunding activities so that they left the lower part of their farms fallow during construction. 2.12 Good progress was made in bunding construction in the first year, but it slowed down in April 1980 due to fuel shortages, security problems and the onset of the wet season. Another delay occurred in March 1981 when the project had exceptionally heavy rains which brought bunding operations temporarily to a standstill. At the end of heavy rains in May 1981, the work was resumed satisfactorily. At the end of 1982, the original year of project completion, the total bunded area was about 12,000 ha or about 53 percent of the appraisal estimate (22,500 ha). The actual number of farmers benefitting, was however higher than the appraisal estimate because the actual size was only 6 ha as compared to 10 ha estimated at appraisal. 2.13 With the extension of the Credit closing date by two years to June 30, 1985, bunding construction continued and by the end of December 1984, the new project completion date, the gross area protected by bunding and the number of project farmers exceeded appraisal targets (Table 1 of Annex 1) . 2.14 Despite the smaller net bunded area and bunded areas under crops, the project benefitted a larger number of beneficiaries (farmers) than estimated at appraisal, and in terms of incremental yields and incomes the project farmers performed better than estimated at appraisal (see para 5.01). The bunding program was enthusiastically received by the farmers, although at first there was minimal farmer involvement in strengthening the ends of bunds or in maintaining them. Shortage of draft animals, scarcity of fodder to sustain them in working condition and lack of tools and implements were reasons given for lack of participation. The attitude of farmers, however, changed when they observed that the newly constructed bunding held up well in spite of livestock damage and some very heavy rainfall, and that the increase in crop yields between bunded and non-bunded became apparent. 2.15 Repair of old bunds was estimated during appraisal to be done on approximately 2,000 hectares (about 10 percent of the area bunded between 1950 and 1964). This was an overestimate. Based on actual field surveys at the beginning of project implementation, old bunds totalling only 500 hectares needed to be repaired, and the work was completed in early 1980. 2.16 The exDerimental gully erosion control measures were planned, during appraisal, to be implemented on 200 hectares in areas to be selected where gully formation was already at an advanced stage. Work started in early 1979 on some selected areas of 50 hectares but was discontinued in early 1980, because the work was badly designed and poor in quality, and was unlikely to remove the cause of gullying. The main reasons were lack of technical knowledge and qualified personnel to do the work. The project concurred with the Bank's recommendations to stop the experimental gully control work temporarily until an overall survey of the catchment areas of the gullies was carried out. This was not done during the project period nor was an effective land use policy to control overgrazing and burning of upland areas formulated and enforced. -8- Development of Small Irrigated Farms 2.17 The original plan was to establish 50 new small irrigated farms averaging 1 ha each alongside stream beds in scattered places throughout the project area where underground water resources existed. The work was scheduled to commence in mid 1980 after completion of the water resources study and when the results of test drilling and pumping were known. The delay in employment of consultants for the water resources study and the problems with the drilling program (paras 2.03 and 2.08) put the implementation far behind schedule. The consultants started their work in mid 1980 but the final test drilling results were not available until April 1983. With no new farm locations for development to be identified for quite some time, the Bank agreed in early 1982 for the project to proceed with the siting, surveying and development of existing farms on the basis of known water supplies and suitable soils. Consequently twenty farms were identified and surveyed by late 1982. The project made the right decision in selecting these farms close to one another to facilitate movement of machinery and construction work. The suitability of these farms was later confirmed by test drilling results with the exception of one farm where the well was dry. Construction work began in early 1983 for five farms with land levelling, construction of wells and irrigation systems, installation of pumps and pipe laying. Another 14 farms were constructed and completed in 1984, making a total of 19 farms. The operation has been a success; crops grown on the farms, mainly citrus, papaya, perennial fruits and vegetables, have generated substantial incomes to the farmers. T'hey have also provided a model for private initiative and a further 100 farms were subsequently developed by other farmers out of their own resources. However, the development of additional farms under the project to meet the targeted 50 in number was not possible due to delays in test drilling, shortage of funds and the approach of the project closing date. 2.18 The appraisal cost estimates for development of a one hectare farm were US$2,000 against US$7,000 in actual construction cost, the latter arising largely because of the high costs of imported equipment and materials like pumps and pipes, and high costs of well construction. Building Construction 2.19 The building program as approved by the Bank included workshops, houses, general stores and school and dormitory facilities for extension training at the three districts in the project area (Hargeisa, Gebiley and Booroma) and the Aburin Experimental Farm for a total estimated cost of So. Sh. 4.9 million. Bid response after local competitive bidding was poor with prices double the estimates (over So.Sh. 9.0 million). The project, therefore, with the Bank's concurrence, divided the work into two parts and negotiated directly with two local qualified contracting firms for a total cost of So.Sh. 5.1 million (US$ 866,000 equivalent). Construction work commenced in late 1979 and was scheduled for completion by the end of 1980. In early 1980, however, the work was held up due to lack of building materials and supplies which, except for local stone and sand, had to be imported. The situation was remedied with the project importing the materials and corresponding adjustments in payments to contractors. Considerable delays, however, were still experienced due to logistic problems and severe water constraints at Aburin. Most residential housing was completed in 1983 but overall work was finally completed in October 1984 with considerable cost overruns. The increased costs and delays resulted in constructing fewer houses for junior and none for unskilled staff. High local inflation during the lengthy construction period and the unforeseen need to construct compound walls added to the cost. The total cost for the reduced building program was So.Sh. 7.0 million as against So.Sh. 2.2 million estimated at appraisal. The estimated and actual building program is shown in Table 2 of Annex 1. 