Nepal - Mahakali Irrigation Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No. 9022 PROJECT COMPLETION REPORT NEPAL MAHAKALI IRRIGATION PROJECT (STAGE I) (CREDIT 1055-NEP) SEPTEMBER 28, 1990 iculture Operations Division ntry Department I a Regional Office it has a restricted distribution and may be used by recipients only In the performance of duties. Its contents may not otherwise be disclosed without World Bank authorization.  CURRENCY EQUIVALENTS 1 Nepalese Rupee (NRs) - US$0.04 NRs 28.0 - U$$1.00 NRs 100 - US$3.57 WEIGHTS AND MEASURES EQUIJALENTS 1 m - meter (3.28 feet) 1 m3/sec - cubic meter (35.3 cubic ft) per second 1 km = kilometer (0.62 mile) 1 ha = hectare (2.47 acres) 1 liter - 61.02 cubic inches M million 1 ton = 2,205 pounds (lbs) ABBREVIATIONS AND ACRONYMS CFRID - Committee for Forest, Resettlement and Irrigation Development in Kanchanpur District DCA - Development Credit Agreement DIHM - Department of Irrigation, Hydrology and Meteorology (renamed the Department of Irrigation 12/87) DOA - Department of Agriculture DOI - Department of Irrigation EC&D - Engineering Construction an: Design ERR - Economic Rate of Return HMGN - His Majesty's Government of Nepal ICB - International Competitive Bidding IDA - International Development Association LCB - Local Competitive Bidding LRD - Land Revenue Department MIDB - Mahakali Irrigation Development Board MIP - Mahakali Irrigation Project (Stage I) PCR - Project Completion Report RWS - Rotational Water Supply SAR - Staff Appraisal Report UNDP - United Nations Development Program WUG - Water Users' Group FISCAL YEAR July 16 - Jutly 15  FOR OFFICIAL USK ONLY THE WORLD BANK Washington. D.C. 20433 U.S.A. Office of D4tarvCmffatal Op.atmu tvahatKm September 28, 1990 MMORAIDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT SUBJECT: Project Completion Report on Nepal Mahakali Irrigation Proiect (Stage I1) (Credit 1055-NEP) Attached, for information, is a copy of a report entitled "Project Completion Report on Nepal - Mahakali Irrigation Project (Stage I) (Credit 1055-NEP)" prepared by the Asia Regional Office with Part II of the report contributed by the Borrower. No audit of this project has been made by the Operations Evaluation Department at this time. Attachment This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its content may not otherwise be disclosed without World Bank authorization.  FOR OFFICIAL USE ONLY NEPAL MAHAYALI IRRIGATION PROJECT (STAGE I) (CREDIT 1055-NEP) PROJECT COMPLETION REPORT Table of Contents Page No. Preface . ................................................................ i Evaluation Summary ..................................................... iii PART I. PROJECT REVIEW FROM IDA'S PERSPECTIVE ........................ 1 Project Identity ..1........................ Background ........ I Project Objectives and Descriptio .. 2 Objectives .................................................. 2 Description ........................... 2 Project Implementation .................................... 3 General .................................... 3 Revision .................................... 3 Construction .................................... 3 Procurement .................................... 3 Agricultural Extension and Training ..... .......... . 4 Fellowships ........................... 4 Reporting ......... ............... 4 Institutional Performance ........................ 4 Coordination: Land Clearing and Settlement ................ . 4 Project Management .. ....... 5 Water Management ......................... 5 Water Users' Groups (WUGs) ................................ . 5 Cost Recovery ........................... 6 Project Results ....................................... 6 Agricultural Production ................................... . 6 Employment and Farm Incomes ............................... . 6 Economic Rate of Return ................................... . 7 Project Sustainability ......................... 7 Bank Performance ......................... 7 Borrower's Performance .................................... . 8 Project Relationship ...................................... . 8 Consulting Services ...................................... . 8 Project Documentation and Data ..... .............. 9 PART II. PROJECT REVIEW FROM BORROWER'S PERSPECTIVE ................... 10 Evaluation of Performance ................. .. .................... 10 Bank's Performance ............................ 10 Borrower's Own Performance ..................................... . 10 Lessons Learned ............................ 10 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.  PART III. SUMMARY OF STATISTICAL DATA Table lt Related IDA Credits ........ . .............................. 12 Table 2: Project Timetable ......... ................................ 13 Table 3: Cumulative Disbursements .13 Table 4: Physical Achievements .14 Table 5: Project Cost and Financing .14 Project Costs .14 Project Financing .15 Table 6: Project Results .15 Direct Benefits .15 Economic Cost and Benefit Streams .16 Studies .16 Table 7: Compliance with Credit Covenants .17 Table 8: Staff Input ..................... . .................. 17 Table 9: Mission Data .18 Table 10: Reassessment .19 Table 11: Physical Works, by Block .19 Table 12: Pilot Scheme Buildings .19 Table 13: Water User Group Formation .20 Table 14: Cost Recovery Performance .20 Table 15: Incremental Crop Production .21 Table 16: Cropping Patterns and Yields ............................... 