Niger - Irrigation Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No. 6275 PROJECT PERFORMANCE AUDIT REPORT NIGER IRRIGATTON PROJECT (CREDIT 851-NIR) June 23, 1986 Operatio!-s Evaluation Department 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. WEIGHTS AND MEASURES Metric British/U.S. Equivalents 1 meter (m) = 3.3 feet 1 cubic meter (m3) 35.31 cubic feet I hectare (ha) - 100 acres = 2.47 acres 1 kilometer (km) = 0.624 mile 1 kilogram (kg) 2.20 pounds (lb) 1 metric ton (mton) = 2,206 pounds (lb) ABBREVIATIONS CCCE - Caisse Centrale de Coopfration Economique (French Aid Agency) CNCA - Caisse Nationale de Cr6dit Agricole (National Agricultural Credit Fund) FAC - Fonds d'Aide et de Coop6ration (French Aid Grant Agency) FED - Fonds Europ6en de D6veloppement (European Aid Agency) GON - Government of Niger INRAN - Institut National de Recherches Agronomiques du Niger (National Agricultural Research Institute) KfW - Kreditanstalt fUr Wiederaufbau (Federal Republic of Germany Aid Agency) NIGELEC - Soci4t6 Nig6rienne d'Electricit6 (National Electricity Authority) ONAHA - Office National des Am6nagements Hydro-Agricoles RMWA - Resident Mission in West Africa UNCC - Union Nigerienne de Cr6dit et de Coop6ration (Nigerien Credit Union) CURRENCY EXCHANCE RATES Name of Currency (Abbreviation): Franc (CFAF) Year Exchange Rate Appraisal Year Average 1978 US$1.00 = CFAF 245 Intervening Years Average 1978 US$1.GO = CFAF 226 1979 US$1.00 = CFAF 213 1980 US$1.00 = CFAF 211 1981 US$1.00 - CFAF 272 1982 US$1.00 = CFAF 329 1983 US$1.00 = CFAF 381 Completion Year Average 1984 US$1.00 = CFAF 437 FOR OFFICIAL USE ONLY THE WORLD 8ANK Washington, D.C. 20433 U.SA. Office of D0rect(r*C.nerul Opfatns Evakltion June 23, 1986 MEMORANDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT SUBJECT: Project Performance Audit Report - Niger Irrigation Project (Credit 851-NIR) Attached, for information, is a copy of a report entitled "Project Performance Audit Report - Niger Irrigation Project (Credit 85i-NIR)" prepared by the Operations Evaluation Department. Attachment This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. FOR OFFICIAL USE ONLY PROJECT PERFORMANCE AUDIT REPORT NIGER IRRIGATION PROJECT (CREDIT 851-NIR) TABLE OF CONTENTS Page No. Basic Data Sheet................ ...................... ii Evaluation Summary................................................ iii PROJECT PERFORMANCE AUDIT MENORANDUM 1. PROJECT BACKGROUND ...............................e....... 1 II. PROJECT IMPLEMENTATION AND OUTCOME ....................... 3 III. FINDINGS AND ISSUES 4..................................... 4 A. Pressure to Lend ............5.................. 5 B. Economic Failure; Financial Success 6.............. 6 C. Civil Works Design and Cost Estimation Shortcomings .. 6 PROJECT COMPLETION REPORT I. INTODUCTION ....................... 11 II. PROJECT DESIGN ..................................... 14 III. PROJECT IMPLEMENTATION .......................... .. 16 IV. IMPACT ON AGRICULTURAL PRODUCTION ....................... 22 V. RATES OF RETURN ..................................... 24 VI. INSTITUTIONAL PERFORMANCE AND DEVELOPMENT ................ 25 VII. GOVERNMENT AND BANK PERFORMANCE ......................... 27 VIII. SPECIAL ISSUES .................................... 28 IX, CHANGES IN REPEATER PROJECTS ............................. 29 X* CONCLUSIONS ........................................ 29 Tables 1-5 Charts 1-2 Map IBRD 12837 (PCR) This document has a retricted distribution and may bo umd by recipients only in the performance of their official dutes. Its contents may not othrwise bo disclsed without World Dank authorization. - i - PROJECT PERFORMANCE AUDIT REPORT NIGER IRRIGATION PROJECT (CREDIT 851-NIR) PREFACE This is a performance audit of the First Irrigation Project in Niger for which a credit of US$21.0 million was approved on October 4, 1978. The credit was closed after a one year extension on June 30, 1984 and was fully disbursed with final disbursement in September, 1984. The project was cofinanced by the German Aid Agency (KfW) for the amount of US$9.5 million. This audit consislas of a memorandum prepared by the Operations Evaluation Department (OED), and a project completion report (PCR) dated June 20, 1985, prepared by the Western Africa Regional Office. The audit memo- randum is based on a review of the Appraisal Report (No. 1682-NIR) dated September 11, 1978, the President's Report (No. P-2379-NIR) of September 20, 1978, the Credit Agreement dated February 15, 1979, and the PCR. In addi- tion, correspondence with the borrower, supervision reports and other internal Bank memoranda on project issues as contained in Bank files have been consulted and Bank staff associated with the project have been interviewed. The audit finds the PCR comprehensive and accurate with respect to the project's principal achievements and shortcomings, but notes that the PCR is regrettably lacking in quantitative data on farmer benefits (see also PPAM paras. 26-27). The issues discussed by the audit have been selected because of their relevance to this and other irrigation projects. The draft audit report was sent to the Borrower for comments on April 25, 1986. However none have been received. PROJECT PERFORMANCE AUDIT REPORT NIGER IRRIGATION PROJECT (CREDIT 851-NIR) BASIC DATA SHEET KEY PROJECT DATA Actual or Actual as Z Appraisal Estimated of Appraisal Estimate Actual Estimate Total Costs (us$ million) 21.0 24.5 117 Credit Amount (US$ million) 15.0 15.0 100 Cofinancing 6.0 9.5 158 Date of Board Approval - 10/04/78 - Date Signing 02/15/79 - Date Effectiveness - 10/12/79 121/b Date Physical Components Completed 12/82 12/83 - Proportion completed by above date (Z) 100 85 - Closing Date 06/83 06/84 119/b Economic Rate of Return ()/a 12.5 3 22 Institutional Performance good good - Technical Performanco good good - Number of Main Beneficiaries (family)/a 3,000 3.159 105 STAFF INPUTl / FY75 FY76 FY77 FY78 FY79 FY80 FY81 FY82 FY83 FY84 FY85 TOTAL Identification/Preparation 15.0 4.3 3.4 - - - - - - - - 22.7 Appraisal 4.7 2.5 59.9 27.7 0.4 - - - - - - 96.2 Supervision - - - - 3.1 .7.5 10.4 17.9 10.6 2.5 9.3 71.3 TOTAL 11.7 63.3 2 7 3.5 17.5 10.4 17.9 10.6 T.5 f7T 190.2 CUMULATIVE ESTIMATED AND ACTUAL DISBURSEMENTS FY79 FY80 FY81 FY82 FY83 FY84 Appraisal Latimate (US$ million) 1.0 5.0 10.0 13.0 15.0 15.0 Actual Disbursement (US$ million) - 3.0 6.8 11.8 14.1 15.0 Actual as I of Estimate - 60 68 91 94 100 Date of Final Disbursement September 1984 Principal repaid to date None MISSION JATA Date No. of Mandays Specializations Performance Types of Mission (mo./yr) Persons in field Represented /d Rating/e Trend/f Problems/g Identification 05,'73 2 7 a,d - - - Preparation 02-09/74 3 - - - - - Pre-appraisal 05/75 2 14 a,c - - - Appraisal 01-02/77 6 21 a,b.c,d,e,f - - - Supervision 1 02/79 1 3 a 2 2 P Supervision 2 1079 3 6 a,a,c 2 1 ,T,F Supervision 3 05/80 3 5 a,a,b 1 1 F Supervision 4 11/80 2 7 a,b 2 2 F,M,T Supervision 5 06/81 3 7 atb,c 2 3 F,M Supervision 6 02/82 3 18 a,b,c 2 3 F,M,T Supervision 7 12/82 2 20 a,e 2 3 T,M Supervision 8 07/83 2 7 a,c 2 2 T,K Supervision 9 03/84 1 5 a 2 1 - OTHER PROJECT DATA Borrower Government of Niger Executing Agency Office National des An6nagements Hydro-Agricoles (ONAHA) Fiscal Year of Borrower October 1 - Paptember 30 Follow-on Project Name Irrigation Rehabilitation / Credit Number 1618-NIR Credit Amount (TIS$ million) 9.3 Date of Board Alproval June 25, 1985 /a Namarigoungou scheme only. 7- Calculated from date of Board approval. 7 Source: Planning and Budgeting Department; input as staff weeks. 7_ a - Irrigation engineer; b - Agronomist; c - Financial analyst; d Economist; e - Soil specialist; f - Sociologist. I 1 * Problem-free or minor problems; and 2 * Moderate problems. I = Improving; 2 * Stationary; and 3 * Deteriorating. F * Financial; M - Management; T = Technical; P - Political (Most critical problem listed first). /h See SAR, Niger-Irrigation Rehabilitation Project, Report No. 5161-NIR, dated May 30, 1985. - iii - PROJECT PERFORMANCE AUDIT REPORT NIGER IRRIGATIC, PROJECT (CREDIT 851-NIR) EVALUATION SUMMARY Introduction Niger is one of the poorest African countries. It faces a number of formidable constraints to development, including poor soils and unfavor- able climatic conditions. Limited opportunities for rainfed agricultural improvements have encouraged the Government to give high priority to invest- ments in irrigation schemes and this IDA and KfW-assisted project was the first major irrig tion project in Niger. Objectives The main objectives of the Irrigation Project were: a) to increase Niger's production of rice through the development of irrigated agriculture and b) to improve the income of participating farmers fr,-m their holdings in the irrigation scheme and in surrounding rainfed areas (PPAM paras. 6-7). Implementation Experience The project was implemented with a delay of about two years by a Project Unit responsible to the newly appointed Project Director of ONAHA's regional office at Tillabery. The Project Unit's staff included a large number of consultant advisors who contributed significantly to the project's implementation. Cost overruns, largely attributable to underestimation at appraisal, gave rise to concern during implementation until KfW agreed to provide additional financing. Project costs at completion were US$24.5 million, 17% higher than estimated at appraisal (PPAM paras. 13-17). Results The project had some notable achievements (PCR para. 4.08) but was economically disappointing owing to high construction costs. The re-estimated economic rate of return at 3% is low due to almosy 100% cost overruns on irrigation works, slow uptake by farmers, lower than expected increase in padi production, and negligible impact on rainfed farmers due to drought (PPAM para. 25). From the social and financial viewpoint of the beneficiaries, however, the project was successful (PPAM para. 26). Irrigated land was distributed equitably, in 0.43 ha average size holdings to 3,159 families, who were obtaining irrigated yields and production close to the appraisal - iv - estimate by project completion. Two very successful cooperatives, which also served as water users associations and are responsible for operation and maintenance of the irrigation system, were established in the two principal villages and their almost 100% cost recovery record in respect of credit and operation and maintenance costs is testimony to the financial viability of the farm enterprises. Sustainability Provided the two cooperatives established under the project continue at their present levels of efficiency, and Government subsidies are maintained, the Namarigoungou scheme should remain fully sustair.able. However, the high per hectare and per farmer development costs coupled with the virtual impossibility of recovering the capital investment makes replication difficult to justify. Findings and Lessons The project has confirmed the social and financial viability of intensive irrigation systems in the flood plain of the Niger River, while at the same time drawing attention to the need for lower capital cost approaches. The suggested approach is to develop smaller schemes at care- fully chosen sites where less infrastructure and land levelling would be required, and this approach is to be supported in a follow up project now being prepared (PPAM para 31). The importance of a comprehensive feasibility study, based on well- founded knowledge of the physical conditions, and supplemented by accurate cost estimation of the detailed design plans, is once again emphasized by this project experience (PPAM paras. 28-30). - 1 - PROJECT PERFOMANCE AUDIT MEMORANDUM NIGER IRRIGATION ?ROJECT (CREDIT 851-NIR) I. PROJECT RACKGROUND Context 1. Niger is one of the largest landlocked countries in Africa, cover- ing about 1.3 million kw2, between latitudes 11* and 23* north. Annual rain- fall, which is scanty and irregular, ranges from as high as 800 mm annually in the south to less than 250 mm in the north. About 10 percent of the total land area is suitable for crop production, but has very low potential, and only 2 percent is cultivated at any given time under the traditional rotation methods of cultivation. 2. Rural activities provide employment and livelihood to the vast majority of the population: 85 percent of the population are engaged mainly in subsistence agriculture and about 7 percent live from pastoral activity. 