9 2.20 The Bank supervision mission agreed that the construction of permanent housing for unskilled workers was inappropriate since most of them lived almost permanently in mobile field camps such as tents or mobile caravans, and suggested in its supervision report to consider, if necessary, the use of prefabricated modular buildings to minimize time and cost overruns. Construction of Water Points 2.21 Appraisal estimates called for the construction of 100 water points for wates for human consumption, each consisting of an underground concrete water tank of about 120 m with hand pumps, silt settling tanks and catcyment ditches, and 50 water points for livestock, each consisting of earth water reservoirs of 6,000 m with catchment ditches. Location of watering points were selected on the basis of the number of families in the districts, the size of family, the number of livestock per family and average distance to existing sources of water. The work for human water points was contracted on a yearly basis beginning in early 1979 in accordance with the DCA's local competitive bidding procedure. Construction of livestock water points started in the same year by force account. Work delays occurred from the beginning in both types of construction due to shortages of cement and diesel. Other problems also occurred, such as reaching agreement with the villagers to obtain adequate sites at proper locations in catchment areas. The construction program was behind schedule each year. As of December 31, 1982, the original date of project completion, only 78 human water points were completed as compared with 100 expected at appraisal, and 20 for livestock as against 50 at appraisal. 2.22 The Bank supervision mission at that time observed that run-on cistern catchment for human water supply was not appropriate due to high cement and labor costs, high evaporation and seepage and difficulty in controlling usage and pollution. The mission also discouraged the ongoing development of permanent livestock water points in areas of agricultural development, which tended to concentrate livestock herds and thus damage farmers bunds. Recommendations were made that water supplies, both for humans and limited livestock use, should be based on bores (tubewells), with hand pumps to discourage over-consumption and pollution, but this was not implemented. 2.23 With the extension of the Credit closing, the project continued the construction of water points with improved designs and after better communication with regional and local administrators and farmers which resulted in agreement on mutually acceptable sites. The new design called for installation of corrugated roofs on top of human water concrete tanks to reduce evaporation, and lining the livestock water reservoirs with polyethylene sheets to prevent percolation. The result of these improvements was satisfactory and by October 1984, the project had completed altogether 101 human water points and 51 livestock water points. Crop Trials at Aburin Dry-Land Demonstration and Experimental Farm 2.24 The crop trial program, as envisaged during appraisal, was to be improved at the Aburin Farm through the introduction of improved varieties, crop husbandry techniques, pest control, rotations and alternative crops including fodder crops. Investigations were also to be carried out to select improved ox-drawn implements suitable for the small local oxen, particularly for sowing sorghum in rows and for inter-row weeding. The farm previously belonged to the Afgoi Research Institute (ARI) and was handed over to the project in March 1978. At the beginning, only 15 ha of land out of a total of 150 ha were under cultivation and the facilities needed rehabilitation particularly staff housing, water supply and electricity. In the first year (1978) fifty ha were planted with sorghum - 10 - with an estimated low yield of 300kg/ha compared with 600kg/ha estimated at appraisal. Simple trials were conducted on some 10.5 ha of which 4.5 ha were bunded with one local variety each of sesame, peanuts, wheat and maize but the trials failed with no tangible results due to low rainfall, bird and locust damage as well as the shortage of staff. 2.25 Following the arrival of the expatriate extension and training officer in mid 1979, trial programs were improved. Land was prepared in bunded and non-bunded areas for comparison of yield results. Germination tests were completed with satisfactory results. Agronomic trials were carried out in 1979 on different crops: sorghum (varietal trials), maize (fertilizer (N) and bunding trials), sorghum (fertilizer trials), peanut (varietal and spacing trials) and cowpea (varietal and bunding trials). The yield figures were however so widely disparate as to be of very dubious value and the 1979 results were considered inconclusive. There were also 13 "demonstration" plots of different sizes from 0.5 to 4.4 ha on the land of small farmers. They included sorghum, maize, sesame and peanuts. Results were equally erratic and no conclusions could be made. 2.26 In 1980, a year of very low, late and discontinuous rainfall, the trials on soil moisture conservation demonstrated their importance. Virtually all crops except sorghum failed on the unbunded areas due to the moisture stress following germination. The failed crops included maize, peanuts, millet, cowpeas and mung beans. On bunded areas sorghum yielded from 10.7 to 23.3 qt/ha with the local white Baidoa out-yielding all other varieties of which 16 were tested. In trials on plant spacing and seeding rate, the application of 8kg of seed per ha at 75 x 30 cm spacing gave the highest average yields of 16.6 qt/ha. Due to lack of water and subsurface water no training was conducted at Aburin during that year. In August 1980, the expatriate in charge of the program since mid 1979 resigned and was replaced temporarily by a Somali graduate with no research experience. The work at the station, however, proceeded slowly with the assistance of the expatriate horticulturist and the resident farm manager. 2.27 During 1981, still without an expatriate manager, the project staff concentrated on improvement of existing local cultural practices for which livestock and tractor drawn equipment was purchased. Varietal trials continued together with multiplication of promising varieties of sorghum, peanut, mung and green beans. Results, however were not properly recorded and inconclusive. 2.28 A qualified replacement (extension and training specialist) was recruited in late 1981 and took up his post in February 1982. The major work was concentrated on land preparation at Aburin and on demonstration farms for trials and seed multiplication. Basically, as in 1981, efforts focussed on the improvement of local cultural practices including tillage, weed control, seed dressing, pest control and high quality seed selection and multiplication. The expatriate, however, stayed for only one year and resigned in February 1983. The work continued from 1984 until the project completion with the same work program as in 1983 with more emphasis on time of sowing, plant densities, cultural practices, weed and pest control. The results of 1983 and 1984 were quite promising. The 1984 (drought year) results also showed marked increases in yields of sorghum between bunded and non-bunded areas (64 percent). 