Nepal - Birganj Irrigation Project

Document of FILE COPY The World Bank FOR OFFICIAL USE ONLY Report No. 4267 PROJECT PERFORMANCE AUDIT REPORT NEPAL - BIRGANJ IRRIGATION PROJECT (NARAYANI ZONE) (CREDIT 373-NEP) December 30, 1982 Operations 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. PRINCIPAL ABBREVIATIONS USED ADBN - Agricultural Development Bank of Nepal AIC - Agricultural Input Corporation DADO - District Agricultural Development Offices DIHM - Department of Irrigation, Hydrology and Meteorology ERR - Economic Rate of Return FMSD - Food and Marketing Services Division GOI - Government of India HMGN - His Majesty's Government of Nepal HYV - High Yielding Variety ICB - International Competitive Bidding ICP - Integrated Competitive Bidding JT - Junior Technician JTA - Junior Technician Assistant NEC - Nepal East Canal NFC - Nepal Food Corporation NR - Nepalese Rupee NZIDB - Narayani Zone Irrigation Development Board O&M - Operation and Maintenance PLAA - Panchayat Level Agricultural Assistant REC - Rice Export Company SAR - Staff Appraisal Report SCF - Standard Conversion Factor SMS - Subject Matter Specialist T&V - Training and Visit (System) GLOSSARY Panchayat - Representative Council at Village, District and National Level Sajha - Village Cooperative Society Terai - Part of the Gangetic Plain south of Hills in Nepal FOR OFFICIAL USE ONLY PROJECT PERFORMANCE AUDIT REPORT NEPAL - BIRGANJ IRRIGATION PROJECT (NARAYANI ZONE) (CREDIT 373-NEP) TABLE OF CONTENTS Page No. Preface ............................................................. i Basic Data Sheet ................................................... ii Highlights ......................................................... iii PROJECT PERFORMANCE AUDIT MEMORANDUM I. SUMMARY .................................................. 1 Project Objectives ...................................... 1 Project Design .......................................... 1 Major Points Raised by Board Members .................... 2 Cost Recovery ........................................ 2 Institution Building .................... ............ 3 Water Management ......... 3 Project Implementation .................................. 3 Project Results ................... 3............3 II. MAJOR ISSUES ............................................. 4 A. Institution Building .............................. 4 B. Civil Works Contracts ............................... 6 C. Tubewell Irrigation . 7............... ..7 D. National Project Performance Audit Function ......... 8 E. Suggestions from the Field .......................... 8 Operation and Maintenance ......................... 8 Project-Bank Communications ....................... .8 Authority of Project Management Unit .............. 9 Operation and Maintenance and Water Charges ....... 9 III. LESSONS FOR THE BANK ................. .................. 10 Annex 1 - Comments from the Irrigation Department .................. 11 Annex 2 - Comments from the Ministry of Food, Agriculture and Irrigation ..................................... . 13 PROJECT COMPLETION REPORT I. Sectoral Background ...................................... 17 II. Formulation .............................................. 17 III. Implementation ........................................... 24 IV. Agricultural Impact ...................................... 34 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. Table of Contents (continued) Page No. V. Economic and Financial Evaluation ........................ 40 VI. Institutional Performance and Development ................ 45 VII. Special Issues ........................................... 59 VIII. Lessons Incorporated in Repeater Projects ................ 59 IX. IDA Performance .......................................... ........ 61 X. Conclusions ....................... ............................ 62 Annex Tables 1 - 23 Chart PROJECT PERFORMANCE AUDIT REPORT NEPAL - BIRGANJ IRRIGATION PROJECT (NARAYANI ZONE) (CREDIT 373-NEP) PREFACE This is a performance audit of the Birganj Irrigation Project in Nepal, for which Credit 373-NEP in the amount of US$6.0 million was approved in March 1973. The loan was closed in December 1981, after the cancellation of US$41,000, three years later than scheduled. The audit report consists of an Audit Memorandum prepared by the Operations Evaluation Department (OED) and a Project Completion Report (PCR) dated June 29, 1982. The PCR was prepared by the South Asia Regional Office. The Audit Memorandum is based on a review of the Appraisal Report (No. PA- 146a) dated March 1, 1973, the President's Report (No. P-1197-NEP) of March 6, 1973, the Credit Agreement dated April 18, 1973 and the PCR. Correspondence with the Borrower and internal Bank memoranda on project issues as contained in relevant Bank files have been reviewed. Bank staff associated with the project have been interviewed. An OED mission visited Nepal in July 1982. The mission held discus- sions with the project field staff; farmers; the Department of Irrigation, Hydrology and Meteorology; the Department of Agriculture and the National Planning Commission. The information obtained during the mission was used to test the validity of the conclusions of the PCR and permitted discussion of project design and Bank procedures. A copy of the draft report was sent to the Borrower on October 28, 1982. Comments received from the Irrigation Department and the Ministry of Food, Agriculture and Irrigation are included as Annexes 1 and 2 to the Project Performance Audit Memorandum (PPAM). Suggested changes have been taken into account. The PCR is factual, although some differences exist between this audit and the PCR on the interpretation of those facts. The issues discussed in the Audit Memorandum were selected on the basis of their relevance in the design and implementation of similar projects in the future. The assistance provided by Government, the Project Implementation Unit and Project Staff in the preparation of this report is gratefully ac- knowledged.  PROJECT PERFORMANCE AUDIT REPORT NEPAL - BIRGANJ IRRIGATION PROJECT (NARAYANI ZONE) (CREDIT 373-NEP) BASIC DATA SHEET KEY PROJECT DATA Appraisal Item Estimate Actual Actual as % of Estimate Total Project Costs (US$ million) 10.1 11.0 109 /a Credit Amount (US$ million) 6.0 6.0 1OU Date Board Approval 03/20/73 Date Effectiveness ------------- -- 07/09/73 Date Physical Components Completed ----- 06/78 12/81 167 /b Proportion then Completed --- --- 100% 98.4% Closing Date ----------------------- 12/31/78 12/31/81 143 L Economic Rate of Return (%) 21% 21% 100 Institutional Performance moderate worse Agronomic Performance satisfactory worse /c Number of Direct Beneficiaries (1980/81) ---- 20,000 farm families 95 /d CUMULATIVE DISBURSEMENTS FY74 FY75 FY76 FY77 FY78 FY79 FY80 FY81 FY82 Appraisal estimate (US$ million) 0.7 2.0 3.4 4.2 5.75 6.0 - - - Actual (US$ million) 0.1 0.6 1.3 2.2 4.0 4.8 5.5 5.65 6.0 Actual as % of estimate 14 30 38 52 70 80 92 94 100 Date of final disbursement: 12/31/81 Principal repaid (11/30/81): not yet due MISSION DATA Date No. of Mandays Specializations Performance Types of Mission (mo./Yr.) Persons in Field Represented/c /d Trendle ProblemsLf Identification (FAO/CP) 03/70 - Preparation (FAO/CP) 05/70 1 4 d - Preparation (FAO/CP) 12/70 2 10 d - Appraisal 12/71 6 60 a, b, d - Postappraisal 1 02/73 2 14 d - Total Supervision 1 08/73 1 5 d 2 M Supervision II 04/74 1 21 d 2 M Supervision III 09/74 3 27 a, d 2 M Supervision IV 04/75 2 24 a, d 2 M Supervision V 10/75 2 44 d 2 F, M Supervision VI 03/76 2 30 d 2 F, M Supervision VII 03/77 2 42 d 1 F Supervision VIII 08/77 4 12 a, b, d 1 M Supervision IX 10/78 2 10 d 2 M Supervision X 02/79 1 3 d 2 M Supervision XI 10/79 3 3 a, b, d 2 M Supervision XII 04/80 2 4 a, d 1 Supervision XIII 11/80 1 3 d 2 F, M Supervision XIV 05/81 2 2 d 2 F, M Total 324 OTHER PROJECT DATA Borrower Kingdom of Nepal Executing Agency Narayani Zone Irrigation Development Board Fiscal Year of Borrower July 16 - July 15 Name of Currency (abbreviation) Rupee (NR) Currency Exchange Rate: (NRs per US$1.00) Appraisal Year Average: 1972 - NRa 10.15 Intervening Years Average: NRa 11.65 Completion Year Average 1981 - NRs 12.00 Follow-on Project: Name Narayani Zone Irrigation Developmenmt Stage 11 Project Credit Number 856-NEP Date Board Approval 10/10/78 /a Note that the project was scaled down in November 1977 from a command area of about 31,400 ha to about 18,700 ha, in order to accommodate a cost overrun. Actual cost in US$ million after conversion of total cost at exchange rate of NRa 12/USS. /b Delays due to late start-up, slow performance of local contractors, and a number of institutional/ managerial problems. /c Estimated production shortfall by about 20% relative to appraisal, due to lower than expected yields at full development, low tubewell irrigation intensities. /d Actual number of beneficiaries relates to final product, estimated to original pro-ect. /e a - agriculturist, b - economist, d - engineer. 7- F - Financial, M - Managerial.  - iii - PROJECT PERFORMANCE AUDIT REPORT NEPAL - BIRGANJ IRRIGATION PROJECT (NARAYANI ZONE) (CREDIT 373-NEP) HIGHLIGHTS The Birganj Irrigation Project was designed to improve the irriga- tion water supply and, thus, help in the introduction of double-cropping of paddy in the Bara and Parsa districts of the Narayani Zone. The project provided for improvement of surface irrigation facilities and the initia- tion of the development of underground water. The basic objectives were t% increase crop production, farm incomes and on-farm employment. The project succeeded in improving productivity and production. Its impact on farm income and employment has been positive. As the first project supported by the Bank in Nepal, this project has served as a learning ground for both the Bank and the country. At the time of this audit, the major problems facing the project related to the lack of means for operation and maintenance of the facilities constructed, the lack of clarity in the sharing of responsibility, authority and accountability between project management and the central ministries, and the lack of motivation for active farmers' participation in managing the irrigation facilities in a manner that would ensure their long-term viability. The project's economic rate of return was recalculated at 21%, the same as at appraisal. The major lessons from this experience for on-going and future projects are: (i) the need for increased Bank input in technical supervision in environments with limited qualified manpower resources, (ii) the need to give priority to training local staff, and (iii) the need in the field of institution building to follow up on temporary stop-gap type solutions (use of Boards as a coordinating mechanism) with planned actions to avoid the continued recourse to such solutions. The following additional points may be of particular interest: - the development of underground water ran into problems due to faulty tubewell designs and to the approach being selected for underground water development in the project area (PPAM, paras. 17-19, PCR, paras. 3.12-3.15); and - operation and maintenance of project works was poor because farmers have not been motivated to pay water charges and project management has not been motivated to collect such charges (PPAM, para. 25).  PROJECT PERFORMANCE AUDIT MEMORANDUM NEPAL BIRGANJ IRRIGATION PROJECT (NARAYANI ZONE) (CREDIT 373-NEP) I. SUMMARY Project Objectives 1. The Birganj Irrigation Project was the first agricultural project financed by IDA in Nepal. It was designed as a five year program to help intensify crop production with particular emphasis on the introduction of double cropping of paddy in the Bara and Parsa districts of the Narayani Zone. The objective was to increase crop production, farm incomes and on-farm employment. To that end, the project provided for the improvement and con- struction of irrigation facilities to serve an area of 28,700 ha in the Terai with water from the Gandak River, and an area of 2,700 ha with tubewell irrigation. It also provided for service roads, for strengthening agricul- tural extension and research services in the project area, for assistance to farmers to establish on-farm drainage facilities, and for technical assistance to help implement the program. Project Design 2. The project was designed on the basis of feasibility studies under- taken over the period 1965-69, identification/preparation missions over the period 1969-71, and an appraisal mission in 1972. The major shortcoming in the feasibility studies identified as a result of Bank review was their silence on the state of local institutions. Aside from a brief reference to a pilot demonstration project to train Nepalese technicians in methods of irrigated farming, the studies provided no evaluation of Nepalese administra- tive machinery and personnel for the implementation and subsequent operation of the proposed irrigation system. During identification/preparation, an attempt was made at correcting this omission, and the following recommen- dations were among the major ones made: (a) Concerning tubewell irrigation, preparation missions recommended setting up a pilot scheme around three existing wells to try out alternative organizations before possible appraisal in two years; (b) concerning project implementation, it was suggested that arrange- ments should be made to ensure a high degree of local participation, i.e. people and authorities already functioning in the project area as opposed to Kathmandu-based bodies; -2- (c) in the same context, the setting up of village multi-purpose cooper- atives and irrigation service units was recommended as a prerequi- site for proper provision of irrigation water, extension services, agricultura 1 inputs and marketing services; (d) moreover, it was agreed that because the project was regarded essentially as a pilot development, the training functions of project personnel should be a vital part of the project; and (e) the principle was established that, first, water charges would be collected to the limit of 25% of net incremental benefits accruing to farmers as a result of the project and, second, that industrial rates for off-peak electricity supply would be utilized in the tubewell irrigation sub-project. 3. Almost seven years of preparation and continuous dialogue between Government and the Bank have led to a project design which was cognitive of the major constraints facing agricultural development in the area: insuffi- cient irrigation and drainage facilities, low level of credit and marketed input use, inadequate agricultural services (research, extension, education, input distribution) and, most importantly, shortage of trained manpower. The appraisal report recognized the difficulties in simultaneously addressing these constraints within the limited time frame of the project period and attempted to provide workable solutions. For example, because of the lack of trained manpower, the appraisal report suggested that "in order to phase construction with the training of Nepalese engineers and extension workers and with on-farm development, civil works would be spread over five working seasons." In addition, the appraisal report called for collaboration between agencies serving farmers in order to make the best use of existing manpower in credit, inputs, supply, extension and research organizations. 4. Information in the project files indicates that there was little economic or technical evaluation for the road component.!/ Moreover, no attention was given to road maintenance (cost, organization and material resources required). Information in Project files attribute this shortcoming in the appraisal process to the fact that no input was sought from the Bank Transportation Projects Department during discussions at the preparation or appraisal stages. As a result, cost considerations were the major factor determining the final scope of the road component and the quality of the roads constructed. Major Points Raised by Board Members 5. Cost recovery. A Board member noted that the proposed annual project charges for water use might be high for the beneficiaries, considering their low level of income. He suggested that a flexible cost recovery plan 1/ Regional staff are of the opinion that roads were an integrated part of the irrigation system and as such did not require separate technical or economic justification. - 3 - might be more conducive to encouraging farmers to make use of project facili- ties. In response, the staff suggested that the Association would monitor the benefits from the project facilities and water charges could be revised as necessary. 6. Institution building. A Board member praised the project for providing assistance to help in building a Nepalese capability for project preparation and planning. He suggested that assistance to the Project Board should concentrate on developing management systems and operating procedures for the project implementing agency and on developing appropriate job descrip- tions for various positions. 7. Water management. A Board member expressed doubt about the feasi- bility of introducing farmers to proper water management practices within the project time frame, which corresponds to the construction period. The staff explained that the necessary precautions had been taken to ensure that on- farm development works were undertaken within one year of the construction of project facilities. Project Implementation 8. The project, as initially defined, was estimated to cost US$10.1 million and was to be implemented over a five year period--July 1973 to June 1978. In practice, because of cost overruns estimated in 1977 at 50%, the project had to be scaled down in order to bring it in line with the financing available. Moreover, the implementation period for the revised project had to be extended from the,initial 5 to 8.5 years. Cost overruns were due to inflation, underestimation of quantities of inputs needed, and delays in implementation. Inflation came to be a reason for cost overrun mainly be- cause price contingencies were underestimated and their reflection in the appraisal estimates was deferred by two years. Quantities of inputs needed exceeded appraisal estimates by 15% compared to an appraisal provision of 10% for physical contingencies. This was due to the preliminary nature of project design at the appraisal stage and to the relative unfamiliarity of Bank staff with Nepal due to their new involvement in this country at that time. Delays in implementation were due to difficulties in procuring goods and services for the project: identifying and recruiting consultants, recruiting local staff, limited local contracting capability, limited interest by outside contractors, and shortages of construction materials. Project Results 9. The project succeeded in its objective of improving productivity and production. The project had a positive impact on employment, albeit below appraisal expectations. There are also indications that project impact on farm income is positive, although the increase may not be equitably distri- buted over the entire farming population. Institutionally, the project served as a learning ground for the design and implementation of later irrigation development projects. However, although the lack of local skilled manpower was a major concern at appraisal and during implementation, project contribu- tion to remedying this constraint was limited. The project has also provided an opportunity for starting the development of a local contracting capability and for introducing self-help through farmers' organizations in the project area. II. MAJOR ISSUES A. Institution Building 10. Among the central issues discussed during project design and apprai- sal was the limited institutional capability in Nepal for project implementa- tion. Bank concern centered around the need for staff training, the need for a high degree of local participation in project implementation, and the need for establishing working relationships between agencies involved in complemen- tary development activities. 11. Concerning staff training, the solution retained was on-the-job training by consultants selected to help in project implementation. This solution did not prove satisfactory for four reasons. First, consultants were selected on the basis of their capability to perform specified technical tasks and not their training ability. In fact, training was given marginal atten- tion during contract negotiation. Thus, even if consultants were given a chance to train Nepalese staff, it is not certain that they could have done a proper job. Second, the responsibilities of consultants, under the difficult conditions which prevailed during project implementation, were of such magni- tude that they could not possibly have devoted the necessary attention to training without causing further delays in performing their technical tasks. Third, although the project was expected to lead to future lending, the country and the Bank did not fully adhere to the principles which guided project design (project as a pilot scheme) and ended up more concerned with the immediate problem of how to implement the project at hand than with the longer term need to increase the absorptive capacity in Nepal. Fourth, on-the-job training by itself may not have been the right approach to in- creasing the manpower capability in Nepal at that time because of a shortage of trainable people (on the job). For this reason, the consulting firm had to remedy the lack of junior Nepalese staff through employing their own junior staff. The shortage of trainable staff was particularly critical in the case of conceptualization and design functions (as opposed to construction super- vision). These functions were assumed by consultants during the totality of the project implementation period. The consulting firm brought this anomaly to the attention of project management and the Bank, but no corrective action was taken. Thus, although the project training needs were recognized prior to and during project implementation, they were addressed in a narrow and short- term perspective. 12. The limited attention to training in the context of this project is consistent with Government reluctance to use credit funds for formal training. For this reason, Government sought bilateral and UNDP assistance to meet the - 5 - training needs of this and other projects. This assistance was, however, slow to materialize (PCR, paras. 6.16-6.18). Moreover, except for the case of extension, no agency was given responsibility for identifying present and future training needs or for establishing a program to meet them. In addi- tion, project management was not motivated to improve the skill standards of project staff due to two factors. First, project staff were "deputed" to the project for a non-specified period during which they remained officially responsible to and accountable to their original departments. Thus, staff could be removed or replaced without the approval of the project General Manager. Similarly, staff could be appointed to the project without the request of the project General Manager. Second, the frequent staff rotation and the centralized decision making concerning staff appointment left no role to project management in manpower planning and development in the project context. 13. Concerning local participation in project implementation, the project provided for establishing a Project Board and for setting up village multi-purpose cooperatives and water user groups. Accounting for limitations due to shortage of qualified staff and administrative constraints, the Board was initially successful in ensuring coordination between ministries involved in the project (PCR, paras. 6.01, 6.02).!/ The effectiveness of the Board was, however, eroded over time. The end result was an unclear allocation of responsibility between the Board, the project's General Manager, the technical departments and the Ministry of Finance. A few examples illustrate this situation. First, the Board had initially the authority to prepare and authorize the project budget. For the last three years, this function was, for all practical purposes, taken over by the Ministry of Finance. The result was an imbalance between project requirements (particularly for operation and maintenance) and allocation of funds. Second, the General Manager was originally given authority for transfer of funds between categories of ex- penditures (up to 10%). This authority was taken away. Third, the project called for the preparation of an accounting system and its adoption by the project entity. Such system was prepared, approved by the Board, and for- warded to Government for comment. So far, no action has been taken. More importantly, despite the emphasis on institution building during project design, the Board did not address itself to building a local capability for project preparation and planning and to establishing management systems suitable for development projects in Nepal. Both of these functions were specifically addressed by Bank staff during project appraisal and stressed during discussions with the World Bank Board (PPAM, para. 6). 1/ The PCR states that over the years the Board has become increasingly effective in carrying out the Project (PCR, para. 6.02), then goes on listing elements which show the opposite. Field staff interviewed by the audit mission agree that the effectiveness of the Board eroded over time. - 6 - 14. Farmer participation in the project was sought through the village cooperative societies and water user groups. Setting up these cooperatives and societies was initially viewed as a prerequisite for the supply of water, extension services, agricultural inputs and marketing services. The emphasis on cooperative societies was 1ater abandoned due to the decision to introduce the T & V extension system.1' Documents in Bank files show that the initial experience with the T & V system has demonstrated that budgetary limitations will not permit continuation of this program in the project area. Government suggested a return to cooperative societies and absorbing extension agents already hired as cooperative society managers dealing with various farmer needs besides extension. This suggestion was rejected by the Bank, and a decision was made to finance extension under the second phase project. Thus, cooperative societies and, according to field staff, water user groups re- mained at the margin of the project. This was an unfortunate development as cooperative societies and water user groups could have been used to promote self-help among farmers and to facilitate their participation both in main- taining the project works at the end of the project investment period and in extending technology to farmers. The PCR assessment of cooperatives and the water user groups reflects project intentions but is not consistent with how farmers and field staff interviewed in the context of this audit perceive the relationship between this project, cooperatives and water user groups (PCR, para. 6.04). Such relationship is at worst nonexistent and at best ineffec- tive. The project as implemented had neither the authority nor the means to promote cooperatives. B. Civil Works Contracts 15. The project experienced difficulties in attracting foreign contrac- tors, and the use of local contractors caused delays in implementation due to limited contracting experience, cumbersome administrative procedures, and shortages of construction materials (PCR, paras. 