3. In the early seventies, Niger experienced severe drought that reduced livestock herds and food production drastically, and led to importa- tion of some 500,000 tons of cereals over the three years 1973-75. This drought lent force to the Government's primary objective in the rural sector of restoring and maintaining food self-sufficiency through maximizing the rainfall potential and expanding the area under controlled irrigation, with rice as the main crop. 4. The irrigation potential in Niger is estimated at about 270,000 ha, of which slightly more than half is in the Niger River Valley. Out of this 140,000 ha of irrigation potential in the Niger Valley, only 30,000 ha are lowlands ("cuvettes"), which have generally good soils easily irrigable by low-lift pumping. The rest consists of unland terraces where soils are quite heterogenous and high lift pumping is costly. As of the end of 1983, there were about 10,000 ha of irrigable land developed or under development in Niger, of which 6,000 ha, including the 1,500 developed under the project, are in the Niger Valley. Foreign aid from several sources assisted in the development of these irrigat'ion schemes. 5. Project genesis can be traced back to a request for assistance by the President of Niger to Mr. McNamara during the latter's visit to Niger in March 1972. This request was consistent with Government strategy at that time of seeking IDA assistance for the relatively large Namarigoungoa pro- Posed irrigation scheme and complementing this with aid for rehabilitation of existing schemes from other assistance sources, the details of which had been spelled out in a 1972 consultant report on irrigation potentials, commis- sioned by FED (see also PPAM paras. 21-24). -2- Objectives 6. The main objectives of the Irrigation Project were: a) to increase Niger's production of rice through the development of irrigated agriculture and b) to improve the income of participating farmers from their holdings in the irrigation scheme and in surrounding rainfed areas. 7. In addition to its main objectives, the project was intended a) to contribute to Government's goal of developing some 1,000 ha of irrigated land annually through the creation of the Cifice National des Am6nagements Hydro- Agricoles (ONAHA), which was to integrate and strengthen construction and management activities associated with irrigation schemes; b) to introduce a variety of crops other than rice; and c) to apply modern technology to the irrigation schemes and, to some extent, improved methods to the rainfed hold- ings of project participants. Design 8. The project, which was the first major irrigation project in Niger, was based on a feasibility study prepared by consultants in 1974 with FAC financing. It received the highest priority of the Niger Government from its early conceptualization stage. It involved the proposed construction of a 1,550 ha pump fed, double crop, polder irrigation scheme at Namarigoungou, on the left bank flood plain of the Niger River, about 130 km upstreatu from Niamey, together with support services comprising land allocation, extension, research, credit, marketing, processing and health improvements for 5,500 participating farmers, including 2,500 farmers on existing irrigation schemes in the immediate neighborhood. The project included substantial institu- tional support, mainly in the form of consultant services, to a newly estab- lished irrigation agency -- the Office National des Am6nagements Hydro-Agri- cole (ONARA) -- which was responsible for project implementation, under the general direction of the Ministry of Rural Development (MRD). The project also included financing for a feasibility study of a proposed irrigation scheme on an adjacent 3,300 ha site (PCR para. 2.05). 9. The irrigation scheme allows for full water control. It comprises a perimeter dike, two pumping stations for irrigation and drainage, complete irrigation and drainage systems, and land levelling. As such it is an inten- sive, relatively sophisticated scheme which proved costly to construct and which incurs relatively high operational and maintenance costs. Finance Plan 10. KfW which offered to cofinance the project shortly after appraisal and IDA agreed to finance 100% of the net project costs as appraised on a 30:70 basis, while Government contributed indirectly through financing investment and operating costs of about US$2.9 million for two regional units of ONAHA outside the project area. The cost overrun of US$3.5 million was also financed by KfW. -3- Pre-Implementation Processing 11. Substantial preparation and appraisal efforts were needed extending over a period of four yeirs. The end result was that the project basically remained the same as originally identified. Negotiations finally took place in July 1978 after engineering design had been completed, and Board presen- tation took place in October. The Board had three questions: the first question related to why the project had taken so long to prepare, to which the explanation was that pre-appraisal investigations had advised against financing because of technical and financial risk, which was only lessened after the promising results of the FED financed Toula scheme (see PPAM para. 23) became available, and that processing of the project was further delayed until detailed engineering design had been completed. The second question requested an explanation of the 100 percent external financing (Bank US$15.0 million, KfW US$6.0 million), the answer to which hinged on the poverty and limited growth prospects of Niger and the fact that Government had already financed the engineering studies, and had agreed to exempt the projIct from import duties, and to finance the establishment and running costs of two other regional offices of ONAHA. The third question wanted to know why water charges were so high, to which the answer was that they were justified by the high cost of irrigation development and the expected benefits. They were also in conformity with existing regulations. 12. Difficulties in meeting conditions of credit signing and effective- ness resulted in delayed project start-up, and the credit was finally declared effective in October 1979, more than six years after the original identification. II. PROJECT IMPLEMENTATION AND OUTCOME 13. Actual construction, by force account, eventually commenced in April 1980 and civil works were completed by December 1983, after a slow start up, but with no major problems. Construction cost overruns, however, were considerable and were largely due to underestimates and other misjudgements, either during final design or at appraisal (PCR paras. 3.05, 3.15-3.17, 7.02). 14. The local project manager adapted well to his assignment, and the relatively smooth implementation process can in large part be attributed to his leadership (PCR para. 6.01). Land allocation was handled most equitably ani two cooperatives, which also served the dual role of water users associations, functioned well as credit sources, and attained unusually good cost recovery records (PCR paras. 6.02, 6.07). 15. The technical assistance staff, provided through a cinsultancy con- sortium, established good rapport with the project manager and his staff, and provided crucial support, including training, in several disciplinary fields. They could be faulted, however, on grounds of inflexibility and lack of cost saving consciousness in regard to civil works construction (PCR para. 3.20). 16. The irrigated area actually attained (1,304 ha) was 97% of appraisal target. An efficient, locally recruited, agricultural extension service was created (PCR para. 6.08). Yields at 4-5 ton/ha for wet and dry season crops were generally up to expectations, allowing for seasonal differences, and production through the implementation stage was about 95% of appraisal target, which is commendable (PCR para. 4.02). On the other hand, rainfed farming impact was negligible, due to drought. 17. The disappointing economic rate of return (3%) is primarily due to the high capital cost of the irrigation system (a risk element which was fully appreciated by the Bank from the identification stage, and which was the main reason why the project was initially turned down after pre-appraisal), and to slower incremental production than estimated at appraisal (PCR paras. 5.01-5.02). However, socially and financially it was a success, with'a particularly good record of technological transfer in the field of irrigated rice farming. Notable achievements in this area included introduction of animal traction and foot-operated threshing machines, effective seed multiplication at the farmers' group level, and improved fertilizers and pest control practices (PCR para. 4.08). ITI. FINDINGS AND ISSUES Overview 18. In purely economic terms, the project must be judged a failire, but in social terms the project has undoubtedly brought considerable advantage to beneficiaries of the irrigation improvements, who are now enjoying improved incomes from their new holdings. Related Experience 19. As with many other high cost irrigation projects, experience in this project highlights the dilemma, common to many developing courtries, in justifying major capital investments intended to benefit a comparative few while so many others are unaffected and remain in poverty.1! 20. A test of Government commitment to the principle of continuing rural development will be the extent to which an attempt is made to monitor the financial situation of the project beneficiaries and to recover a proportion if the capital costs of the irrigation system from these beneficiaries in order to generate funds for further investments designed to assist other, less fortunate, rainfed farmers. 1/ See also PPAR, Cameroon: Second SEMRY Rice Project (Loan 1512-CM/Credit 763-CM), OLD Report No. 5156 dated June 25, 1984 (para. 53), and PPAR, Chad: Lake Chad Polders (Credit 592-CD), OED report under preparation (PCR para. 30). -5- Issues Arising A. Pressure to Lend 21. The project was notable for the very long (65 month) period between identification and negotiations. The course of events during that time is instructive for the way it illustrates the considerable care and attention given by the Bank in its attempt to determine whether or not the project was worthy of support. Serious doubts first arose with the release of a 1973 Bank identification report, prepared prior to completion of the consultant feasibility study but based on a similar design concept, with an estimated 12% economic rate of return. This report prompted several Bank staff to question the viability of the project as proposed on technical (double cropping complexity; canal lining; individual land holding size), and economic (cost and yield) considerations. Receipt of the feasibility report about six months later, which estimated the ERR as 18%, while reassuring from the economic viewpoint, did not reduce the Bank's technical concerns, and prompted a pre-appraisal in May 1975 to review and evaluate the project. The mission concluded that the re-evaluated rate of return (7.7%) would be marginal at best, due to the high cost of civil works. 22. Release of the final pre-appraisal mission report (in April 1976) led to a further careful deliberation within the Bank regarding the merits of proceeding with the project. The decision of the Bank at the conclusion of this round of discussions was to withdraw support from the project, and the Government was so informed in September 1976. 23. Bank withdrawal prompted a high level appeal for reconsideration from the President of Niger in November 1976 to which the Bank, in order to maintain good relationships, responded favorably with a proposal to re-examine the project. Positive results reported for the smaller, but similar, Toula project made available to the Bank about that time were used to justify this re-examination which led to appraisal in February 1977. The project as appraised was conceptually similar to that originally identified almost four years earlier and had an almost identical rate of return -- 12.5%, but this time around Bank support was justified on the grounds of high priority for food self-sufficie-cy and the fact that the project would initiate an institutional reform through establishment of a new irrigation agency. 