2.29 With the arrival in March 1983 of a new expatriate team leader (a dryland crop agronomist) the quality and quantity of work conducted at the station and on demonstration farms improved. The work concentrated on seed introduction, trials and seed multiplication. Intercropping and cultivation with ox-drawn equipment was also pursued. - 11 - 2.30 In retrospect, the existing Aburin station should not have been chosen for the dry land crop trials. The remoteness of the farm, which made access difficult in the dry season and almost impossible in the wet season combined with the lack of water severely limited its usefulness as a research/demonstration/training farm combined with the lack of adequate housing and training facilities. All these factors affected recruitment of qualified personnel to work at the station. The trials programs suffered due to discontinuity of staff with for instance three different expatriates working at the station during the project period: one between July 1979 to August 1980, the second between February 1982 to February 1983 and the third from March 1983 to October 1984. It was a consensus among the Bank mission and the project management that an alternative site be sought, at or close to Gebiley, the center of the project area but no suitable site had been identified by the time of project completion. Extension of Geed-Deeble Horticultural Station 2.31 The main objectives set out at appraisal were to expand the existing 10 ha citrus farm and nursery at Geed-Deeble to a total area of 30 ha under irrigation in year 3 (1979). The existing 10 ha were to be utilized for variety trials of citrus, peaches, plums and other fruits and the additional 20 ha for a nursery for production of grafted seedlings of citrus and other fruits for sale to farmers (5 ha); and for a date nursery (5 ha) and the multiplication of selected sorghum seed (10 ha). Most of these objectives particularly the extension of the area were not met since the station was without a reliable and adequate source of water. At the beginning, 3 ha were under cultivation including orchards and nurseries and strenuous efforts were made to maintain or save the citrus stocks even to the extent of delivering water and introducing foliar spraying. The station proved to be inappropriate. Access to the location, about 40 Km north east from Hargeisa, was along a rough track road which made it unsuitable as a demonstration center and for the distribution of planting material and was also unacceptable for the stationing of trained staff. In addition, there was no chance of secure future water supplies as later test drillings proved negative. In April 1980, the Bank mission recommended that a new site with good potential water supply be sought. 2.32 In spite of these problems, the expatriate horticulturist who assumed his duties in late 1979 with the aid of two local extension assistants was very active in the field in arranging advisory meetings with farmers on a T & V basis. A survey was also conducted in the project area to collect core data such as farm size, irrigated areas, range of crops, state of husbandry, pest and diseases, water supplies, production and input costs and incomes. Pamphlets were prepared in Somali and a radio broadcast program was established to deal with current production problems. Since seed multiplication could not be carried out at the station as originally intended it was done on contact farmers' fields with good results. 2.33 Test drilling for water was carried out again at the station in September 1982 without success. The new hole was dry and could not be deepened further as the aquifer below was the main supply for Hargeisa and any further drawdown for horticultural purposes would jeopardize the water supply for the city. The rehabilitation of the old bore hole provided water only to irrigate about 3.5 ha which was even insufficient to sustain the citrus budding program and carry out investigation with other fruits and vegetables. Minimum stock maintenance was achieved by hauling 9,000 liters of water a day which proved uneconomical. Since 1983 the Geed-Deeble has been maintained as a holding operation supplying rough lemon rootstocks, budded plants and budding materials to the 19 irrigated farms developed under the project. The station was finally handed over to the Jojoba project - 12 - when the project closed in June 1985. A new site for the research station is being established under the Northwest II project. 2.34 The objective of this project component was not met. The plans to rehabilitate the existing station should have been subject to verification of water availability. ExDansion of Extension Services and Training Facilities 2.35 The main objective at appraisal was to rehabilitate, strengthen, and expand the existing extension service for dryland farming to a total of 23 extension agents (the ratio of 1 agent per 600 farms at full development) mainly directed towards (but not restricted to) farmers on bunded lands. It was to concentrate on simple concepts - including timeliness of cultivation and planting, improved seed, row sowing, weeding, the use of improved implements and plant protection - and was to utilize demonstration plots on farmers' land. A small horticultural extension service (three professionals) was also to be created to provide advice to farmers on the production of fruit and vegetables under irrigation, and was to concentrate on the fifty small irrigated farm to be developed under the project. 2.36 Because of staffing constraints there was little extension activity in the first two years of the project. The farmers initially enquired about the supply of seedlings and advice was given to them on planting techniques. Farmers were contracted and some trials were carried out on participating farms' fields. Despite the small number of seven extension agents at the beginning (1978) with inadequate transport and equipment, work progressed reasonably well. Credit must be given to the Somali Senior Extension officer who did his best under difficult conditions. In 1979, the number of extension agents was increased to 13. Pre-season training of extension agents commenced in April 1980 to prepare a technical package based on seed treatment for smut, seed bed preparation, sowing times and plant population densities. In early 1981, some of the new recruits resigned due to harsh living conditions and some were sent for training abroad. The number of extension agents in 1981 therefore decreased to only six. 2.37 In late 1981, with active participation of the expatriate horticulturist, extension work on a modified T & V system was carried out with satisfactory results (see para. 2.40). 2.38 For dry land farming, there were only six extension workers in 1981 operating from six centers in the three districts of the project area as compared to 23 scheduled at appraisal. The number of extension agents increased to nine in 1982 and the activities conducted on 24 plots demonstrated the benefits of bunding, row cropping, seed dressing, seeding rates etc. The work on the demonstration plots was carried out by farmers themselves which created better participation and understanding. Multiplication of selected seed was also undertaken at the farmers' fields. The shortage of petrol, and the absence at times of a full time extension training officer restricted extension services activities. 2.39 In 1983, the number of extension agents decreased to seven, but their work performance under the guidance of the expatriate agronomist was commendable. The agents operated out of the 3 main towns in the region and covered 12 villages and 50 demonstration farms serving about 6,000 farms. Simple techniques were shown with emphasis on a cultivated fallow for weed suppression and moisture conservation, row planting for ease of interrow cultivation, plant spacing and density, seed dressing for smut control and control of stalk borers. Radio talks and pamphlets in Somali were also used. The extension agents received training by the Headquarter's subject matter specialists every - 13 - two weeks, discussed specific problems and planned their work for the next two weeks. Although the expatriate agronomist left the project in late 1984, the extension activities continued to be carried out satisfactorily with the assistance of the expatriate horticulturist until the project completion. 