21 Table 17: Incremental Benefits to Farm Families .22 Table 18: Incremental Agricultural Employment .22 Map IBRD 14899R  - i - NEPAL MAHAKALI IRRIGATInN PROJECT - STAGE I (CREDIT 1055-NEP) PROJECT COMPLETION REPORT PREFACE This is a Project Completion Report (PCR) for the Mahakali Irrigation Project, Stage I (Cr. 1055-NEP) in Nepal for which a Credit of US$16.0 million was extended to His Majesty's Government of Nepal (HMGN) in September 1980. The Credit was closed December 31, 1988. The undisbursed balance of the Credit, US$27,042.98, was cancelled on July 17, 1989. Information sources used for the PCR include the Staff Appraisal Report (SAR), Development Credit Agreement (DCA), Supervision Reports, correspondence with the Borrower, Quarterly Project Progress Reports as submitted by the Borrower, and internal Bank memoranda on project issues as contained in relevant Bank files. Parts I and III of this PCR had been sent to the Berrower for their review and comments. IDA received their comments in January 1990 and incorporated them in the PCR. The Borrower contributed Part II of the report. The PCR was prepared by the Agriculture Division, Asia Technical Department. A specific field trip in connection with the PCR was not made.  - iii - NEPAL MAHAKALI IRRIGATION PROJECT (STAGE I) (CREDIT 1055-NEP) PROJECT COMPLETION REPORT EVALUATTON SUMMARY i. Objectives. The Mahakali Irrigation Project (Stage I) was to improve and extend an existing irrigation system on the far western Terai. The objectives of the project were to: (a) improve the reliability of water deliveries, in the quantities required, to farmers within the existing system (3,400 ha); (b) extend irrigation facilities to an additional 3,200 ha; (c) improve communications and operating efficiency by providing all-weather roads; and (d) accelerate agricultural development and thus increase farm incomes. The project was designed to: (a) improve and extend irrigation and drainage works; (b) construct access roads, model farm and necessary buildings; (c) strengthen agricultural extension activities; and (d) provide fellowships for Nepali students to study engineering and agriculture. ii. Imr]ementation. Initially, project implementation lagged far behind schedule. There were changes in project scope and delays in procurement of a main civil works contractor. By January 1984, the project was two years behind schedule. In late 1984, the project was reassessed and the project areai was reduced from 6,600 to 4,600 ha (Table 10) without reducing the estimated project costs. The area reduction resulted from: (a) a 1981 Government ban on forest clearing and (b) errors in the maps used at appraisal. The ban on forest clearing precluded clearing of over-mature trees, brush and stumps from about 900 ha. The mapping error resulted in an additional 1,100 ha reduction in project area. The fellowship component was removed from the project ,t reassessment. HMGN had resisted the use of credit proceeds for training, and difficulties had been encountered in placing students in appropriate institutions. Through a grant to HMGN, UNDP assumed responsibility for a fellowship program in engineering and agriculture. iii. After a civil works contract was signed in late 1984, implementation proceeded rapidly, and civil works were completed by June 30, 1987. Because of the delay in award of contract, however, the project closing date was extended twice. It was further extended through December 31, 1988, to allow formation of a Water User Group (WUG) in the command area, and to provide continuity of agricultural extension activity until the November 1988 effective date of Cr. 1924-NEP (for the follow-on Stage II project). iv. The project was unable to achieve physical targets set at appraisal, but all revised targets were fully met (Table 4), and the civil works constructed are of internationally acceptable quality. The pilot farm has accomplished much in the areas of agricultural training and cropping trials (para. 13). The Project Organization performed well and developed strong institutional capacities under Stage I (para. 18). A most noteworthy characteristic of this project is the level of cost recovery achieved. Water charge collection rates in the  - iv - Mahakali Irrigation Project command area are the highest of any HMGN-managed irrigation system (para. 23). v. Results. The primary results of this project are: (i) increased water management efficiency; (ii) increased cropping intensities; (iii) increased crop yields; and (iv) increases in incomes as well as in training and employment opportunities. Installation of water flow control structures and organization of WUGs have resulted in increased water management efficiency (paras. 