24. As the above sequence of events show, the delay in processing was largely due to the fact that the Bank had strong economic and technical reservations, and it is unlikely that the project would have reached the Board without strong persistence from the Government side. In the light of subsequent events, one could argue that the Bank should have resisted the pressure to re-examine the project and instead more strongly pursued efforts either to rehabilitate existing irrigation schemes or to support smaller, simpler and lower cost new schemes. - 6 - B. Economic Failure; Financial Success 25. In contrast to the poor economic rate of return (3%), which will come as no surprise to those Bank staff who expressed skepticism from the start in regard to the project's viability on the basis of high development costs, the financial return to beneficiary farmers from their irrigated plots appears to be very favorable. 26. Although the PCR does not present any direct, quantitative evidence of farmers' benefits, there is abundant indirect evidence which points towards the fact that farmers' incomes have improved considerably since they commenced cultivation of their allocated plots. Perhaps the strongest sup- porting evidence is the fact that cost recovery through the two project cooperatives has been almost 100% (99.3% for Djambala cooperative and 96.7% for Namarigoungou cooperative in April 1984). This recovery rate is particu- larly worthy as it includes command area operation and maintenance charges, energy, and pump maintenance and some provision for eventual pump replace- ment, as well as input supplies (but not capital costs of the irrigation system). Other indirect evidence of the benefits is the fact that yields are more or less comparable with appraisal estimates (PCR Annex 1, Table 4), and that the project experienced no shortage of applicants and very few benefici- aries have had to be suspended because of debt repayment delays. The provision of subsidized fertilizer and agricultural equipment, which continued to be heavily subsidized throughout the implementation period, contrary to Bank intent, provide a further boost to farmers' incomes. 27. The absence of quantitative data on farm incomes in the PCR is regrettable, not only because the improvement of farm incomes was one of the two principal project objectives, but also because of the importance to Government of recovering at least a sizeable proportion of capital investment costs in such high cost schemes as this. The absence of accurate income information based on representative farm budgets inhibits Government from developing a fair and equitable cost recovery policy. C. Civil Works Design and Cost Estimation Shortcomings 28. The PCR (para. 3.15) draws attention to several serious cost under- estimates in the staff appraisal report which are variously attributed to inadequate surveys (resulting in a greater than expected amount of land levelling, and to deeper and more extensive drain construction to combat alkalinity); poor judgements on the part of either the design consultants or the appraisal mission (cement quantities, numbers of buildings, force account efficiency, and construction scheduling); and increased excavation costs resulting from a change in construction technique following appraisal (compacting as well as complete canal lining). 29. The above factors largely explain the approximately 100% construc- tion cost overrun. One or more of these items, if correctly forecast, could have lowered the estimated ERR below the threshold figure considered accept- able for Bank operations at that time, and led either to Bank rejection of the project for a second time, or to a more critical reassessment of the design parameters with a view to effecting further reductions in unit area development costs. - 7 - 30. In this regard, the question of optimum size of individual command area units within the total perimeter may be important. The identification report envisaged four separate irrigation zones of about 400 ha each, each to be served by its own pumping station. This concept was based on a nearby project successfully undertaken in the early 1970's with Taiwanese assis- tance. The feasibility study proposal prepared by consultants, and accepted by the appraisal mission, was based on two separate zones rather than four, which could have contributed towards the high construction costs due to the need for larger and stronger structures to accommodate the higher flow volumes, including the higher embankments to support the pump heads necessary for effective water distribution over the larger command areas. It is not clear why the two zone approach was adopted rather than four, but the adverse consequences of this large scale approach seem incontestable, and the lesson of experience has been learned to the extent the current irrigation philo- sophy in Niger and other countries in the region is one of "small and inten- sive" pump schemes which are expected to prove less costly to construct and operate. 2/ The audit feels that the Bank may have lost an opportunity to influence favorably a developing trend towards smaller irrigation command units by not pressing the consultants in that direction at the design stage, or to introduce such modifications at appraisal.3/ Findings and Lessons 31. The low economic rate of return inherent in the comparatively large scale, high unit area cost approach adopted in the Namarigoungou scheme and the even higher unit costs determined for the neighboring Gabou-Bonfeba pro- posed scheme, which was feasibility studied under the project and rejected (PCR para. 3.08), led to a reconsideration of irrigation strategy on the part of the Government and IDA in favor of smaller, more simply designed schemes in the most opportune areas where unit area development costs are less. A small rural operations project incorporating this approach is currently being prepared for Bank support. 32. Furthermore, given the favorable financial returns from double cropping in this and other intensive irrigation schemes, IDA is now assist- ing with an Irrigation Rehabilitation Project4/ designed to improve the facilities on 16 existing irrigation schemes to enable them to support two high yielding rice crops on an intensive system similar to that successfully applied under the project. 2/ See also Senegal: Small Rural Operations Project (Loan 991-SE). 3/ The Agriculture and Rural Development Department comments that the issue of project size remains contestable as it is not clear that small pump schemes would be less costly to construct and operate. The adoption of the two zone approach rather than four has increased the civil works for embankment but on the other side the number of standby pumps and the staff operational requirements for pump operation have decreased. 4/ Staff Appraisal Report No. 5161-NIR, Niger: Irrigation Rehabilitation Project, May 30, 1985. 一9一・ がkZ分介2細ぐ NIGER IRRIGATION PROJECT (Cr. 851 - NIR) PROJECT COMPLETION REPORT June 20, 1985 Western Africa Projects WAPAC 4 A dà - 11 - NIGER IRRIGATION PROJECT (Credit 851-NIR) PROJECT COMPLETION REPORT I. INTRODUCTION A. Background 1.01 Government agricultural policy since the droughts in the early 1970s has been to give the highest priority to food security and to develop strategies to exploit the potential of rainfed and irrigated agriculture. 1.02 In controlled irrigation, Government aims at a modest rate of development of 1,000 ha per year. The Niger river region offers the largest potential for irrigated agriculture in Niger: some 140,000 ha are irrigable 1/ However, this area cannot be fully developed for double cropping until the flood of the Niger river is regulated by a dam (proposed by Government at the Kandadji site). Under the present hydrological conditions, a total of some 16,000 ha could be developed for double cropping irrigation along the Niger river. At the current rate of development, the goal of 16,000 ha would -ake another 10 years to attain, since the present area under irrigation is about 6,000 ha, including the 1,500 ha developed under the Irrigation Project. 1.03 The project is located in the Sahelian part of Niger, in the floodplain on the left bank of the Niger river, about 150 km northwest of Niamey (see map). The Association approved a credit of US $15.0 million on October 4, 1978 (Cr. 851-NIR) to finance it. This was the Bank Group's fifth operation in Niger's agricultural sector and its first irrigation project. The other Bank Group operations in agriculture were: (a) 1970: Cr. 207-NIR (US $0.6 million) for agricultural credit, technical assistance to "Caisse Nationale de Cridit Agricole" (CNCA) and training of extension staff (see also PPAR, OED Report no. 2055, dated May 10, 1978); (b) 1973: Cr. 441-NIR (US $2.0 million) for a rdlief fund to help re-establish the autonomy of people in drought-stricken areas by helping them to redevelop and improve their farms and herds (PPAR, OED Report no. 2588, dated June 29, 1979); (c) 1976: Cr. 608-NIR (US $10.7 million) for the Maradi Rural Development Project to provide farmers with production improvement packages, farm roads, wood fuel plantations and improve livestock services, and to study the irrigation potential of the area (PPAR, OED Report no. 3992, 1/ Of this total, only 30,000 ha are lowlands ("cuvettes") easily irrigable and suitable for rice cultivation. The rest consist of highland terraces with heterogeneous soils that would be too costly to irrigate under present conditions. - 12 - dated June 21, 1982); and (d) 1978: Cr. 800-NIR (US $4.5 million) for a Forestry Project to help Government establish pilot irrigated and rainfed tree plantations, improve the rural forestry activities of the Forestry Department; finance an extension training program, and to provide assistance to research and a workshop maintenance unit (PPAR; OED Report no. 4569, dated June 21, 1983). 1.04 The results obtained from ongoing area-based rainfed productivity projects such as Maradi (Cr. 1026-NIR) and Dosso (Cr. 967-NIR) have been below those originally expected. The second Maradi project has been substantially reformulated, while Dosso will not, as Government had requested, be extended beyond its closing date. The Zinder Seminar of November 1982, a national seminar held to determine Government's strategy for intervention in rural areas, recognized this unsatisfactory performance and came up with the following recommendations: (a) regionalization of project administration, whereby small but mobile teams of high quality field staff would operate at the district level; (b) introduction of mobile, better trained extension teams able to adapt to changing local demands; (c) promotion of farmer participation and self-management; and (d) small-scale, village-level tests and demonstration activities to promote "bottom-up" technical packages. B. Identification 1.05 In February 1973, a mission from the Bank's Regional Mission in West Africa (RMWA) discussed with the Government of Niger (GON) a study, prepared by a consultant firm under FED financing, that would deal with current rice development schemes along the Niger river, the feasibility of improving existing schemes, as well as the establishment of new rice irrigation projects. A Government Working Group reviewed this study and prepared detailed proposals for possible projects. Government requested that the Bank finance the Namarigoungou Irrigation project; other donors were asked to finance the rehabilitation of existing projects. 1.06 In May 1973, a joint RMWA/HQ mission visited Niger to identify the project. Although the ERR was expected to be marginal, the mission concluded that a pumping irrigation scheme, with full water control, would be the most viable solution. The draft Identification Report was discussed in September 1973. The Bank accepted the mission's conclusions, but realized that in view of the strained economic situation in Niger, especially given the continuing drought, GON's contribution to the project would be minimal. A final decision was postponed pending the outcome of a feasibility study financed by FAC and carried out by a consultant firm. 1.07 The feasibility study was prepared in the first half of 1974 and submitted in July 1974. As presented in this study, the technical conception of the irrigation scheme was similar to that proposed in RMWA's identification report: a net area of 1,500 ha of fertile alluvial soils along the Niger river would be protected against flooding and would then be cropped twice a year under full irrigation through pumping. After review of the document, the Bank concluded that, - 13 - although the study was not presented very clearly, it contained all the information necessary for project appraisal. C. Appraisal 1.08 The project was preappraised in May 1975. The mission concluded that the economic viability of the project was marginal due to the high cost of civil engineering works. It was also the mission's assessment that not enough water would be available for double cropping and that the cohabitation of two conflicting tribes in the project area would create too many problems. The initial Bank position was therefore not to accept the project for financing. 