2.40 For horticultural extension, activities improved considerably following the arrival of the expatriate horticulturist in late 1979. From 1982 three extension agents working under an expatriate specialist were in charge of more than 600 farms in the project area. Regular visits were made to these farmers and demonstration of fertilizer use and irrigation layout were held. Plant protection methods were also demonstrated and potato production trials conducted. These demonstrations had a positive production impact on the contact farms and also resulted in savings in water usage and on fertilizer application. Details emphasized were improved water use, citrus budding, spacing, hole preparation, planting methods, stock raising from seed and pruning. Vegetable production demonstrations were conducted on 29 farms to improve production and expand the range of vegetables (onions, tomatoes, potatoes and salads) in the area. The work of the horticultural extension was considered satisfactory throughout the project implementation period, and the work performance of the expatriate horticulturist was commendable and outstanding. The combined number of staff at Aburin and Geed Deeble stations for the extension services of both dry land and horticulture is shown in Table 3 of Annex 1, 2.41 The original objective during appraisal of having 23 extension agents for dry land farmers was not met due to difficulties in recruiting or transferring qualified people to work in the Northwest Region where working conditions and facilities were harsh. From time to time, national extension workers were attached to the project unit but with limited success. The sharp drop of the number of extension agents in 1981 was due to temporary leave of these people for overseas and local training programs provided by the project. After completion of training, very few returned to the project. Consultancy Services and Feasibility Studies 2.42 Consultancy services were envisaged during appraisal and stipulated in the project's DCA as a condition of Credit effectiveness. The original idea was to have consultants from a single firm or joint venture to provide services which would include (a) a Project Management Team to assist the project manager in the execution of the project; (b) Feasibility Studies for further development of soil and water resources in the Northwest and Bay Regions; (c) Technical Assistance to the PIU with regard to the development of small irrigated farms; and (d) Establishment of bunding costs and criteria for cost recovery. The idea of having one firm provide all of the above services did not materialize after the first appointed consulting firm left the project in February 1978 (para 2.03) and the project had to resort to the recruitment of consultants on an individual basis. These consultants arrived at the project from late 1978 onwards. They formed a project management team and were found to be competent in their respective fields. There was, however, some discontinuity of the services of certain specialists which hampered project implementation. This was due to the harsh nature of the Northwest Region and hard living conditions in Hargeisa which made it difficult to maintain a complete team all the time. Details of various positions of these consultants and their respective period of services are shown below: 1. Team Leader : Soil Conservation Specialist (8/78 to 12/79), Dryland Agronomist (3/83 to 10/84), - 14 - 2. Extension and Training Specialist: The first one (7/79 to 8/80), the second one (2/82 to 2/83), 3. Horticultural Specialist: 9/79 to 6/85 (continued in the second phase project), 4. Financial Controller: The first one (5/78 to 7/79), the second one (4/80 to 6/85 and continued in the second phase project), 5. Chief Mechanic : 8/78 to 6/85. 2.43 During late 1980 and early 1982 when the positions of team leader and extension and training specialist fell vacant, the three remaining expatriates did a commendable job in assisting project management to carry out the work as effectively as they could in order to avoid serious delays in project implementation. 2.44 The feasibility studies and technical assistance to irrigated farms development also suffered long delays. After the departure of the first consultants, the project had to select a new firm out of the three short-listed ones in accordance with the Bank's guidelines. In November 1979, a three year contract was signed with a new selected firm for a total cost of French Francs 9,214,000 (US$ 1.04 million). The contract called for the firm to carry out the feasibility study of the Northwest Region Phase II project and to provide technical assistance to the PIU with regard to development of 50 irrigated farms. The feasibility study for the Bay Region was omitted as it was separately carried out under the funding of the Bank's Project Preparation Facility (PPF). The firm started work in January 1980 and produced the first (Inception) report in May 1980 on schedule. Subsequently, work slippage occurred over the delay in the test drilling which was beyond the control of the consultants. As a result, a draft report for the feasibility study of a Phase II project was not available until September 1983 and the final report was completed in November 1983, one year behind schedule. The overall performance of the consultants was however considered satisfactory by both the project and the Bank missions. The consultants also provided good technical services to the project with regard to the development of small irrigated farms and the study in establishment of quantum and criteria for recovery of costs of land bunding as required under the contract. Staffing 2.45 At the beginning of project implementation in 1977, there was a shortage of staff particularly at the senior and middle level positions. After considerable efforts by the project manager, the staff situation improved in 1978 but there were still some middle level staff positions which remained unfilled such as a deputy financial controller, storekeepers and qualified mechanics for tractor and vehicle repairs. Also the position of a deputy project manager to assist the project manager in his responsibilities was initially not filled as no suitable candidate could be found. With the appointment of a deputy project manager in early 1980 and an increase in salaries by the Government in 1981, the above middle level positions were filled and the project staffing since 1982 was considered well above the Somali average and good by most project standards. At the end of 1983, the project manager resigned from the project to undertake advanced study abroad in soil and water management, and his successor who stayed with the project until its completion and continued into the implementation of the Phase II project proved to be competent and did a satisfactory job throughout the remaining project period. The actual number of staff at the end of project implementation as compared with the number estimated at appraisal is shown in Table 4 of Annex 1, - 15 - 2.46 Generally, there was not much difference in the number of the senior and middle level staff between the actual and appraisal estimates. The only significant difference was the number of the extension agents for the dryland farm which was 23 (appraisal) versus 7 (actual). This was due to difficulty in finding qualified persons to be trained as extension agents. III. Project Costs. Disbursements and Procurement 3.1 Total costs were estimated in 1976 at US$ 13.9 million (So.Sh 86.8 million) over a 6-year period. With the delays, the actual project implementation commenced in April 1977 and took almost 8 years with the total expenditures amounting to $ 14.1 million (So.Sh 148.6 million). 