20-21). vi. Average cropping intensities in the command area increased by 40% and average yield of improved paddy and wheat rose 48% and 64%, respectively, between 1983 and 1989 (Table 15). Programs designed to train farmers in optimal cultivation practices were instituted under Stage I. and contributed to the large increases in agricultural production. Farm incomes rose by a minimum of 100%, (Table 6(a)). Agricultural employment in the project area rose by 76% (Table 18). Both of these increases are higher than was projected at appraisal. The appraisal estimate of the economic rate of return (ERR) was 14%; the estimate at the time of reassessing the project was 17%. The present estimate of the ERR is 20%, which is particularly impressive in light of the substantial reduction in project scope. vii. Sustainability. The achievements of this project have had substantial positive impact on agricultural development in the project area. Two factors will contribute significantly to continuing the project's objectives after closing: (a) the institutional development that occurred during implementation (which is continuing under Stage II); and (b) the high rate of water charge collection efficiency. viii. Findings and Lessons. The project achieved its goals even though the physical targets were revised during project implementation due to changes in both the scope of the project (para. 10) and the procurement mode for the main civil works contract (para. 12). The weaknesses idencified in project implementation were: (a) lack of structured incentives for farmers to be more involved in the project cycle from an early stage; for example, formation of effective WUGs and construction of field channels by farmer groups were less effective before a timely and reliable water supply was provided (para. 22); and (b) inadequate project preparation as evidenced by the mapping error (para. 10) and lack of a realistic assessment of the relative feasibility of Local Competitive Bidding (LCB) and International Competitive Bidding (ICB) civil works contracts (para. 12). The lessons derived from this experience have been incorporated in the formulation of the follow-on project.  NEPAL MAHAKALI IRRIGATION PROJECT (STAGE I) PROJECT COMPLETION REPORT PART I. PROJECT REVIEW FROM IDA'S PERSPECTIVE Proiect Identity Name : Mahakali Irrigation Project (Stage I) Credit No. : Cr. 1055-NEP RVP Unit : Asia Region Country : Nepal Sector : Agriculture (Subsector, Irrigation) Background 1. At the time of project appraisal in 1979/80, farm production and related activities comprised more than 60% of Nepal's GDP. The sector engaged almost 90% of the labor force and provided more than 80% of all exports. Of the 24,500 sq km (17% total land area) available for cultivation, at most 15% was irrigated. About 90% of the cropped area was sown with foodgrains, which then largely determined agricultural growth. Duri.ng the period 1970/71 to 1977/78, the trend in production of major foodgrains, except wheat, was rather disap- pointing. Production grew at about 1.1% annually due to expansions in acreage, while yields, except for wheat, fell at a rate of about 0.8% per year. Cash crops, however, fared much better partially due to favorable prices. Production growth in cash crops, too, was derived from expanded cultivation and not from increasing yields. 2. In the late 1970s, HMGN strategy was to develop the Terai for generation of domestic and exportable foodgrain surpluses as a resource for further development. Since possibilities for area expansion are limited in the Terai, emphasis was given to increasing yields, mainly by expansion and upgrading of existing irrigation infrastructure and inputs. Initial results from the first IDA-assisted irrigation project in Nepal - Narayani Zone Irrigation Project (Stage I) (Cr. 373-NEP) - indicated that substantial yield improvements could be expected from improved irrigation water supply combined with increased agricultural inputs and effective extension services. HMGN's strategy stressed diversification into cash crops whose performance had been promising and whose marketing prospects were good. 3. The Bank Group's strategy in Nepal placed major emphasis upon the directly-productive sectors (particularly agriculture) and the development of complementary infrastructure. Since IDA's first involvement with irrigation in Nepal in 1973, there has been continuing IDA assistance to the irrigation s^ubsector in Nepal (See Part III, 1). The main objective of Bank involvement in irrigation in the Terai was to improve the reliability of water distribution to the farm through both technical improvements and more effective organization of water users. 4. Improvement of the existing Mahakali irrigation system in the Far Western Region of Nepal was initiated by HMGN in the early 1970s. In 1976, the Mahakali Irrigation Project (MIP) was identified by IDA and in 1978, its preparation for IDA finance was undertaken by an international consulting firm.  At appraisal in January 1980, the system had three major problems: (a) there was severe siltation; (b) the number of tertiary outlets and the length of tertiary canals were insufficient for proper water d'istribution; and (c) field chalnnls serving individual farms were lacking. 5. The SAR estimated total project costs for system improvement and extension at US$19.5 M. Negotiations were held in Washington, D.C. on May 20- 23, 1980. An .DA Credit (1055-NEP) in the amount of US$16.0 M (15.1 M SDR equivalent) was approved by the Board on June 9, 1980. The DCA was signed September 29, 1980, with an effective date of December 30, 1980. The project had a five-year implementation period, with a completion date of June 30, 1985. Project Objectives and Description 6. Objectives. The objectives of the project were to: (a) improve the reliabif.ty of water deliveries to farmers within the 3,400 ha existing system; (b) excend irrigation facilities to an additional 3,200 ha; (c) improve communications and operating efficiency by provic.ing all-weather roads; and (d) accelerate agricultural development and thus increase farm incomes. To accelerate agricultural development, the project was also designed to: (a) improve water control through installation of additional control structures, and through the extension of the distribution system to outlets serving 7-8 ha parcels of land; (b) intensify agricultucal extension and training; and (c) develop technical staff by providing fellowships for Nepali students. 