1.09 Because of the continuing drought and the resulting unreliability of rainfed agriculture, Government's policy of seeking self-sufficiency in food through irrigation development was reinforced and became high priority. In a letter dated November 26, 1976, the President of Niger personally asked the Bank's President to reconsider the Bank's position on the proposed project. Meanwhile, excellent results had been obtained on the FED-financed Toula perimeter (about 10 tons of paddy per year in two crops). Based on these favorable results, the Bank decided to go ahead with the proposed irrigation project, and to send an appraisal mission without delay. 1.10 Appraisal took place in January/February 1977. The mission was composed of a financial analyst, an agro-economist, a civil engineer, an agronomist, a sociologist and a soils specialist/drainage engineer. Except for the irrigation component, which was based on the feasibility study prepared in 1974, the project was prepared by the mission. Although the mission considered the project to be expensive and its return low, it recommended that it be executed for the following reasons: (a) it was given the highest priority by Government, which believed that only irrigation could help the country to regain some degree of food self-sufficiency; and (b) the project would help Government to bring about major institutional reforms by (i) creating a national agency responsible for the construction of all new irrigation schemes and for providing organizational and technical services on all schemes; (ii) equipping this agency with the staff and material to build new schemes at lower costs than in the past by introducing force account methods; and (iii) introducing the principle of cost recovery from farmers and consumers (see issues paper, March 1977). 1.11 Three issues remained to be settled with Government: (a) concrete lining of all primary and secondary canals: the mission recommended that concrete-lining be done only to the extent required by soil conditions (about 50% of total length); this ran counter to Gfnie Rural engineers' conviction that all primary and secondary canals should be lined; (b) IDA's level of financing: IDA funds available for the project were insufficient to cover all project costs and Government would have to seek additional financing; and (c) capital cost recovery: Government was reluctant to introduce the capital levy proposed by the appraisal mission. - 14 - 1.12 A decision meeting was held on March 18, 1977. The meeting accepted the arguments (para 1.10) in favor of the project and discussed the principal issues remaining (para 1.11). It was decided that: (a) canals should be concrete-lined only when technically and economically justified; (b) as IDA financing (US $15.0 million) would cover only about 75% of total project costs, the Bank would impress on Government the need to seek other sources of financing; and (c) the Bank would support the mission's recommendations concerning the capital levy on farmers and an increase in the consumer price of rice. It was further decided that processing beyond yellow cover stage should await completion of the detailed engineering study, in order to have more precise cost estimates. 1.13 In July 1977, the German KfW expressed interest in co-financing the project, because the cancellation of its intended financing of two smaller irrigation schemes released funds for the Namarigoungou project. 1.14 The draft yellow cover appraisal report was reviewed in July 1977, with no major changes in project design. The results of the final engineering study became available in February 1978 and total project costs were estimated at about US $21.0 million. This estimate was incorporated into the green cover appraisal report issued in April 1978. D. Negotiations, Board Presentation and Signing 1.15 Negotiations took place from July 12-15, 1978. The following major changes were made to the draft Credit Agreement: (a) ONAHA (Office National des Am6nagements Hydro-Agricoles) would be responsible for carrying out project construction and agricultural development, as originally planned, while the Ministry of Rural Development would maintain an overall supervisory role; (b) because of increases in the consumer price of rice to the level of the import parity price, no further increase of the consumer price was considered justified and the corresponding articles in the draft agreement were thus deleted at the delegation's request; and (c) it was agreed that concrete lining of canals would be undertaken based on criteria mutually acceptable to the Borrower and the Association. 1.16 Conditions for Board Presentation were: (a) that ONAHA be established, and (b) that a firm be selected to provide the internationally recruited staff. However, because the bids received for the provision of technical assistance were very similar and thus difficult to evaluate, Government was unable to award the contract before the date set for Board Presentation, nor were ONAHA's decree and statutes approved by the Council of Ministers. The Bank, however, considered that the conditions were substantially fulfilled and the project was presented to the Board on October 4, 1978. Signing of the Credit and Project Agreements were then held up subject to fulfillment of the above conditions. The agreements were signed on February 15, 1979, and the Credit became effective on October 12, 1979 (para 3.02). - 15 - II. PROJECT DESIGN A. Project Objectives 2.01 The objectives of the project were twofold: (a) to increase Niger's production of rice through the development of irrigated agriculture, and (b) to improve the income of participating farmers from their holdings in the irrigation scheme and in rainfed areas. The project was to reach full development at completion. 2.02 In addition to its main objectives, the project was intended (a) to contribute to Government's goal of developing some 1,000 ha of irrigated schemes annually through the creation of ONAHA, which was to integrate and strengthen construction and management activities hitherto carried out on the irrigation schemes by the "Service du G6nie Rural" (GR) and the "Union Niggrienne de Cr6dit et de Coopgration" (UNCC), respectively; (b) apart from producing rice, to produce a variety of vegetables, coarse grains and fodder crops; and (c) to apply modern agricultural technology to the irrigation schemes and, to some extent, improved methods to the rainfed holdings of project participants. 2.03 Full water control under the project was to make possible the replacement of 500 ha of floating rice existing before the project (yield about 1.0 ton/ha) by 1,350 ha of improved rice, yielding 4.8 tons/ha per season. In addition, about 150 ha of irrigated upland crops would be grown, including sorghum, yielding 2.5 tons/ha; cowpeas, yielding 1.2 tons/ha; and tomatoes, yielding 25.0 tons/ha. 2.04 The main beneficiaries of the project were to be the 3,000 farming units (16,000 people) who would take up cultivation in the Namarigoungou polder (average farm size 0.5 ha). Other beneficiaries were to include the 2,200 farming units already cultivating on the existing schemes. B. Project Description 2.05 The project was to consist of: (a) construction of the Namarigoungou polder 2/ (dike, irrigation and drainage systems and on-farm works); (b) provision of services to farmers who would take up cultivation on the Namarigoungou polder (1,500 ha of net area) and on the Lossa III scheme (162 ha of net area); and to the farmers already cultivating on 1,215 ha under irrigation development. Such services were to include: delivery of irrigation water, assistance in organization, extension, applied research and 2/ See IBRD Map No 12837 (PCR) - 16 - seed multiplication, procurement of inputs, agricultural credit and crop marketing; (c) improved health services; (d) training for project staff, farmers and artisans; (e) provision of the necessary physical infrastructure and equipment to carry out project activities; and (f) preparation of a feasibility study for a nearby po_er (Gabou-Bonfeba, 3,300 ha net area) north of Namarigoungou. 2.06 Total project cost was estimated at US $21.0 million. (CFAF 5.1 billion), net of taxes. The IDA credit totalled US $15.0 million or 71% of total costs. KfW was to finance DM 12.5 million (US $6.0 million). The economic rate of return over 25 years was estimated at 12.5%. 2.07 Irrigation water was to be provided by pumping from the Niger river, which has water of excellent quality. Annual incremental production after project completion was projected to consist of: 12,500 tons of paddy, 400 tons of sorghum and 950 tons of various vegetables, and, on the rainfed holdings of the Namarigoungou farmers, 1,600 tons of cereals. In addition, production of wood poles was to have an annual value of some CFAF 16 million, and the livestock sub-project was projected to generate some CFAF 6 million annually. Production increases on the existing scheaes were estimated at 900 tons of paddy in the polders, and 700 tons for cereals under rainfed conditions. III. PROJECT IMPLEMENTATION A. Effectiveness and Start-up 3.01 The legal documents required the fulfillment of three conditions before the project could become effective: (a) the KfW had to advise the Association that all conditions for the initial disbursement of the KfW loan were fulfilled; (b) an initial amount of CFAF 250 million had to be deposited by Government in a special account; and (c) the Borrower had to contract with a firm, under terms and conditions acceptable to the Association, for the provision of a technical advisor and other key personnel for the Regional Director at Tillabery. 3.02 It was expected during negotiations that Government could meet these three conditions within two months, and the date of effectiveness was scheduled for April 16, 1979. The contract for technical assistance was signed on February 15, 1979, but, because houses were not completed and delivery of equipment was delayed, the conaultant engineers' team did not arrive until October 1979. Regarding the initial deposit of CFAF 250 million into the special account, which was also an effectiveness condition, CNCA opened an overdraft facility of up to CFAF - 17 - 398 million for the project in May 1979; this overdraft facility was replaced by the Government's deposit of CFAF 250 million into the special account in March 1981. On this basis, and once KfW advised IDA that all its conditions had been fulfilled and the Agreement was published in the Nigerien Government Gazette of August 16, 1979, the Credit finally became effective on October 12, 1979, about six months later than expected. 3.03 Initial delays in Credit signing and in the award of contracts for engineering services, construction and procurement of equipment resulted in a one-year delay in the construction of irrigation works. Although earthmoving equipment arrived in mid-1979, construction could not start until March 1980 because work on the protection 14ke, which was essential to the construction of other works, could not proceed between July 1979 and February 1980, when the level of the Niger river was at its annual peak, preve.ting access to the works. B. Revisions Early in Implementation 3.04 Before construction works began, a Bank mission and technical staff from ONAHA carried out a desk study to compare the merits of concrete lining of main canals and laterals vs. an entirely unlined irrigation network (section 3.05 of the Credit Agreement). The study showed that the lower investment costs of the unlined alternative were offset by the discounted value of the additional water losses and maintenance costs. In addition, the mission considered it inadvisable to increase ONAHA's responsibilities in its initial period of existence through the creation of a strong maintenance unit that the unlined alternative would have necessitated, and therefore opted to line all primary and secondary irrigation canals. 