1/ A breakdown is as follows: Table 1: Appraisal vs. Actual Costs Appraisal % of Actual Activities Estimate Actual to Estimate ................ 000. So.Shs U$ So.Sh US$ S h US 1. Construction of Field Bunds and Water Points 51.9 8.2 86.5 8.5 167 104 2. Small-Scale Irrigation development .7 .1 3.5 .2 500 200 3. Dry-land Crop Trials 1.1 .1 5.9 .6 536 600 4. Geed-Deeble Horticulture Station 3.4 .6 4.8 .4 141 67 5. Extension and Training 14.5 2.3 16.8 1.4 116 61 6. Feasibility Studies and Technical Assistance 15.2 .IU 31 3 205 3115 86.8 13.9 148.6 2/ 14.1 171 101 3.2 At the very beginning of the project GOS provided the project with a revolving fund of So.Sh. 2.0 million to facilitate operations. The fund was regularly utilized though faced difficulties in replenishment. Despite timely submission of annual budget requests, GOS never fully met the projects financial requirements nor did the funds come forward in a timely manner. Extraordinary efforts had to be made by project management and Bank staff to secure funds from the Ministry of Finance. Details on the causes of cost overruns of the various components are provided in Chapter II (Project Implementation and Operating Performance). 1/ Additional amounts of $550,000 and So.Sh. 4,651,000 were funded under Cr. 821-SO for the Groundwater Drilling Program. 2/ This increased amount over the SAR estimate is largely due to the numerous devaluations of the Somali shilling. - 16 - Disbursements 3.3 The project resorted to the Bank's direct payment procedures for all imports after initial experiences in opening letters of credit with the Central Bank caused delays and problems with the insurance of imports. Reimbursements to the project for local currency expenditures made by IDA through the Somali Central Bank were credited to the project's accounts with delays of up to 4 - 5 months and usually only after intervention by project staff who had on hand the Association's payment advice. Disbursements were prolonged as a result of the extended project completion period but were above the regional disbursement profile. By and large disbursements lagged about 20% behind appraisal estimates (Table 5 of Annex 1). 3.4 The credit was reallocated to provide for an increase of US$ 1.35 million in the category for vehicles, tractors, equipment and supplies. This is largely accounted for through the higher than originally anticipated purchase of spare parts to build up stocks ($330,000), the purchase of building materials for the civil works ($530,000) and the purchase in 1983/84 of machinery, vehicles and equipment for $370,000 to ensure continuity of project activities until Phase II became operative. 3.5 Overspending for the Feasibility Studies and Technical Assistance by $605,000 was mainly due to the delay by two years in the start of the water resource study and preparation of Phase II as well as an underestimation of consultants' time by 23 man-months. Additional overspending of $550,000 for the drilling under the water resource study was caused through appraisal underestimation. The funds came from IDA Credit 821-SO (Technical Assistance). 3.6 In comparison, estimates for operating costs for bunding and water points including extension and training were underspent by $1.5 million. This was mainly a local currency expense and the savings were achieved through prudent financial management and from the Somali Shilling devaluations. Disbursements for civil works were $370,000 less than anticipated. This was compensated through over disbursements for building materials. Accounts and Audits 3.7 The audit for the cumulative project accounts as of June 30, 1985 (Credit closing date) reported that the project accounts were maintained in Somali currency; foreign currency items were converted into Somali shillings at the official rates of exchange prevailing on the date of individual disbursements by IDA. The differences in items of US$ liabilities to IDA arising out of the official devaluation were charged to an account "Devaluation Difference" and written off in the Statement of Net Operating Expenditures. The financial statements were prepared under the historical cost convention and inventories were stated at the lower of cost and net realizable value. Fixed assets were depreciated at rates calculated to amortize the cost of the assets over the approximate period of their estimated useful lives. 3.8 The IDA Credit contributions to the project amounted to US$ 9,949,447 with the balance of $50,553 being canceled. In addition, an amount of US$ 550,000 was provided from the IDA Technical Assistance Project (Credit 821-SO) to meet the foreign currency cost of the groundwater drilling program. The total Government of Somalia contribution up to June 30, 1985 amounted to So.Sh 49.8 million. USAID PL480 funds disbursed through the Ministry of Finance contributed So.Sh. 4,651,000 to meet the local currency cost of the groundwater drilling program. - 17 - 3.9 Detailed stock registers for inventories of spares, stocks and supplies were maintained and reportedly physically verified by project management. The overall impression of the state of the accounting records and the system of internal controls was that they were adequate and functioning satisfactorily. The long form audit report does not include any qualifications or unusual items. Procurement 3.10 The procurement procedures as prescribed in Schedule 3 of the Development Credit Agreement guided procurement matters and caused no undue problems. There were a few exceptions which were agreed to by the Association in the interest of speedy project implementation. The variances were: (i) at the very beginning, the procurement of six 4WD vehicles through direct purchase arrangements with the manufacturer's supplier in Kenya and UK; (ii) the direct procurement of pumps and pipes from abroad for the 19 small scale irrigated farms at a value of $47,000; and (iii) the procurement of building materials under the civil works building contract since the local contractor was unable to import the goods directly. 3.11 Machinery, vehicles and equipment were largely procured under ICB with a few direct purchases from abroad. No local purchases were made under this category. Requirements for operating activities such as fuel were procured through the Somali Petroleum Agency, and building materials for the water points were obtained locally. Civil works for buildings and water points for humans were let under LCB. Force account was used for the bunding operation and livestock water points. Consultants services were obtained through short listed competition under criteria agreed to by the Association. 3.12 There was no misprocurement and no recorded evidence of diversion of procurements for purposes other than project related activities. The amount procured under each procedure is given in Table 6 of Annex 1. IV. Institutional Performance and Development 4.1 Overall management of the project has been good, particularly in view of the enormous constraints imposed on the project by outside factors such as security measures which have at times tested the determination of project staff, curfews, logistical problems due to the remoteness of the project area and restrictions on movement of people in rural areas with disruptive effects on project work. The financial management was sound and physical performance was good despite all the logistical and staff problems. 