3.20-3.22). Several sugges- tions were made by Bank staff and by the country to address this issue. The Government requested a review of the applicability to Nepal of IDA's regula- tion that a contractor be paid in his own national currency. The Government argued that this regulation inhibited Indian competition. The example was cited of an arrangement with the ADB whereby Korean contractors were paid in freely exchangeable currency to encourage their participation. The Bank did not concur with this suggestion. Instead, the Bank chose to address the 1/ The Region is of the opinion that cooperative societies were not a factor in Project design. The decision to introduce the T & V system of exten- sion did not cause a deemphasis on cooperative societies because the emphasis was never there. - 7 - shortage of construction materials and the limited participation by local contractors. It recommended that Government offer to supply cement and steel to all contractors and to design contracts to make them attractive to small contractors. Only the latter part of the proposal was retained. 16. In retrospect, it is clear that the local contracting capability and the suitability of project works for international competitive bidding were poorly assessed at appraisal. Thus, neither the Government nor the Bank had any plans to address the civil works contracting problems in a systematic and orderly manner. Instead, the issues were addressed on an ad hoc basis, and it is doubtful that Nepal will face fewer problems contracting for future projects than it did for this project. Some of the difficulties encountered by Government and by the Bank could have been avoided if existing limitations had been more realistically assessed and recognized. Provisions could have been made to broaden the project objective to allow the country to acquire the capability to implement subsequent construction programs. This, obviously, has a cost both in terms of time needed to implement the project and in terms of financial resources. The PCR recommendations concerning the local contrac- ting industry are worth considering in the context of on-going or future Bank-supported projects (PCR, para. 7.01). C. Tubewell Irrigation 17. Information gathered in the field indicates that the consultants given responsibility for the design of the tubewells under the project had (at the time the project started) limited prior experience in underground water development technology. The initial design of the wells was faulty. This was aggravated by poor construction supervision and subsequent poor operation and maintenance. The end result has been that some wells cannot be used and others yield an unreliable irrigation water supply. This has led to a dimin- ishing use of this facility., 18. The poor performance of the groundwater development component of the project raises two questions. First, was the approach selected for under- ground water development (tubewell) cognitive of the area's limitations (physical and institutional) or should an alternative approach have been selected (for example, wells operated by individual farmers)? This audit found no evidence in project files that the issue of which approach to follow was raised and discussed. Second, why did the failure of wells continue over time? This indicates that the supervision. of the implementation of this component was not handled properly by the Borrower and the Bank. 19. This experience points to the need to strengthen the technical supervision by both the Borrower and the Bank in on-going groundwater develop- ment activities in Nepal. It also points to the need to reconfirm that tubewells represent the appropriate approach to underground water development in this area. Furthermore, this experience reaffirms the need for the Bank to scrutinize the technical designs and any changes therein during implementation and to insist on proper operation and maintenance as a prerequisite for any further tubewell development. D. National Project Performance Audit Function 20. In 1978, it was suggested that the Bank encourage Nepal to develop a national project performance audit function. The strategy was for the first few audits to be carried out jointly by the Bank and the National Planning Commission, which was responsible for project monitoring and auditing. The Birganj Irrigation Project was considered a possible first candidate for such a joint effort. Project files show no follow-up on this suggestion. The PCR was prepared by regional staff with basic data provided by the project authority. In fact, not only did the Bank not encourage involvement of the National Planning Commission, but Bank staff concerned with the timely prepa- ration of the PCR even suggested using consultants for compiling the informa- tion in case the project authority had problems doing it. This was an unwar- ranted development given the number of on-going projects in Nepal which could benefit from both an internal and an external assessment of what was feasible to achieve at Birganj under the conditions which prevailed in the country and from identifying which among those conditions were or could have been changed within the the project's time frame. Within its limited means, the Planning Commission is actively seeking to help technical ministries establish the functions of project monitoring and evaluation as a control mechanism and as a basis for evaluating the effects of individual projects and the impact of various projects in relation to development objectives. These efforts deserve Bank support, which can be provided through (i) closer contacts between the Bank missions and the Planning Commission at the design stage of various projects in order to establish the basis for monitoring, evaluation, and project auditing; (ii) joint preparation of Project Completion Reports; and (iii) training of Nepalese staff. E. Suggestions from the Field 21. Operation and maintenance. Maintenance of project works could have been facilitated if equipment needed for maintenance had been financed under the project and if the project period had been extended to include two or three years of "pure project operation" (as opposed to construction). This would have allowed local procedures for operation and 1aintenance to be worked out, both from a technical and a financial viewpoint.. 22. Project-Bank communications. There exists a gap in communications between the project unit and the Bank due to two factors. First, field staff perceive the Bank as dealing mostly with the central agencies. As a result, project management is often not informed in a timely manner about project matters. A suggestion was made that project-related correspondence, regard- less of its source in the Bank (Projects, Programs, Legal, Controller), 1/ Regional staff indicated that the Bank continues to monitor actively the 0 & M activities in the Stage I project as part of Bank involvement in the Stage II project. - 9 - be systematically copied directly to the project unit. A suggestion was also made that routine clearance matters be done through the Bank Resident Mission to save time and allow direct exchange of views. In this regard, it was also suggested that supervision missions should, in the early stages of construc- tion, have the capability to deal with procurement, engineering, and agricul- tural issues. Moreover, they should be given authority to clear straight- forward tender documents. This would eliminate the need for project manage- ment to wait two o three months (and, at times, six months) for confirmation from Washington 23. Authority of the Project Management Unit. The Project Board has proven an inadequate tool for project management. It was suggested that the project manager should have the authority to manage and should be held ac- countable for his actions. Presently, Board members, who sit on several boards, do not have the time to study project problems and make decisions to solve them. At the same time, their role as defined in the project precludes the project manager from making decisions without calling a Board meeting. This has been a source of repeated delays in project implementation. 24. A related matter concerns the coordination among agencies serving farmers (extension, input supply, credit, marketing, research, operation and maintenance). The lack of such coordination has proven a constraint in this project due to the lack of authority on the part of project management to influence their actions and to the lack of motivation for people running these agencies to align their services with project requirements. It was suggested that provisions (mostly financial) could have been made in the project for direct participation by these agencies. This would have given them a direct interest in the project and made them accountable to the project management unit. 25. Operation and maintenance and water charges. There is a need to motivate farmers to participate and to motivate the project management unit to collect water charges. Under the present circumstances, farmers do not feel pressed to pay water charges or to contribute to the maintenance of the irrigation system because they feel that doing so is unlikely to improve the quality of the services they receive, including timely water supply. Simi- larly for the project management unit, there is no correlation between the amount of water charges collected and the availability of funds for them to maintain and operate the irrigation system. Moreover, project management has no power to collect water charges as there is no legislation to enforce collection. It was suggested that the issue of cost recovery could be pro- gressively resolved if (i) the project management unit were given the respon- sibility to provide reliable services to farmers (in order to motivate them to 1/ Regional staff indicated that although a gap in communication between the Bank and field staff may have existed in the early stages of the project, this gap has been bridged. The audit suggests that further bridging may be needed since field staff perceive that gap as still existing. - 10 - participate), and (ii) simultaneously, funds collected as water charges were made available for discretionary use by the project management unit, which would be held accountable to the Ministry of Finance for the use of such funds. III. LESSONS FOR THE BANK 26. The training needs in the context of this project were not well identified at the design stage. The responsibility for training was given to expatriates whose ability to train was not assessed and whose primary duties were of a technical and executive nature. In addition, because of their overriding concern with physical progress in project implementation, neither Government nor the Bank gave training the needed attention during project implementation. A training component should be designed and costed at least as carefully as other project components during project formulation. 27. The Bank's strong concern with strengthening the extension service may have negatively affected the extent of farmers' participation in the project by indirectll hindering the development of farmers' cooperatives and water users groups.- Institution building should be designed with due regard to various project needs and the means available to meet them. 28. The concept of a Board as a coordinating agency and a facilitator of decision-making has not worked well in the context of this project due to unclear allocation of responsibility and accountability between the Board, the Project Manager, the technical departments and the Ministry of Finance. In view of the multiplicity of Boards coordinating all major projects in Nepal and in view of the limited number of people serving simultaneously on all of these Boards, the concept of Project Board has ceased to be an operationally suitable mechanism for coordination. Proper selection of project managers and reinforcement of their authority and accountability may be an alternative to the Boards. 1/ The Region disagrees with this conclusion. -11- ANNEX 1 RCA DEC 1? ?4< 22p I DEVA iP . H I .!TO, j L0 29 DECEISEE 13, 12 ? RTHi'ANDU li.EPAL 732 FOR rAF-l, 0E0 RE iEPAL PPAR ORAFT FOR clRrCA,J IRRIGATIOi PhOJECT CD 6HATT, DG FUR IRRIGAT.ION DEPARTMENT HAS ASKED ME TO PASS FOLLO'WIG COilHENTz TO YOU: AAA PARAPRAPH 11, PAGE 7, THE PHRASE QUOTE EXPATRIATE TECHJ4CIA,bS FRO i'DIA UN'UOTE SHOULD bE DELETED. THE CONSULTANT, vi'PPONx kOEI, DID N CT HIRE ANY INDIAN TECHNICIANS. THERE 1WERE SOME IfDIAh OVERSEERS HIRED BY HMG AT THE BEGINNING OF THE PROJECT FOR AbCUT A YEAR OR SC BUT THESE wERE REPLACED SY 1EPALE6E. BE- PARAGRAPH 14, PAGE 10. THE STATEMENT THAT'THE PROJECT HAD AO AUTHORITY TO PROhOTE COOPERATIVES AND wATER USER OROUPS Is NOT FACTUAL. THE PROJECT DID HAVE AUTHORITY TO PROMOTE AATER USER GROUPS. REGARDS SLADE  ANNEX 2 - 13 Page kone 215 Ministry of . B,oeg 5Varavani ent 'Board Narayani Zon opment Project Head Offics Ref No:- ü < <> K hwik Birganj(Nepal) Ref No:- / t/p Date:--a' 9 kr. Shiv ;. Kapur )irec tor Operations Ejvaluati,N Department South Asia Projecte Department Internaticial Develo,ment Association 1818 h, btreet, 1\.W. Washiri,.,ton D.G. 204,31. Dear kr. ä.apur, .Mef; - kj ies Audit rt on Ieal Biý-pnj l rriigation Pro jep C_rcdljt'f-F 'bith reference to your letter dvted October 28,1982. 1 ar, sendin¿· herpvith our veiws a. crmmnts. I hore it will be holful to fins-li,e the Project performriene Audit report. with retards. Sirc y. ours ,. Thakur ) General ange General Manäget - 14 - COMMENTS ON PROJCT PER)ORMANCE AUDIT REPORT ANNEX 2 NEFAL BLKGAXJ ItIGATION PROJECT Page 2 (NARA"ANI ZON) CREDIT 375 . 4, a W OCot.25.1982 a0 . Pare-14 a- Ministry of Land Reforms did not agree to transfer co.operative section under Project managements. Pagce-15 Para-24 :- Project feels that by introducing extension services, the Water User Group has not been affected adverely, as the people participation has increased through water user group specially in water management. Pare-4.04 s- Now 40% of area is 4 covered by improved paddy (Normal or ead 2nd.paddy) wich ripens in Nov. soit does not require irrigation in Nov. previously more than 90% were local paddy which ripenas late and require irrigation in Nov. Para 4.08 s- Actually in Bara & Parea districts produce major portion of wheat and maise seed supplied by AIC throughout Nepal. So coneiderTble *ount of improved seeds (wheat, maize) find their uay from seed growers directly to farmers. In case of paddy, every year 2-3 NT seeds of improved newly released varieties are distributed to 2000-3000 farmers wh6ch in ncxt years covere substantial area. Para 4.11 a- Most of maize planted in pro,ect area is winter maize, which yields not less than 4 K:/hs. Para 6.06 a- At present Masuli is most pupular variety and is growr. in more than 40% area of the project. TABL-2 PART-A PbYsical Achievements 1.2 Check structure 3 Noe. 1.3 Foot bridge 2 Nos. 1.5 Drainage Inlet 9 Nos. - 15 - NEPAL BIRGANJ IRRIGATION PROJECT (NARAYANI ZONE) Credit 373-NEP PROJECT COMPLETION REPORT South Asia Projects Department Irrigation II Division October 14, 1982 1 1 - 17 - NEPAL BIRGANJ IRRIGATION PROJECT I. SECTORAL BACKGROUND 1.01 The development of Nepal's agriculture has been and still is severely constrained by a number of physical and institutional factors. The population is growing fast and is leaving increasingly the densely populated hill areas with their thin resource base. Agricultural production, even in the more favored plains of the Terai, has been at best stagnating. Per capita consumption as well as exports of foodgrains have been declining. This was true when the Birganj Irrigation Project was started and is even more so today. A sustained growth of agriculture could only be expected from a more intensive exploitation of the limited agricultural land base. Irrigation, under the given climatic conditions, is a key towards this objective, although by no means the only one. Irrigation facilities in the early 1970s covered about 7% of the cultivated land, mos.tly in the Terai and were generally restricted to supplementary irrigation of paddy in the rainy season only. Construction and completion of existing irrigation infrastructure therefore had high priority in past development plans of Nepal. 1.02 The present project was designed to create improved irrigation facilities, not only capable of perennial and more reliable water supply, but also linked to the coordinated provision of crucial support services, such as extension and research, agricultural input distribution and credit supply. Another important cons-traint, the shortage of skilled managers, engineers and agricultural officers, was addressed by foreign consultants through intensive on-the-job training of such staff during project implementation. The project fitted into the strategy of intensive development of agriculture in the Terai which had a higher productive potential, promised quicker returns, and was less difficult and costly to develop than agriculture in the Hills. 1.03 This IDA project was also the first agricultural project to be supported by the Bank Group in Nepal. It is also the first IDA irrigation project in Nepal to reach completion. The project was favored by its rela- tively easy location in the Central Region of Nepal, its proximity to a major urban center at the Indian border, Birganj, and its accessibility by road from Kathmandu and from India. Implementation difficulties to be expected were thus less than in most other locations of Nepal. II. FORMULATION Origin 2.01 In 1959 an agreement was signed between HMGN and the Government of India for the execution of the Gandak Irrigation and Power Project. Under this agreement GOI was to finance and construct a barrage on the Gandak river and irrigation systems on its left and right banks commanding a total area of about 1.2 M ha in India and 60,000 ha in Nepal. In Nepal the irrigation - 18 - 3 system would only be constructed down to secondary canals of 0.57 m /sec (20 cusecs) capacity. t was agreed that the construction of the distributary system below 0.57 m /sec capacity would be the responsibility of HMGN. Under this agreement, India started construction of the Nepal Eastern Canal (NEC) in 1968. 2.02 In 1969 Nippon Koei Company (Japan) was contracted under the FAO/ UNDP NEP 7 Project to prepare a Master Plan for the Central and Eastern Terai, including the preparation of feasibility studies for selected areas. The Birganj Irrigation Project was identified for inclusion in the studies by an FAO/IBRD Cooperative Program mission in early 1970. Initially the Bank decided against participation in an ongoing Indian aid project and Nippon Koei com- menced the preparation of a feasibilily study for a 10,000 ha groundwater development project for 300 tubewells in the Birganj area. However, when it became apparent that the Indian aid project would be completed by 1972 and that surface water development had a higher priority than groundwater develop- ment the Bank agreed to finance a joint surface and groundwater development project. HMGN, FAO/UNDP and the Bank subsequently agreed that Nippon Koei should prepare a feasibility study for a combined surface and groundwater project. A December 1970 FAO/IBRD Cooperative Program project preparation mission suggested that the surface irrigation scheme should be limited to 14,000 ha. However, as the NEC would be constructed up to the Bagmati river and command some 40,000 ha gross it was decided that the feasibility study should cover a larger area, some 28,700 ha. It was also agreed that the groundwater component should be limited to the re-equipping of 14 existing tubewells (also constructed under Indian aid) and the drilling of 14 new tubewells. The total groundwater area was thus limited to 2,730 ha; it was also intended to function as a pilot operation for future large scale ground- water development. 1/ Preparation, Appraisal, Negotiation and Approval 2.03 Several FAO/IBRD/CP missions assisted HMGN with the preparation of the project in 1970. The main issues identified at that time were: (i) lack of adequately trained Nepalese staff to implement and operate large irrigation systems; (ii) the need to set up an efficient organization for project implementation and operation; (iii) the lack of provisions in the 1959 agreement between India and Nepal on: (a) specific amounts of water Nepal would be entitled to receive from the Don Branch Canal 2/'; (b) the operation of the regulators in the Don Branch Canal for supplying water to Nepal as these structures were located in India; and (c) ensuring adequate maintenance of the Don Branch Canal by India. 1/ Mainly to test and strengthen HMGN's institutional capabilities as there was no doubt about the groundwater potential as such. 2/ The Don Branch Canal links the Nepal East Canal with headworks on the, Gandak River; it also feeds another main canal in India. - 19 - 2.04 The feasibility study for a combined surface and groundwater project in the Birganj area was completed by Nippon Koei in August 1971. 1/ The study was based on: (i) feasibility study designs for one surface irrigation block (Block V, 2,000 ha) and a groundwater development for 200 ha with 3 tubewells, both with topographical,maps prepared on scale 1:2,500 with 25 cm contour intervals; (ii) semi-detailed soil surveys with irrigation capability classi- fication carried out to USDA standards for 22,000 ha of the surface water project area and 20,000 ha of the groundwater service area, including the groundwater pilot project area (detailed soil surveys were prepared for the two model areas mentioned in item (i)); (iii) detailed geohydrological studies including a geophysical survey, test pumping of 11 existing wells and 5 wells drilled under the project, and the preparation of a digital ground- water model; (iv) an economic survey covering four villages in the project area and covering some 1,100 ha and 1,100 farm households; data were obtained regarding family size, cropping patterns, labor input, output, marketing, animal power etc.; (v) crop projections were based on experience available from three research stations in the Terai (the Parwanipur, Hardinath and Janakpur farms), and existing farm practices in the Terai; (vi) economic farm gate prices for crops and inputs based on projected world market prices; and (vii) construction rates similar to those used for construction in the Birganj area. 2.05 Following IDA's request additional discussions were held between HMGN anq GOI in October 1971. The parties agreed that Nepal would be entitled to 24 m /sec (850 cusecs) at the point of entry of the Don Branch Canal in Nepalese territory. 2.06 Project appraisal took place in November/December 1971. The appraisal mission found that the inlet transition of a major acqueduct in the Don Branch Canal had collapsed in the monsoon of 1971. It was also discovered that some narcotic drugs were grown in the project area. 2.07 During appraisal, it was agreed that HMGN would set up a special Project Board, the Narayani Zone Irrigation Development Board (NZIDB), to be responsible for the execution and operation of the project. This was to ensure efficient coordination of the various activities in one organization instead of being spread over several departments and institutions. The Board would appoint a General Manager who would be the Chief Executive of the organization. It was further agreed that an expatriate consulting firm be recruited to provide the necessary training, technical assistanee and advisory services to the Board's staff. 2.08 Subsequent processing of the project encountered considerable delays due to: (i) the uncertainty as to the date of completion of ongoing works in the project area by GOI (through the Government of Bihar); (ii) delays in deciding on how to deal with the issue of ensuring adequate maintenance of the Don Branch Canal in India; and (iii) internal delays within the Bank due to staff reassignments. In September 1972 the Bank decided that loan negotiations be delayed until: (i) an agreement had been reached within the framework of the Gandak Coordination Committee, consisting of representatives of GOI and HMGN, that the Don Branch Canal should not be closed for regular maintenance - 20 - and repair for more than 30 days a year; and (ii) receipt of a firm schedule for completion of the project and a detailed statement of the steps intended to be taken by HMGN and the Government of Bihar to ensure the timely completion of the works. It was also decided that the setting up of a Project Board. together with the appointment of a General Manager having qualifications and experience acceptable to IDA would be a condition of presentation of the Credit to the Board. Further that, as a condition of negotiations, HMGN would have to provide assurances that it has the authority to collect water charges from users of publicly constructed surface and groundwater irrigation works. During October and December 1972 satisfactory agreements were reached between HMGN and the Government of Bihar on the maintenance and operation of the Don Branch Canal and the completion of the irrigation works by the Government of Bihar. This was confirmed during loan negotiations in January 1973 at which time HMGN also reaffirmed its decision taken in December 1972 to exercise strict control on the production and distribution of narcotic drugs in the project area. The Nepalese delegation also presented the Association with a legal opinion that HMGN had the legal authority to levy water charges. Further, a draft agreement for the establishment of the Project Board was submitted. The Board was formally established before the project was sub- mitted to the IDA Board of Directors. However, the appointment of a General Manager with qualifications and experience acceptable to IDA became a condition of effectiveness. The recruitment of a consulting firm acceptable to the Association became one of the conditions of the Credit Agreement. The project was approved by the Board of Directors on March 20, 1973. Project Objectives 2.09 The project's main objective was to raise agricultural production, incomes and employment by intensification of cropping in an area of about 31,400 ha. This was to be achieved by completing and improving the existing irrigation works so that a switch to higher yielding varieties of paddy in the rainy season and to irrigated upland crops in the dry season would be possible. By completing the Indian-built surface irrigation system, the IDA project was to provide year-round reliable water supply to outlets serving units of 30-40 ha each. The same objective was pursued by rehabilitation of existing and construction of new public tubewells north of the Nepal East Canal. This component had as its main purpose the testing of the feasibility of groundwater development for groups of farmers with a view toward future expansion of this pilot scheme. In addition, the project aimed at providing a coordinated package of agricultural support services (agricultural extension and research, cooperation services, service roads, etc.). Since the project was one of the first of its type and scope in Nepal, institution building and demonstration effects were considered important objectives, too. Project Description 2.10 The project is described in paras 4.01-4.10 of the Staff Appraisal Report (SAR) dated March 1, 1973. Specific components were: - 21 - (a) Surface-Water Irrigation (i) Improvement of the Nepal Eastern Canal including the construction of additional control structures and the improvement of existing structures. (ii) Improvment of about 50 km of secondary canals above 0.57 m /sec (20 cusecs) including the addition of control structures. (iii) Construction of about 1,150 km of secondary and ter- tiary canals with control structures and field turnouts. (iv) Construction of about 800 km of drainage networks. (v) Construction of about 290 km of canal service roads and surfacing of about 110 km of these roads to link the project area with the Kathmandu-Birganj road and with cooperative stores. (b) Groundwater Irrigation (i) Testing of existing tubewells and replacement of inadequate pumps, motors and control devices. (ii) Drilling and equipping of about 14 new tubewells. (iii) Construction of about 18 km of 11 kV transmission lines to service existing and new tubewells. (iv) Construction of related storage tanks and irrigation distribution systems, consisting of lined main canals, an unlined secondary canals to the turnouts commanding 30-40 ha service units each, and drainage networks in the command of each tubewell. (c) Project Facilities (i) Construction of offices, workshops, stores and living quarters for staff and consultants. (ii) Installation of a communications system for the irri- gation system operations. (iii) Installation of hydrological telemetry equipment for the Nepal Eastern Canal. (iv) Construction of about 12 cooperative stores. - 22 - (d) Technical Assistance A consulting firm was to be retained to provide the project with training, technical assistance and advisory services in engineering, construction, general administration, personnel finance, accounting and auditing, agricultural extension and cooperative services and in the operation and management of the project as an integrated agricultural project. (e) Feasibility Studies Feasibility studies were to be carried out by consultants for: (i) A pumping scheme to lift water from the Nepal Eastern Canal to irrigate areas north of said Canal. (ii) A further irrigation project to be selected. (f) On-Farm Development The project was to provide technical assistance to plan, design and layout on-farm watercourses and field drains and to indicate area requiring landshaping. The project was also to organize farmers to complete construction of these works them- selves. Because holdings in the project area are small,'most of the farmers were supposed to be able to carry out the works with their own labor. Farmers with holdings larger than 3 ha not able to undertake the work with their own labor could hire NZIDB to carry out the work for them. The Agricultural Deve- lopment Bank of Nepal (ADBN) was to provide credit to these farmers to help finance the works. Assurances were obtained that under the project, farmers would receive irrigation only after they had completed all on-farm development works on their lands. Should a farmer fail to complete the construction of the watercourses, field drains and landshaping as indicated in the Plan within one year after construction of the tertiary canal and drain serving his area, NZIDB would perform such construction and recover the expenses from the landowners. 