3.05 Actual costs for concrete-lining exceeded CFAF 1.0 billion, or about 70% of the total cost for irrigation canals (excluding structures). This represented about 20% of total cost for irrigation works construction, and about 15% of total project cost. The construction technique used by the project consisted of excavating the future canal body, in the case of permeable soils, to 0.20 m below the planned canal bottom and replacing it by compacted fill of selected impermeable material. This method was maintained by the project engineer for safety reasons throughout the construction phase. It is believed that this method rendered concrete lining superfluous, and that if this unnecessary concrete-lining had not been done, an estimated 50% could have been saved on the investment costs for the irrigation canals. 3/ 3/ Actual savings (50%) would have represented less than the 70% mentioned earlier in this paragraph, because earth canals require a larger cross section than concrete lined canals to convey the same volume of water. - 18 - 3.06 Because of the great demand for irrigated plots, the production of fodder crops on 25 ha under irrigation, foreseen at appraisal, was omitted. Although not included in the appraisal, drinking pools for cattle were constructed on the late-al canals at the terrace side of the project at the Bank's suggestion. These pools supplanted the original river water supply, which had been disrupted by the project. This solution proved to be a good alternative to the traditional cattle passages. 3.07 Maintenance problems with pumps in the perimeters that were already developed caused the irrigation water supply to those perimeters to be unreliable, and the continuing drought seriously affected the rainfed areas. This made the provision of services to these irrigated perimeters and the rainfed holdings (para 2.02 c) less effective and less intensive than foreseen at appraisal. 3.08 The feasibility study for the nearby Gabou-bonfeba area was only partially implemented. A preliminary engineering study showed that development costs would be unrealistically high (over US $30,000 per ha). After discussions with consultants and Government, the former were requested to try to find an alternative and cheaper solution by increasing the net/gross land ratio through the elimination of unfavorable areas and by reviewing technical aspects, such as concrete-lining of canals, hydraulic characteristics, plot sizes, etc. The resulting revised unit price was still about US $22,000 per ha -- too high to economically justify execution of the project or even further engineering studies. 3.09 it would probably have been possible for consultants to select a number of small perimeters where soils and topographic conditions are favorable. By simplifying design, a viable alternative project could have been developed. However, as the disbursements for this component had already exceeded the category by 10% and, in the light of the emerging priority to the rehabilitation of existing perimeters, the Bank suggested the study be discontinued at the preliminary design level. C. Implementation Schedule 3.10 The actual implementation schedule of works as compared to the appraisal forecast is illustrated in Annex 1, Table 1. Anticipated and actual completion dates of construction of irrigation infrastructure, buildings and power lines are given below: Appraisal Actual Delay (months) Access roads, buildings, stores July, 1979 June, 1980 11 Dike construction December, 1979 July, 1982 31 Irrigation infrastructure - upstream district July, 1980 July, 1982 24 - downstream district May, 1981 December, 1983 31 High tension line Tillabery - February, 1980 June, 1981 16 Namarigoungou - 19 - 3.11 Major delays occurred during project start-up. Once the works got underway, however, the project caught up, since: (a) 50% more equipment was procured when force account implementation proved much less efficient than anticipated at appraisal; (b) additional equipment was hired from ONAHA's central equipment pool; (c) part of the earthworks were contracted out; and (d) the project gained experience over time. This improved performance can be shown as follows: Total Area Operational (ha) Incre- mental Rice Nurseries Upland crops Forest 4/ Total area (ha) June, 1981 212 12 - - 224 - June, 1982 587 38 15 - 640 416 June, 1983 1,142 56 30 50 1,278 638 June, 1984 1,304 79 41 70 1,494 216 3.12 By the original target dates, works on the transmission line and on the dike had not yet started, and only 15% of the rice perimeter was operational. All planned works were finally completed, however. It is clear that the required start-up time was seriously underestimated, that the efficiency of force account implementation was overestimated and that the original time schedule was unrealistic. In terms of costs, however, force account implementation in Niger compares favorably with contractors' work; this is one reason that this type of intervention has been retained under the proposed Irrigation Rehabilitation project. The main factors causing delay were discussed in para 3.03. D. Reporting 3.13 Satisfactory quarterly progress reports were prepared by Project Management. In the course of project execution, the time elapsed between the end of the reporting period and the submission of the progress reports to Headquarters decr.ased from about 4 months to 2 months. E. Procurement 3.14 Major contracts for goods and civil works (pumping stations) were awarded following international competitive bidding (ICB) for a total value of about US $4.0 million (as foreseen at appraisal). The total amount awarded through local competitive bidding (LCB), mainly the construction of buildings, amounted to about US $4.0 million (as against only US $1.5 million estimated at appraisal). This included about US $2 million for goods and services for which the bids exceeded US $50,000 and which, according to the Credit Agreement, should have been procured 4/ Excluding 15 ha wind breaks. - 20 - through ICB, i.e., (a) the electrical power line, which was negotiated directly with NIGELEC, the authority with a monopoly for constructing power lines in Niger; (b) fuel and lubricp-nts (free of tax), for which contracts were negotiated between GON and local oil companies; and (c) cement, the justification given for waiving ICB being that only small lots could be accepted at a time due to lack of storage facilities. Even so, deliveries of fuel and cement were erratic (para 3.22). F. Quantities and Costs 3.15 Quantities and costs are based on the information presented in the Project Completion report 5/ (March 1984) prepared by ONAHA. In Table 1, actual and estimated quantities of main works are compared. The main deviations from appraisal are: (a) excessive cut and fill for irrigation canals, which can be explained in part by the construction method, i.e., the replacement of sandy soils by impervious materials (para 3.05), which was not foreseen at appraisal; (b) the amount of compacted fill for the dike, which greatly exceeded the appraisal estimate, as the project decided to widen the dike body to facilitate the passage of construction equipment for compaction; (c) the quantity of concrete used, which was much larger than foreseen, partly due to an underestimate in the appraisal report, and partly to excessive use of concrete, especially on some of the structures (canal crossings, concrete lining, etc., necessitated by the irregular shape of the bottom and side slopes of the canals, which was not accounted for at appraisal); and (d) heavy land leveling on some 20% of all plots (over 2,OO ms/ha), which was, because of insufficient survey work, not foreseen at appraisal. In addition to increasing costs, this heavy leveling removed topsoil in the borrow areas, and resulted in low agricultural productivity there. 3.16 Table 2 in Annex 1 compares actual and estimated project costs. Actual costs, expressed in local currency, amounted to CFAF 7.5 billion (US $24.5 million) or about US $15,000 per hectare 6/ , 46% higher than the appraisal estimate of CFAF 5.1 billion (US $21.0 million). Cost overruns based on US dollar currency equivalents, however, amount to only 17%. This lower figure is explained by the favorable overall exchange rate during the project implementation period (average of CFAF 275 per US dollar during implementation, as against 245 at appraisal). 3.17 The cost overrun on the irrigation works construction component was almost 100%, which accounts for the global project cost 5/ Available in West Africa files, consisting of a main report (150 pages) and 4 annexes: 1 - Agronomy, 2 - Irrigation, 3 - Management of Cooperatives, and 4 - List of Beneficiaries. 6/ Average cost for the whole construction period. However, efficiency improved over time, and construction costs during the last project year were reduced by about half. - 21 - overrun. Reasons for the high costs of the irrigation component were mentioned earlier in this report (para 3.05) and can be summarized as follows: a) underestimation of quantities and unit costs at appraisal; b) excessive use of concrete and earthworks; and c) overestimation of the efficiency of force account implementation. The cost overruns were financed from additional funds made available by KfW (DM 10.0 million or about US $3.5 million), bringing the total to US $24.5 million or CFAF 7.5 billion. IDA's share in project costs was thus reduced from 71% *o about 61%. 3.18 Cost savings occurred because some of the agricultural development services and agricultural equipment components were not executed. As mentioned earlier (para 3.07), unreliable irrigation water supply and erratic and insufficient rainfall made investments in the existing perimeters and in the rainfed areas unattractive. G. Disbursements 3.19 The IDA credit of US $15.0 million and the KfW credit of DM 22.5 million (about US $9.5 million) were disbursed between October 1979 and April 1984. Due to initial implementation delays, disbursements started one year later than anticipated at appraisal. However, once project implementation started, actual disbursement was in line with the appraisal forecast. For practical purposes, the final time discrepancy between actual and appraisal estimates was only 6 months (Annex 1, Table 3). By the original target dates for project completion and full disbursement, 91% and 94% of the funds, respectively, had been disbursed. Actual disbursements compare favorably with the standard profile for irrigation projects in the East and West Africa regions (Annex 2, Figure 2). H. Performance of Consultants, Contractors, Suppliers and Borrower 3.20 The lack of experienced local staff was a major constraint, and Government agreed to the recruitment of a relatively large number of expatriates. Consultants provided an expatriate technical advisor to the Project Director, in addition to ten other expatriate specialists. The performance of the consultants was satisfactory, except for the following points: (a) project design and subsequent modifications followed standard practices in similar projects in Niger; no effort was made to find alternative solutions in order to reduce project costs; (b) practically no design adjustments were introduced during implementation, despite the high development costs, i.e., heavy leveling in certain areas, etc.; and (c) a combination of difficult working conditions and personal problems caused seven different persons to occupy the post of expatriate financial manager. Most of them did not remain in post for very long and the performance of those who did remain was generally poor. Fortunately, a qualified candidate was selected towards the end, and that person managed to clear the accounts before the closing of the project. 3.21 Technical personnel, as well as students from the technical school of Kollo, received on-the-job training from the expatriate staff. The results were positive and local staff is now available to replace - 22 - part of the expatriate staff under the follow-up project. Efficient training caused performance to improve during project implementation. Since no alternative employment was available at the close of the project, most of the newly trained staff were released from their duties. Some found jobs in the private sector, but others are awaiting the start of new project activities. 3.22 International contractors were employed for the construction of 2 pumping stations; the electric line was built by NIGELEC. Delays in implementation were caused by local contractors. The construction of buildings, offices and workshops was contracted locally, but had to be completed on force account when contractors defaulted, and there were difficulties in the delivery of cement and fuel. IV. IMPACT ON AGRICULTURAL PRODUCTION A. Incremental Output 4.01 Production targets (para 2.07) for vegetables and livestock were not achieved, because these components were scaled down (para 3.07). Due to continuous drought, frequent pump failure, deterioration of the irrigation systems resulting from lack of maintenance, and an unreliable supply of irrigation water to existing rice perimeters, production of rainfed crops was negligible and that of irrigated rice on existing perimeters below target. Project management concentrated its efforts on the new rice perimeter of Namarigoungou, the only one for which reliable production data are available. The European Aid Agency (FED) provided technical assistance to the Toula perimeter, which was originally included in the project. 