4.2 With the active support and encouragement of senior staff of the Ministry of Agriculture in Mogadishu (some 2,000 km away) the Project Manager assembled a committed management team and instilled discipline and responsibility at all staff levels. In the appraisal report it was assumed that a Somali national as project manager would be supported by a management team provided through a consulting firm. In fact, project effectiveness was delayed in order to appoint such a firm. Due to the isolation of the project area, the poor social and administrative structure of the region, personality problems and security reasons, the consulting firm withdrew from the project in 1977 leaving the project effectively without any senior technical staff. 4.3 With IDA encouragement the project recruited individual experts for the senior positions of Team leader, Extension and Training officer, Horticultural Extension officer, Chief Mechanic and Financial - 18 - Controller. The position of Soil Conservation Officer was never filled but the task was handled by either the project manager or team leader. As per project design, the position of Land Use Planning Engineer was to be filled by the team leader, but the project never benefitted from such an expert. The Team Leader position was filled by three different persons during the periods 1977, 1978-79 and 1983-84. The Extension and Training officer position was filled in 1979-80 and 1982-83. During the other times the Horticultural Extension Officer handled the work. The Chief Mechanic remained the same from 1978 to the end of the project. The Financial Controller position was not provided for in the project as appraised. However, it became obvious that a qualified person was required. The first candidate was directly recruited and remained with the project for one year in 1978-79. In mid 1980 an ADS staff member joined the project and only left the project after the start of the Phase II project (Cr. 1538-SO). 4.4 Despite some high turnover of senior staff and even more so of middle level and lower level staff, the dynamic project manager maintained a core team which carried the project through. Middle management staff was very difficult for the project to attract and retain. This was largely due to an inadequate salary and wage structure which - although about 50 percent above the civil service scales - did not provide sufficient income to maintain a reasonable standard of living. Staff transferred to the project through the Ministry of Agriculture had low academic qualifications, little or no practical experience and almost invariably a dislike of being transferred to the Northwest Region where working conditions were known to be hard with few facilities, a high cost of living and little opportunity for additional incomes. From time to time the Agricultural Farm Management and Extension project provided extension staff but with limited success. 4.5 The consultancy component has suffered from discontinuity due to local security breakdowns resulting in the departure of the consultants at an early stage in the project. The engagement later of a consulting firm to carry out feasibility studies had some initial difficulties, largely due to staffing problems, but the final report of the Water Resource Study and Phase II preparation report were good. The first drilling contractor departed in bankruptcy after a year of little progress. The Chinese drilling team thereafter performed very well. Despite all these difficulties, the project management utilized the consulting services in a cost effective manner. 4.6 After being with the project for seven years the project manager was sent abroad in April 1984 for studies funded under the project. Immediately, the Ministry of Agriculture appointed the project manager designate, who had joined the PIU in late 1983, as new project manager and who was still in post at the time of project closing. 4.7 The project is to be credited with instituting a large number of local training courses for farmers and its own staff. Project staff benefitted from training as mechanics, tractor/heavy equipment operators, drivers and in administrative skills. Fifteen staff received extensive on-the-job training while assigned to support the consultants engaged with the water resource study and Phase II preparation. The project manager was abroad for eight months on project related studies including dryland farming, and the deputy project manager for two months on a Monitoring and Evaluation course. One person went for training in France and one in Germany as mechanics. Of eight graduate students in the USA only three returned. 4.8 The project management established effective recording and reporting systems and submitted comprehensive and detailed quarterly and annual progress reports as well as annual work programs, but cumulative progress report would have been very useful. - 19 - V. Project Impact Dryland Farming 5.1 The main activity of the project was soil conservation and moisture retention through field bunding to raise the production of sorghum, the dual purpose crop providing grain for human consumption and stover as fodder for livestock. However, the impact of bunding on crop production remains inconclusive because of conflicting and inadequate data. According to project management, bunding caused a yield increase of 75 percent, or 38 percent over the appraisal estimate of yield increase resulting from the project (Table 7 of Annex 1). 5.2 Other consultant surveys, however, questioned the project management's methodology of data collection and gave a more guarded assessment on the impact of bunding. A survey carried out in 1982 by a consulting firm preparing the second phase project showed that the effectiveness varied for different terrain. It found that on lowland areas, yields could be as much as 43 percent higher on bunded fields, while on sloping areas (2-3 percent gradient) and plateau areas (< 2 percent slope) as well as on flat depressions, no significant difference in yield could be detected. Since lowland areas account for only some 16 percent of the project area, and since this 43 percent yield increment is offset by a loss of farm area of 15 percent due to the bund itself, 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. 5.3 A survey in 1987 done by another consulting firm employed under Northwest II showed that the impact of bunding on yields was minimal. However, it recognized that the survey was only based on one season's crop cut data. It was to conduct another survey the next year, but in 1988, the civil war in the north intensified and all project personnel were evacuated. The survey findings also indicated that there was a need to examine issues of bund design, bund maintenance and agronomic practices in order to realize the full potential of bunding. 