2.11 Physical targets of the project were later modified to accommodate a cost overrun. These changes are discussed in para 2.15 below. Project Cost Financing and Procurement 2.12 Total project cost (including civil works, equipment, land acquisi- tion, administration, technical assistance and on-farm development) was estimated at NRs 102.4 M (US$10.12 M at the then prevailing exchange rate of NRs 10.125/US$). Local cost was estimated at NRs 62.4 M (61%), and foreign exchange cost as NRs 40.0 M (39%). The cost estimate included physical con- tingencies of 9.5% over base cost (except for on-farm works and land acquisi- tion), and price contingencies based on an assumed local price escalation of - 23 - 7% per year and an escalation of foreign prices of 5% per year. 1/ Dis- aggregated cost estimates are shown in Table 1. Project scope and cost were later revised (para 3.03). 2.13 The Credit of US$6.0 M was on standard IDA terms. It was to finance about 59% of total cost (100% of foreign exchange cost and 33% of local cost). The Credit was expected to be disbursed over five years (September 1973 to December 1978). The cost of land acquisition, administration and on-farm development, and the recurrent costs of operation and maintenance were not eligible for IDA reimbursement. 2.14 Procurement of civil works and equipment was required to be under international competition bidding procedures. Civil works contracts were to be let for one or two irrigation blocks at a time, and only after land had been acquired for the works. Targets and Goals 2.15 As a result of the project, controlled irrigation water supply was to be provided for about 28,700 ha of surface water command and 2,700 ha of tubewell command. Benefits were to accrue in form of increased production of paddy (through higher yielding varieties), wheat (through dry-season irriga- tion), pulses, oilseeds, vegetables and sugarcane. These aims, together with a strengthened system of agricultural support services, were expected to result in an increase of cropping intensity from about 135% to 170%, and in an incremental production of individual crops as shown below: Production Targets for the Original Project Scope ('000 tons) At Time of Future With?ut At Full fa la Appraisal Project - Development- Incremental Paddy 40.8 48.0 90.4 42.4 Wheat 7.9 11.3 47.6 36.3 Pulses 0.8 0.8 1.3 0.5 Oilseeds 0.5 0.6 1.2 0.6 Vegetables 8.6 9.4 18.8 9.4 Sugarcane 23.6 28.3 125.6 97.3 Total Value (NRs M) 47.0 51.0 122.0 71.0 /a Estimated at ten years after completion, i.e. 1988. 1/ This is the assumption stated in the SAR, however, SAR cost tables did not exactly reflect this assumption (para 3.25). - 24 - 2.16 Yields of major crops were expected to double (paddy) or triple (wheat). The total value of production was expected to increase by about 140% over the "without project" production. The project was to reach about 20,000 farm families, of which about two-thirds were considered to be subsistence smallholders. Incomes of small, medium and larger farms were expected to increase by about 93%, 106% and 128%, respectively, in the surface water area, and by 80% on average in the tubewell area. On-farm employment generated by the project was estimated to exceed "without-project" employment by about 75% at full development. Project construction itself was estimated to create about 7,000-8,000 man-years of employment. 2.17 The economic rate of return was estimated both including and exclud- ing the cost incurred under the India-built project, i.e., before the IDA project. Excluding sunk cost, the rate of return amounted to 24% for the surface water and 14% for the tubewell area. It was estimated as 21% for both components combined. Including sunk cost, the rates of return were 14% and 11%, respectively (14% combined). III. IMPLEMENTATION Effectiveness and Start Up 3.01 The project became effective on July 9, 1973 after the Project Manager had been appointed. After project appraisal UNDP continued its work on the project under a "bridging operation". Under this operation contract documents were completed in June 1972 for: (i) the civil works in Block I; (ii) construction of irrigation, drainage and roads systems and grain stores in approximately 1,000 ha of the existing tubewell scheme (Phase I) and the electrification of 7 existing wells; and (iii) construction of the project headquarters. Additional semi-detailed soil surveys to cover the entire 40,000 ha gross command area under the NEC were also carried out, as well as drawdown tests for the existing tubewells. A shortlist of consulting firms was approved by IDA in February 1973 and it was anticipated that project construction could start in October 1973. However, the consultants contract with Nippon Koei was only signed on December 28, 1973. In January 1974, an implementation schedule was prepared as shown in Chart 1. However, immediate difficulties were experienced with the tendering for the civil works. Only one tender was received for the civil works to be constructed for the existing tubewells (Phase I) and the project's headquarters. There was also a lack of interest for the supply of pumps and electrical facilities required for the groundwater scheme Phase I; the contract had to be retendered. In addition, the tender documents prepared for the civil works of Block I could not be issued as it was discovered that a secondary canal that was supposed to have been constructed under Indian aid had in fact not been constructed and required an important crossing under the Uriya River. It was decided to subdivide Block I into two Sub-Blocks, IA and IB. - 25 - 3.02 Because of lack of interest from international contractors to parti- cipate in tendering for the small scale irrigation works under the project it was agreed in September 1974 to tender only among local civil works contrac- tors. Initial delays were also due to a serious shortage of qualified depart- mental staff for survey work and design, the high turnover of staff as staff positions were only on a temporary basis, the shortage of vehicles and survey equipment and the frequent changes in project management (there were four different project managers until project management stabilized in early 1976). Due to these difficulties the tender documents for Blocks II, III, IV, V and VI were only completed in mid 1976. Revision 3.03 The project as appraised was estimated to cost US$10.1 M equivalent and was to complete a surface water irrigation scheme to serve 28,700 ha (in 12 Blocks) and a pilot tubewell scheme to irrigate 2,700 ha, as well as improvement of 60 km of the NEC and 50 km of existing subsidiary canals, con- struction of secondary and tertiary irrigation and drainage channels, service roads, project buildings, technical assistance and the preparation of feasi- bility studies for a second project. However, after initial starting up delays and rapid increases of construction cost, it became clear that the project faced a cost overrun of approximately 50%. 1/ As the Government found itself unable to provide from its own resources the additional funds required to complete the project as originally defined, it requested IDA that the size and scope of the project be reduced so as to bring it in line with the finan- cing available. A revised project was submitted to the Executive Directors of IDA in November 1977 and subsequently approved. Under the revised project the surface water irrigation area to be developed was reduced from 28,700 ha to 16,000 ha, thereby eleminating from the project 12,700 ha of net irrigable area. The 2,700 ha groundwater scheme was left unchanged. It was considered that the revised project would remain technically feasible and, at a rate of return of 18-20%, economically justified. Its total cost was estimated at US$9.7 M (US$4.6 M in foreign exchange) compared with the original project cost estimated in 1972/73 at US$10.1 M. It was anticipated that the revised project would be completed by July 1979 and the closing date of the Credit was postponed to December 31, 1979. The Government also requested the Association to include the works eliminated from the revised project in a second stage project for irrigation development. The second phase project was appraised by IDA in September 1977 and a US$14.0 M credit for the Naryani Zone Irrigation Development Stage II Project was approved by the Board on October 10, 1978 (Credit 856-NEP). Physical Progress 3.04 The project works undertaken by India were completed in two phases. The first 30 km of the NEC command area was handed over to HMGN in 1975 and the balance in 1977. The award of civil work contracts to be financed from Credit 373 continued to be difficult. Only one tender was received for the 1/ This was the estimate of cost overrun in 1976/77. - 26 - NEC I works; it was rejected and the works had to be retendered. Also ten- dering for Blocks II, IV and V was not successful in 1976 as there were not enough local contractors who could carry out civil works of even this limited scale. International contractors had been prequalified but had submitted no bids. The tender documents were therefore simplified to meet better the need of the local contractors and the work for Blocks II, IV and V was sub-divided into smaller contracts to encourage local bidding. Most of the construction work was finally underway by March 1977. However, the implementation capacity of the local contractors appeared to be very limited and although their contracts were small it generally took more than three years to complete one Block or Sub-Block. Construction progress was also affected by a shortage of cement in the 1977/78 and 1979/80 dry seasons and the shortage of fuel in the 1979/80 construction season. 3.05 An assurance had been obtained during negotiations that no civil works tenders would be issued until the lands had been acquired by HMGN. Although this is HMGN policy, in practice it could not be accomplished because of time consuming procedures. However, as the construction work progressed only slowly, this did not contribute significantly to the delays. 3.06 The works constructed under the project have been detailed in Table 2. In addition, farmers have constructed a considerable number of farm channels within the irrigation sub-units. The irrigation and drainage systems developed under the project area may be summarized as follows: - 27 - Area Covered by Field Net Irrigable Channels within Area Developed 7.5 ha Sub-units /a (ha) (%) A. Surface Irrigation Scheme Block IA 1,200 35 Block IB 1,910 100 Block II 2,300 65 Block III 2,500 60 Block IV 2,100 45 Block V 2,570 65 Block VI 3,400 70 Sub-total A 15,980 B. Groundwater Scheme Groundwater Irrigation Scheme Phase I 1,050 /b 100 Groundwater Irrigation Scheme Phase II 1,680 /b 100 Sub-total B 2,730 Total A+B 18,710 /a By February 1981. 7T This is the design net command area. Actual irrigation covers much less area (see para 3.14). Design Standards and Quality of Works 3.07 Surface Irrigation Scheme. For water distribution purposes the project area has been sub-divided into irrigation units of 30-40 ha receiving water from one gated outlet. Water measuring structures (Parshall flumes) have been installed at the heads of the branch secondary canals (commanding areas of some 500 ha). Tertiary canals have been constructed within the irrigation units to provide water to sub-units which are on average about 7.5 ha. The tertiary canals have been provided with the necessary division boxes, which can be closed with concrete stop- logs. The feasibility study recommended a design discharge of 0.94 1/sec/ ha. This was soon increased to 1.0 1/sec/ha for the major system and to 1.5 1/sec/ha for tertiary canals. Recently prepared water rotation schedules suggest that a capacity of 2.0 1/sec/ha for the tertiary canals would have been more appropriate, but by making use of existing freeboard the required capacity can be accommodated. - 28 - 3.08 Drainage. The 1972 design report recommended a drainage design discharge of 5.8 1/sec/ha. In view of anticipated cost overruns and the very large, unmaintainable drainage channels resulting from this coefficient, this design discharge was reduced to 1.0 1/sec/ha which proved to be insufficient. In 1978 the design discharge was again increased to 3.0 1/sec/ha and this appears to be quite adequate so far. 3.09 Roads. The feasibility study recommended an all weather road net- work consisting of metalled roads along the NEC, the main and secondary branch canals and metalled roads linking every village to the canal roads and to Birganj. However, during appraisal it was decided, mainly for reasons of reducing costs, to restrict metalling to the NEC road and the secondary canal roads that would connect with cooperative input stores. The remaining road network would consist of 3.5 m wide earth roads along the main secondary and branch secondary canals. In 1975 it was decided, due to shortage of funds, to delete the roads leading to the cooperative input stores as the levels of the existing village roads that would be used for this purpose needed to be raised considerably which was not included in the project cost estimate. The roads along the branch secondary canals have in practice been terminated at the nearest village road while for the remaining length of the branch secondary canal an embankment width of 1.5 m was adopted, i.e., suitable for motorcycle traffic. Also, the embankments along the tertiary canals were made 1.0 m wide for use by motorcycles. 3.10 It appears that projects such as this necessarily have to develop the basic rural infrastructure, not only the irrigation system itself, espe- cially in a rather undeveloped country as Nepal. Experience has already shown that the project needs a better road system than anticipated during project appraisal. The earth roads along the canal embankments are only suitable for light vehicles and are closed for bullock cart traffic to prevent damage to these roads. However, as the existing village to village road system is in very poor condition, it is difficult to enforce this rule, especially during the monsoon season. In retrospect the project should have been designed to provide at least: (i) a road suitable for heavy truck traffic along the main canal and connections from this road to the input stores; (ii) improvement of existing village tracks; and (iii) the metalling of the canal operation roads to provide better access into the project area by O&M and agricultural extension personnel. 3.11 Works included in the Stage II Project (Credit 856). As a result of operational experience gained with the project, deficiencies in the design of the irrigation, drainage and road networks were recognized prior to the reformulation of the project in a Stage I and Stage II Project. Additional improvement works for the Stage I area have therefore been included in the Stage II credit. These works include, inter alia: (i) adding 10 cm gravel to the NEC service road; (ii) improving a second low level earth road along the NEC for bullocks and trucks by raising this road by 60 cms; (iii) improvements to the village road-system by providing cross drainage structures and bridges; (iv) lime stabilization tests for low cost roads; (v) construction of raised earth roads between the NEC and the input stores; (vi) increasing the capacity of existing drains to 3 1/sec/ha where these had been designed at 1 1/sec/ha; (vii) some improvements to the Tilawe irrigation system; and (viii) the improvement of seven drain outfalls. - 29 - 3.12 Groundwater Scheme. The project included the rehabilitation of 14 existing wells which had been constructed under Indian aid from 1969 through 1971 and the construction of 14 new wells. In 1975 three of the existing wells caved in and it was decided to construct three new wells to replace them. Thus 17 new wells were constructed under the project, most of them with fiberglass for casings and screens, except for the pump housing which remained mild steel. However, the jointing of these two different materials was difficult and the mild steel for the pump housing was also replaced by fiberglass in the design stage. In 1979 it was found that another four of the existing wells had to be replaced, three due to corrosion and one due to considerably reduced discharge, and also that three of the new fiberglass wells constructed under the project had failed due to cave-in. Already during construction one of the fiberglass wells had caved in but was replaced by the contractor at his cost. The reasons for the fiberglass failures were inves- tigated in 1979 by Nippon Koei's expert who had taken part in the design of the wells and also had supervised construction. Nippon Koei's report was not very conclusive but it seems that the major reason for the failures was the weak joint between the different sections of the casing pipe. In one case of failure a contributing factor may have been that the gravel pack pipe was not designed to extend down to the acquifer. Three more new wells have sub- sequently failed (one in 1980 and two in 1982). In retrospect, it was a mistake to try a relatively new technology in Nepal that requires high standards of construction. A new drilling contract was signed in 1980 to replace seven of the failed.wells. They were to be constructed of mild steel and are now virtually completed (June 1982). 3.13 Although farmers' response to the availability of water was initially very good, the scheme has not been successful due to unreliable service, not only because of the well failures, but also because of irregular power supply, stolen transformers, delays in completion of electrical connections, and the lack of an adequate stock of spare parts at the site, leaving wells non-operable for extended periods. The latter is partly due to HMGN's difficult procurement procedures and the small budget available for operation and maintenance. For example, seven wells needed to be repaired in 1980 and could only be kept operational by cannibalising pumps and motors of wells that went out of operation in 1979. 3.14 The groundwater scheme was planned to irrigate 2,730 ha (net), but indications are that under normal conditions, i.e., no power failures, no major breakdowns, etc. the actual irrigable area would be less than 2,000 ha because the adopted design capacity (0.53 1/s/ha) 1/ for pumps and canals is less than the actual required peak capacity of about 1.0 1/s/ha. All new wells have a capacity of about 80 1/s and are able to irrigate about 80 ha only instead of the 120 ha planned. Considering that paddy was included in the cropping pattern, the design capacity was underestimated. Unfortunately, due to the low water demand as a result of the unreliable water supply, the actual irrigated area has not reached more than about 500 ha in a single crop season. 1/ Over 24 hours, actually pumps were sized for about 0.67 1/s/ha in order to limit operating hours to 19 hrs/day. - 30 - 3.15 The groundwater scheme was also reviewed by a Bank consultant familiar with the area and the original Feasibility Report. His report is attached as Annex. On Farm Development 3.16 Good progress has been made by farmers with the construction of field channels below the outlets. Initially farmers were opposed to field channels as they were accustomed to field to field irrigation and did not want to give up part of their land for the construction of these channels. The first step in the process of constructing these channels was that the farmers, in order to gain confidence in the reliability of the system, had to see first that irrigation water was reaching their division box. Then farmers took time to realize that the field channels were very useful for growing a winter crop and an early paddy crop. Also it took time for all farmers in the sub-unit to agree to the construction of the field channels as this is a communal effort and it was possible for a few farmers at the upstream end to obstruct construc- tion. The whole process of field channel construction took about three years after water became available at the turnout. So far, farmers have utilized their own labor to construct the field channels. Credit from the Agricultural Development Bank of Nepal has not been utilized for this purpose. 3.17 The credit agreement included the condition that if a farmer should .fail to carry out on farm development works within one year after construction of the tertiary canals and drains serving his area the NZIDB would then undertake to construct these works and recover the costs from the land owner. Strict adherence to this condition has not been found practical as the NZIDB had neither the budget nor the legal farmework to undertake this type of work. It was found that enforcement of this condition was not really needed as it was shown that persuasion worked much better. Construction Standards and Quality Control 3.18 The structures for the irrigation, drainage and road works were mostly designed in brick except where standard concrete pipes could be used or where reinforced concrete elements were needed. One advantage of construction in brick was that it saved cement which was often in short supply, but the quality of the bricks manufactured in local kilns was mediocre. Quality control of the works was difficult because many small works were scattered over a large area, and the local contractors proved to be very inexperienced and lacked sufficient supervisory staff. Continuous support was available at site from a Nippon Koei construction supervisor, but his impact was naturally limited. As the contractors had virtually no construction equipment, compac- tion of the earth embankments was not very good. Despite these constraints the quality of the construction works is reasonable. One way of improving quality and increase output could have been to prefabricate the smaller structures. Special Studies 3.19 Feasibility studies have been financed out of the credit to prepare the Bhairawa-Lumbini Groundwater Project and the Sunsari Morang Irrigation & - 31 - Drainage Development Stage I Project. These projects were subsequently financed by IDA in 1976 and 1980 under Credits 654 and 812, respectively. A feasibility study for,a pumping scheme to lift water from the NEC to irrigate an area north of this canal was also initiated but was subsequently dropped in favor of a proposal to develop a Narayani Zone Stage III Project commanding some 10,000 ha located between the Arwa and Bagmati rivers. This would com- plete the 40,000 ha development originally envisaged in the Gandak Agreement between Nepal and India. The improvement of the NEC Stage III which commands this area is included in the Narayani Zone Irrigation Development Stage II Project (Credit 856). Procurement and Construction 3.20 After the initial difficulties had been overcome in the setting up of an effective project organization, the main constraint in implementing the project on schedule proved to be the weakness of the local construction industry. During negotiations, assurances were obtained from HMGN that civil works would be awarded after international competitive bidding and that the civil works contractors would be prequalified. It was also intended that contracts would be let for two Blocks each. However, after several interna- tional contractors had been prequalified it became evident that these were not interested to tender for such small scale civil works in a remote area. Surprisingly, Indian contractors also did not show no interest in these works. Although IDA was initially reluctant, it was eventually agreed that each Block should be split up in four Sub-Blocks in order to encourage more parti- cipation of local contractors. The contractors were allowed to submit bids for either one or more of the Sub-Blocks. Also for the benefit of the local contractors it was decided to simplify the tender documents and to make cement and steel available to them. For prequalification purposes the contractors were divided in classes A, B, C and D. Class A contractors were allowed to bid for any number of Blocks, Class B contractors for works totalling NRs 10 M or two Blocks, Class C contractors for works amounting to NRs 5 M, or one Block, Class D contractors for works to NRs 3M and Class D contractors for work up to NRs 1.5 M. The civil works contracts awarded under the project have been tabulated in Table 3. 3.21 The performance of the contractors has been poor by international standards. The performance of the National Construction Corporation of Nepal (NCCN), which was awarded contracts for the construction of the project headquarters and the groundwater area (Phase I) has been especially disappointing. Generally speaking, all contractors have been hampered by the lack of trained technical staff, (which resulted in poor management and organization), the lack of financial resources and access to bank financing, difficult banking procedures in procuring bid bonds and performance bonds, and the lack of basic construction equipment such as compaction equipment, concrete mixers, trucks, etc. Despite these constraints the contractors' capabilities have developed somewhat during the course of the project and they are now able to handle somewhat larger works. 