4.02 Paddy production on the Namarigoungou scheme for the period 1980/84 increased from 500 tons 7/ to 6,260 tons (1984 wet season), or about 95% of appraisal target (Table 4). This slight production shortfall is due to the lower than expected cultivated area: 1,304 ha instead of 1,350 ha, although the gross area developed for rice was 1,383 ha. Constraints in the cropping calendar, mainly causee by a cold spell in December-January and low river discharges in the dry season, made it impossible to cultivate the nursery areas (79 ba) which need to be established before harvesting the previous crop; this was not foreseen at appraisal. Yield levels fluctuated from one season to the next, the main reason being serious pest attacks (para 8.03) in some seasons. B. Technology Transfer 4.03 As far as irrigated rice farming is concerned, most technical targets were acbieved. Due to drought, however, the impact on rainfed 7/ "Without project" situation. - 23 - farming was negligible, as farmers had no choice but to concentrate their efforts on their irrigated plots. Traditionally, however, farmers have, because of higher return to labor, given rainfed crops priority over irrigated rice. It is nevertheless hoped that, under the present price structure, farmers will continue to cultivate irrigated rice, even if rainfall returns to normal. 4.04 Land allocation was handled by the advisory body to the Sous-Prifet of Tillabery, the population, and the Nigerien head of the project's agricultural development services. Although the project's land was traditionally managed by the villagers of Namarigoungou and Djambala, villagers from 15 outside villages also received land on the irrigated perimeters. The main selection criteria foreseen at appraisal (priority for original land users, and distance between the villages and the project area) were respected. Of the 20 villages originally selected, 7 were dropped because they were too distant and their inhabitants not interested. On the other hand, two villages not included in the original project, but which were located near the project area, did receive land. 4.05 Irrigated plots became available only gradually and it was therefore decided that the different villages would receive temporary plots to help them all start out on an equal footing. Once settled, however, farmers were reluctant to move from their plots when more land became available, and efforts made to consolidate farmers from the same village into a single service unit were unsuccessful. Different village groups are thus spread over the project area, although this has had no apparent negative effect on organization and management. 4.06 Due to excessive local demand, the project has not been able to accommodate migrants from other regions. Indeed, even local demand has outstripped the project s capacity. 4.07 The numbers of families and male adults settled are shown below: Cooperative Namarigoungou Djambala Total Number of families 1,683 1,476 3,159 Number of male adults 3,151 2,724 5,975 Number of plots 3,003 2,815 5,818 Area - rice (ha) 673.8 635.0 1,308.8 - nurseries (ha) 28.6 27.5 56.1 Average area per family (ba) 0.42 0.45 0.43 Average area per male adult (ha) 0.22 0.24 0.23 Average area per plot (ha) 0.23 0.24 0.23 - 24 - The number of families settled slightly exceeds appraisal estimates: 3,159 versus 3,000. However the average area allocated per family is slightly lower: 0.43 instead of 0.50 ha. 4.08 The project compares favorably with other projects in the region. Its main achievements are: (a) the creation of two cooperatives that serve as models for others; (b) the successful introduction of animal traction and foot-operated threshing machines. Farmers took over all land preparation activities, which in other perimeters had been done by ONAHA with tractors and with large subsidies; (c) a good record of cost recovery 8/ in addition to the cost of inputs, beneficiaries bore the full cost of energy and pump maintenance, and set aside a provision for pump replacement* (d) effective seed multiplication at the farmers' group level (the project even produces seeds for other perimeters); (e) effective applied research, fertilizer recommendations, also accepted on other perimeters, and successful treatment of a serious pest attack; (f) acceptable yields (despite serious pest attacks, yields average 4.0 tons/ha of paddy) and double cropping on the entire developed area. 4.09 As the construction period was longer than anticipated at appraisal, the period devoted to agricultural development was judged insufficient to sustain the high yields achieved at project completion. To maintain the present high performance, further support will be necessary. The project has therefore been included in the proposed Irrigation Rehabilitation Project for such aspects as credit, follow-up training, agricultural extension, assistance to cooperatives, and monitoring of alkaline soils. V. RATES OF RETURN 5.01 The recalculated rate of return of the project is 2.7%, compared to 12.5% at appraisal. The rate of return was recalculated taking into account all the actual costs, converted into 1983 terms, except for those associated with the health component, and the audit and studies. Benefits are also in 1983 terms, using the Bank's price 8/ As of April 1984, cost recovery was 99.3% for the Djambala cooperative and 96.7% for the cooperative of Namarigoungou. - 25 - forecasts as of January 1985 for the computations of farmgate prices for paddy and fertilizers (see Annex 1, Table 5 for details on the assumptions used for th2 economic analysis). 5.02 The decrease in the rate of return is the result of (a) lower than expected incremental paddy production (total cultivated areas at completion were slightly lower than appraisal estimates), and a slower than anticipated expansion in the area cultivated (ERR would be 8.2% with incremental paddy production at the pace of area expansion estimated at appraisal); (b) the project's negligible impact on rainfed farming and livestock (ERR would be 4.5% if benefits from rainfed agriculture and livestock estimated at eppraisal were included); (c) cost overruns on construction of irrigation works (ERR would be 4.3% if costs estimated at appraisal were used); and (d) additional support activities to be financed under the proposed irrigation rehabilitation project to sustain high performance (ERR would be 3.0% if these were not included). VI. INSTITUTIONAL PERFORMANCE AND DEVELOPMENT A. Institutional Design and Growth 6.01 The project was carried out by ONAHA, which was created in 1979 as a financially autonomous public enterprise. ONAHA's regional director at Tillabery was responsible for day-to-day implementation of the project and for the liaison with other public agencies in the region. This formula worked well and the project was satisfactorily executed; special credit should be ziven to the Nigerien project director for his excellent performance. The project was implemented without much input from ONAIA's central office at Saga, which was weak in administrative and financial management. These problems will be tackled under the proposed follow-up project (Irrigation Rehabilitation). 6.02 The project established two farmers' cooperatives, covering the areas served by each of the two main pumping stations. The first cooperative (Namarigoungou) was established in May 1981, just before the distribution of land to the first group of farmers. The second cooperative (Djambala) was founded in June 1983, when the first service units of the second pumping station became available. 6.03 The cooperatives are organized at three levels of responsibility: (a) at the base are all the farmers/beneficiaries, who are obligatorily members of the cooperatives, and who meet 2-3 times per season to decide how the agricultural campaign will be executed; (b) beneficiaries cultivating in the same service unit (tertiary unit) are grouped together into "Groupements Mutualistes Villageois" (GMV), preferably according to the villages to which they belong. The farmers of each GMV elect their Management Committee (president, secretary and treasurer) as well as delegates for each 10 ha block in the GMV. These committees transmit important messages from the cooperative level to the farmers; and (c) a Management Committee at the cooperative level, - 26 - consisting of three persons (president, secretary and treasurer) elected by the General Assembly, and responsible for running the cooperative. 6.04 The management committees for the cooperatives receive daily advice from ONAHA's staff, although responsibility rests with the committees themselves. Meetings of the General Assembly are difficult to organize as each .ooperative has about 1,500 members. In practice, therefore, only the representatives of the GMV's Committee including delegates participate. The discussions of these General Assemblies focus on repayment conditions, the marketing of paddy and technical issues. 6.05 The number of families and male adults in each cooperative are shown in para 4.07. Each village has a waiting list for vacant plots. This puts pressure on the current beneficiaries to comply with the regulations to avoid losing their plots, and makes it relatively easy for the cooperatives to achieve satisfactory agricultural campaigns, maintenance of the infrastructure and especially cost recovery, 6.06 The cooperatives are responsible for input supply, operation and maintenance of the irrigation system, as well as for cost recovery, which includes provisions for pump replacement, repairs and maintenance. Capital investment costs are not recovered. The way in which cooperatives are organized and are responsible for running the project, together with the assistance they receive from ONARA, has worked well and has been a crucial element in the success of the project. It will serve as an example for the proposed Irrigation Rehabilitation project. B. Cost Recovery 6.07 Cost recovery, including charges for repayment of agricultural inputs, operation and maintenance of the perimeters and depreciation of the pumps, has been almost 100% throughout the project period (para 4.08). A few beneficiaries who had not paid their dues for 3 consecutive campaigns were suspended, and their plots taken over by the cooperatives in order to recover the outstanding debts. These plots will eventually be reallocated to new users. C. Staff and Training Issues 6.08 Agricultural extension was provided by the project. Extension workers were recruited among young adults in the project area who had not lost touch with their rural environment. They were trained in weekly work sessions over a period of 30 months; training was based on the activities to be carried out simultaneously in the field. The recruitment policy paid off; the project's extension workers did a satisfactory job of assisting beneficiaries. The local plant protection specialist succeeded in making the extension workers aware of phyto-sanitary problems. Literacy programs and cooperative training were successfully administered to farmers with the aid of audiovisual equipment. - 27 - D. Accounting and Reporting 6.09 The accounts were kept in accordance with generally accepted standards, although coordination between general accounting and cost accounting left something to be desired and no reliable figures on the different categories of expenditures are available. The fact that seven different expatriate financial managers were employed (para 3.20) certainly contributed to the lack of co-ordination and continuity. In accordance with the Credit Agreement, accounts for each fiscal year have been audited and certified by independent auditors. VII. GOVERNMENT AND BANK PERFORMANCE 7.01 Government complied with most of the major Credit conditions, with the following exceptions: (a) the special account was not regulary replenished, so that there were frequent shortages of project funds; (b) recovery of capital costs 9/ of the irrigation works: this was not done because it would have removed the incentives for farmers to participate in the project, whereas at appraisal it was expected that at least part of these costs could be recovered. However, charges amounting to about CFAF 90,000 per ha and per year were recovered on agricultural inputs, operation and maintenance charges, and depreciation of the pumps. 