5.4 The Aburin Research Farm did not produce any useful results, largely due to the lack of water availability. Adaptive research with varietal and agronomic trials, although well planned and executed, suffered from such factors as bird attacks, smut infestation and water stress -- all of which are prevalent throughout the project area -- with the results so widely (and wildly) disparate that they were of dubious value. It also failed to produce adequate quantities of seeds for distribution. Because of these difficulties, the PIU staff concentrated its efforts on demonstration plots. Although not backed by any proven research, the extension advice given and the demonstration plots developed on farmers fields have, according to project management, had a beneficial impact. The advice given related to moisture conservation, seed dressing, short season sorghum varieties, early planting, seed selection from highest yielding heads, legume introduction into rotation, plant density, and weed and pest control. This was supplemented through radio talk programs and the distribution of pamphlets prepared by the PIU. - 20 - Small Irrigated Farms 5.5 The improvement in fruit crop and vegetable production is noteworthy despite the constraints of meaningful applied research. The Geed-Deeble horticultural research station, provided to the project by the Ministry of Agriculture, proved to be in an unsuitable geographic location and suffered from the lack of water. This water problem was never fully overcome and at the end of the project the station was discontinued. Again, the technical messages which became available derived not from local centralized research but practical demonstration of low cost techniques at the farm level. 5.6 Due to the delays in the water resources survey which was to identify 50 new small farms for irrigation, the project decided in 1983 to proceed with the development of 19 existing but underutilized farms with proven water resources. This was a successful operation which reportedly produced 100 percent iscrease in yields of some crops (Table 8 of Annex 1), 5.7 At full development the irrigated farms provided incomes of about So.Sh. 450,000 per farmer per year which compares with appraisal estimates of So.Sh 490,000. Inhibiting factors for development were identified as fruitfly infestation coupled with shortages of insecticides and sprayers. Marketing of the produce was often hindered due to lack of fuel for transport. The production of alfalfa, groundnut and groundnut hay which was originally projected did not materialize. Human Water Points 5.8 The project successfully accomplished the planned construction of water points. Despite delays in the start-up period due to logistical, staffing, equipment, materials and fuel shortages, the project completed 101 water tanks of 250m3 each (plan: 100 tanks @ 120m3). An estimated 5,300 farm families are now assured of secure water provision throughout the year contributing to a better and healthier life and environment. Livestock Water Points 5.9 The project also constructed 51 livestock water reservoirs of about 6,000 m3 each (plan 50 points). The building of the reservoirs suffered from the same delays as the human water point construction, but additional delays were experienced due to the difficulty in locating suitable sites without occupying valuable arable land and destroying bunded areas through the gathering of large numbers of animals and stock trails. Gully Erosion Control 5.10 The gully erosion control works in areas where the gully formation was already in an advanced stage was abandoned early in the project because it proved to be prohibitively expensive and could not be maintained until at least some of the following conditions were met or addressed: an effective land use policy to control overgrazing, burning of upland areas, discontinuation of arable farming on slopes without adequate soil and water conservation methods, and prohibition of indiscriminate tree and scrub destruction without a large-scale reforestation program. The Credit - 21 - Agreement included a covenant requiring GOS to safeguard surface and underground waters. A draft water law was prepared in the early 1980's, however it was not finalized. Project Design Problems 5.11 The project did not take sufficiently into consideration the implication of the geograpbic isolation of the Northwest Region with its inherent logistical problems and the substantial risks involved in dry land farming on so large a scale in an area with low and erratic rainfall, including the poor state of applied research and agricultural technology. At the design stage, the need for administrative facilities were underestimated and it was not fully realized how few qualified and senior middle management staff would be available. The expectation that the existing research farms could be fully utilized did not materialize since the severe water constraint was not recognized. Transfer of Technology 5.12 The poor state of applied research and lack of technical information provided little new agricultural technology. However, the efforts of the project specialist staff to establish both dryland and irrigated demonstration plots on farmers fields and to disseminate extension advice is commendable. The demonstration effect of the small irrigated farms was impressive (para. 2.17). Bunding also had an impact on yields but the extent needed to be further studied. 5.13 The results from the Central Agriculture Research Station at Afgoi in the southern part of the country were not applicable to the project area; meaningful applied research should be developed on a new research station in the north. VI. Economic Re-evaluation 6.1 Due to the conflict in production data (paras. 5.1, 2, 3), the economic rate of return could not be calculated with reasonable confidence. From a long term point of view, the project could be considered satisfactory since it had laid the groundwork, and provided an infrastructure and administrative framework on which to continue improvements of food production and water supplies and to lessen the reliance on food imports and/or relief donations. The experience gained under this project should help to encourage the settlement of nomads and improve dryland farming. Furthermore, the project helped to fill many gaps in the knowledge about the region through the resource inventory undertaken by the National Range Agency and the test drilling for groundwater studies and quantification of such resources. VII. Bank Performance 7.1 From the very beginning the Bank supported the Borrowers' National Development Plans of which this project was only one of many priorities. To ensure the implementation of this project the Bank responded with flexibility and increased the IDA allocation from $ 4.0 million to $ 10.0 million when co-financing support failed. 7.2 The Bank supervised this project through its Nairobi office and made special efforts in supporting the project management in its difficult tasks to overcome the various constraints - 22 - highlighted in earlier chapters of this report. The large number of supervision missions with different staffing specialties and yet staffing continuity attests to this. The annual project implementation reviews in Mogadishu were also used to support project management to obtain actions from central ministries. No supervisions were possible from late 1984 to mid 1985 as a result of a severe cholera outbreak in the region which denied access to the project area by any outsiders. Continuity of Bank staff for supervising the implementation stage proved very effective and has been of utmost importance in building mutual respect and trust. 