3.22 Also, the procurement of equipment and materials has been difficult as most had to be imported from abroad. Because of the limited nature of the tenders there was generally little interest and long delays were generally experienced in delivery because of the landlocked nature of the country. - 32 - Consultants 3.23 The recruitment of a consulting firm on terms and conditions satis- factory to the Association was a condition in the Credit Agreement. On December 28, 1973, a contract for consulting services was signed by the NZIDB and Nippon Koei Company Ltd. (Japan). The main tasks of the consultants were: (i) to prepare topographical maps of the project area; (ii) to provide engi- neering services for the project including planning, design and construction supervision and the preparation of tender documents; (iii) the provision of administrative and agricultural advisory services; (iv) the preparation and implementation of a training program; and (v) the preparation of feasibility studies for other projects (para 3.19). Approximately 400 manmonths of con- sulting services for the Stage I project have been provided. Of these, 365 man months have been financed out of Credit 373 while another 35 manmonths have been financed from Credit 856. Approximately 25 manmonths have been used for agricultural services and 3.5 man months for accounting while the balance of more than 370 manmonths have been used for engineering services. The initial input of consultancy services in the project was considerable; during the first two years, from December 1973 through December 1975 approxi- mately 250 man months have been used. The consultants' input during subse- quent years was considerably less as the Nepalese staff on the project was gradually able to take on more work. The services provided by the consultants have generally been satisfactory, except that little formal training was carried out. Also, the impact of the consultants on the agricultural develop- ment in the project area has been minor as the number of manmonths allocated for this purpose was negligible compared to the marmonths spent on engineering. On the basis of their experience with the project the consultants and IDA staff agree that NZIDB would still need assistance mainly because of the limited number of experienced Nepalese staff available. Such assistance would be required for work scheduling and project budgeting, staff planning, tendering and evaluation, adjustment of design and construction schedules, commissioning and maintenance of equipment, preparation of other than standard designs, testing of soils and concrete, quality control, preparation of water distribution plans, and the monitoring of irrigation requirements, canal losses, and the collection and analysis of data on-farm production. Cost, Disbursements and Financing 3.24 The original project was to command 31,400 ha and was estimated to cost US$10.1 million equivalent (NRS 102.4 M, see Table 1). The cost share of the revised Stage I project only, commanding 18,700 ha, would have been approximately US$6.8 M. However, at completion, the Stage I project will actually cost approximately US$11.0 M (Tables 1 and 4). 1/ Thus, a cost overrun of approximately 60% in dollar terms 2/ has been experienced over and above the (revised) appraisal cost estimate. The appraisal cost estimate 1/ This includes an estimated US$0.25M for on-farm works, for which there was no project expenditure. The latest estimates, received from project management after completion of this report, result in a total of US$10.7 M. Within this on-farm works amourt to US$0.75 M, which is, however, believed to be overestimated. 2/ About 90% in local currency. - 33 - could have been updated at the time of credit negotiations, since detailed designs and estimates had by then become available for Block I, which showed that construction costs had already increased considerably. If this had been done, the cost overrun could have been reduced to some extent. Also, price contingencies entered in the appraisal cost tables were actually much lower than implicit in the assumed local and foreign inflation rates (5% and 7% per annum, respectively). It appears that at appraisal escalation factors were applied only from FY 75 onwards. Physical contingencies (about 10% of all cost except for land aquisition and on-farm development) were set at a low level in the appraisal report, given the preliminary nature of the designs at the time; 20% would have been more appropriate. The most important factors contributing to the cost overrun, however, were a delay of about three years until construction actually started (institutional start-up problems and tendering difficulties), the realignment of exchange rates with the Nepalese Rupee after the dollar devaluation in 1973, general inflation after the petroleum crisis, 1/ and finally the overall slow progress in construction because of which the Stage I project could only be completed in mid-1981, while the original project (31,400 ha) had been scheduled for completion in 1978. A contributing factor to the cost overrun has also been the increase in the number of anticipated structures to be built under the project from about 1,200 at the time of appraisal to an actual number of about 3,800. Disburse- ments of the credit have been much slower than anticipated (Table 7). 3.25 Financing. Due to the cost overrun, HMGN had to fihbnce a larger part of total cost than estimated at appraisal (about 44% vs. 31%). IDA financed about 55% of total cost, whereas farmers contributed about 2% of the total (estimated) through their own labor. A split-up between local and foreign exchange cost has not been attempted, due to an uncertain proportion of implicit foreign exchange cost in civil works. Total foreign exchange cost, however, has probably not exceeded 35% of total project cost. % of total cost Actual Appraisal IDA - US$ 5.96 M 54 59 HMGN - US$ 4.79 M 44 31 Farmers - US$ 0.25 M 2 10 Total - US$11.00 M 100 100 1/ It is estimated that the price index for "development expenditure" in Nepal rose by about 150% from 1973/74 to 1980/81. See Table 5. - 34 - IV. AGRICULTURAL OIPACT Impact Area 4.01 As mentioned in para 3.03, the project scope was considerably reduced in 1977: instead of 31,400 ha, the Stage I project encompasses only about 18,700 ha. For the purpose of evaluating the agricultural impact of the project, this reduced size is taken as the basis of comparison with agricultural production targets as indicated in the appraisal report. The targets were prorated to reflect the smaller area. In the case of tubewells, an additional reduction in impact area is taken into account, since the wells are actually not irrigating 2,700 ha as envisaged at appraisal, but not more than about 500 ha due to technical failures. Preproject Cropping Patterns, Yields and Production 4.02 Preproject conditions prevailed, for all practical purposes, until the year 1976/77, i.e. crops were mostly rainfed in the project area. For this report, estimates of preproject cropping patterns and yields have been slightly revised relative to appraisal estimates, based on additional informa- tion available since then. New estimates are shown in Table 8. 1/ Overall cropping intensity was around 140% on a net cultivated area of about 17,900 ha. In the wet season, about 85% of the area was under paddy, mostly under local varieties, but also including some high yielding varieties (HYVs) recently introduced in the area. Less than 30% of the surface irrigation area may have received some supplementary irrigation in this season, mostly from the Tilawe System. In the dry season, wheat, pulses, oilseeds and vegetables were grown on about 50% of the area. No irrigation was available in this season. In addition, some mostly rainfed sugarcane was cultivated, more so in the tubewell area than in the surface water area. Yields were generally low and virtually stagnant: rainfed paddy produced about 1.4 t/ha for local varieties, and about 1.9 t/ha for HYVs. Wheat yields were about 1.1 t/ha. Sugarcane yields had been particularly poor, about 22 t/ha on average. Overall production of food grains was about 36,000 tons, oilseeds about 450 tons, vegetables about 3,200 tons, and sugarcane about 18,100 tons. The value of preproject production, with the revised estimates, is only slightly higher than the value given in the SAR, when using the same prices (Table 10). Future Without Project 4.03 Assumptions in the appraisal report on availability of irrigation in the future without project are unclear. It projected a small increase in paddy and wheat areas, and a 1% per year increase in yields. Presumably that implied a negligible impact of any potential irrigation water supply through the rudimentary NEC system as built by India. This is also the basis of evaluation taken in this report. Whereas no increase in cropping intensity is assumed in the future without project, a moderate shift from local paddy 1/ For a comparison with appraisal estimates of cropping pattern and yields refer to Table 9. - 35 - varieties to HYV was nevertheless expected to occur, in continuation of a past development. Some yield increases would have occurred as a result of improved cultural practices through continuing extension efforts. These increases are assumed as roughly 10% by 1985/86 for most crops. "Without-project" yield levels are generally in agreement with appraisal projections. Projected Cropping Patterns and Yields at Full Development 1/ 4.04 It is now expected that the project may reach full development in about 1985/86. At the time of the PCR mission, most of the cropped area during the wet season was irrigated, while only half of the same area received irrigation water during the preceding dry season, because construction of irrigation facilities was still ongoing at that time. Even under these conditions, the total cropping intensity in both seasons appeared to have already reached 172% (surface irrigation scheme), as against 170% projected in the SAR for full development. This high cropping intensity is attributable to the availability of assured perennial irrigation water, thereby enabling some advanced farmers to pass from double cropping to triple cropping. A sample survey conducted by the Integrated Cereal Project (USAID) during 1976-78 on the impact of irrigation on the change of cropping pattern in the project area indicates that about 8% of the farmers have adopted a triple cropping system, paddy-paddy-wheat (see table below). This practice, however, requires a strict adherence to the cropping calendar, careful selection of suitable (short duration) varieties and skilled farm management to cope with labor shortages likely to occur during the peak period of land preparation and harvest. One setback farmers could face under this system is the yield of the third crop, wheat, which is reduced when the optimum time of sowing (mid- November) is missed. Also, the second paddy crop could not receive water for irrigation when the canal system is closed for maintenance (generally in November). Moreover, as there is also a canal closure period in March/April it may be better to encourage the farmers to concentrate on one paddy crop only, with the canal operating reliably from the end of April to allow trans- planting to take place over a six-week period from beginning of June-mid-July with harvesting from mid-September through October. This would enable a month to elapse before sowing of rabi wheat at its optimum time in mid-November. In any event, a triple cropping system in excess of about 30% of the area would create irrigation water shortage at full development of stages II and III of the project. Thus, the implementation of programmed water allocations should be strictly observed. Distribution of Crop Sequences Among Farmers (%) Cropping Before After Sequence Irrigation Irrigation Paddy-Wheat 49 42 Paddy-Other Crops 27 26 Paddy-Fallow 22 18 Paddy-Paddy-Wheat 0 8 Perennial 2 6 Total 100 100 1/ Refer to Tables 8 and 11. - 36 - 4.05 Taking into account the availability of water, farmers' attitude toward an intensified cropping system and strengthened support services, it can be expected that the cropping intensity in the surface water area will reach about 180%, which is about 10% higher than the projection made during appraisal. In the groundwater scheme, about 190% would be attained in the future with-project situation (mid 1980s). Yields 4.06 Yields of paddy and wheat projected at the time of appraisal were somewhat higher than those now visualized; for other crops the yield levels are well in line with the SAR. 1/ Yields for HYV paddy and local paddy, both irrigated with canal and groundwater, are now projected at 3.2 and 2.3 t/ha respectively, at full development (mid 1980s), compared to the estimates in the SAR of 3.5 t/ha for HYV and 2.7 t/ha for local paddy (see Table below). At the time of appraisal, the HYVs had just been introduced (Parwanipur-1 and Masuli were released in 1973; IR 20 and IR 22 were recommended for cultivation in 1972) for the Terai regions and much optimism about their yields prevailed. The projected yields of paddy are based on increased fertilizer application, controlled water management and development of varieties even better suited to the project area. Yields of Major Crops (at full development) SAR New Estimate -----------(t/ha)------------- Irrigated HYV Paddy 3.5 3.2 Local Paddy 2.7 2.3 Vegetables 12.0 12.0 Wheat 3.1 2.6 /a Oilseeds 0.7 0.8 Pulses 0.9 0.9 Maize - 2.0 Sugarcane 40.0 40.0 Residual Moisture oilseeds - 0.5 Pulses 0.6 /a In some areas where a high watertable has developed or may develop project management estimates now that yields.of wheat may be lower. On the other hand, project management feels that maize yields may be substantially higher. 1/ See Table 12 for assumption on yields. - 37 - Use of Modern Agricultural Inputs 4.07 No survey or other data are presently available to assess the project's impact on farmers' use of modern production inputs, except for seeds. The only indication of the development of fertilizer use in the area is given by sales statistics of the Birganj Branch of the Agricultural Input Corporation (AIC), which is the sole distributors of fertilizers in Nepal. According to their data (see Table 13), AIC distributed to cooperative and farmers in Bara and Parsa districts 1/ about 40% more fertilizers (in nutrient terms) in 1979/80 than in the preproject period 1974/75-1976/77. It can be assumed that a large part of the increase of about 670 t (nutrients) went actually to the project area, since here the incentives to use it were clearly much greater than in the other (mostly rainfed) areas of the two districts. Most of the fertilizers (75%) consist of nitrogen, whose sales also have the highest growth rate of the three major nutrients. In general, there seems to be no reluctance on the farmers' side to apply increasing amounts of fertilizers; the problems are rather to be found with the national supply situation (all fertilizers are imported, imports are erratic and often untimely), with distribution to the'villages and with the short term credit system. (paras 6.13-6.21). It is recommended that a future survey of actual achievements include an investigation of (a) crops actually fertilized, (b) the proportion of these crops in the cropping pattern, (c) the level of fertilizer application on crops, and (d) farmers' perception of optimal levels of application and of constraints. 4.08 The use of plant protection chemicals is still low (less than US$1.0 per ha). A correct figure is difficult to obtain, since in addition to AIC the private sector is also allowed to sell these inputs. Most of the chemicals are applied to paddy for seed treatment and to control occa- sional outbreaks of paddy diseases. The adoption of improved seeds has made rather slow progress, mainly due to supply problems (wheat) and, in case of paddy probably due to a lack of farmers' awareness of the need to periodically renew their paddy seed. The sale of maize seeds has grown rapidly in recent years. Distribution of improved seeds through AIC from 1975/76 to 1978/80 to the project area was as follows: 2/ Paddy Wheat Maize (t) (t) (t) 1975/76 16 70 0.5 1976/77 33 60 2 1977/78 21 52 4 1978/79 20 90 4 1979/80 23 110 10 1/ The surface water scheme is located in Parsa district, the tubewell scheme in Bara district. Both districts have a net cultivated area of about 104,000 ha of which the project covers 18,700 ha. 2/ Some improved seeds find their way from seed growers also directly to farmers. - 38 - The total improved seed requirement for cereals, assuming a one in three year replenishment, is estimated at about 470 t per year in the future, necessitating a considerable step-up in seed production, distribution, and quality control. The estimated use of inputs, "with" and "without project", is shown in Table 14. Incremental Production from the Project 4.09 As mentioned in para 4.03, incremental production is estimated relative to a "without-project" situation which does not include any irri- gation releases from the incomplete NEC system. Apart from slight yield increases, production without the project would therefore not have changed very much from the preproject production. Projected production "with" and "without project" for individual crops, is presented in Table 10 and summarized below: 1/ Production Estimates for Surface Water Scheme - Appraisal vs PCR ('000 tons) Future Without Full Project Development Incremental SAR PCR SAR PCR SAR PCR Cereals 30.2 31.7 70.2 64.9 40.0 33.2 Pulses 0.4 1.0 0.7 1.8 0.3 0.8 Oilseeds 0.3 0.4 0.6 1.1 0.3 0.7 Vegetables 4.8 3.0 9.6 7.7 4.8 4.7 Sugarcane 14.6 13.6 63.9 44.7 49.3 31.1 The total value of production, taken at prices used in the SAR, would have reached NRs 27.5 M in the future without project, and is now estimated to reach about NRs 56.5 at full development, an increase of about 105%. This compares to appraisal estimates of NRs 25.9 M and NRs 62.2 M, for "without" and with project, respectively, implying an increase of about 140%. The value of incremental production (NRs 29.OM) thus falls short of what was expected at appraisal (NRs 36.2) by about 20%; however, this is more than compensated for by the increase in crop prices that occurred between appraisal time and present. The shortfall is due to a somewhat higher "without project" cropping intensity and (iii) lower yields at full development. 4.10 Over recent years, irrigated wheat yields have averaged about 1.8 t/ha, whereas the SAR estimated that under conditions of full development irrigated yields could attain a level of 3.1 t/ha. Unless farmers apply adequate amounts of fertilizers and adhere strictly to an early planting date crucial to increase the yield of wheat, the SAR projections will be difficult to attain. A revised estimate of about 2.6 t/ha, seems a more realistically attainable figure. 2/ 1/ A comparison for the tubewell scheme was omitted, due to the uncertainty regarding actually irrigated areas, cropping patterns and yields. It is estimated that only about 20-30% of expected incremental production was achieved. 2/ See also footnote to table in para 4.06. - 39 - 4.11 At the time of appraisal, maize was not considered in the projected cropping pattern. However, since the introduction of new promising varieties in the project area, the area planted to maize is rapidly increasing, specially in the tubewell irrigated area. Although maize is a relatively new crop, a yield of about 1.8 t/ha obtained by farmers at present would increase to about 2.0 t/ha at full development. Employment and Equity 4.12 Employment. In the absence of an in-depth study focussing on changes brought about in the local farm and non-farm economy as a result of the project, reliable estimates of the project's impact on employment cannot be given. Such analysis would have to differentiate between em- ployment generation for farm family labor according to farmsize, both on the family's own farm and on other farms, for landless laborers and for migrant labor both from Nepal's hills and from India. From the present and projected cropping patterns, with and without project, and the cor- responding crop labor requirements, one may conclude, however, that over- all farm employment will have increased by about 40% at full development relative to the without project situation, from about 2.45 M to about 3.45 M mandays. (This compares to an appraisal estimate of an increase of about 75%, from 2.8 M to 4.9 M mandays, when adjusted for the reduced project scope.) The full development labor requirement will still be far from exhausting the estimated farm family labor force of about 11 M mandays. Hired labor nevertheless plays an important role in the local farming economy, (i) to meet peak labor demand mainly for transplanting and harvesting, (ii) throughout the crop season on larger farms, say larger than about 3 ha, and (iii) even on smaller farms presumably due to reasons related to religion and caste. The project authorities stated that no specific wage increase was observed for the project area as a consequence of the project. This seems plausible in view of the avail- ability of a substantial pool of underemployed family labor and of land- less laborers both in Nepal and in the border areas of India. (Actual remuneration of hired labor is mainly in kind, either in the form of small tenancy rights for "ploughmen" and "spademen", or in the form of a certain share in the crop harvested, or as a fixed amount of cereals plus a meal (presently about 3 kg of paddy per day).) In addition to direct farm employment there is probably a substantial increase in related activities and employment (transport, milling, processing), part of which, however, may be carried out by the farm households themselves. No quantification of this effect was attempted. 4.13 Equity. The appraisal report did not state any specific equity objective except that practically every farm would be better off under the project. This latter statement is probably realistic. Some qualitative conclusions on the project's equity impact can, however, be drawn from the nature of the project in combination with the prevailing socio-economic characteristics of local agriculture. First, the project was designed to allocate water to all farmers in the command in an available manner, i.e. in equal amounts per unit area. It is a matter of good water management and system maintenance to which degree this will actually be achieved. The - 40 - establishment of Water Users Groups for service units and subunits (para 6.04) may help a more equitable distribution of water within these units. (That some "tailender" problems may still arise at certain times cannot be fully avoided.) By definition, however, an even distribution of water reinforces the already present structure of landholdings: larger farms receive more water. While such practice may appear neutral with respect to equity, and furthermore, the only technical and politically feasible way of water alloca- tion, there is a presumption that larger farmers can reap proportionately more benefits per unit of water than smaller ones, due to easier access to credit (both short-terms and other) from all sources and to other services, and due to the nature of prevailing tenancy relations. About 54% of farmers in the project area are working as either pure tenants (18%) or as owner-cum-tenants (36%). Most of the tenants are to be found in the smaller farmsize classes, and most of the tenancy arrangements provide for sharecropping of the main crop (50:50 is customary and the maximum allowed by the law). 1/ Since input costs are not shared, landlords collect an even larger share of the net returns from crop production on tenants' land. While this is already a disincentive for the small tenant farmer (research in neighboring Bihar, India, has shown that sharecropped plots are less intensively farmed), he is also in a disadvantage vis-a-vis the lending institutions where a land- owner's certificate is needed to establish collateral, quite apart from the general difficulties he experiences when dealing, often as an illiterate, with bureaucratic credit application requirements. Informal credit (from landlords and money lenders) is often easier to get, but much more costly. Equity issues raised in connection with tenancy relations are national in scope and require a strong political will for their solution. Close scrutiny should be given to the rules and procedures governing access to cheap institu- tional credit from ADBN (para 6.21), so as to offset some of the disadvantages facing the small farmer. It remains to be seen what beneficial effect the formation of Water Users Groups may have in this context, in as far as they function also as credit and input procurement intermediaries. 4.14 Environmental Impact. At the time of appraisal an increase in the incidence of malaria due to perennial irrigation was recognized as a risk. The project authorities informed IDA that at present there is no known case of malaria in the project area. The campaign for malaria eradication in the Terai was apparently successful. V. ECONOMIC AND FINANCIAL EVALUATION Economic Analysis 5.01 The economic feasibility of the project has been reassessed, taking into account the actual implementation schedule, actually incurred costs and better forecasts of full development benefits. (As to the groundwater com- ponent, the recent rehabilitation of wells has not been taken into account with its benefits and costs.) While this is still not an analysis "ex post", due to the uncertainty regarding final benefits, it covers a critical phase of the development. The assumptions made to arrive at an estimate of the rate of return are spelled out in the following paragraphs. - 41 - 5.02 Input and Output Prices. All inputs and outputs of agricultural production have been valued in constant terms, i.e. in 1980/81 Nepalese Rupees. 1/ Economic prices of major traded outputs and of fertilizers have been derived from actual and projected world market prices (in constant 1980 US Dollars), for each year between 1976/77 and 1985/86, with due adjustments for shipping, local transport, handling and processing. 2/ Dollar prices were converted into Nepalese Rupees at a constant rate of NRs 12.00/US$. 3/ Prices of pulses and vegetables (non-traded) were taken at an average market value. Farm labor was valued at NRs 7 per manday throughout, a weighted average of going peak and off-peak wages. Too little information is presently available to discern real wage increases due to the project, and to assess a possible overvaluation of labor's opportunity cost in the going wage rates due to unemployment and underemployment. While it is believed that a rate of NRs 7 per day might be a rather high valuation of labor, it can be shown that changes in the level of economic wages would not have an appreciable effect on economic viability. Animal labor was valued at the going rental rate of a pair-day, multiplied by the assumed standard conversion factor (SCF) for Nepal of 0.9. Seeds and chemicals were valued at their financial cost; and as in the cse of labor, deviations from their "true" economic cost are not critical for economic analysis. Total input cost was furthermore increased by a contingency factor of 5%. Prices used in economic analysis are presented in Table 15. 5.03 Direct Investment Cost. Actual project expenditure on civil works, force account works and equipment from 1973/74 to 1980/81 was used to derive economic project cost. In addition, estimated cost of on-farm development (mainly field channel construction by farmers), for which there is no corresponding project expenditure, was distributed over the years 1978/79 to 1982/83 in accordance with observed and projected progress of these works. The total cost of direct investment is NRs 89.5 M in current prices, of which NRs 69.0 M was for the surface irrigation scheme and NRs 20.5 for the tubewell scheme. 