7.02 The appraisal report clearly underestimated project costs as follows: (a) the efficiency of force account implementation was less than expected and 50% more equipment had to be procured (para 3.11); (b) volumes of work were higher than calculated at appraisal (para 3.17); (c) the annual requirement for spare parts was 32% of new equipment value, but provision was made for 18% only; (d) the quantity of cement was miscalculated at appraisal by about 50%; and (e) the number of houses and other buildings was about double the appraisal estimate, as were their costs. 7.03 The project was closely supervised because of its importance for the sub-sector. A total of 9 supervision missions took place between February 1979 and March 1984, or about 2 per year. All supervision missions enjoyed close collaboration with the project, which was fortunate to have the same competent manager throughzut the implementation period. The bank was represented by 4 different project officers during the 5-year project period. 9/ Set at appraisal at CFAF 66,000 per ha and per year. - 28 - 7.04 Two joint supervisions with the co-lender, KfW, took place. KfW also supervised the project separately, and informed us of its major findings. VIII. SPECIAL ISSUES 8.01 Weather conditions were very unfavorable during the project period. Below average rainfall (less than 400 m) made the cultivation of such traditional rainfed crops as millet, sorghum and maize impossible and farmers concentrated on their irrigated plots for a livelihood (paras 3.07, 4.01, 4.03). 8.02 Because of insufficient soil studies at the design stage, the serious risk of alkaline soils was discovered only during Bank supervision. Although the alkali problem is not yet widespread, it may eventually result in crop losses unless the basic cause -- poor drainage -- is eliminated. It is estimated that improved subsoil drainage facilities 10/ may be needed in from 15 to 25% of the project area, if yield levels in the affected areas are to be preserved. To prevent alkalinity, drains have been dug 0.20 m deeper than the original designs in part of the project area. One reason for including the Namarigoungou perimeter in the proposed Irrigation Rehabilitation project is to monitor the development of alkaline soils; the services of an expert in alkalinity and drainage have been included in the costs of this latter project. 8.03 Repeated white fly attacks (Aleurocybotus indicus) were a major problem during project implementation, and caused paddy yields to drop from about 5.0 to about 3.0 tons per ha. In coordination with INRAN, the project successfully combatted these attacks by treating the nurseries with an effective insecticide. This was a notable accomplishment for the project's research workers. White flies have also been observed in other rice perimeters along the Niger river, but damage has been much less severe than in Namarigoungou. The project agronomist speculated that the rice variety commonly used in Namarigoungou (IR-1529-680-3) may be more susceptible than others. Trials are underway to verify this. 8.04 Fertilizer trials carried out over five cropping seasons under the Namarigoungou project demonstrated that a different application rate (400 kg of compound fertilizer 15-15-15 and 150 kg of urea) yields an incremental value of production 5 to 6 times greater than the incremental fertilizer costs with the official extension service formula (100 kg of compound fertilizer 15-15-15 and 200 kg of urea) under the existing subsidized price structure. On the basis of the project 10/ The economic implications have not been considered in the ex-post ERR calculations. - 29 - results, the new formula will be introduced in the proposed Irrigation Rehabilitation project. 8.05 The project description provided for a dialogue with Government on subsidies for agricultural equipment and inputs. Because of its limited scale, the project had no noticeable policy impact on subsidies; the subject has been taken up again in the context of the structural adjustment program currently under preparation which, because of its larger scope, has a better chance of influencing Government's thinking. IX. CHANGES IN REPEATER PROJECTS 9.01 Costs per farmer and per hectare were too high to justify replication (para 3.17). Before new irrigation schemes are constructed, it is essential that the existing perimeters be upgraded, in order to assure a reliable supply of irrigation water (para 4.01). A number of institutional changes should also be carried out to ensure sufficient incentives for the farmers to continue operations. These elements are included in the proposed Irrigation Rehabilitation project. 9.02 One possibility we should investigate in the future is that of simplified projects, in the form of small village perimeters, which would require less infrastructure and land leveling, and could be operated independently. This would lower investment costs without sacrificing production and hence increase economic viability, but would at the same time demand a greater commitment from the beneficiaries for maintenance of the infrastructure. The small village perimeters implemented under the Small Rural Operations project in Senegal (Cr. 991-SE) are a successful example. The potential for this type of development should also be explored in Niger, and small village perimeters could eventually be included in the proposed Small Rural Operations project. X. CONCLUSIONS 10.01 The project attained its major objectives of a) increased food production; b) the creation of ONAHA as a vehicle to integrate and strengthen the construction and management activities of irrigation schemes; and c) the introduction of improved agricultural technologies. 10.02 The project's main additional achievements are: (a) the successful introduction of animal traction and foot-operated threshing machines; (b) seed multiplication at the farmer's group level; (c) applied research (fertilizer and pest control); (d) acceptable yields and double cropping on the entire area developed; (e) the establishment of two cooperatives that function satisfactorily; and (f) a good record of cost recovery. - 30 - 10.03 The main lessons learned from the project will be applied to the follow-up project and are summarized below: At the design stage: (a) the importance of adequate soil and topographic maps; (b) the value of livestock/agriculture integration and of adequate stockwater supply and grazing facilities; (c) the advantages of force account implementation of earthworks which, although not ideal, remains the best option under the present conditions in Niger because of its cost effectiveness and flexibility; and (d) for cooperatives' management the importance of having as great a degree of responsibility as possible; During Implementation (e) adherence to the cropping calendar to avoid problems in critical periods (cold weather, unavailability of irrigation water); (f) the wisdom of recruiting extension agents from the project area; and (g) the necessity of allowing sufficient time to reach and sustain project performance. - 31 - TABLE 1 NIGER IRRIGATION PROJECT PROJECT COMPLETION REPORT Actual and Estimated Quantities of Main Works Actual quantities Appraisal as a percentage of estimate 1/ Actual 2/ appraisal estimate 1. Earthworks on Irrigation and drainage systems Excavations, ms - Irrigation canals 58,000 156,000 270 - Drainage canals 225,000 204,000 90 Earthfill, m3 - Irrigation and drainage canals 164,000 534,000 330 - Dike 340,000 556,000 160 Roads along canals, ml 79,000 105,000 130 2. Concrete works Concrete, mS 1,750 4,250 240 Reinforced concrete, m3 350 500 140 Steel for reinforcement, T 65 85 130 Concrete lining, is, 3/ 16,500 18,900 120 3. On-farm works Clearing, ha 1,750 1,830 110 Deep plowing and surfacing, ha 1,350 1,400 100 Leveling, m /ha between 0-750 850 680 80 750-2000 700 590 80 over 2000 - 224 n.a. Bunding, ml 568,000 528,000 90 Tertiary 4/ irrigation canals, ml 75,000 77,000 100 Tertiary 4/ drains, ml 84,000 99,000 120 1/ Report No. 1682-NIR (September, 1979), Annex 7, Table 1. 2/ Rapport Final, Amnagement hydro-agricole de Namarigoungou, mars 1984. 3/ 165,000 m2 X 0.10 m _ 16,500 M, or at 300 kg/m equivalent to 4,950 tons cement. Z/ Field ditches and drains. NOTE Excluding quantities of materials for buildings, as these were given as units in the appraisal report. TABLE 2 - 32 - Page 1 NIGER IRRIGATION PROJECT (Credit 851-NIR) PROJECT COMPLETION REPORT Actual and Estimated Project Costs Appraisal Actual Actual Costs as Estimate 1/ Costs 2/ a percentage of million CFAF appraisal estt.ate Irrigation Works Construction Equipment and vehicles 639 1,048 160 Operating expenses 646 1,693 260 Labor and local staff 355 3/ 1,082 300 Expatriate staff salaries 500 3/ 558 110 Construction of buildings 45 / 230 510 Supplies 320 563 180 Pumping stations 225 325 140 20 kv line from Tillabery 168 161 100 Sub-total Irrigation Works 2,890 5,660 195 Direction "ONAHA" Local staff salaries 41 18 40 Expatriate staff salaries 210 153 70 Constructions 21 30 140 Vehicles 5 15 :0o Sub-total Direction 277 216 80 1/ Report No. 1682-NIR (Sept. 1978), Annex 9, Table 1 2/ Rapport Final, Aminagement Hydro-agricole de Namarigoungou, mars 1984 3/ Local and expat. staff salaries separated in line with Annex 9, Table 2 1 Construction of buildings for Irrigation works construction and Admin. and Financial Services separated in line with Annex 9, Table 14. TABLE 2 - 33 - Page 2 Appraisal Actual Actual Costs as Estimate Costs a percentage of million CFAF appraisal estimate Administrative and Financial Services Local staff salaries 90 109 120 Expatriate staff salaries 258 139 50 Construction of buildings 11 4/ 57 520 Equipment 17 20 120 Vehicles 16 6 40 Administrative and indirect expenses 5/ 170 313 180 Sub-total Administrative and Financial Services 562 644 120 Agricultural Development Services/Support Unit 6/ Local staff salaries 142 51 40 Expatriate staff salaries 415 244 60 Construction of buildings 108 177 160 Equipment 138 31 20 Vehicles 34 29 90 Sub-total Agricultural Development Services/ 837 532 60 Support Unit Health Component and Training Wells and two villages 4 3 80 Village health kits 4 3 80 Training of First Aid Volunteers 37 20 50 Sub-total Health Component and Training 45 26 60 Agricultural Equipment and Incremental Inputs Rice development 171 100 60 Rainfed development 23 8 40 Livestock sub-project 32 - 0 Sub-total Agricultural Equipment and 226 108 50 Incremental Inputs 4/ Construction of buildings for Irrigation works construction and Admin. and Financial Services separated in line with Annex 9, Table 14. 5/ Including 56 million CFAF finance charge / Cost of Agricultura' Development Services and Support Unit combined (deviation from App, tisal Report) TABLE 2 - 34 - Page 3 Appraisal Actual Actual Costs as Estimate Costs a percentage of million CFAF appraisal estimate Other Agricultural Projects Coordinator 99 101 100 Audit of project accounts 9 29 320 Specialized consulting services 15 14 90 Gabo-Bonfeba feasibility study 177 194 110 Sub-total other 300 338 110 TOTAL PROJECT COSTS 5,145 7/ 7,524 150 7/ Including 15% physical contingencies on irrigation works, and 10% on the rest, as well as price contingencies as follows: 1978 = 8%; 1979 - 7.5%; 1989-82 - 72 N.B. Due to rounding, the figures differ slightly from those presented in the appraisal report. - 35 - TABLE 3 NIGER IRRIGATION PROJECT (Credit 851-NIR) PROJECT COMPLETION REPORT Cumulative Disbursement Status Profile Actual as percent of Fiscal Appraisal Appraisal Year Quarter Estimate Percent Amount Actual Estimate Profile (US$ M) (%) (US$ M) (US$ M) (%) (%) A B C (C/A) (C/B) 1979 3 - 0.5 0.1 - - - 4 1.0 2.0 0.3 - - - 1980 1 1.5 3.8 0.6 - - - 2 2.0 6.6 1.0 1.5 75 150 3 3.5 9.4 1.4 2.2 63 157 4 5.0 13.1 2.0 3.0 60 150 1981 1 6.5 16.8 2.5 4.5 69 180 2 8.0 21.3 3.2 5.5 69 172 3 9.0 25.8 3.9 6.4 71 164 4 10.0 30.9 4.6 6.8 68 148 1982 1 11.0 36.0 5.4 7.3 66 135 2 12.0 41.4 6.2 10.0 83 161 3 12.5 46.7 7.0 10.7 86 153 4 3.0 52.0 7.8 11.8 91 151 1983 1 13.5 57.2 8.6 12.0 89 140 2 14.0 62.1 9.3 12.8 91 138 3 14.5 67.0 10.1 13.5 93 134 4 15.0 71.3 10.7 14.1 94 132 1984 1 75.6 11.3 14.6 97 129 2 79.3 11.9 15.0 100 126 3 82.9 12.4 15.0 4 85.9 12.9 15.0 1985 1 88.9 13.3 15.0 2 91.3 13.7 3 93.7 14.1 4 95.5 14.3 1986 1 97.3 14.6 2 98.7 14.8 3 100.0 15.0 1/ neference Annex 1, Table 3a, IDA-Irrigation, Adler's Memo of June 9, 1982 2/ First quarter ending for the profile: 12/31/78; Board date 10/04/78 3/ Rounded, only US$ 23,000 outstanding 4/ Last disbursement was made in September 1984 TABLE 4 - 36 - NIGER IRRIGATION PROJECT (Cr. 851-NIR) PROJECT COMPLETION REPORT Agricultural production 1/ (actual and appraisal estimates) Actual Appraisal Estimate Cropping Area Yield Prod. Area Yield Prod. Season (ha) (T/ha) (t) (t/ha) (t/ha) (t) 1980 (WS) - - 500 750 4.5 3,375 1981 (DS) - - - 750 4.6 3,450 1981 (WS) 212 4.5 954 1,350 4.6 6,210 1982 (DS) 317 4.0 1,264 1,350 4.8 6,480 1982 (WS) 587 4.3 2,524 1,350 4.8 6,480 1983 (DS) 790 4.3 3,397 1,350 4.8 6,480 1983 (WS) 1,142 5.0 5,710 1,350 4.8 6,480 1984 (DS) 1,278 3.4 4,345 1,350 4.8 6,480 1984 (WS) 1,304 4.8 6,259 1,350 4.8 6,480 (WS) = wet season (DS) - dry season 1/ For paddy only. 2/ Excluding nurseries. 