7.3 The Bank proved flexible in amending the allocation of proceeds to accommodate the recognized implementation progress; in adjusting disbursement procedures to help the project overcome constraints outside the control of project management; and in extending the project completion dates and Credit closing dates to make up for delays at the start of the project and to facilitate an easy transition into the second phase project. Also, when it became apparent that there would not be sufficient funds to execute the drilling programs, the Bank was very constructive by agreeing to provide funding of $550,000 through the IDA Technical Assistance Credit 821-SO. VIII. Lessons Learned 8.1 The lessons learned are summarized as follows: - The project's focus on an appropriate technology was the key to its success despite very difficult circumstances. Although the impact of bunding is still inconclusive, the project has yielded useful findings about the technology. Building on this experience, the second phase project was to extend bunding to areas where it was appropriate and complement it with other water and soil conservation measures. - Adapting a technology to a region requires a long time. After an eight-year effort to develop rainfed farming in the Northwest Region of Somalia, the project had reached a stage where it should be possible to determine the effectiveness of the bunding technology. Unfortunately, in large measure because of inadequate monitoring and evaluation, it was not yet possible to determine the technical effectiveness and the economics of bunding. It was therefore important for the follow-up project to ensure that: (a) the experiences made under the project were not dissipated; and (b) the cost/benefit of bunding was properly monitored and the results suitably assimilated into the strategy for developing agriculture in the country. These lessons were incorporated in the second phase project. - Strong Government commitment and a dynamic project manager are essential to project success. Despite the difficulties with security and the project's remote geographic location, the project was satisfactorily carried out. - Demonstration effects are important. If appropriate technology is introduced, farmers can find ways to develop with their own resources even in the least developed areas. - Covenants aiming at ensuring the availability of certain essential project requirements (such as the responsibilities of WDA) should not have been deleted and merely referred to in the Agreed Minutes of Negotiations. - 23 - - The difficulty of recruiting consultants to a geographically isolated area should be recognized in the project design. Provisions for housing and other support should be adequate. The thorough review by the various Bank departments prior to finalizing a project proved extremely valuable in identifying important issues and resulted in adjustments in the project design. The regular supervision of the project assisted in maintaining the momentum of project implementation through a flexible application of Bank guidelines and procedures to overcome specific project related difficulties. -25 - Ann I (Page 1 of 5) Table 1: Bunded Areas and Beneficiaries Actual Appraisal Target as % of at Proj. Completion Actual Estimate Project farmers (No.) 2,500 5,336 209 Gross area protected (ha) 25.000 32.315 129 Net area bunded (ha) 22,500 1' 16,330 Z' 73 Cropped bunded area (ha) 15,000 10,778 72 Average farm size (ha) 10 10 60 I/ 90 percent of farm area changed to about 50 percent. 2/ 66 percent of bunded area, 33 percent fallow. Table 2: Building Program Appraisal % Actual Housinz Estimat Actual to Estimate Expatriates and Project Manager 8 6 75 Senior Staff 5 8 160 Junior Staff 39 10 26 Unskilled workers 12 - - Dormitory/Lecture Hall 1 1 100 Compound wall 1 - Appraisal % Actual Others Estirate Acal to Estimat Offices 4 5 125 Store 1 1 100 Workshop 1 1 100 Implement Shed 1 1 100 - 26 - Annex I Table 3: Extension Staff (Page 2 of 5) Dry Land (Aburin Station) Position 1977 1978 1979 1980 1981 1982 1983 1984 1985 Extension & Training Officer 1 - 1 1 - - Deputy Training Officer - - - 1 1 1 1 1 1 Group Leader - - 3 3 3 3 1 1 1 Farm Manager 1 1 1 1 1 1 1 1 1 Assistant Farm Manager 1 1 1 1 1 1 1 1 1 Skilled Labor 7 7 7 7 7 7 7 7 7 Typist 1 1 1 1 1 1 1 1 1 Tractor Driver 2 2 2 2 2 2 2 2 2 Vehicle Driver 1 1 2 2 2 2 1 1 1 Extension Agents - 7 13 13 6 9 7 7 7 Horticulture (Geed Deeble) Position 1977 1978 1979 1980 1981 1982 1983 1984 1985 Horticulture Officer (Expat) 1 - 1 1 1 1 1 1 1 Deputy Horticulture officer 1 1 1 1 1 - - - - Extension Agent - - - 1 1 3 3 3 3 Farm Manager 1 1 1 1 1 1 1 1 1 Assitant Farm Manager 1 1 1 1 1 1 1 1 1 Skilled Labor 7 7 7 7 7 7 7 7 7 Typist 1 1 1 1 1 1 1 1 1 Vehicle Driver 1 1 2 2 2 2 2 2 2 Tractor Driver 1 1 1 1 1 1 1 1 1 - 27 - Annex I Table 4: Project Staffng (Page 3 of 5) Appraisal Estimates Actual (Yr. 6-1983) (Yr. 1985) Remarks 1. Soil & Water Conservation Decartment General Manager 1 1 Somali Management Teom Leader 1 - Expatriate Deputy Mgt. Team Leader - 1 Somali Soil Conservation Officer 1 - Expatriate Deputy Soil Cons. Officer 1 1 Somali Plamning Officer 1 1 Somali Monitoring & Evaluation Officer 1 Somali General Supervisor 1 Somali Unit Leader 6 Somali Unit Deputy Leader - 6 Somali Levellers 6 10 Somali Leveller Assistants 12 - Somali Chief Mechanic 1 1 Expatriate Deputy Chief Mechanic 1 1 Somali Mechanics/Electricians 9 8 Somali BulLdozer Operators 18 19 Somali Agric. Tractor Driver 8 6 Somali Vehicle Drivers 12 26 Somali Asst. Plant Operators 18 17 Somali Artisans 4 3 Somali Messengers - 1 Somali Watchmen _ 14 Somali Total 94 f == ... 2. Administration Department Financial ControlLer 1 1 Expatriate Accountants & Asst. Accts. - 6 Somali Secretary 1 2 Somali Security Officer 1 Somali Cashier General 1 Somali Personnel Officer 1 Somali Typist 5 6 Somali Tracers 3 Somali Storekeepers - 2 Somali Watchmen 16 Somali Messengers 8 a Somali Total 10 44 3. Dryland Croo Trials and Extension Deoartment Ext. & Training Officer 1 - Expatriate Deputy Ext. & Trg. Officer 1 1 Somali Extension Agent (Aburin) 1 1 Somali Farm Manager (Aburin) 1 1 Somali Unit Leader - 2 Somali Extension Agent 23 7 Somali Aburin Farm Staff 4 9 Somali Typist 1 1 Somali Agric. & Vehicle Drivers 4 4 Somali Total 36 26 4. Horticulture Deoartment Horticulture Ext. Officer 1 1 Expatriate Assistant Farm Manager - 1 Somali Senior Ext. Agent 1 4 Somali Horticultural Ext. Agent 3 11 Somali Farm Workers 6 2 Somali Driver 1 1 Somali Total 12 20 =Z ax -28 - Table 5: Credit Disbursement (Page 4 of 5) Categora Credit Actual % Actual to Allocation Disbursemeni Allocation ---------US$- Vehicles, tractors, equipment and mat 3,610,000 4,960,000 137 2. Operating costs for bunding and water points 2,900,000 1,390,000 48 3. Technical Assistance incl. Feasibility Studies 2,830,000 3,430,000 121 4. Civil Works 540,000 170,000 32 5. Unallocated 120.00 _ _ Total 10.000.000 9.950.000 99 Table 6: Procurement Procedures and Amounts US Dollars % of Credit (million) International Competitive Bidding 2.6 26 Local Competitive Bidding 1.7 17 Direct Procurement abroad 1.6 16 Consulting Services L 3 TOTAL ED ,2 -29 - A 1 Table 7: Sorghum Production in Bunded Area (g 5 of 5) Without With Project With Project Actual as % of Project (Appraisal (Actual) Estimate Est.) Cropped Area (ha) 16,700 15,000 10,778 72 Annual Sorghum 5,000 9,000 8,900 99 Production (tons) Annual Incremental - 4,000 3,900 98 Production (tons) Production per ha (kg.) 300 600 825 138 Table 8: Small Irrigation Farms - Crop Production % Increase Without Project With Project with Project Citrus fruits 3,815 tons 7,628 tons 100 Tomatoes 322 tons 970 tons 201 Potatoes 54 tons 162 tons 200 Onions 161 tons 485 tons 200 44. *' S O M A L I A NORTH WEST REGION AGRICULTURAL DEVELOPMENT PROJECT 12 Ai. 12_ . _PRIORITY AREAS FOR CONSTRUCTION ROADS 1.A..4 12 OF FIELD BUNDS ....... WATERSHED BOUNDARIES .TJ? OA 7 . il'~J FEASIBILITY STUDY AREA FOR SECOND MAIN WADIS '/ PHAse OF PROJECT ---- DISTRICT BOUNDARIES 7.. 17
Группа Всемирного банка · Project Completion Report
Somalia - Northwest Region Agricultural Development Project
Открыть оригинал документа
Полный текст размещён на сайте публикующей организации. lawenc.com индексирует метаданные и ведёт на официальный источник.
Полный текст
Основные сведения
Организация
Группа Всемирного банка
Тип документа
Project Completion Report
Страна
Сомали
Источник
Всемирный банк