4/ (See Table 4), 5.04 Sunk Cost. The Staff Appraisal Report mentions a figure of US$5 million (NRs 50 M) as expenditures on the project up to appraisal time, without further details. In order to derive an economic rate of return on the project as a whole, an attempt was made to estimate the portion of sunk cost attributable to the area of the revised Birganj Irrigation Project (16,000 ha net). First, an estimated NRs 3 M was deducted as sunk cost into the existing tubewell scheme, leaving NRs 47 M for the surface scheme. To this, a 50% price increase over the years 1968/69 to 1972/73 was added, although it is not known for certain whether the figure given in the SAR and allocated to the year 1973/74 was already inflated to that date. The result- ing NRs 70.5 M were then apportioned to the Stage I (revised project) area according to its share in the length of the main canal (43%) and assuming a 1/ The Staff Appraisal Report used (presumably) a 1972 price basis. 2/ IBRD price projections have changed considerably in May 1982. Economic prices in US$ would now be lower. On the other hand, the exchange rate was changed from NRs 12.0 to NRs 13.2 per US$. 3/ Therefore, prices in Rupees reflect also a basis of constant 1980 US Dollars. 4/ Latest estimates received after completion of the report indicate that these costs were actually less. - 42 - hypothetically reduced capacity. 1/ The sunk cost for the surface scheme is thus assumed to be NRs 22M in 1973/74 prices. No interest (backward discounting) was added for the period up to 1973/74. Existing tubewells were assumed to carry a sunk cost of NRs 2.2M in 1973/74 prices, accounting for eleven wells only,and not for pumping equipment. 2/ 5.05 Project Overhead. Establishment cost, consultant services and headquarters buildings were considered to be project overhead. It amounted to NRs 44 M in current prices, or 49% over direct investment cost, higher than assumed at appraisal (34%). 5.06 Adjustment for Inflation. Since benefits are valued in constant 1980/81 Rupees (or US Dollars) at border prices, project cost streams had to be brought to this price level. It was considered more correct from a theoretical viewpoint to use the Bank's International Price index for this purpose rather than a domestic one such as the National Urban Consumers Price Index. Also, the former gives a more conservative (higher) estimate of "real" costs. It increased by about 120% between 1973/74 and 1980/81, whereas the domestic consumer price index rose only by about 70%. The inflator used for economic analysis is shown in Table 5, together with other indices for comparison. 5.07 Conversion of Financial Cost to Economic Cost. Further adjustments were made to financial project cost in order to take account of distortions in domestic prices vis-a-vis international prices due to taxes and 'subsidies, multiple foreign exchange rates and due to non-tariff restrictions. For Nepal in recent years, a standard conversion factor (SCF) of 0.9 has been estimated, equivalent to a shadow exchange rate of about NRs 13.30/US$. The SCF is applied to non-traded goods and services, in the absence of specific conversion factors. Cost of civil works in the surface irrigation scheme, which excludes departmental (force account) works, was adjusted by a conversion factor of 0.9, as a weighted average factor when assuming certain portions of traded and non-traded goods and unskilled labor in the total (see Table 16). To all other civil works, a conversion factor of 0.84 was applied, reflecting a higher content of unskilled labor. Costs of force account and on-farm works were converted by a factor of 0.68, since they contain predominantly unskilled labor employed in the slack season. Cost of equipment, virtually all imported, and of consultant services, has not been converted. To all other cost items the SCF was applied. 5.08 Separate Evaluation of Surface Scheme and Tubewell Scheme. In order to evaluate the surface water and the tubewell component separately, an allocation of overhead had to be made. In the absence of more detailed information, economic overhead cost was distributed over the two schemes in 1/ 55% of actual capacity, costing about 70% of the actual canal per km. 2/ Note that no "sunk benefit' has been entered for the existing wells. Benefits from production before rehabilitation seem to have been low at any rate. - 43 - accordance with their shares in direct investment cost (74.5% and 25.5% respectively). Force account works and the small amount of land acquisition have been allocated entirely to the surface scheme as direct investment cost. While part of the overhead, particularly in years 1978/79 and 1979/80, may have to be charged to the Stage II project, no such adjustment was made; no overhead cost dabitable to Stage I was entered in 1980/81. 5.09 Replacement Cost. An economic life of 20 years for wells drilled under the project is assumed. Pumping equipment would have to be replaced every 7 years. The re-drilling of wells would cost NRs 0.54 M per unit (NRs 0.50 for the existing smaller tubewells), and re-equipping with pumps and motors would cost NRs 0.12 M per unit (NRs 0.10 M). 5.10 Economic O&M Cost. As shown in Table 17, annual O&M cost for the surface irrigation scheme is estimated as NRs 210/ha, including NRs 45/ha for amortization of necessary equipment. Since this equipment is already included in the investment cost, a financial O&M cost of NRs 165/ha was assumed. By applying appropriate conversion factors, the corresponding economic cost is NRs 145/ha. An estimate of O&M cost for tubewells is also detailed in Table 17. This refers, however, to a tubewell operation at the projected intensity. Without further investments (redrilling, reequipment) and increased efforts towards higher performance standards, the originally projected inten- sity will not be achieved. Actual O&M cost has therefore been scaled down by a factor proportional to the estimated shortfall in pumping hours. The eco- nomic cost of power was estimated in accordance with an analysis of such cost made for the Bhairawa-Lumbini Groundwater Project, Stage II. 5.11 Economic Benefits. The economic benefit streams estimated for the surface water component are shown in Table 18. They reflect, as approximately as presently possible, the actually irrigated areas and yields for past years and updated projections for the future. For the tubewell scheme, less com- plete and reliable information on irrigated areas is available; a summary of some performance indicators is given in Table 21. For simplicity, and since the result of the economic analysis would be unsatisfactory at any rate for this component, a full-scale performance was projected for this component and was then scaled down by factors approximating the ratio of actual to full irrigation intensity. It was assumed that the tubewell scheme would generate only about 22% of the incremental benefits that would have occurred under a full performance scenario. 5.12 Economic Viability. Economic rates of return (ERR) were calculated over a project horizon of 40 years, under the abovementioned assumptions, with the following results: 1/ Without Sunk Cost With Sunk Cost Surface Water Scheme 26% 19% Tubewell Schemes negative negative Entire Project 21% 16% A sensitivity analysis for the surface scheme is given in Table 19. 1/ If current prices had been used, the ERRs would have been approximately 30-40% higher. -. 44- Comparison with Appraisal Estimates 5.13 The ERR for the original project scope was estimated at 21% at appraisal, and at 14% when sunk cost was included. As to components, the surface irrigation scheme was evaluated at 24% and the tubewell component at 14% 1/. A rough re-evaluation of the project in its reduced scope in 1977 also gave an ERR of 18-20%. In comparison, therefore, the project as a whole has performed according to expectations. This has come about, however, despite considerable cost overruns, and due to substantial shifts in the relative value of project benefits, since increases in economic output prices (in 1980/81 currency) were able to compensate for a substantial benefit shortfall (25-30%), a slower benefit build-up and for the cost over- run. Unfortunately, the SAR does not give detailed assumptions on the derivation of economic costs and benefits, so that an analysis of the effects of changes in relative prices and of deviations of benefits in real terms from those originally projected are difficult. It also appears, in compari- son, that the surface scheme performed even better than projected, whereas the groundwater component remains obviously far short of expectations. While the rehabilitation of failed or non-operable wells recently undertaken, together with further efforts to strengthen supervision, operation and mainte- nance of the groundwater scheme, appear certainly worthwhile, it is not clear whether a higher overall performance in the future would raise the ERR of this component over its entire life, i.e. including past years with their costs and benefits. As mentioned earlier, such a more optimistic scenario for the future was not evaluated. Impact on Farm Budgets 5.14 A survey of farm households to evaluate the impact of the project on the budgets of benefitting farm families is not available at this stage. Also, monitoring data do not include any information leading to an assessment of the changes in income and wealth/indebtedness of the beneficiaries. It is strongly recommended that an appropriate survey of such a nature be undertaken as soon as the project reaches full development, which should include data on a control group of non-benefitting farmers outside the project area. 5.15 In order to obtain an approximate estimate of changes in income of farm households, average cropping patterns, crop production and farm in- puts with and without project at the prospective time of full development were evaluated at financial farmgate prices prevailing in 1980/81. No attempt was made to distinguish between different farm sizes. Instead, a representa- tive farm of 1.0 ha in the surface irrigation area was assumed. 2/ (Detailed calculations in Table 20.) As a result, it is estimated that incomes from farming (crop production) will be about 130% higher than what they would have been without the project 3/, more than estimated at appraisal (para 2.16). 1/ 14% and 11% respectively, when sunk cost was included. 2/ The average farm size is very close to 1 ha. 3/ NRs 7,180 vs. NRs 3,120 "without project." - 45 - This does not include potential effects on livestock production and on off- farm employment (which may be reduced due to higher on-farm family labor use). It is not clear whether the increase per unit area would be higher or lower for smaller farm units than for larger ones. Small farms tend to employ proportionately more family labor and may thus incur less labor costs. Empirical surveys elsewhere also show that in dryland farming small farms generally show higher cropping intensities. The basically even water allocation (equal volumes per net ha), however, would not suggest widely differing irrigation intensities. The increase in net benefits per ha may therefore be less for small farmers. Moreover, as indicated in the discus- sion of equity aspects (para 4.12), small farmers may be at a disadvantage concerning access to credit, inputs and services. In as far as they operate plots under sharecropping or other tenancy arrangements, they will have to transfer a major part of the net return to landlords. In summary, it is felt that increases in incomes may not be "equitably" distributed over the entire farming population. VI. INSTITUTIONAL.PERFORMANCE AND DEVELOPMENT The Narayani Zone Irrigation Development Board 6.01 For purposes of project implementation HMGN established in early 1973 the NZIDB, to be responsible for the execution of the project and for its operation and maintenance. The main purpose for setting up this semi- autonomous agency was to ensure proper coordination between the various ministries involved in the implementation of the project. The rules of the Board were approved by IDA in 1974. The NZIDB's functions include inter alia: (a) approval of annual budget proposals for executing the project; (b) appoint- ment of senior staff; (c) approval of contracts for materials over NRs 100,000 and civil works contracts over NRs 2 M and consultants' agreements; (d) semi- annual reviews of implementation programs and progress; and (e) coordination of policy matters in programming, budgeting and finance. The powers of the NZIDB were vested in an eight member board of directors including the Secretary of the Ministry of Food, Agriculture and Irrigation as Chairman, and as members respresentatives of the Ministry of Water and Power, the Ministry of Land Reform, the National Planning Commission Secretariat, the Department of Agriculture, the Department of Irrigation, Hydrology and Meteorology, and as Secretary the General Manager of the Narayani Zone Irrigation Project. The organization of the Board has basically been implemented as originally antici- pated. The field organization is controlled by the General Manager who has four divisions under him: an Engineering Division, an Agricultural Division, an Administrative Division and a Finance Division. 6.02 Over the years NZIDB has become increasingly effective in carrying out the project which was in large measure due to improvements and to the continuity in management since 1976. The project organization has nevertheless been subject to certain bureaucratic constraints. This resulted in relatively slow award and signing of contracts although procurement has no doubt been faster than would have been possible if it had been undertaken within the - 46 - Ministry. The main constraint, however, (properly identified during project appraisal), has been the shortage of experienced engineering and agricultural staff. In practice, all staff working on the project have been seconded from the DIHM and the Department of Agriculture and staff appointments and promo- tions had to take place within the rules of the HMGN public service commission and this limited salaries and other incentives that would be required to attract competent staff. As a result, staff have not been adequately motivated to work, mainly because of the lack of financial incentives, even with the 30% project allowance that has been paid. Productivity also generally dropped when work was not supervised by the consultants. Typically, in 1978, the engineering staff under the General Management consisted of two Executive Engineers, 14 Assistant Engineers and 30 Overseers, not only for Stage I but also for Stage II implementation, a number which is barely adequate. This number has recently deteriorated and at present there are only one Executive Engineer and 12 Assistant Engineers. Recently, the General Manager and Senior Agricultural Officer have been replaced due to promotion. All major irriga- tion projects under implementation in the Terai are now being governed by separate development boards. It should be possible for HMGN to consider the unification of control over these projects under one Terai Development Board. Also, the Agricultural Division was generally inadequately staffed. Until recently, there was a lack of agricultural subject matter specialists. The required number has now been recruited but the staff posted are young graduates without the necessary experience. The continuing success of the project will much depend on the ability of HMGN to provide the necessary engineering and agricultural staff for the operation of the project. Coordination of Agricultural Development Activities 6.03 During appraisal no special arrangements had been made for the coordination of agricultural development activities on the project level which would involve coordination of research and extension, input supply, agricultural credit and marketing. In theory such coordination should be realized through existing district and regional coordination committees. However, in practice these committees involve themselves only with policy matters and are not really designed to meet the day to day requirements of a specific project. Therefore, informal meetings have been arranged by the Senior Agricultural Officer of the project, about five to six times per year, when he meets with representatives of the ADBN, the AIC, the Rice Export Company and the District Officer who represents the Sajhas (local cooperatives). Water Users Groups 6.04 As was intended during appraisal, the farmers in the project area have been organized in water users groups (WUGs) for the purpose of construct- ing and maintaining field channels as well as for water distribution within the farm units. The formation of such WUGs in the project area has been completed for about 95%. Each WUG is headed by a leader, while every 7.5 ha sub-unit is headed by a sub-leader. The organization of the farmers in WUGs was carried out by the cooperative section of the NZIDB which is headed by a cooperative officer who has a staff of six cooperative assistants, one - 47 - for each block. For the purpose of obtaining the farmers cooperation more than 3,500 pamphlets were distributed among the farmers. All landholdings located within each service or sub-service unit have been registered and the leaders have been given detailed job descriptions which include inter alia: (i) the distribution of water according to the advice of the engineering and agricultural divisions; (ii) the organization of the farmers for constructing field channels and for the introduction of improved agricultural practices and for arranging the maintenance of field channels. Project's overseers (for O&M) will provide technical assistance to the leaders in water management. Agricultural Supporting Services 6.05 Agricultural Research. The National Rice Improvement Program at the Parwanipur Agricultural Station near Birganj is actively assisting the dissemination of improved cultural practices of paddy, testing better cropping systems in farmers' fields and training extension workers and farmers. In a coordinated effort with the project, they set up field trials to demonstrate the performance of different paddy varieties to extension workers and farmers. The station has directly assisted the project by: (i) training extension workers on rice growing and on rice based cropping systems; (ii) providing a rice minikit to the farmers in the project area through extension workers. The minikit served as the medium to disseminate the improved rice varieties in the farmer's field. Each packet contains improved seed and information regarding varietal characters and planting instruction. The coordination between the station and the project office is well maintained. The project provided the station with about Rs 270,000 to .pave the 1.5 km access road from the Birganj highway. 6.06 At about the time of appraisal, the station released two promising paddy varieties, Parwanipur-1 and Masuli. The former is known for its short duration (120 days) with high yield, and the latter for its fine grain quality but long duration (165 days). As the farmers in the area intensified their cropping systems after receiving irrigation water, Parwanipur-1 was rapidly disseminated to suit the double cropping system. Unfortunately, the variety possesses some undesirable characteristics, namely non-shattering and pre- harvest grain sprouting, which may cause damage to the standing grain and reduce yields. 6.07 Recently, the station released four new varieties which are high yielding, of short duration and moderately resistant to diseases and insects. Among these, Laxmi is gaining in popularity among farmers for its high yield. This variety is likely to replace the existing Parwanipur-1 for intensified cropping systems. 6.08 Under the project, the research program at Parwanipur was to include development of an efficient water management scheme for the area and testing of the profitability of alternative crops such as sugarbeet, cotton, ground- nuts and pulses. As no funds and staff were provided for such testings, the station conducted only limited irrigated paddy trials and small scale sugarbeet varietal trials. The practicability of growing other crops was not assessed. - 48 - 6.09 Considerable research progress has been made under the ongoing Integrated Cereals Project (ICP) assisted by USAID. The program, initiated in 1977, has shown the benefits of a coordinated approach to paddy, wheat and maize research and production. During the last four years, paddy-based cropping system trials have been carried out under the ICP in block IB, in coordination with project staff. Trials have just been completed and the findings are to be demonstrated at a tertiary unit in each of the six blocks in the project area jointly with the project staff. This so-called "compact production program" is scheduled for three years; funds and staff would be provided by USAID. 6.10 In view of low sugarcane yields due to relatively low winter temper- atures and non-availability of suitable varieties available to cope with them, the SAR specifically suggested that some sugarbeet varietal trials be conducted as a possible substitute for sugarcane. Accordingly, the Jitpur Sugarcane Research Farm located about 20 km from Birganj conducted small scale varietal trials. Because of difficulties involved in collecting germplasm and obtaining technical information, the trials were discontinued after two years without success. No follow-up trials are being considered at this time. The Farm currently concentrates its efforts on varietal trials, improved cultural practices including ploughing depth, proper time of planting and fertilizer application. 6.11 The IDA-assisted Agricultural Extension and Research Project (Cr. 1100-NEP) would upgrade research capabilities by providing the Parwanipur Agricultural Station with farm facilities, machinery, staff housing and additional research staff. Agricultural Extension 6.12 At the time of appraisal, there was only one extension worker for every 2,000 farm households, and the extension effort itself was weak because of inadequate training and poor logistic supports. With the assistance of a Bank consultant in 1975, the then existing extension services in the project area have been changed to the Training and Visit (T&V) system. It was the first instance of this system to be introduced in Nepal as a pilot scheme. Originally, it was planned to confine the T&V system to the project area only, but later it was extended to the entire Bara and Paraa districts upon the suggestions of a 1975 IDA supervision mission. 6.13 Although the system experienced some setbacks due to the slowness in recruiting the required staff, the extension activities have progressed satisfactorily. An evaluation of the pilot scheme carried out by the DA indicates that over 90% of the farmers interviewed felt the program has helped to increase production. Under the scheme, the extension service was reorganized and strengthened. The number of field staff was increased, enabling each extension worker to attend about 400 farm families to carry out the regular fortnightly training for Panchayat Level Agricultural Assistants (PLAAs), and to visit contact farmers on a fixed day of the week. The exten- sion program in the area is administered through six sub-centers, a Junior Technician (JT) is in charge of each. The project area now has a relatively - 49 - sound, well-directed extension effort. The scheme has proved to be effective in allowing each extension worker to make more frequent contacts with farmers and concentrate on solving the farmers' problems. 6.14 Encouraged by the favorable results obtained from the pilot scheme, HMGN has further extended the T&V aystem to four additional districts, cov- ering three ongoing IDA assisted projects in the Terai. This program is financed from the Narayani Stage II credit. 6.15 Bottlenecks impeding effective extension work are lack of competent Subject Matter Specialists (SMSs) and frequent transfer of DADOs. Recruitment of the lower level staff was completed by 1978 (55 PLAAS, 14 JTAa, and 6 JTs) but the appointment of the SMSs had been considerably delayed. Of the five SMSs sanctioned for the Narayani Stage I and II, only two had been positioned in 1979 and the remaining in 1980. As these SMSs are recent agricultural graduates, they lack practical farm experience. Although the extension acti- vities are in full swing at present, it seems that it will take some more time to upgrade the quality of training given by SMSs. Another serious constraint which weakened routine extension activities is the frequent transfer of District Agricultural Development Officers (DADOs). Since the inception of the project, there were about six changes of DADOs. Often the position is vacant due to extended period of study leave taken by DADOs, and no replacement is provided. Training 6.16 The first pre-service training for PLAAs to be assigned to the project area under the T&V system was carried out at the Parwanipur Agricul- tural Station in 1975. Since 1975 a total of about 190 PLAAs attended one week inservice training courses on rice. In addition to the regular fortnighly training under the T&V system some selected PLAAs were sent for a one-month course to Janakpur Agricultural Station to acquire farming techniques for other cereal crops. A special one-week in-service training for JTs and JTAs was also carried out as yearly routine activity at Parwanipur for rice; at the Bhairawa Agricultural Station for wheat; at the Rampur Agricultural Station for maize and at the Jitpur Sugarcane Farm for sugarcane cultivation techniques. About 20-30 JTs/JTAs attended each training. 6.17 The quality of training has been generally inadequate, mainly due to the lack of experienced instructors and poor facilities. As the IDA-assisted Nepal Agricultural Extension and Research Project (Cr. 1100) will provide the necessary training facilities and place emphasis on upgrading the quality of trainers, most of the constraints are expected to be removed. 6.18 Construction of the six sub-training centers for the project area was not planned under the project. The Narayani Zone Irrigation Development Stage II Project (Cr. 856-NEP) provided for their inclusion. Presently, two sub-centers are being completed, one is under construction and three would be constructed by early 1982. As an interim measure, the training of PLAA and farmers is being carried out at rented facilities. - 50 - Agricultural Input Supply and Credit 6.19 Input Supply. Procurement and wholesale distribution of fertilizers in Nepal is a monopoly of the Agricultural Input Corporation (AIC). All fer- tilizers are imported, about 50% by means of foreign aid (grants). Birganj is the major entry point. AIC has a regional office and a main branch office here, from where fertilizers, seeds and chemicals are supplied to Parsa, Bara and Rautahat districts, hence also to the project area. This vicinity is an advantage for the project. AIC has large storage facilities for the three districts (11,000 tons existing, 5000 tons under construction) in Birganj, more than sufficient to accommodate supplies for one year at present. Nationally, but particularly in the central Terai and the Narayani Zone, AIC has an impressive record of growth in fertilizer sales. Its management is satisfactory considering the difficult (geographical) conditions. Th main problems at the wholesale level are: - the insecurity and erratic arrivals of shipments from abroad - difficult and costly trans-shipment from Calcutta to the border at Birganj - insufficiency of the truck fleet owned by AIC - insufficient supply in view of demand - passive sales attitude, no sales promotion FAO has recently prepared a Crop Intensification Program for ADB financing, 1/ under which certain bottlenecks in agricultural support services would be addressed in four districts (Parsa, Bara, Rautahat, Kailali). Inter alia, the program would provide for import and distribution of 6,200 t of additional fertilizers to Parsa and Bara districts over three years, beginning 1982/83. No provision is made to strengthen AIC directly, pending the outcome of a study on this subject. 