3/ Report No. 1682-NIR (September 1978). Annex 6, Table 6. (Note 6). Severe white fly attack. TABLE 5 -37- Page 1 -- migi R-lg a a . r4 4 - 44%lggEilp!3 An 2 9 22 enleg -,- . - a - --nu- 2"24 = 1C a .0 * ýd 2. 3 -- - 5 " .. q. - f 82 i%fz Aa 22 galg> e g än8 P,l 3@ @ -"- =n e 4%x n- -l s . ä n a - - - . rA -ll - -4 - mg g: Aga evig fta-•~. g %e - - fg f - *fl =RE6f..d - m-4 -i04 --- 9. CR*"EU gn 2 2g 2 Sj 0.o -- - - 4 n.4 ~ 0. 4 : C-- n- o e E -- ----f~. 0' o -- m a x -n - - - - 1~ -n-- n °w - g-2 8 aw -- - e* de 4e - fl- - a. m - -g fl- no 0f 0 l --. f - *f «- *fla e --Eg 4 - 3E-. - - - g- fl l ===*f0 - f gl= e se g a e fssi llfi ls -1 -, g * mi a - E 52 NIGER IRRIGATION PROJECT ECONOMIC ANALYSIS (Unit: million CFAF) 1979 i980 1981 1982 1983 984 1985 1986 1987 1988 189 5990 1991 1992 1993 1994 59% 5996 5997 5998-2003 COS's PM03ECT CosiS IRRIGATION #M CMNIMUCION mEm t m miCLES o I50 377 257 192 63 7 -165 0 0 0 0 0 0 0 0 0 0 0 0 0 OM TINS EPISES 8 58 414 lå8 48 23 0 0 0 0 0 0 0 0 0 0 0 0 0 0 LhM A8 LOC <TOFT El l 207 293 323 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 EIPATBIE SF SALMiS 16 524 54 160 503 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 CINMOI N F 8UIL8N6SF0 21 36 6 å i 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 SIPLIES 2l 2 48 144 586 99 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 PUtINSlIAIIS 19 88 141 37 0 40 0 0 0 0 0 0 0 0 0 0 0 0 0 0 20K9LIKEFUMILLAERI 46 98. 16 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 SM-IITA 262 924 1339 1338 1058 70 -165 0 0 0 0 0 0 0 0 0 0 0 0 0 DERECTION ~Ma LOEAL SIF SLMIES E0 2 3 3 3 3 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 EAIRIAE SIF SaE 1 31 32 34 40 6 0 0 0 0 0 0 0 0 0 0 0 0 0 0 COMSTRUCIKIGUF 6 5 0 5 5 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 WICLES 4 0 0 7 4 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 SIM-10TAL 23 35 35 45 48 6 0 0 0 0 0 0 0 0 0 0 0 0 0 0 WIIISiMIRE Mi FIMlIUCAL SME9CES LWAE STAFF SKAIES E 0 16 20 26 28 4 0 0 0 0 0 0 0 0 0 0 0 0 0 0 ErPatIaIE STOFF SALMIES 7 22 28 32 36 14 0 0 0 0 0 0 0 0 0 0 0 0 0 0 CMSIRWCIIICIF 8UILDNGS F 3 12 0 9 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 EIP1Il1 1 14 2 2 l 0 0 0. 0 0 0 0 0 0 0 0 0 0 0 0 MEICLES 2 2 0 3 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 AMIISTIRATIE 6 IIBMI EIPEISS 6 i 38 59 59 67 26 0 0 0 0 0 0 0 0 0 0 0 0 0 0 StG-Iea 04 109 122 132 44 0 0 0 0 0 0 0 0 0 0 0 0 0 0 46ICUIURL E9ELOlTI SER9WCES LoC SIF SALOIES El 0 0 10 t6 17 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 EPalteiaE SIF SALOIES 0 37 58 62 80 7 0 0 0 0 0 0 0 0 0 0 0 0 0 0 COMSIIECIN Ni 9 34 14 16 2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 EDNINENT 0 0 8 6 17 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 ENICLES 0 8 16 5 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 SU8-101A 9 79 106 105 116 7 0 0 0 0 0 0 0 0 0 0 0 0 0 0 180UM 0 0 2 10 8 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 SPECILIJED CSIIlNB SRVCES 5 5 5 3 0 8 0 0 0 0 0 0 0 0 0 0 0 0 0 0 IDTA PIlJEIRCCOSIs 355 1144 1593 1623 1363 135 -165 0 0 0 0 0 0 0 0 0 0 0 0 0 UOP EFLAER 65.4 76.4 84 92.6 100 lo8 11IAL P0ECT Csis lm983PCES 481 1498 1896 i72 1363 525 -165 0 0 0 0 0 0 0 0 0 0 0 0 0 03 QQ- 00 L- NIGER IRRIGATION PROJECT ECONOMIC ANALYSIS (Unit: million CFAF) 5979 l9m0 1981 1982 1983 19184 95 19 1987 198 1989 9 99 m 992 1993 1994 1995 19% 1997 1998-2003 IRR6ATION 06 COSI$ D5 9.18 34.89 74.58 99.67 500.6? 100.67 100.6? 100.67 100.6? 100.6 100.6 6 100.6 6 10.6 6100.6 .00.67 10.67 100.67 100.67 tvUl mKA 2008/M 10NS3 42.4 180.8 386.4 556.4 521.6 521.6 521.6 521.6 521.6 525.6 521.6 521.6 521.6 521.6 521.6 525.6 521.6 521.6 ECK~1DHIC PICE CFWS 1t4 152 556 144 154 165 t70 174 579 184 184 184 184 184 184 184 184 184 414E CFF NIt.50IN 13.00 t2.54 13.46 20.74 23.72 27.23 28.% 3:.20 32.04 33.86 33.86 33.86 33.86 33.86 33.06 33.86 33.86 33.86 DP I00i8/mT 31 21.2 %.4 193.2 258.2 260.8 260.8 260.8 260.0 260.8 260.8 260.8 260.8 260.8 260.8 260.8 260.8 260.8 260.8 ECOOMIC PRICE CFM/K8 504 l59 134 153 166 574 580 187 194 202 202 202 202 202 202 202 202 VALUE CFO BItLION 2.29 10.76 25.89 39.50 43.29 45.38 46.94 49.77 50.60 52.68 52.68 52.68 52.68 52.68 52.68 52.68 52.68 52.68 ~URT9 UnEA 2008/m 91 2.5 10.8 24.2 31.0 31.3 31.3 31.3 31.3 31.3 31.3 31.3 31.3 31.3 31.3 31.3 31.3 31.3 31.3 MALUE CFAF RILLION 0.29 1.2 2.69 4.46 4.82 5.16 5.32 5.45 5.60 5.76 5.76 5.76 5.76 5.76 5.76 5.76 5.76 5.76 w 110016/5 im5 K 1.3 5.4 t1.6 15.5 15.6 15.6 15.6 15.6 15.6 15.6 15.6 15.6 15.6 15.6 15.6 15.6 15.6 15.6 VMLUE CFOF 1 0..191 0.14 0.65 5.55 2.37 2.60 2.72 2.82 2.93 3.04 3.16 3.16 3.16 3.16 3.56 3.16 3.16 3.16 3.16 PESTSC5IE 19Ma 2.85PACETS/m 604 2576 5506 7359 7433 7433 7433 7433 7433 1433 7433 7433 7433 7433 7433 1433 7433 7433 VAlUE CFRF HILLION WAF60~/PACET 0.04 0.15 0.33 0.44 0.45 0.45 0.45 0.45 0.45 0.45 0.45 0.45 0.45 0.45 0.45 0.45 0.45 0.45 5N1PU 101AL 15.75 25.32 43.92 7.51 74.87 80.94 84.43 87.86 91.72 9. 5.0 95.90 95. .0 95.90 955.. 95.0 95.10 EOUIPIET IMPLEXENTS EL MiEN L# 135 150 150 150 0 0 0 0 0 135 150 550 150 0 0 0 0 0 ALUE 1 CFOM38286 5.57 5.14 5.74 5.74 0.00 0.00 0.00 0.00 0.00 5.5? 5.74 5.74 5.74 0.00 0.00 0.0 0.00 0.00 lu9EsaE* 53 573 180 57 0 53 573 180 57 0 53 173 580 57 0 53 l3 80 ALUE 1 CFAF M5000 6.10 19.90 20.70 6.56 0.00 6.50 19.% 20.70 6.56 0.00 6.50 19.90 20.10 6.56 0.00 6.50 19.90 20.0 IRRI6ATI0N REMEISLItATION PMMC Kl 0 0 0 0 40.91 39.07 33.36 19.54 17.22 6.52 6.52 6.52 6.52 6.52 6.52 6.52 6.52 6.52 FMILY LAMOR 9955911 N) 2%80 126560 270480 361480 36520 35120 365120 365120 36520 365120 365120 365120 365120 365120 36520 365120 5520 365120 vagE At CFAF4SOl1AY 01 J3.36 56. 121.72 162.67 64.30 164.30 164.30 644.30 *64.30 564.30 P64.30 t64.30 164.30 164.30 164.30 164.30 164.30 64.30 101AL CUSIS 481.45 1497.51 1944.51 1895.25 1629.55 467.33 255.75 391.07 402.66 393.06 380.47 372.56 379.23 393.03 393.84 373.95 367.39 373.49 387.29 388.09 HET TCEKTALt BENEFITS -506.45 -1526.55 -1906.43 -1672.65 -979.97 366.7? 556.19 372.43 404.30 521.34 577.32 644.30 681.26 641.28 623.02 686.54 666.92 643.37 673.20 646.22 Al EmPCS, JAIIW1 i985, 1983 Cumawl NICE. 81 OFFICIAL PRICE I982/83. C SU1C: tIGER FORESFR 51 fi8mEtt SAR 01 RESIONAL VALKU USE FOR iffl AiTIDN REM11ITATION PMOJECT El *hPlaI5 KF OF 0.85 Fl APLIIIB CF eF 0.85 A558 ASUn18t IM 9 e TIR50 19 Du0 FOR IE PRECI. 6 12CLUDING FIlM IL CD~ES. Ni *KPLVNS BCF OF 0.85 =D Ak^551B 1 ON M IAF 159 0 FOR 1HE P ECW. 11 9NLIENS ELECTRICITY, MINIElT~E OF PUffS AM INFRASTROCTIRE, M 8EPLCEI OF P~OFMPS. 41 DU 10 90CERA1lIIES OF FERI1LIZER (URE6 M 15-15-55) FROM NIGERIA, I6015 THAI ALTERNA5TIV FERTILIZER F"lLA 9810 8E A00PIED 4200M 5F UREA AM 100K8 OF WIP IffiEf0 OF 15088 OF UREA 4M 4008 DF 15-95-15. Kl om AMTU5 6 OF TmE TAL mEA. LU ASSONING 18111 5/2 OF EISTIN1 FAl INPLEREXT UlUfS MERE RE0 FOR IRRIGATION, 8 AT 1101-SU188101 PRICES in 5983. 9) 203 OF PRUJECI COWS FOR ASSISTA CE TO COOPERAIVES iFOlCTINsL LITERAC O 9RAKN, UA ~fGENENT TAINiNGY,TCk, FOR AFPLIED RESEARCN, = NUI5ORIN 6 EVALMION. d) 1409AYS/M 8> 0P91MIT COSI OF .2801 IS REVOCE 10 CF4F450/HANOT>Y FROB 600115A1 ESIMIt0 FOR IR1ATI0N REILITA1IN PMU3ECT, RIE 10 KORE LUNITED VUPORIMITIES 0111I1 OF IR8RIAT1ON 19 THE P80IECT MEA 9110 L O L E ME L O F R A I FA L L S . Ö 0 TABLE 5 - 40 - Page 4 Explanatory Note Key Parameters Used in Economic Analysis I. Appraisal: 12.5% over 25 years. 1. Benefits (a) incremental production accruing only to the Namarigoungou farmers, although the project was expected to benefit all farmers on the existing irrigation schemes in the "Arrondissements" of Tillabery and Tera; and (b) incremental production consisting of (i) incremental production of paddy (on 1,350 ha at full development), of mixed cropping (sorghum, cowpeas, and vegetables on 300 ha), of firewood and building poles (eucalyptus on 25 ha), of livestock (animal fattening from fodder crops on 25 ha), and (ii) improvement on rainfed holdings of the Namarigoungou farmers. 2. Costs (a) all project costs, except for those associated with the health and training components, audit and studies; (b) operating and maintenance costs on irrigation schemes; and (c) incremental agricultural inputs. II. Re-evaluation: 2.7% over 25 years. All costs and benefits are in 1983 terms. The Bank's price forecasts as of January 1985 were used for the computations of farmgate prices for paddy and fertilizers. 1. Benefits (a) no benefits from livestock were considered, since 25 ha of fodder crops for livestock were dropped from the project during implementation; (b) no benefits from rainfed farming were considered, since continuous drought in the project area has caused project agricultural services to concentrate on irrigated perimeters rather than on rainfed farming; and (c) no benefits from by-products (straw) were included; the assumption is that they would have been cancelled out by losses of benefits from pastureland occupied by the irrigation perimeters. 2. Costs (a) all project costs, except for those associated with the health component, audit, and studies are included. The project costs TABLE 5 - 41 Page 5 for training are included, due to the importance of training in the success of viable operations of irrigation schemes; and (b) 20% of costs of cooperative assistance (functional literacy, management training, etc.), applied research, and monitoring and evaluation components under the proposed irrigation rehabilitation project are included. 一7〞一 戶勾、＇〞合多本祈 NIGER IRRIGATION PROJECT (CR 851-NIR) Construction Schedule (Actuat & as Forseen at Appralsal') 1978 1 1979 1 19801 1981 1982 1983 l 2 1 31114 f l. 4 3 41 1 2 -1.3 J 4 1 2 1 3 4 l 2 3 4 l 2 3 4 ORGANZAMN I Ctecition of ONAHA -4 &8 p a IT All 2. Conwilanis a sal att en amillginnanneftei I AT A CONSIRUCTION BY FORCE ACCOUNT D 3. Procuren>enl of Equiprnent mig en led 4. Access Roacis. Bulldings, Stofes 7 7t 1 ål&6 5 Dike Corwirucilon guinean§§ 6 ttttg<311<>rl/Dlot~ /<3n-F<3tm w~ 0 l Lj - Upstreorn District - Downstfeorn Dilsttlet CONfRACIED ~S t,] i IA D 7. Pumping Slotions 11 gegga# 111L 11i1 egiten, (CMI Engiticeting & Equipment) i. I Al t 8. 14.1. tirte lillobefy - Nornarigoungc>u TO, k a aing a ii l = Tendet CONSTRUCT110N SCHEME A = Aword l = trnp4~.enllollon a a a m a m n- Apptolsol Es~ te o = C«ivefy V leffidge=ý Actual 0 = Opetation (6/81: 210 ha: 1182.316 ha: 6182:587 ha. 1183: 790 ha. 6183 1,100 ha; 1/84:11.366 ha), Rice Only Worid B~ - 27496 - Report No. 1682-NR (Sept.. 1978). Annex 14 NIGER IRRIGATION PROJECT (CR 851-NIR) US $ MILUON Disbursement (Actual, Appraisal Estimate & Profile) 30 Actual ........ Total Disbursement IDA 25 - .--- KfW Ap...-sa, Total Appal .. . . IDA Esimate .......... KM/ 20 Profile --------- Total - East & West Africa Region irrigallon (05 03.82) 15 10dn-7 ÄrD 5 0 FY 1979 FY 1980 FY 1981 FY 1982 FY 1983 FY 1984 FY 1985 FY 1986 WordBank -27497 N I G E R NAMARIGOUNGOU IRRIGATION SCHE Genaral Plan of the Proposed Project in the Feasibility Report, S BRANCH OF THE NIGER CUT OFF BY PROJECT P3 -- FLOOD GATE THIS PCR MAP IS BASED ON IBRD 12837 MAY 1977 THE BASIC PROECT CONFIGURATION HAS NOT CHANGED, 10 HOWEVER, SOME IRRIGATION AND DRAINAGE CANALS HAVE BEEN REALIGNED. IBRD 12837PCR) ME ogreah 1974 River - - 9 s 1.9N O R T ý ~p i 4 -- -- - -P3T G -3:- - Nomarigoungou Diambala ftS , fet ffibe p~~,e j>, nIeIIb.* w~ e,7 ~ ~MØ k,e swee55. V~ DIKES of u* 8"W0* #wre F~« C«~'e o MAIN PUMPING STATIONS i 0° joo .LIBYA PRIMARY AND SECONDARY IRRIGATION CANALS ALGERIA DRAINAGE CANALS PERIMETER AND SECTOR BOUNDARIES -f?E RIVER -' LATERITE ROADS ~ VILLAGES N G E R AREASHOWN ON MAP CHAD 0 NIAMEY.-. 0 1 2 3 BURKINA KILOMETERS N I G E RIA CAMEROON.• ______ BENIN 10 MAY 1986