7. Description. Nine project components were specified in the SAR: (a) improvements to the existing irrigation and drainage system on about 3,400 ha; (b) extension of irrigation and drainage systems by 3,200 ha to 6,600 ha, including about 30 km of new distributaries and 115 km of tertiary canals; (c) construction of an interconnected network of all-weather public roads (about 40 km) and canal service roads (about 65 km); (d) construction of buildings for engineering and agricultural activities and staff quarters; (e) strengthening agricultural extension, research and training activities, including construction of a pilot demonstration farm-cum-training center; (f) procurement of materials, equipment and vehicles for construction, survey and laboratory activities and project operation and maintenance; (g) provision of technical services, including foreign and local consultants for project implementation and aerial photography and mapping; (h) fellowships for Nepali students to study civil and mechanical engineering and agriculture; and (i) undertaking agricultural development activities in the Bhairawa-Lumbini Groundwater Project (Cr. 654-N'P). 8. Consistent with the development plan for Kanchanpur District, nmGN assumed responsibility for two activities integral to the MIP. First, the Timber Corporation of Nepal, by authority of the Forest Products Development Board, was to complete clearing work on about 2,200 ha of over-mature trees, brush and tree stumps from within the project area. Second, in order to unite two sections of the Suklaphanta Wildlife Reserve and reduce river bank erosion, HMGN planned to move farm families living along the Mahakali and Chaudar Rivers to the project area. Responsibility for moving the farmers rested with the Resettlement Department of the Ministry of Forests. Both activities were to be coordinated through the Committee for Forests, Resettlement and Irrigation Development in Kanchanpur District.  Proiect Implementation 9. General. The Credit became effective February 27, 1981. Initially, project implementation lagged behind schedule. This was mainly due to changes in project scope and delays in procurement of the main civil works. Actual project area available for development was lower than estimated at appraisal for two reasons. First, a 1981 Government ban on all forest clearing precluded clearing the wildlife reserve area originally included in the project. Second, the maps used at appraisal proved to be inaccurate when checked against fully- controlled project mapping completed in 1982. By Janu<y 1984, the project was two years behind schedule; the contract for civil works construction had yet to be finalized, and disbursements were low: only 10% of the Credit had been disbursed, as compared with an appraisal estimate of 70%. After the signing of a civil works contract in September 1984, however, implementation proceeded rapidly. Following reassessment (para. 10), and with two extensions of the closing date, civil works were completed by June 30, 1987, si.x months ahead of the agreed contract schedule. A third extension through December 31, 1988 allowed: (a) completion of WUG formation in the Stage I area; (b) continuity of agricultural extension activities through the model farm and training programs; and (c) finalization of preparation and design work for Stage II (Cr. 1924-NEP, which became effective 11/18/88). The project was conmpleted without cost overrun because: (a) the ci'il works contract was let at 78% of the engineer's estimate; and (b) funds initially allocated to fellowships were re-allocated to the civil works category. 10. Revision. In light of implementation delays and reductions in scope, the project was reassessed in February 1984. Reassessment was carried out to revise the project scope, and to determine the economic viability of a reduced- scope project. The forest clearing ban and mapping error resulted in reduction of overall project area by about 2,000 ha net (900 and 1,100 ha, respectively). The amcnded physical targets reflected the 30% reduction in overall project area (Table 10). Work on improvement to the main canal including the construction of sediment ejectors, was dropped because the original design capacity was adequate to del.Jver the required water to the Stage I area. It was determired that the project was still viable (estimated ERR of 17%), and an amendment to the DCA revising physical targets was issued by IDA in October 1984. The amendment was formally agreed to by HMGN in November 1985. 11. Construction. Although the project was unable to meet the initial physical targets set at appraisal, all revised targets were fully met. Both the area of extension and the number of pilot scheme buildings constructed exceeded targets (Table 11). 12. Procurement. Initially, there was some difficulty in securing a contractor for main project works. It was agreed at appraisal that LCB would be utilized for the civil works contract. However, there were no suitable responses to the LCB tender and it became clear that local contractors possessed- neither the expertise nor the experience. In mid-1982, a decision was made to use ICB. Contractor performance was satisfat-tory, and works were constructed to internationally acceptable quality. All pilot facilities were constructed by local contractors with satisfactory results. 