6.20 Input Distribution. Serious constraints exist for input distribution, especially of fertilizers, from AIC to farmers. While farmers may occasionally buy their requirements directly from AIC, most of the distribution is handled by the District Cooperative Union as an intermediary and its affiliated Village Cooperative Societies (Sajhas). Sajhas function also as retailers for basic consumer goods and as marketing agents. 2/ More than 80% of their turnover, however, is made up of input sales, of which 90-95% con- sists of fertilizers. In addition, Sajhas are intermediaries for loans be- tween the Agricultural Development Bank of Nepal (ADBN) and member farmers. Previously ADBN was responsible for direct management of cooperatives. In 1978, however, management was turned over to respective village panchayats 1/ FAO Investment Centre, Report of the Nepal Crop Intensification Program Preparation Mission, Report No. 7/81 DDC, NEP-16, February 1981. This report gives a good analysis of the input and credit problems. 2/ There are also five private dealers retailing in seeds, pesticides and organic manure. - 51 - to create more participation of villagers in the affairs of their coopera- tive. It is a commonly accepted fact that management has deteriorated con- siderably since then, both financially and in terms of business efficiency, since appointed managers are generally not well prepared for their job, and local politics tend to interfere with the cooperative's business. Timely and sufficient supply of imports to farmers suffer under this constraint. Efforts are, however, made by HMGN's Cooperative Department to impart more and better training to local managers. Management problems are compounded by - lack of transport facilities, at least in peak periods. (Most fertilizers are transported by trucks, on often difficult roads during the wet season.) This is, however, less a problem in the project area than elsewhere, since dis- tances are short and the road network is better. - loading capacity at AIC godown - increasing ineligibility of Sajhas for ADBN loans, which cuts the membership off from the source of institutional short term credit and restricts demand for inputs. There are 9 Sajhas in the project area, with a membership of about 23,000. Less than half of the members participate actively, e.g., through input purchases, less than 15% apply for, and less than 10% receive loans. 6.21 Credit. ADBN is virtually the only source of institutional credit to farmers. It provides short, medium and long term loans to indivi- dual farmers, sajhas and recently also to Water Users Groups or other "joint borrowers". Interest rates for short term (crop) loans are 14% for farmers and cooperatives. A cooperative becomes ineligible for loans if its over- dues exceed 50% of its outstanding loans, which may exclude otherwise eligible farmers from access to credit. Except for natural hazards, a farmer may not be delinquent in order to obtain a loan. Direct access to ADBN credit is only customary for larger farmers and now for joint borrowers. Loan applications are cumbersome for often illiterate farmers, the approval process is slow, and actual delivery of the credit (in kind) is often delayed through the interme- diary role of the district Cooperative Union and the Sajha. Sajhas are not very successful in enforcing collection of loans at the end of the season, partly also because their crop marketing system is not well developed and integrated with their credit business. Loan recovery by ADBN from individual borrowers, however, has in general not been better. The ratio of overdues to outstanding loans, which in 1978/79 and 1979/80 had reached levels well above 50%, has recently been reduced, but at the expense of new loans. There is a generally stagnant or even declining trend in the local ADBN's lending volume. 1/ If input sales have gone up in the project area, which is not 1/ Available statistics for Parsa district are not quite consistent and not complete, but they allow this conclusion. - 52 - doubted, farmers must have taken more recourse to their savings or to private moneylenders to finance them. Processing, Storage, and Marketing 6.22 The area is well served by a number of rice mills with an estimated daily capacity of about 300 t. In addition, there are some small rice hullers operating in the villages. A flour mill of about 6,000 t annual capacity is operating in the area. There are also two to three small size wheat mills in eadh village. A sugarmill located in Birganj is presently operating below capacity. In general, processing capacity is adequate to meet present require- ments and the increased production expected from the project. 6.23 The AIC is storing fertilizers at wholesale level and the village cooperatives at retail level. AIC has built up a considerable storage capacity in Birganj for Bara and Parsa districts; about 11,000 t capacity is completed and another 5,000 t capacity under construction. This is con- sidered sufficient for the foreseeable future. For management by cooperative societies, the project has provided three 50 t and three 100 t capacity input supply storage facilities under the project; one for each irrigation block in addition to 3 existing godowns of 75 t each. A 500 t storage facility has been constructed in the tubewell area in Bara district. These storage capa- cities are now in use by village cooperatives as main input distribution points. Total cooperative storage (about 1200 t) will, however, not suffice, if the project's demand reaches an annual level of about 7,000 t of fertilizer material at full development, and if half a year's supply should be accommodated. 6.24 There are about 600 privately owned grain storages ranging from 85 to 340 t each with a total capacity of about 76,000 t. The Nepal Food Corporation (NFC) and the Birganj Rice Export Company (REC) own a total of about 8,000 t capacity of storage. This is adequate to accommodate the production procured by these buyers from the project. 6.25 Four weekly village markets (mandis) trade mainly in cereal grains, pulses and vegetables. NFC and REC purchase grains through local cooperatives for export or distribution to the Hills. Farmers usually sell their wheat to either private grain dealers who visit the area or directly to flour mills. Sugarcane is brought to the Birganj Sugar Factory by the growers on bullock- cart or tractors. In general, traditional marketing channels are well- developed; millers in Birganj and in the neighboring Indian State of Bihar have a widespread network of traders in the villages. The cooperatives' role in marketing of members' produce, however, is still far from satisfactory, due to management and storage constraints. Operation and Maintenance 6.26 During appraisal there had been concern about the adequacy of maintenance of the Don Branch Canal and the desirability to keep canal closures to less than 30 days per year. The NEC was first operated during the monsoon of 1977 when rainfall was so sporadic that farmers' demands for - 53 - irrigation became incessant. Because of the imminent danger to the rainfed paddy crop in the command area, and although the NEC restoration works and distribution canals were only partly completed, the project management decided to provide irrigation water to the system on an emergency basis. The rain deficit was still severe in September 1977 during an IDA mission's visit. At that time information was received from the Indian authorities that the Don Branch canal would be closed for maintenance for about one month starting on October 8 and not on October 31 as had been assumed previously. This closure would have most seriously affected the standing crop of more than 20,000 ha which needed full water supplies up to at least October 15 from the NEC. Upon suggestion by the IDA mission, HMGN immediately contacted the Indian authorities and a deferment of the Don Branch Canal closure date was agreed upon. Following this experience, communications between the project authorities and the irrigation authorities from Bihar have steadily improved. Regular meetings are now being held in September to discuss experiences during the monsoon season, in November to discuss water releases for the rabi season (normally on November 30) and in February to decide on the closure period during March/April. Furthermore, the Bihar irrigation authorities have now stationed a Junior Engineer at the tailend of the Don Branch Canal and will install a telephone line from the Gandak barrage to the Executive Engineer's and Junior Engineer's quarters. During the 1978 monsoon season the Don Branch Canal breached at six places and was closed for a 15-day period for repairs in July. Because of adequate rainfall during this period the standing crops were not affected. Another breach occurred in late August 1981 and the canal was closed for about 5 weeks. 6.27 In recent years canal openings for the rabi season have been very late: in 1979, December 23, in 1980, December 18, and in 1981, December 24. This must have affected the wheat crop seriously and the Bihar irrigation authorities should be persuaded to release water earlier. Provided early rabi water can be ensured, an extended paddy season should be discouraged; it could be considered to stop water deliveries to the blocks at the end of September. 6.28 The areas irrigated since 1977 are shown in Table 21. The irri- gation was unsystematic so far (except in Block IB) and basically it was continuous flow. A draft operation and maintenance manual was completed by the Consultants in 1979 but detailed water delivery schedules had not yet been prepared. It was therefore agreed that a Nippon Koei water management specialist (consultant) would assist the project staff to prepare such schedules and that his services would be financed from Credit 856. Unfor- tunately, recruitment of this consultant was delayed, but by January 1981 the first water delivery schedules had been prepared. The system adopted was that each block would be subdivided into three areas of equal size, but only one area would be entitled to receive water for early paddy in a particular year. For a second area irrigation water would be guaranteed for middle paddy while a third area would only receive supplementary irrigation water as far as available. An annual rotation would be implemented among the three areas. An alternative water delivery schedule is being considered whereby the available water supply would be equally distributed over the entire block and the farmers would share the available supply. It was agreed that both alternatives -54 - would be tried out in separate blocks or parts thereof. At this time, opera- tional data are not yet available. However, it is of the utmost importance that the irrigation schedules be implemented carefully as without it there may not be enough water to irrigate the Stage II and III areas when they are completed. It is foreseen that the implementation of proper water management procedures will be a very difficult task. It seems that additional legislation is needed to establish appropriate rules and regulations for systems operation and maintenance, to ensure proper functioning of the irrigation associations and to provide the project authorities with sufficient powers to enforce the rules and regulations. At present the project staff does not have sufficient power to enforce the scheduling of irrigation water and the necessary discip- line among the irrigators. 6.29 The Don Branch Canal carries considerable quantities of silt into the NEC. The silting up of the canals is therefore a main concern and regular annual maintenance will be crucial. Certainly adequate budget allocations will be required for this purpose. The problem also is that the closure period is too short to do all required survey work, the tendering and implementation of desilting works. The desilting work will therefore be more mechanized. 1/ Suggestions have also been made to use an echosounder in the canals prior to closing and to study the feasibility of a desilting basin at the head of the NEC. 6.30 For effective communications along the NECI, a telephone .line has been established connecting the projects headquarters with its branch offices at Pipra and Tilawe. For operation and maintenance purposes one Assistant Engineer will be positioned in Pipra and Tilawe, each being in charge of three blocks. In addition, there will be one Overseer for each block. Under Credit 856 additional O&M equipment is being procured and eight irrigation sub-centers for O&M personnel are being constructed. A list of equipment procured under Credit 373 is provided in Table 22. 6.31 An Assistant Engineer (part time) is in charge of the operation and maintenance of the 14 existing wells rehabilitated under the project and the 14 new wells. He is assisted by one Overseer, two electricians and one mechanic. One pump operator is in charge of each pumping station and he also distributes the water according to a distribution schedule prepared by the Junior Technician. A major constraint in the operation of the tubewell scheme has been the well failures (para 3.12), the lack of spare parts for pumps and motors and the frequent interruptions in the power supply. 6.32 Some workshop equipment and tools have been procured under the project but a proper functioning workshop has not yet been established in the project area, mainly because of lack of Nepalese expertise to run such 3 3 1/ Recently 6 hydraulic excavators (two of 0.4m and four of 0.2 m capacity) have been procured for this purpose from Cr. 856. - 55 - a workshop. In 1978, it seemed that the Overseas Development Ministry (UK) was interested to provide technical assistance to the IDA financed projects in the setting up of workshops and staff training. This assistance did not materialize and funds were made available under the Technical Assistance Credit for this purpose, as the additional workshop equipment is under procurement from Credit 856. However, the required workshops are still to be established. So far, the pumps and motors of tubewells have been repaired by private workshops in Birganj. 6.33 Information to determine reliable operation and maintenance cost of the system is not yet available. Because of delayed commissioning of the works maintenance during the early years of project implementation has gen- erally been carried out under the construction budget. However, budget allocations during recent years have been insufficient. For example, an O&M budget of NRs 3.0 M had been requested for FY 1980/81 but only NRs 1.5 M was allocated. Tentatively, the operation and maintenance costs for the surface irrigation scheme have been established at NRs 210 per ha and for the ground- water scheme at NRs 340 per ha (Table 17). Cost Recovery 6.34 The credit covenants required HMGN to recover from the beneficiaries by means of water charges (i) the full cost of operation and maintenance and (ii) in progressive steps over a reasonable period of time, the investment cost of the project, having due regard to the incentives of farmers-to use the irrigation water and to their capacity to pay. Water rates were to be reviewed, and, if necessary, to be revised not less than every five years. (In the Stage II project, Cr. 856-NEP, this was subsequently reduced to three years). Charging was to begin one year after the completion of a particular block for irrigation. So far charging and collection have been made for Block IA and the tubewell area only. The rate is the same for both areas, NRs 100/ha per irrigated crop. Assessment and collection is by the project authority (NZIBD) through its Accounting and Finance Division. NZIBD intends to charge beneficiaries in the entire Stage I area beginning with the 1981/82 agricultural season. 6.35 Procedures for assessment and collection of water charges are laid down in the Operation and Maintenance Manual of NZIDB (July 1980). It is important to note that those charges are not taxes or levies imposed by the State by virtue of its taxing power, but user fees that a licensee 1/ (NZIBD) charges against the services it provides to its consumers (the farmers) under Section 7 (1) of the "Irrigation, Electricity and Related Water Resources Act, 1967". This makes the Board's position as to enforcement of collection somewhat weak. The O&M manual is silent on the issue of collection of charges from users that simply refuse to pay. The short collection history so far is not very encouraging with respect to the surface irrigation scheme. Collection continuously declined from 6.6% of dues in 1977/78 to 0.6% of dues in 1979/80, 1/ Licensed for development of water resources. - 56 - whereas it increased for operable tubewells from 12% to 74% in the same period (Table 23). The Irrigation Department is aware of NZIDB's enforcement problems and considers to propose changes in the legal basis for assessment and collection of charges. It is also intended, at a later stage, to involve Water Users Groups in the collection of dues. 6.36 A rate of NRs 100/ha for irrigated crops would result in a total annual charge of about NRs 170/ha at full development. Currently O&M cost for the surface water scheme is estimated as NRs 210/ha per year including NRs 45/ha for amortization of equipment. Given full collection, this would therefore be an 80% recovery of 0 & M cost. It would be less for the tubewell scheme where 0 & M cost is estimated at over NRs 500/ha, but where the same rate of NRs 100/ha is paid by farmers. Recovery of capital cost, although foreseen in the O&M manual, is at present not attempted. It is an open question whether this may be done in the near future, given the political implications of higher rates in general, and vis-a-vis lower rates in other government irrigation projects. Theoretically, the amortization of project cost over 50 years at 10% interest would amount to about NRs 625 per ha for surface irrigation and NRs 1300 per ba for the tubewell scheme. (NRs 720 per ha on average). This would take up about 18% of estimated average incremental income (para 5.14), and together with presently assessed charges for O&M about 22%. While this may appear to be a reasonable level of recovery, it has to be kept in mind that "marginal" farmers are hardly able to pay such amount, and, that a large part of incremental income of tenants would go to landlords, whereas the watercharge will be paid by the tenants. Monitoring 6.37 Monitoring consisted of: (i) maintaining construction cost data through normal accounting practices; (ii) the measuring of groundwater levels in 65 observation wells; water levels are being read monthly in the dry season and bi-monthly in the monsoon season; so far there is no indication of water- logging danger, but it is imperative that measurement continue; (iii) measuring flows in the NEC; the head regulator has been calibrated and one out of three automatic water level recorders as been installed; in May 1979 water losses in the NEC were determined at 0.5 m /sec only; (iv) the collection of meteorologi- cal data at two stations, in Pipra and at the project's headquarters; (v) for the groundwater scheme the maintaining of pump records, static and drawdown water levels, and irrigated area records; unfortunately, only three wells have orifice meters which are in operating condition but additional flow meters will be installed under Stage II. 6.38 Monitoring of agricultural developments was conceived as an impor- tant project activity in the appraisal report, to be started at project inception and to be continued for some years after completion. The experi- ence with agricultural monitoring has been disappointing. At the time of appraisal, it was agreed with the NZIDB that the project office, in collabor- ation with the Food and Marketing Services Division (FMSD) of the DA, would conduct a field survey, collect agricultural data and carry out an evaluation study of the project results. The main objective was to measure the economic - 57 - benefits and changes in farmer's income as a result of the project. Accord- ingly, FMSD dispatched one JT to the DADO's office in Parsa district in 1977 to undertake the studies. The JT was assisted by 10 enumerators, usually JTAs and PLAAs serving in the area. A monitoring and evaluation unit was formally established in early 1980 at the project office. The unit is headed by an Agricultural Officer (SMS) and assisted by two JTs. They work closely with FMSD. So far, FMSD is mainly concentrating its effort on interviewing farmers on crop yields while staff provided by the project is engaged in crop cutting surveys of paddy, wheat and maize. Interviewing farmers to assess yields bears a risk of mis-representation because farmers generally estimate their yields which leaves the reliability of the figures in some doubt. Crop cutting survey data do not always tally with those of ICP and FMSD. The survey could be improved if the sampling were broad enough to fully represent crop per- formance in the project area (stratified sampling). A sufficient number of replications should also be made to upgrade the quality of this survey. 6.39 Monitoring, so far, is limited to crop yields and no systematic assessment of the overall impact of irrigation and of agricultural supporting services has been carried out. Although the project maintains a small unit for monitoring and evaluation, its activities are confined to a routine crop cutting survey on a small scale. It needs substantial strengthening by augmenting qualified staff. Contracting out crop and other surveys, e.g., to the Agricultural Project Services Center (APROSC) or a similar institution, may be an alternative worth considering. In a more general way, a logical framework for all monitoring activities has to be developed since monitoring is not an objective in itself. It should be derived from medium and long-term goals set for the project, and should be able to answer questions relating to the degree of goal achievement and to the reasons for deviations from goals. This would include monitoring not only of the effects of the project, but also of critical inputs (technical, institutional, material, personal). More awareness and knowledge of what is actually happening in the project area are preconditions for sustained development. Staff Training 6.40 While working with the project staff during the first year of project implementation the consultants identified serious deficiencies in the knowledge and experience of the Nepalese staff. It was found that the Assis- tant Engineers and Overseers had very little experience in carrying out the required survey, planning and design work. It was also expected that their capability to supervise the construction works would be poor. The extension workers were also found to be poorly trained and their knowledge was consi- dered inadequate to satisfy the requirements of the extension program. In accordance with their terms of reference the consultants prepared by November 1974 a comprehensive training program for the project staff. This program consisted mainly of on-the-job training in survey work, planning and design, construction supervision, operation and maintenance, workshop management, extension services and cooperative services. To the extent possible the consultants have continued their on-the-job training although it may not have been as comprehensive as was intended due to other work and time pressures. - 58 - Still the achievements in staff training must have been considerable as the implementation of the Stage II project is at present only being supported by two or three expatriates. 6.41 A provisional training program was included in the Narayani Zone Stage II project (Credit 856) and HMGN committed itself to the implementation of this program. Since.HMGN did not accept credit funds for training programs it requested UNDP to finance this program. UNDP has agreed to finance this program, but the signing of the Irrigation Training Project (NEP/79/035) is long overdue. This project would not only provide training programs for the Narayani Zone Project but also for the IDA financed Bhairawa Lumbini Ground- water Project (Credit 654) and the Sunsari Morang Irrigation and Drainage Development Project (Credit 812). A training coordinator is being recruited. Accounting and Reporting 6.42 HMGN has its own specific accounting system which unfortunately did not prove to be very suitable for external aid financing and, initially, accounting documents could not be submitted to IDA without appropriate modi- fications. Nippon Koei's accounting advisor and later several Bank missions comprised of an accountant (consultant) or disbursement officer have assisted the project authorities in improving the system and in providing guidance in disbursement procedures. In 1976, an accountant manual was prepared by the project staff and was reviewed and commented upon by IDA's consultant in accounting. The Manual was subsequently submitted to the Auditor General's office for review but has not yet been finalized. Regular annual audits have been undertaken by a private accounting firm working under contract with the Auditor General's office. HMGN had given assurance that audit reports would be submitted within four months after the end of the fiscal year. However, this did not prove to be possible; a period of approximately one year should be allowed for submission. Project reporting has been satisfactory and quarterly progress reports have been submitted regularly. Institution Building 6.43 During project appraisal the need for a strong project organization (a Board), the shortage of qualified engineering and agricultural staff in Nepal and the need for a strong consultants input were well recognized. Still the local capabilities were over-rated and project implementation took con- siderably longer than anticipated. There will be a continued need for a strong project organization during the operation and maintenance phase, especially between the engineering aspects of operation and maintenance and the aspects of agricultural development. Staff training in these disciplines will be very much needed. It is expected that the UNDP financed Irrigation Training Projects will meet most of this need. Since IDA is involved in the financing of the Stage II project, it will continue to follow the developments in the Birganj Irrigation Project Stage 1. A strong HMGN support for the project will be a continuing requirement, especially in providing continuity of management, well qualified staff and adequate budgetary support. - 59 - VII. SPECIAL ISSUES Local Contracting Industry 7.01 Experience with the Birganj Project has proven that it is difficult to attract foreign contractors (including Indian) to Nepal for relatively small civil works contracts. Project implementation through local contractors has shown that the local construction industry is still in its infant stage and this limits the pace of development in the irrigation sector. There are only a few relatively large civil works contractors in Nepal and these are mostly specialized in building construction. The other contractors are small and inexperienced and lack the necessary financial and technical re- sources and managerial skills. The construction industry faces a number of constraints, including inadequate financial incentives, which are caused by: (i) long delays in the award of contracts and delays in receiving payments from Government; (ii) Government procurement and bank procedures are so demanding and antiquated that they impose a dampening effect on the growth of Nepal's construction industry; (iii) chronic shortages of basic construction materials such as steel, cement and fuel; (iv) lack of financial support resulting in lack of construction equipment; (v) lack of spare parts (govern- ment regulation and red tape effectively prohibit Nepal based equipment dealers from stocking any spare parts at their facilities). These constraints would have to be removed if the contracting industry is to develop. Conditions could be improved through (a) the provision of financial support; (b) the improvement or removal of government procedures that now hamper the construc- tion industry; (c) an equipment rental pool; (d) adequate stocks of spare parts and basic construction materials; and (e) additional technical assistance required over and above the ongoing UNDP/ILO Project/NEP/78/015 which would, inter alia, support a management training cell for local contractors at Tribuvan University. Certainly the strengthening of the local construction industry will not be an easy process. For the time being therefore a dual approach in implementing irrigation projects seems to be in the best interest of Nepal: the large and complicated new projects could be implemented by international contractors, possibly with the participation of local contractors or sub-contractors and the smaller projects could be implemented entirely by local contractors. The joint participation of local contractors with foreign contractors in one project would also considerably contribute to the transfer of technical knowledge. VIII. LESSONS INCORPORATED IN REPEATER PROJECTS 8.01 Following the approval of the IDA credit for the Birganj Irrigation Project the following other credits were approved for irrigation projects in Nepal (i) the Bhairawa-Lumbini Groundwater Project, Credit 654 (US$9.0 M) for 7,000 ha of groundwater development; (ii) the Narayani Zone, Stage II Irrigation Project, Credit 856 (US$14.0 M), which was formulated after revision of the original Birganj Irrigation Project (paras 3.03); (iii) the Sunsari Morang Irrigation and Drainage Development Project, Credit 812 - 60 - (US$30 M), which is to rehabilitate and complete a surface irrigation scheme commanding 66,000 ha; (iv) the Mahakali Irrigation Project, Credit 1055 (US$16 M), which is to rehabilitate and extend an existing scheme to 6,600 ha; and (v) the Babai Engineering Credit, Credit 1093 (US$3.5 M) which is to provide engineering designs and tender documents for a new irrigation scheme commanding 13,500 ha. 8.02 A very good design feature of the Birganj Irrigation Project has been to bring irrigation water as close to the individual farm as possible by providing water through project constructed irrigation canals to irrigation sub-units of 7.5 ha and to have the farmers construct their own field chan- nels within these sub units through their Water Users Groups. This important principle has been adopted in all later irrigation projects. 8.03 The pilot groundwater scheme in the Birganj project could have been successful was it not for a number of well failures, unreliable power supply and poor maintenance. The design and construction of the deep wells under the Bhairawa Lumbini Groundwater Project have been of higher standards and such well failures are not expected here. In this project, all casings are of non-corrosive mild steel.. 8.04 The road network constructed under the Birganj project 2onsists mainly of earthen canal operation roads at a density of 0.9 km/km not accessible for public transport which has proved to be a constraint. The road system designed for the Bhairawa Lumbini Groundwater Project where metalled village roads are constructed at a density of 1.2 km/km2 was a considerable improvement. Unfortunately, the lesson learned from the Birganj project was not adopted for the Sunsari Morang Project where the density of the proposed canal operation roads (0.4 km/km2) is even less than for the Birganj Project. However, road standards have been considerably improved in the Mahakali and Babai Irrigation 5rojects where metalled canal roads at a density of approx- imately 1.6 km/km will also be accessible for public transport. 8.05 The most important finding from the implementation of the Birganj project has been the very slow rate of construction by the local contractors. Not counting the initial start up delays, the actual construction period of the 18,700 ha scheme has been more than four years. Unfortunately, in the Sunsari Morang Project, it was assumed that the 66,000 ha command area could be rehabilitated in five years. This has already proven to be overly opti- mistic. Anticipated construction schedules for the Bhairawa Lumbini and Narayani Zone Stage II Projects were much more realistic but still too opti- mistic. For the relatively small Mahakali Irrigation Project an implementation period has been assumed of five years which seems realistic also in view of additional constraints such as forest clearing, settlement and the remoteness of the site. Both for the Babai and the Sunsari-Morang projects it was decided to attempt to award a construction contracts to international civil works contractors. Designs and tender documents are being prepared with this objective in view. 8.06 The proceeds of Credit 654 were partly to be used for operation and maintenance equipment, but did not provide for procurement of construction - 61 - equipment. The lack of basic construction equipment was a serious constraint in the performance of the local contractors. Starting with the Sunsari Morang Project the provision of construction equipment has therefore been included in the IDA credits. Because of chronic shortages of cement and steel the procure- ment of these materials was also included in the Credit for the Mahakali Irri- gation Project as well as the provision of storage facilities for fuel oil. 8.07 The T&V extension system so successfully introduced in the Parsa and Bara districts was extended to 6 districts under the Naryani Zone Irriga- tion Stage II Project (Credit 856) and subsequently to 9 other districts under the Nepal Agricultural Extension and Research project (Credit 1100, 1981). 8.08 The initial experience with the organization of Water Users Groups headed by farmer-leaders for the purpose of field channel construction, operation and maintenance has been good. The same concept has therefore been repeated in other IDA irrigation projects. IX. IDA PERFORMANCE 9.01 The issues raised-during project preparation and appraisal were gen- erally well identified although in retrospect IDA was too concerned with the adequacy of maintenance of the Don Branch Canal. This canal also serves some 85,000 ha in India and the Indian authorities have therefore all reason to en- sure its operation. Damages to this canal due to breaches in the monsoon season are more critical but are largely beyond the control of the Indian authorities and have been repaired as a matter of urgency. 9.02 The covenants included in the Credit Agreement, such as the recruit- ment of a consulting firm and the setting up of a Project Board as well as the condition for project effectiveness, the appointment of a Project Manager, were all appropriate and necessary. Two covenants may have been too severe. The first is IDA's insistence on ICB, a requirement which was later dropped when foreign contractors did not show any interest to tender for the small scale works to be constructed under the project. The second exception relates to the time frame for on-farm development works, whereby it was stipulated that the farmers should carry out the on-farm developments within one year of the con- struction of the tertiary canals and drains serving their area and that, if this was not done, NZIDB would undertake to construct these works and recover the cost from the land owner. This covenant was not enforced as it takes the farmer about three years to carry out these works while NZIDB neither has the necessary authority nor the funds to carry out these works by itself (para. 3.16). 9.03 During the more than eight years of project implementation, IDA supervision missions have regularly visited the project area. Continuity of staff participating in these missions has been satisfactory. These missions, sometimes assisted by individual consultants, have helped the project authorities in supervising the work of the consultants, solving procurement issues, formulating work programs, identifying constraints and - 62 - finding solutions for these. The missions often acted as a channel of communication between the project authorities, the consultants and the DIHM, the Department of Agriculture and the Ministry of Finance to resolve certain problems, to request for additional assistance and staff, etc. The project management and the HMGN authorities have been very appreciative of the Bank's assistance. Where problems developed the IDA staff have assisted as much as possible, sometimes by creating new programs such as the following: (i) an agricultural extension program was prepared by an IDA consultant based on the T&V system that had proved so successful in India, not only for the two districts covered by the project but for an additional four dis- tricts as well. This extension program was later in- cluded in the Narayani Stage II Project and then further extended under the Agricultural Extension and Research Project (Cr. 1100). (ii) assistance in the setting up of appropriate mechanical workshops for all IDA financed irrigation projects. IDA was able to interest the ODM of the UK to provide such assistance. However, this did not materialize and such assistance is now included in the Technical Assistance Credit 659. This component may only be implemented during the Stage II project. (iii) a small study was formulated to examine in more depth the needs of the local construction industry and to formulate a project for possible financial, managerial and technical assistance. The study, to be undertaken by a local con- sultant, is still to start. (iv) the need for more staff training was identified by IDA and suggestions were made to UNDP to finance such a program. The implementation of the Irrigation Training Project for staff on alll IDA financed irrigation projects is now expected to commence soon. (v) detailed.water delivery schedules have been prepared for the project by an expert of Nippor Koei, which activity was financed out of the Stage II credit. X. CONCLUSIONS 10.01 The main conclusions to be drawn from the project are: (a) The implementation of irrigation projects in Nepal is slow due to major constraints such as the shortage of competent engineering and agricultural staff, chronic - 63 - shortages in construction materials, and the limited construction capacities of local contractors. Conse- quently, Nepal has a limited capacity to implement ir- rigation projects. Future decisions on the level and extent of investments in irrigation projects should give serious consideration to these constraints. (b) Groundwater development in Nepal will not be easy. A major constraint is the lack of mechanical engineers and technicians to maintain the facilities. The best avail- able technology, but appropriate to local conditions, will be needed for the design and construction of wells. Other prerequisites for successful operation are an insured electricity supply, adequate workshop facilities for maintenance, and an adequate stock of spare parts and budget support for O&M. (c) The irrigation system design should include appropriate road networks which are indispensable for the provision of agricultural support services and the transportation of agricultural inputs and production, and for 0&M. (d) Consideration should be given to include the financing of fertilizer as a project component in any future Bank- financed irrigation project in Nepal. (e) Certain constraints that have been noticed in the agri- cultural development such as inefficiencies in the coop- erative societies and the AIC organization, are difficult to be solved on a project level but need to be tackled on a national level as they require high level policy decisions. (f) Adequate cost recovery appears possible. However, a pre- requisite is that the farmers will be provided with reli- able water supply. It will further require a strong com- mitment from HMGN and the project authorities to enforce the collection of water charges. Recommendations 10.02 The ultimate success of the project depends very much on the ability of HMGN to undertake certain actions or provide services which are of great importance for the development of the scheme. These would include: (a) the implementation of detailed water delivery and rota- tion schedules and the training of Assistant Engineers and Overseers in water management scheduling; - 64 - (b) provision of adequate staff and training for operation and maintenance and of adequate maintenance budgets especially for desilting work and spares for pumps and motors; (c) the setting up of properly equipped and staffed workshops for operation and maintenance equipment; (d) the drafting of the necessary legislation for the opera- tion and maintenance of irrigation projects which would provide the O&M staff with the necessary powers to under- take their work, and for the proper functioning of the irrigation associations; (e) advising farmers on on-farm water management through the extension service; (f) providing the project with an efficient vehicle for the coordination of agricultural development activities such as extension, research, input supply, agricultural credit, marketing, etc. (g) improvement of the project monitoring system, especially monitoring of agricultural and economic development; (h) drafting of legislation enabling irrigation development boards to enforce collection of watercharges and to impose sanctions on delinquent farmers. As this is the first IDA-financed irrigation project reaching the operation stage in Nepal, it would be important that IDA continue informal supervision of the project in the coming years. This can be done simultaneously with the supervision of the Narayani Zone Stage II Project which is still under imple- mentation. The lessons still to be learned during the operational phase of the project will be extremely useful for the projects that are still under implementation. - 65 - ANNEX Page 1 NEPAL BIRGANJ IRRIGATION PROJECT Groundwater Component General 1. As envisaged at appraisal, the groundwater component of the project was to include: (a) Testing of 14 existing wells provided in 1969-1971 under Indian Aid and re-equipping them with appropriate pumps, switchgear and ancillary electrical equipment (these wells were each expected to yield about 50 I/s and serve 75 ha). (b) Construction of 14 new wells with a fibre glass casing and screen design and equipping with pumps, switchgear and ancillary electrical equipment (these wells were each expected to yield about 80 1/s and serve 125 ha). (c) Construction of about 18 km of 11 kv transmission line and provision of 11/0.4 kv tranformation at the well head. (d) Construction of pump houses, night storage reservoirs, lined channel (about 1,000 m) and unlined channel conveyance to turnouts commanding chaks of 30-40 ha and drainage net- works in the command of each tubewell. 2. Testing of the 14 existing wells during the design stage showed that three of the wells pumped sand in excessive amounts and these were replaced by new wells to yield about 80 1/s; their command areas were increased accordingly. Of the remaining 11 wells, seven were equipped with pumps yielding in the range of 54 to 58 1/s and would serve nominal command areas of 57 ha; four wells proved capable of yields of about 75 1/s and would serve nominal command areas of 100 ha. Four of the existing wells subsequently became unserviceable (one in 1977, three in 1979) due to sand pumping and are now being replaced. 3. Of the 14 new wells, three pumped sand and gravel from the outset (commissioning attempted in 1979) and were never put into operation, and three more subsequently failed for the same reason (one in 1980 and two in 1982). As the wells were pumping gravel larger than the screen slot size, it must be concluded that the casing/screen assembly had failed, probably at the key joint on the fibre glass pipe constructions. All the new wells have nominal command areas of 120 ha. ANNEX -66 - Page 2 4. It was originally considered that vertical shaft turbine pumps driven by vertically mounted electric motors would be the most suitable pumping units as the motor (the element believed to be most successful to failure) would be easily accessible for repair or replacement. However, as in the meantime cheaper and good quality submersible pumps had become available in India, 20 of the wells were fitted with electric submersible pumps (14 on the new wells and six as replacements on old wells) and it is planned to re-equip four additional old wells with this type of unit. 5. Until recently, power availability for the project has been unreliable. The switchgear was designed (in conjunction with the night storage reservoir) to maximize availability of power and make maximum use of off-peak pumping. It was believed that this would encourage HMGN to set a preferential energy price for off-peak hours. The switch relays therefore included a timing switch for selection of off-peak pumping and no-float switches on the storage reservoir so that pumping could be automatic, if necessary, during the night hours. 6. The night storage reservoir of earth construction provided storage for 8-10 hours of pumping during the off-peak period. It also eliminated the necessity to irrigate at night except during periods of peak water demand. 7. Two types of lined channel were constructed - the first, brick lined in situ in earth fill and the second, precase, lightly reinforced channel sections set in earth fill with sealed joints. The geometry of the command areas proved to be such that outlets could generally be provided to chaks of 10-12 ha. Outlet boxes are of brick construction and were provided with wooden slide gates or stop logs. The unlined channel sections were built by the farmers. 8. Field drains were provided as necessary with outlets to nullahs. Performance of the Systems 9. After a joint mission in September 1977 to supervise the Birganj Irrigation Project and to appraise the second Stage project (Narayani Zone Irrigation Development, Stage II Project), it was possible to state in the Appraisal Report (September 1978): "Nine existing tubewells were restored and 18 new deep wells (130,m) are being installed in the groundwater area of Stage I. .. The nine restored wells began to deliver water in 1976 and,-with strengthened agricultural extension services, iave helped to produce encouraging farming results. The average cropping intensity has already reached 180%. The remaining 18 wells are expected to be in operation by.late summer, 1978. Operation and maintenance, competently carried out by project staff, is satisfactory and the last nine wells have achieved an operating factor of over 90%. 67 - ANNEX - 67 - Page 3 10. The performance of the systems has, with one notable exception, been generally retrograde in the years subsequent to the supervision/ appraisal mission of 1977 through to the present. Performance summaries in terms of areas irrigated by season and annual operating hours are given in Table 1. These reveal that, with the exception of Old Well No. 2 - Parwanipur, utilization of the systems operating in 1977 has tended to decrease through the years (often to less than half of 1976-77 levels), the newly commissioned systems never achieved high utilization and commissioning of the new wells and systems was delayed until the start of the kharif season of 1979. 11. The enigma of the performance of Parwanipur Well No. 2 remains. This area continued to maintain high levels of utilization and irrigation intensity (149% in 1981-82 and averaging 132% from 1977-78 through 1981-82). In fact the limits of this command are essentially set by the water supply and farmers must plant crops such as gram or mustard without irrigation on the area. The cropping intensity is very high. Various vegetables and hemp figure significantly in the cropping pattern. Early rice is sown in the largest area possible and use of fertilizer is general. 12. Most of the remaining tubewell areas present a dismal agricultural picture. With few exceptions, the crops appear little or no better than those on adjacent lands lacking irrigation. Vegetables and other cash crops are rarely seen. 13. It is believed that no single cause can be attributed to the deterioration or failure to achieve performance in the tubewell irrigation systems. The condition of the wells has been discussed above (see paras 2 and 3). The Indian Aid wells were suspect as possible sand pumpers from the outset, though the test pumping should have eliminated those which would present immediate problems. The development of sand pumping problems after several years suggests deterioration of the screens due to corrosion and it is possible that these wells are approaching the end of their useful life, though wells of similar construction and in similar water chemistry environments commonly last up to 20 years in adjacent Gengetic Basin sediment areas. It was perhaps a mistake to introduce the use of fibre glass casing and screen to contract drillers who had no previous experience with this material. The Feasibility Report well design called for mild steel casing and 'Johnson' type wire wrapped screen but this was changed at Appraisal to the fibre glass construction. The new replacement wells are being constructed with mild steel casing and slotted pipe screen with gravel packing, a technique and materials with which the contract drillers are fully acquainted. 14. The pump units have given little trouble except in the cases of sand pumping wells. Starter relays and switch gear have presented more problems but these units are easily accessible and should be swiftly repaired. However, the initial stock of spares for pumping units and accessories has been utilized and has not been replenished. It has been noted that a pump unit could be out of commission for one to three months awaiting repair. The project electricians and mechanics have very limited facilities and depend mainly on the local market for spare parts and repairs. - 68 - Page 4 15. The 11/0.4 kva tranformers are the property of the project (they are more usually on the charge of the Electricity Company) and there have been some problems with pilfering these units for their copper content. For example old well No. 6 - Chorni was out of action for two years (1977-78) after the transformer was stolen. A decision to use transformers with aluminum winding appears to have discouraged pilfering. 16. None of the no-float switch gear on night storage reservoirs are now working though they are simple devices. Given the low level of utilization of most of the systems, the present high availability of power 1/ (nearly 24 hours/day supply) and the fact that preferential rates for off-peak energy use were not granted, automation for night pumping is not essential any more. 17. Only Parwanipur Well No. 1 continues to use the storage facility provided and this is in line with the high utilization of this system. There is some belief that leakages from the reservoirs are significant. However, this claim is not backed up by data, though a simple ponding experiment would be easy to arrange. 18. The lined channels have not presented maintenance problems. The brick lining is preferred to the precast units, being easier to construct and maintain and not being subject to breakage in transport to site. The adequacy of the unlined conveyance varies between sites. The farmer-built conveyance at Parwanipur No. 2 is of relatively high density and well maintained. Staffing for Tubewell Operation and Maintenance 19. The present staffing for operation and maintenance of tubewells is one Assistant Engineer (part time), one Overseer, two Electricians, one Mechanic and one Operator at each well plus a guard on wells remote from villages. Two Amans are provided to collect the water rates. 20. The tubewell operation is run from an over-elaborate office facility (given the small size of the project) but has only one jeep for transport which must be shared with the Assistant Engineer's other duties. There is a small workshop capable of minor repairs only. 1/ Mainly due to increased power supply from India and completion of the Kulekani hydro-electric project. - 69 - ANNEX Page 5 21. Nevertheless, this small staff should be adequate to serve the tubewells. Its main limitations are lack of spares and transport. The record keeping is aimed mainly at accounting and does not adequately monitor the development. However, log books are kept on the wells and it would be possible to abstract information on areas of crops irrigated and amounts and timing of waterings. 22. The NZIDB retains the services of a Junior Technician (Agriculture) with the duties of organizing water distribution on the tubewells and advising on water requirements. The tubewell commands were originally organized for rotational water supply on a fixed day basis. This has little importance in most commands at present due to the low demand for water as a result of the unreliable water supply. 23. The Operators may, in some cases, also be a constraint on the proper utilization of the facilities. A number of them have been recruited from distant villages from the well and are not required to reside at the well. Thus, with the automatic systems out of action and the Operator possibly absent; a cultivator may not obtain water in a timely fashion if he requires it. Agricultural Extension 24. Until 1977, the project was responsible for its own agricultural extension, had a keen and energetic Extension Officer and was apparently successfully applying extension methodology. In 1978, the project came under a regionalized Extension Service. Thus now, the tubewells are serviced by two part-time Junior Technicians (with much larger regional responsibilities in terms of area) and nine Panchayat Level Agricultural Assistants (Extension). 25. The effects of extension are presently notable by their absence in the tubewell systems and the success of Parwanipur No. 2 well cannot be distributed to extension services presently available. In fact the extension advice (and orders) may in some cases be detrimental. For example, the Sugar Cane Specialist insists that this crop requires only two or three waterings to establish it after planting in October-February and thereafter does not require irrigation. This may be true if the farmers are to continue to use poor quality planting materials, lack fertilizers and protection chemicals and be content with very low yields. Certainly the cane in the irrigated areas appeared little or no better than the adjacent unirrigated cane. Nevertheless, an attempt is made to collect a double water rate for cane while simultaneously denying farmers access to water on demand for this crop on the basis that irrigation is not required. - 70 - ANNEX Page 6 26. Availability of seeds and fertilizers presents a continuing problem. Apparently Nepal depends to a large extent on aid grant gifts of fertilizer from donor countries. Prices are usually kept in line with Indian cross border retail prices to discourage smuggling. In case the gifts of fertilizer are inadequate, imports must be arranged, usually from India, and are often insufficient in quantity and late on arrival. The 1981 rabi season posed a particular problem of short supply of fertilizer which the Extension Service attempted to alleviate by rationing with no preference given to farmers with irrigation. By the time the bureaucracy of this attempt at equity had been worked out and ration cards issued, all the farmers received too little fertilizer too late. Water Rates 27. A water rate is presently charged on an area per crop basis of NR 100/ha except for sugarcane for which NR200/ha is charged. The same charges are made (but not so effectively collected) in the Gandak surface water irrigation service area. 28. There is evidence that the water rate charged should not be a constraint on utilization of tubewell water if the supply is secure. The Agriculture Department retained one of the Indian Aid wells for 'its Water Utilization Project. The well equipped with a 55 1/s diesel motor driven turbine pump delivers water through 2.3 km of brick lined channel and well laid-out earth channels to a command area of 50 ha. The water is sold for cash payment before delivery of 16 NR/hr. This (1981-82) rabi season has 40 ha irrigated, mainly of wheat but including some maize, mustard, potatoes and other vegetables. In the 1981 kharif the entire surface was to rice with a high proportion of early variety. Three irrigations are given to wheat and four to rice - three to four inches being applied at each soaking. At these charges, the equivalent water rate is three to four times higher than levied from project wells. The system is run by one Junior Technician and an Operator. Extension service is provided by the project which uses 5 ha of the farmers fields for replication experiments, supplying seeds, fertilizers, chemicals and water free to the experimental areas. There are 60 farmers in the command and most of the land parcels have been consolidated and levelled. Performance Summary of Improved and New Tubewell Irrigation Systems Well Nominal Area Irrigated by Seasons (ha) Annual Hours Operations Location Serial Well System Diadharge CCA 77-78 78-79 79-80 80-81 81-82 77-78 78-79 79-80 80-81 81-82 No. No. Commissioned (1/a) (ha) K R K R K R K R K R Old Wells Improved Kalaya 1 1 1976 54 57 24 13 12 21 15 25 26 15 ? ? ? 794 898 751 ? Parwanipur 2 2 1975 56 57 50 25 37 31 37 41 38 33 40 45 ? 2202 2377 2228 ? Rampur 3 6 1977 57 57 36 18 31 26 16 24 28 22 ? ? 7 835 650 831 ? Parsaum 4 7 1976 58 57 39 25 22 17 23 18 18 18 18 11 ? 907 1007 700 ? Khutwa /b 5 8 1976 58 57 23 12 4 11 6 --Unserviceable-- ? 456 Chorni 6 9 1977 77 100 --Unserv- 8 7 26 28 28 10 23 -- 128, 608 1250 ? Kalaya 7 10 1976 57 57 33 15 15 8 9 9 13 10 10 5 488 609 532 7 Patauda 8 12 1976 80 120 ---- 29 - 17 29 29 22 - - ----- 477 585 727 - Inarwa /b 9 15 1976 77 100 44 27 20 22 ----Unserviceable-- 7 ? Bhataura 10 16 1976 88 100 40 27 22 28 24 33 24 39 ? 7 7 957 846 619 ? Rhaluhi /b 11 17 1976 83 100 20 39 42 33 -----Unserviceable---- 7 ? Bhawanipur /b 12 18 1976 57 75 14 -----Unserviceable----- - 7 Bishanbarpur 13 19 1978 80 100 8 24 13 25 25 27 ? ? 685 699 ? Lal Palsa 14 20 1978 73 75 7 5 4 4 15 --Unserv- 648 205 - New Wells Rampur /c 1 N-1 1979 80 120 35 55 43 39 Unserv 655 759 - Bairiya /c 2 N-2 1979 80 120 ---Never worked-- Bhataura7c 3 N-3 1979 80 120 - -Never worked- Sitalpur 4 N-4 1979 80 120 36 36 9 46 ? ? 551 1063 ? Sitalpur 5 N-5 1979 80 120 20 46 58 46 ? ? ? 974 T Shishwania /c 6 N-6 1979 80 120 23 18 --Unserv--- 769 Balerampur 7 N-7 1979 80 120 18 45 22 32 ? ? 751 838 ? Motisar 8 N-8 1979 80 120 5 22 10 14 10 16 249 338 ? Motisar 9 N-9 1979 80 120 15 46 25 28 7 7 455 679 ? Dohan 10 N-10 1979 80 120 20 23 3 24 ? ? ? 415 7 Buniyad 11 N-11 1979 80 120 27 29 3 28 ? ? 833 705 ? Awadhapur /d 12 N-12 1979 80 120 16 33 20 4 Unserv 485 360 -- Barewa 13 N-13 1979 80 120 16 18 17 8 ? T 526 625 ? Khutwa /c 14 N-14 1979 80 120 --Never worked-- Totals 323 201 249.234 402 605 454 483 ? ? /a Recorded in fiscal year to July. b Old well being replaced due to sand pumping - will be replaced with 80 1/s capacity and extended command area. /c Original well being replaced due to structural failure of fibre glass casing and screen. /d Requires replacement well. - 72 - TABLE 1 NEPAL BIRGANJ IRRIGATION PROJECT Project Cost: Appraisal Estimates vs. Actual Cost Appraisal 1/ Appraisal, Prorated 2/ Actual 3/ (NRs M) (NRS M) (NR M) I. Surface Irrigation Scheme 1. Civil Works - Nepal East Canal 4.2 2.4 8.4 - Irrigation Blocks 18.4 10.3 48.3 2. On-Farm Development by Farmers 10.1 5.7 -2.4 3. Equipment and Spares for O&M 6.5 3.6 4.1 Sub-Total 39.2 22.0 63.2 II. Tubewell Irrigation Scheme 1. Drilling of Wells 2.8 2.8 6.8 2. Civil Works 2.9 2.9 8.4 3. On-Farm Development by Farmers 1.0 1.0 0.3 4. Pump Equipment and Electrical Connection , 2.9 2.9 4.3 Sub-Total 9.6 9.6 19.8 III. Project Overhead 1. Establishment 11.1 7.4 22.9 2. Consultants 18.5 12.3 18.2 3. Buildings (Headquarters) 5.8 3.9 2.9 Sub-Total 35.4 23.6 4/ 44.0 IV. Land Acquisition 12.1 7.2 0.2 5/ V. Other Project Costs 1. Cooperative Godowns (Grain Stores) 0.4 0.4 0.6 2. Agricultural Training Centers - - 0.1 3. Feasibility Study 5.7 5.7 4.1 Sub-Total 6.1 6.1 4.8 Total 102.4 68.5 132.0 (US$10.12 M) 6/ (US$ 6.77M) 6/ (US$10.96 M) 7/ 1/ Appraisal cost already contains orice and nhysical contingencies. 2/ To take account of reduced scope of project: Surface scheme covers only 16,000 ha under the project, as compared to 28,700 ha at appraisal. 3/ Most of the figures are actual expenditures; exceptions are items 11.4 (partly based on contract amounts), 1.2 and 11.3 (estimated). Expenditures on force account were allocated to "civil works" of surface irrigation scheme. 4/ Taken as 2/3 of original appraisal estimate, since reduced scope would not have implied proportionally smaller overhead. 5/ This amount appears somewhat small, but was given by project authorities as the actual amount. It was allocated (in economic analysis) to the surface scheme only. b/ The appraisal report used a rate of NRs 10.125/US$. 7/ Based on average exchange rate of NRs 12.04/US$ over the project period. TABLE 2 73 PART A NEPAL BIRGANJ IRRIGATION PROJECT Physical Achievements 1) Nepal Eastern Canal Improvements km 0 to km 26.5 (NEC-1) 1.1 Check cum Escape structures 2 Nos. 1.2 Checx structure 1 No. 1.3 Foot bridge 1 No. 1.4 Side Spillway 2 Nos. 1.5 Drainage Inlet 1 Nos. 1.6 Gravelling of operation road 26.5 km 1.7 Left Bank Strengthening 2.0 km 2) Surface Irrigation Area 2.1 Area, net 15,980 ha 2.2 Main Secondary canals 76.1 km 2.3 Branch Secondary Canals 109.1 km 2.4 Tertiary Canals 408.5 km 2.5 MSC operation road t3.5 m, earth) 76.1 km 2.6 BSC roads (3.5/1.5 m, earth) 109.1 km 2.7 Structures for irrigation canals 3,792 Nos. 2.8 Drains 289.4 km 2.9 Structures for drainage 256 Nos. 3) Ground Water Area 3.1 Area, net 2,730 ha 3.2 Canalswith brick and concrete lining 43.7 km 3.3 Related structures 568 Nos. 74 - abNa PAT a BuISANJ IuRIGATION PRDJMCT Physte41 Aabtbvadang -a astall* Bllpn'Block-2 1ko ck- Sk- ock-5 Stoc-6 Total 1) Nepal Eastern Canal Improvements PM 0 to Em 26.5 (AEC-1) 1.1 Check cum Escape structures 2 2 1.2 Check atructure 1 1 1.3 Foot Bridges 1 1 1.4 Side Spiliway 2 2 1.5 Drainage Inlet 1 1 1.6 Gravelling of Operation Road 26.5 km 26.5 km 1.7 Left Bank Strangthening 2 km 2 km 2) Surface Irrigation Area 2.1 Area, net (ha) 1,360 2,080 2,310 3,090 2.140 2,700 2,80u 16,480 he 2.2 Nain Secondary Canals 0.5 ia 10.4 ta 10.50 km 12.3 ka 17.2 km 14.2 km 11.0 km 76.1 km 2.3 Branch Secondary Canals 16.3 11.7 23.20 18.9 3.2 14.5 21.3 109.1 ka 2.4 Tertiary Canals 37.8 48.4 52.00 70.3 56.0. 67.2 00.8 408.5 km 2.5 MSC Operating Roads (3.5 m. earth) 0.5 10.4 10.50 12.3 17.2 14.2 11.0 76.1 la 2.6 BSC Roads k3.5/1.5 m. earth) 16.3 11.7 23.20 18.9 3.2 14.5 21.3 L09.1 km 2.7 Structures for irrigation canal (No.) a) Turnouts 34 46 70 76 44 67 48 385 b) Intake Structures 4 2 2 8 c) Check 10 5 11 8 3 15 15 67 d) Culverts 62 72 130 137 66 82 84 633 e) Spillways - 5 1 3 1 2 4 16 f) Bridges - - - 2 - - 1 3 g) Parshall flumes 3 3 6 6 4 6 8 36 h) Drops 11 10 7 7 10 - 2 47 i) Division boxes 199 280 382 452 310 417 468 2,508 J) Aqueducts - - - 1 3 - - 4 k) Syphono 1 6 - 2 - - 1 10 1) Crosa Drains - - - - j 22 37 62 m Foot bridges - - - - - - 13 13 TOTAL STRUCTURES 320 427 607 694 448 613 683 3,792 2.8 Drains 17.0 38.7 66.1 62.1 7.3 48.3 49.9 289.4 Ia 2.9 -Structures for drainage canal (No.) 13 28 35 45 4 37 36 198 a) Culverts b) Drops 1 8 - 9 1 12 3 34 c) Aqueduct - - - 1 - - - 1 4) Curve Protection - - - 14 - 8 - 22 e) Drainage Inlet - - - - - 1 - 1 TOTAL STRUCTURES 14 36 55 69 5 58 39 256 3) Ground Water Area ELUSTIG WL.S VE.S 3.1 Area. net (ha) 790 1,680 2,470 ha 3.2 Canal with brick and concrete lining 11.2 32.5 43.7 Ia 3.3 Reiated Structures (1o.) a) Division boxes 72 191 263 b) Culverts 18 40 58 c) Farshall flumes 11 25 36 d) Foot Paths 139 42 181 e) Cross drains 1 16 17 f) Drops 7 7 g) Syphon 6 6 TOTAL STRUCTURES 247 321 568 TABLE 3 - 75 - NEPAL BIRGANJ IRRIGATION PROJECT Credit 373 Awarded Civil Works Contracts Works Name of Contractor Contract Amount (in million NRs) A. Surface Irrigation Scheme Block IB Chitawon COECO 4.43 Block LA Chitawon COECO 2.98 Block II-1 Ram Jan Ansari 1.38 Block 11-2 Ram Jan Ansari 1.06 Block 11-3 Bashiruddin Ansari .95 Block 11-4 Bashiruddin Ansari 1.31 Block III Sharma & Company 3.58 Block IV-1 General Const. Co. .87 Block IV-2 General Const. Co. 1.21 Block IV-3 General Const. Co. .86 Block IV-4 General Const. Co. 1.29 Block V-1 Reliable Builders 1.26 Block V-2 Reliable Builders 1.31 Block V-3 Reliable Builders 1.13 Block V-4 Reliable Builders 1.27 Block VI Modern Const. Co. 5.69 NEC I Bashiruddin Ansari 4.72 B. Surface Water Irrigation Groundwater Phase I (civil) NCCN 3.27 Groundwater Phase II(civil) Himal Iron & Steel 6.44 Tubewell Construction (17 wells) Pioneer Tubewell Industries 5.12 C. Others Project Headquarters NCCN 3.47 Grain Stores .57 76- TABLE 4 NEPAL BIRGANJ IRRIGATION PROJECT ACTUAL PROJECT COSTS (NRS MILLION) YEAR 73 74 75 76 77 78 .79 80 81 82 TOTALS. SURFACE WATER SCHEME CIVIL WORKS - - 1.9 6.2 14.5 7.6 3.4 1.2 - - 34.8 FORCE ACCOUNT WORKS - 0.9 4.7 1.0 4.3 6.2 4.8 - - - 21.9 EQUIPMENT 0.5 - - - 1.5 1.9 0.2 - - - 4.1 ON-FARM WORKS ./ - - - - - 0.2 0.4 0.7 0.8 0.3 2.4 LAND ACQUISITION - 0.1 - - - - 0.1 - - - 0.2 SUBTOTAL 0.5 1.0 6.6 7.2 20.3 15.9 8.9 1.9 0.8 0.3 63.4 EXISTING TUBEWELLS, DRILLING - - 0.9 - - - - 1.0 - - 1.9 CIVIL WDRKS - 0.3 1.0 0.9 - - - 0.3 - - 2.5 EQUIPMENT / COfffCTION - - 1.3 - - - - 1.3 ON-FARM WORKS __ - - 0.1 SUBTOTAL - 0.3 3.2 0.9 - - - 1.3 - - 5.8 NEW TUBEWELLS DRILLING - * 0.1 4.1 - - - 0.7 - - 4.9 CIVIL WORKS - - - 1.7 3.2 0.2 0.8 - - - 5.9 EQUIPMENT / CONNECTION 2/ - - 1.5 1.5 - - 3.0 ON-FARM WORKS 1/ - - - - - - - 0.2 SUBTOTAL - - 0.1 5.8 4.7 1.7 0.8 0.8 - - 14.0 SUBTOTAL. WELLS - 0.3 3.3 6.7 4.7 1.8 0.9 2.1 - - 19.8 PROJECT OVERHEAD HEADQUARTER BUILDINGS - 0.7 0.6 1.3 - - - 0.3 - - 2.9 ESTABLISHMENT 0.9 3.1 4.2 3.5 3.5 3.8 3.9 - - - 22.9 CONSULTANTS 0.2 4.7 5.9 3.6 2.5 1.1 0.2 - - - 18.2 SUBTOTAL 1.1 8.5 10.7 8.4 6.0 4.9 4.1 0.3 - - 44.0 TOTAL IRRIG. INVESTMENT 1.6 9.8 20.6 22.3 31.0 22.6 13.9 4.3 0.8 0.3 127.2 OTHER PROJECT COSTS TRAINING CENTERS - - 0.1 - 0.1 GRAIN STORES - - - 0.6 - - - 0.6 FEASIBILITY STUDY - 0.3 1.4 1.7 0.7 - - - 4.1 SUBTOTAL - 0.3 1.4 1.7 1.4 - - - 4.8 TOTAL PROJECT COST 1.6 10.1 22.0 24.0 32.4 22.6 13.9 4.3 0.8 0.3 132.0 1/ Estimated (NRs 150/ha) 2/ From contract amounts NOTE: Rows or columns may not add up to respective totals due to rounding. TABLE 5 - 77 - NEPAL BIRGANJ IRRIGATION PROJECT Inflators for Economic Analysis International Price Nepal Investment National Urban Index 1/ Price Index 2/ Consumer Price Index 3/ Year Index Inflator Index Inflator Index Inflator 1973/74 45.4 2.20 39.4 2.54 59.1 1.69 1974/75 54.2 1.85 53.9 1.86 69.0 1.45 1975/76 58.4 1.71 57.9 1.73 68.5 1.46 1976/77 61.3 1.63 60.8 1.64 70.4 1.42 1977/78 69.6 1.44 68.3 1.46 78.2 1.28 1978/79 81.0 1.23 75.4 1.33 80.9 1.24 1979/80 91.7 1.09 89.0 1.12 88.8 1.13 1980/81 100.0 1.00 100.0 1.00 100.0 1.00 1/ IBRD Commodity Price Forecasts, June 1981 update. This index was used in economic analysis of the project. 2/ Weighted index derived from International Price Index and an index representing level of selected wholesale prices in India, in:Nepal- Policies and Prospects for Accelerated Growth, Table 9.3, IBRD 8/1981 (Green Cover Economic Report). 1980/81 level of this index estimated by ASPAC. 3/ Source as above, Table 9.2. TABLE 6 NEPAL BIRGANJ IRRIGATION PROJECT CREDIT 373-NEP ANNUAL SCHEDULE OF DISBURSEMENTS (as of May 21, 1982) IBRD Accumulated Disbursements Fiscal Million US$ Equivalent Actual Disbursements Year & Actual Total Appraisal Revised as a Percentage of Semester Disbursements Estimate Estimate Appraisal Estimate 1974 1st 0 0.30 14 2nd 0.1 0.70 1975 1st 0.3 1.30 23 2nd 0.6 2.00 30 1976 1st 1.1 2.70 41 2nd 1.3 3.40 38 1977 1st 1.7 4.10 41 2nd 2.2 4.20 52 1978 1st 3.2 5.40 59 2nd 4.0 5.75 70 1979 1st 4.7 6.00 78 2nd 4.8 80 1980 1st 5.3 5.5 88 2nd 5.5 5.8 92 1981 1st 5.6 6.0 93 2nd 5.7 94 1982 1st 5.96 1 99.3 Closing Date 12/31/78 6/30/81 1/ Balance expected to be cancelled. Table 7 - 79 - NEPAL BIRGANJ IRRIGATION PROJECT Disbursement by Category (US$ '000) As at Negotiations Actual (as of May 21, 1982) Imported Equipment and Spare Parts 700 758.4 Consultants 1700 1720.3 Civil Works 2500 3480.2 Unallocated 1100 - TOTAL 6000 5958.9 l- 1/ It is expected that the balance of US$41,100 will be cancelled. - 80- TABLE 8 NEPAL BIRGANJ IRRIGATION PROJECT Cropping Patterns and Yields Revised Estimates: Surface Water Area Befdre'Develdpfient 1/ Future Without Project Future With Prolect Area (%) Yield(t/ha) Area (%) Yield (tha). Area (%) Yield(t/ha) Paddy HYV, irr. 20 2.1 25 2.3 68 3.2 HYV, rf. 10 1.9 15 2.1 - - Local, irr. 5 1.7 - (1.8) 25 2.3 Local, rf. 50 1.4 45 1.5 - - Wheat Irrigated - - 45 2.6 Rainfed 34 1.1 34 1.2 - - Maize (irr.) - - - - 7 2.0 Oilseeds Irrigated 5 0.8 Rainfed 5 0.4 5 0.5 3 0.5 Pulses Irrigated - - 9 0.9 Rainfed 10 0.5 10 0.6 5 0.6 Vegetables Irrigated 1 6.5 1 7.0 4 12.0 Rainfed 2 5.5 2 6.0 - - Sugarcane Irrigated 1 25 1 25 7 40 Rainfed 3 20 3 20 - - Cropping 141 141 178 Intensity Irrigation Intensity 27 2/ 27 2/ 170 1/ Base Years 1973-77 (average). f/ Basically from the Tilawe Scheme NEPAL BIRGANJ IRRIGATION PROJECT Cropping Pattern and Yield in the Staff Appraisal Report Before Development Future Without Project Future With Project Area % Yield (%) Area ) Yield (%) Area (%) Yield (%) Paddy, HYV - - - 44.3 3.5 Paddy, local 86.6 1.5 90.0 1.7 49.1 2.7 Wheat 25.0 1.0 30.0 1.2 50.0 3.0 C Oilseeds 3.8 0.4 5.0 0.4 5.0 0.75 1 Pulses 4.8 0.5 5.0 0.5 5.0 0.85 Vegetable (potato) 5.0 5.5 5.0 6.0 5.0 12.00 Sugarcane 5.0 15.0 5.0 18.0 10.0 40.0 130.2 1/ 140.0 168.4 1/ The text of the SAR (para 6.01) states a cropping intensity of 135% before development. - 82 - TABLE 10 NEPAL BIRGANJ IRRIGATION PROJECT Incremental Production Comparison of Appraisal with PCR Estimates Surface Water Scheme Future Without Project Future With Project Incremental ('000 tons) ('000 tons) ('000 tons) SAR 1/ PCR 2/ SAR 1/ PCR 2/ SAR PCR Paddy 24.4 25.2 46.0 44.0 21.6 18.8 Wheat 5.8 6.5 24.2 18.7 18.4 12.2 Maize - - - 2.2 - 2.2 Total Cereals 30.2 31.7 70.2 64.9 40.0 33.2 Pulses 0.4 1.0 0.7 1.8 0.3 0.8 Total Food Grains 30.6 32.7 70.9 66.7 40.3 34.0 Oilseeds 0.3 0.4 0.6 1.1 0.3 0.7 Vegetables 4.8 3.0 9.6 7.7 4.8 4.7 Sugarcane 14.6 13.6 63.9 44.7 49.3 31.1 Total Value (NRs M) - 97.7 100.1 235.5 214.7 137.6 114.6 (25.9) (27.5) (62.2) (56.5) (36.2) (29.0) 1/ Staff Appraisal Report; figures in SAR were pro-rated to account for reduced area (16,000 ha instead of 28,700 ha). 2/ New estimates for this report. 3/ At economic prices; 1985 projected prices in constant 1980 Rupees (these are considerably higher than prices used in the SAR). Figures in brackets give value at prices used in SAR (1980 prices in constant 1972 Rupees). - 83 - TABLE 11 NEPAL BIRGANJ IRRIGATION PROJECT Transition of Cropping Patterns Surface Irrigation Scheme (15,980 ha) Crop Pre- Project 77/78 78/79 79/80 80/81 81/82 82/83-85/86 Etcetera Kharif - ------------------------- %of net CCA ---------------------- Paddy HYV, irr. 20 30 39 52 66 68 68 68 Paddy HYV, rf. 7 - - - - - - - Paddy local, irr. 8 35 25 25 25 25 25 25 Paddy local, rf. 50 29 26 16 - - - - Vegetables, irr. 1 2 2 2 2 - - - Vegetables, rf. 1 - - - - - - - Rabi Wheat, irr. - 12 25 27 29 35 40 45 Wheat, rf. 34 26 20 20 20 10 5 -- maize, irr. - - 2 2 3 5 6 7 Oilseeds, irr. - - 2 2 3 5 6 6 Oilseeds, rf. 5 5 4 4 4 3 3 3 Pulses, irr. - - 3 3 4 6 7 9 Pulses, rf. 10 10 8 8 8 6 5 5 Vegetables, irr. - 2 2 2 2 4 4 4 Vegetables, rf. 1 - - - - - - - Perennial Sugarcane, irr. 1 3 4 4 5 6 7 7 Sugarcane, rf. 3 1 1 1 1 1 - - Cropping Intensity 141 155 163 168 172 174 176 178 Irrigation Intensity 30 84 104 119 139 155 163 172 NEPAL BIRGANJ IRRIGATION PROJECT Crop Yields (t/ha) 1973/74 1974/75 1975/76 1976/77 1977/78 1978/79 1979/80 1980/81 Wet Season Irrigated HYT n.a. 2.0 2.1 2.2 2.0 2.4 2.7 2.7 Local Paddy 1.5 1.7 1.7 1.8 1.6 1.9 1.9 2.0 Vegetables 6.0 6.0 6.0 6.5 6.5 8.0 10.0 11.0 Rainfed Local Paddy 1.3 1.2 1.4 1.4 1.3 1.6 1.4 Dry Season o0 Irrigated Wheat 1.6 1.5 1.7 1.8 1.8 Oilseeds 0.7 0.7 0.8 0.8 0.7 Pulses 0.8 0.8 0.9 0.9 1.0 Maize 1.6 1.7 1.8 Vegetables 8.0 10.0 11.0 Rainfed Wheat 1.0 1.2 1.1 1.1 1.0 1.2 1.3 n.a. Oilseeds 0.4 0.5 0.5 0.5 0.3 0.5 0.5 n.a. Pulses 0.5 0.6 0.6 0.7 0.5 0.6 0.6 n.a. Vegetables 5.5 5.5 5.5 5.5 5.3 5.8 5.8 n.a. Perennial Sugarcane 22.0 25.0 25.0 25.0 25.0 32.0 36.0 38.0 -85- TABLE 13 NEPAL BIRGANG IRRIGATION PROJECT Sales of Fertilizer in Bars and Parsa Districts by AIC 1/ N P205 K 0 Total nutrient tons 1974/75 1173 366 91 1630 1975/76 1170 357 116 1643 1976/77 1477 295 124 1896 1977/78 1667 336 77 2080 1978/79 1630 349 113 2092 1979/80 1783 435 168 2356 The two districts comprise about 104,000 ha of net cultivated area. The project area (18,700 ha) is located in Parsa district, and represents the only significant irrigation development in the two districts. \'E PA. RlR.ANJi IKRIGATIuAN 1.kEtl Input Requirements Assumed for Economic Anaivsis (per hectrare' 1<oP LABOR dITLLOk Ka ulEmil(ALS þ',' _L _P N_ (mandavs) (paLrda's s_ Rpees) _ _ .._ _k __kg___ __ kg BF FWO FW BF FWO FV HF Fit F'1 4 BF FWO V BF FWO FV BF PWO Fi Paddvy HY :rr. 120 125 135 40 40 45 60 zo 150 51t 5<1 50 30 40 70 10 30 - 5 10 Padd\ HYV rfd. 120 122 t 40 40 * 20 25 - . 50 50 .3 5 * - - - - Paddv Local irr. 115 118 122 40 40 43 30 40 -W 50 50 50 13 15 40 - 5 10 - - 5 Padd\ Local rfd. 00 105 - 35 5 * 10 20 50 5 5 10 - 5 - - Wheat irr. ,80)* - 95 -30)* - 32 o) . - 120 '1005* - 100 150)T - 0 (25)* - 40 (5)* - 10 Wheat rfd. 78 80 80 30 30 30 - - - 120 120 120 - - - - - - - - - Maize (85)W * 90 (25)* * 25 (501 * 60 <25)* * 25 (40)* 60 (10)* * 20 (10)* 15 Oilseeds irr. (70)* * 70 (25) 25 (8)* * s (25* * - (20)* * 20 k5)* * 5 Oilseeds rfd. 50 55 55 20 20 20 - - - 8 9 8 - - - - - - - - - Pulses irr. (40)* * 43 (12V' * 12 - - 351* 40 (-1* 10 (-5* * 20 (-** - Pulses rfd. 27 27 27 10 10 10 - - - 35 35 35 - - - - - - \'egetables PoIato) irr 130 135 150 25 25 25 100 150 200 2,500 2,500 2.500 6 70 80 30 35 40 50 60 80 Vegetables (Porato) rfd. 120 120 120 20 20 20 50 80 80 2,500 2.500 2.500 30 30 30 - - - - - - Sugarcane irr. 1/ 91 91 134 20 20 20 50 50 100 2.000 2,000 1.250 60 b0 120 40 40 80 - 60 Sugarcane rfd. 2/ 87 87 20 20 - - * 2.000 2.000 40 40 * 20 20 - - - aiL not grown. Fig,,res in b'rackets serýe as ass,med mno tal i-,els iwards t,sll devel,pment out. HF - Before Developmenr FWO Furlre Without Pro iect FW Future With Project 1/ Average over 4 years (2 ratoons>. 2/ Average over 2 1/2 vears il ratooi. NEPAL BIRGANJ IRRIGATION PROJECT Prices Used for Economic Analysis (NRs per ton) 1976/77 1977/78 1978/79 1979/80 1980/81 1981/82 1982/83 1983/84 1984/85 1985/86 Paddy 2,280 1,740 2,050 1,510 1,870 2,160 2,330 2,430 2,520 2,620 Wheat 2,920 1,790 1,760 2,000 1,970 2,250 2,400 2,480 2,560 2,650 Maize 1,930 1,350 1,170 1,170 1,170 1,410 1,840 1,900 1,960 2,020 Oilseeds 4,770 4,820 3,800 3,490 3,240 3,720 4,640 4,400 4,300 4,140 Sugarcane 606 280 230 248 600 426 372 384 397 410 Pulses 2/ 4,000 4,000 4,000 4,000 4,000 4,000 4,000 4,000 4,000 4,000 Vegetables (potato) 2/ 2,000 2,000 2,000 2,000 2,000 2,000 2,000 2,000 2,000 2,000 Nitrogen 10,800 9,100 8,920 9,100 9,650 9,550 9,680 10,150 10,620 11,100 Phosphate 10,760 8,330 7,780 8,730 9,030 9,330 9,380 9,700 10,010 10,330 Potash 5,770 5,060 4,960 5,220 5,800 5,700 5,620 5,620 5,630 5,640 Labor NRs 7/manday (weighted average of peak and slack season rates) 0 Bullocks: NRs 18/pairday (financial price x standard conversion factor 0.91 Chemicals: at given financial cost (no conversion) Seeds (per kg): Paddy HYV NRs 2.8; Paddy local NRs 2.3; Wheat irrig. NRs 4.0; wheat rainfed MRs 2.3; Oilseeds NRs 4.1; Maize NRs 1.9; Pulses NRs 4.0; Sugarcane NRs 0.3; Vegetables (Potato) NRs 2.5 1/ In constant 1980/81 N Rupees; derived from actual and projected world market prices (in constant 1980 US$) and adjusted for transport and handling. t2l 2/ Same as financial price (price of pulses probably somewhat underestimated). - 813 - TABLE 16 NEPAL BIRGANJ IRRIGATION PROJECT Economic Analysis: Conversion Factors Applied to Financial Prices (a) Civil Works, Surface Scheme: Traded 1.0 x 50% (excluding force account) Non-Traded 0.9 x 30% Unskilled Labor .0.65 x 20% Average 0.9 (b) Civil Works, Tubewell Scheme: Traded 1.0 x 40% Non-Traded 0.9 x 20% Unskilled Labor 0.65,x 40% Average 0.84 (c) Force Account and Non-Traded 0.9 x 10% On-Farm Works: Unskilled Labor 0.65 x 90% Average 0.68 (d) Equipment, Consultants: 1.0 (e) Standard Conversion Factor: 0.9 - 89- TABLE 17 NEPAL BIRGANJ IRRIGATION PROJECT Estimated Annual Cost of Operation and Maintenance A. Surface Irrigation Scheme (Stage I, 16,000 ha) Amortization of Equipment $ 60,000 Equipment Operating Cost 30,000 /a 0 & M Buildings 10,000 Labor and Materials 75,000 Staff Costs 105,000 5280,000 0 & M Cost/ha $ 17.50 (NRs 210) B. Tubewell (80 ha net irrigated) Salaries $ 400 Electricity 2,000 Repair and Maintenance - Pump, motor, switchgear 70 - Structures, Canals 450 Miscellaneous 500 $ 3,420 0 & M Cost/ha $ 43.0 (NRs 510) /a In comments after completion of this report, project management estimated this cost item as US$60,000. This would result in an 0&M cost of about US$21.3 per ha. NEPAL BIRGANJ IRRIGATION PROJECT (A) Streams of Estimated Project Benefits (NRs M) Surface Irrigation Scheme WITHOUT PROJECT WITH PROJECT INCREMENTAL Gross Input Net Cropping Gross Input . Net Cropping Gross Input Net % of Full Year Benefits Cost Benefits Intensity Benefits Cost Benefits Intensity Benefits Cost Benefits Development (%) (%) 1976/77 78.6 39.2 39.5 141 78.6 39.2 39.5 141 - - - - 1977/78 59.0 37.2 21.8 140 72.0 46.1 25.9 155 13.0 8.9 4.1 6 1978/79 76.8 39.8 36.9 141 104.2 61.8 42.4 163 27.4 22.0 5.5 8 1979/80 59.7 38.8 20.9 141 100.8 61.7 39.2 168 41.1 22.9 18.3 27 1980/81 76.4 39.3 37.1 141 139.2 67.6 71.6 172 62.8 28.3 34.5 50 1981/82 82.9 39.9 43.0 141 157.1 72.3 84.8 174 74.2 32.4 41.8 61 1982/83 87.2 40.2 47.1 141 175.5 76.3 99.2 176 88.3 36.1 52.1 76 1983/84 91.8 40.9 50.9 141 191.1 81.4 109.7 178 99.3 40.5 58.8 86 1984/85 96.3 41.5 54.7 141 203.0 84.3 118.7 178 106.7 42.8 64.0 93 1985/86 100.1 42.2 57.9 141 214.7 88.1 126.7 180 114.6 45.9 68.8 100 etc. NOTE: Rows/Columns might not subtract to exact totals, due to rounding. (B) Summary of Economic Benefit and Cost Streams for Surface Irrigation Scheme (NRs Million) Year Benefits Costs 2/ 1 - 2.76 2 - 12.38 3 - 21.36 4 - 19.73 5 4.10 31.01 6 5.50 20.58 7 18.30 11.27 8 34.50 5.13 9 41.80 4.07 10 52.10 3.05 o0 11 58.80 2.30 12 64.00 2.30 13-40 68.80 2.30 1/ Year 1 corresponds to 1973/74. 2/ Excluding sunk cost in year 1. 91 TABLE 19 NEPAL BIRGANJ IRRIGATION PROJECT Sensitivity of Economic Rate of Return Surface Scheme ERR ERR (Excl. Sunk Cost) (Incl. Sunk Cost) Base Case 25.7% 19.2% Cost +10% 24.2% 18.1% +20% 22.8% 17.1% -10% 27.5% 20.5% -20% 29.5% 21.9% Benefits -10% 24.0% 17.9% -20% 22.2% 16.6% +10% 27.3% 20.3% +20% 28.8% 19.2% Benefits -10% 22.5% 16.9% Cost +10% Benefits -20% 19.6% 14.7% Cost +20% ERR would decrease to 10% for a shortfall of "true" benefits relative to estimate of .70%; (59% including sunk cost), or an excess of "true" cost over estimated cost of 240% (144%). TABLE 20 NEPAL BIRGANJ IRRIGATION PROJECT Estimated Typical Farmbudget for an Average Size Farm (one hectare) A) Future with Project (all crops irrigated) Crop Area Yield Price Gross Return per ha input Csat per ha Net Return Total Labor (na) (t/ha) (NRS/t) (NRS) (NRS) (NRS) (M) Paddy HYV 0.75 3.2 2,300 7,360 1,400 4,470 94 Paddy local 0.25 2.3 2,AO 5,29u 1,130 1,040 30 Wheat 0.50 2.6 2,200 5,720 2,050 1,83> 48 Pulses 0.15 0.9 4,000 3,600 1,030 385 7 Oilseeds 0.10 0.8 4,500 3,600 870 270 7 1.75 8,000 186 Labor charges (say Z5% of total labor is hired) 0.25 x 10 = NRS 465 Interest on workirg capital - 0.14 x 1/2 x 2,600 - NRS 180 Water charges (NRs bO ha per crop) 100 x 1.75 - NRe 175 Income from crop cultivation: 8,000 - 405 - 180 - 175 w NRA 7180 B) Future without Project (all crops rainfed) Crop Area Yield Price Gross Return per ha Input Cost per ha Net Return Total Labor (ha) (ha) (NRS/t) (NRS) (NRa) (NRP) (MD) Paddy HYV 0.35 2.1 2,300 4,t30 1,380 1,210 43 Paddy local 0.60 1.5 2,300 3,450 940 1,500 63 Wheat 0.35 1.2 2,200 2,640 920 600 28 Pulses 0.10 0.6 4,000 2,400 610 180 4 Oilseeds 0.05 0.5 4,500 2,250 450 90 3 1.45 3,580 141 Labor charges (say 25% of total labor hired) 0.2 x 10 x 141 NRs 353 Interest on Working Capital: 0.14 x 1/2 x 1,450 = NRes 102 Income from crop cultivation: 3,580 - 353 - 1u2 NRs 3,125 C) Incremental Income from Cron Culltivattnn NRa 7180 - 3125 - 4055 Increase: +130% NEPAL BIRGANJ IRRIGATION PROJECT Summary of Tubewell Performance 1/ 1) Existing Tubewells (n-14) 2) New Tubewells (n-14) Nominal CCA: 1112 ha (average 79.4 ha per vell) Nominal CCA: 14 x 120 ha z 1680 ha Total Discharge: 955 1/sec (average 68.2 I/see per well) Total Discharge: 14 x 80 1tsec : 1120 Ifsec Year 77/78 78/79 79/80 80/81 81/82 79/80 80/81 81182 Kh Rb Kh Rb Kh Rb Kh Rb Kh Rb Kb Rb Kh Rb Kh Rb Area irrigated (ha) 323 201 249 (226+) 165 234 241 214 (78+) (944) 231 371 210 269 (10+) (16+) (No. of wells recorded) (10) (9) (12) (11) (11) (10) (10) (9) (4) (4) (11) (11) (10) (10) (1) (1) No. of wells operational 10 9 12 12 11 10 10 9 8 8 11 11 10 10 8 8 Annual Hours of operation NA (8351+) 8913 8632 NA (5274+) (6756+) NA (No. of wells recorded) -- (11) (10) (10) -- 9 (10) -- Average Annual Hours of Operation NA 759 891 863 NA 586 676 NA for recorded wells Water pumped per ha irrigated NA 4:07 5250 4509 NA 3081 4062 NA (gross) recorded wells (m'/ha) Water pumped per ha of NA (1912+) 2070 1983 NA (904+) (1158) NA nominal CCA (m 3/ha) 1/ Compiled from Table in Annex. - 94 - NEPAL TABLE 22 BIRGANJ IRRIGATION PROJECT OPERATION AND MAINTENANCE EQUIPMENT PROCURED Description No. Purchased Angle Dozer with Loader 1 Angle Dozer with Backhoe 1 Motor Grader 1 Concrete Grouting Equipment I Portable Vibrating Roller 2 Submersible Pump 6 Tamper 4 Rammer 2 Motor Cycle 10 Telephone System I lot Instruments for Meteorological Observation 1 lot Instruments for Water measurements 1 lot Equipments & Tools for Workshop 1 lot Inspection vehicles ( i) Long Chassis 4 wheel Drive 4 ( ii) Regular Chassis, 14 4 wheel Drive (iii) Pick-up truck 3 Spare Parts for Vehicles 1 lot - 95 - TABLE 23 NEPAL BIRGANJ IRRIGATION PROJECT Assessment and Collection of Water Charges A) Surface Water Scheme - Block IB only Year Area irrigated Rate per ha Assessment Collection (ha) (Rs) (NRS'000) (NRs'000) 1977/78 1068 98 104.7 6.9 1978/79 3365 100 334.9 3.7 1979/80 3066 100 305.6 1.8 B) Tubewell Scheme 1977/78 109 98 10.7 1.3 1978/79 155 98 15.3 6.0 1979/80 1007 98 98.8 73.0 Source: Narayani Zone Irrigation,Development Board NEPAL BIRGANJ IRRIGATION PROJECT Construction Schedule and Actual Progress WORK ITEMS 1973 1974 1975 1976 1977 1978 1979 1980 1981 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 CREDIT EFFECTIVE PROJECT HEADOUARTERS + 5 MEM m a SURFACE IRRIGATION SCHEMES NECI(MPROVEMENT wlM A BLOCK IA (3110 hal -A Wa BLOCK IB w A sg BLOCK II 12 300 ha) w- W-A BLOCK III (2 500 ha) WA - F- BLOCKI 12 100 ha V- a* es ag*es BLOCK V (2,570 ha) BLOCK VI 13 400 hal GROUNDWATER SCHEME TUBEWELL DRILLING - mi * .@eg PHASE ] CIVIL WORKS 11,050 ha) PUMPS ELECTRICAL INSTALLATION s PHASE II CIVIL WORKS I1 680 hal w mW PUMPS ELECTRICAL INSTALLATION LEGEND WeAm CONSTRUCTION SCHEDULE AS PER JANUARY 1974 ACTUAL CONSTRUCTION PROGRESS s..age EXPECTED PROGRESS IN 19808 DRY SEASON World Bank-22798 히