13. Agricultural Extension and Training. Agricultural extension activities under the MIP have been important to project success. This component was thus  expanded and strengthened under the follow-on projtct (Cr. 1924-NEP). The pilot farm was established in 1982, but only became fully operable in mid-1986. At IDA urging, a full-time agricultural consultant was hired in May 1986 to guide trai-aing and cropping trials. Performance of the pilot scheme is noteworthy. Cropping intensity in the 13.2 ha farm is very high (225%) and yields of major crops exceeded appraisal estimate. A vetiver grass nursery was started in the command area in 1988 to aid soil and moisture conservation work. Between June 1986 and March 1989, training sessions totalling 224 course-days11, with a total attendance of 1,848a', were held at the pilot farm. 14. Many farmers in the project area had inadequate knowledge of water manag, .ent and irrigated farming practices. Work on the pilot farms and dissemination of the results were thus important factors in improving farming practices in the commend area. Courses in vegetable growing techniques, important because of HMGN's emphasis on diversification into cash crops, gene-ated an unanticipated level of interest, especially by women. While women constituted only 8% of total participants in training sessions not related to vegetable-growing, they comprised 94% of the participants in vegetable-growing courses, 15. Fellowships. A major constraint to implementation of irrigation projects in Nepal has been the shortage of well-trained technical personnel. Fellowships were thus an important part of the project. 1V4GN agreed during credit negotiations to implement a fellowship program. IHowever, shortly after the credit effectiveness, 1'4CGN established a policy of not using Credit proceeds for training, and was reluctant to start the program. The Bank emphasized the importance of the fellowship component, but attempts to arrange for the requisite number of university placements were not successful. In 1983, through a grcnt to HMGN, UNDP assumed responsibility for a fellowship p.ogram in engineering and agricultural sciences. At project closing, UNDP had funded 47 engineering and agricultural fellowships for Nepali students. 16. Reporting. As with other projects in Nepal, the performance of HMGN in fulfill'ng the IDA audit and accounting requirements was weak. The Bank repeatedly stressed the importance of timely reporring, but the required submission of reports within six months after the end of the fiscal year did not prove to be possible. The DCA was revised in 1988 to extend the submission of accounts and audits to 12 months after the end of each fiscal year, while still requiring unaudited accounts to be submitted within six months of the fiscal year end. Institutional Performance 17. Coordination: Land Clearing and Settlement. Land clearing and settlement functions proved to be problematic throughout the project period. The Committee for Forests, Resettlement, and Irrigation Development (CFRID) in Kanchanpur District did not function and was replaced by a higher committee from the Land Settloment Department in December 1985. In spite of this, both tree 1/ Length of duration ranges from one to thirty days. VI Includes farmers (men and women), lead farmers, WUG leaders, and field extension workers.  - 5 - clearing and final settlement proceeded slower than anticipated, and clearing was not completed until 1988. The Timber Corporation of Nepal did not have sufficient capacity for the task, and brush and stump removal was slow throughout project implementation. Settlement and assignment of land titles was delayed. At project closing in December 1988, title to 980 ha of land still remained to be assigned although tLe entire command area has been cultivated by settlers. This affected the level of water charge collections (para. 23). 18. Proiect Management. Overall, the project organization performed successfully. The Engineering Construction and Design (EC&D) unit, although not part of project organization at the appraisal stage, functioned particularly well. The unit was formed during civil works design review and cc.struction processes. Project engineers received on-the-job training by working with supervising engineers from the project consultant. The EC&D unit developed into one of the strongest components in project management. 19. Water Management. The project contributed to improved water management in the commaatd area. New and improved physical works constructed under the project, and standardized operation and maintenance procedures both played a role. The two most significant elements, however, were: (a) implementation of Rotational Water Supply (RWS) in blocks III and IV; and (b) formation of WUGs throughout the command area. 20. In 1985, the Bank advocated a redesign of the water distribution system in MIP to allow the use of RWS. The recommendation was based on review of a wide range of irrigation experience in both India and Nepal, and the results of a system design study conducted in India by a Bank consultant. That review prompted a decision to adopt a structured, gateless fixed supply system, which incorporates elements of the warabandi system which has operated successfully in Northwest India. The cost of these improvements was relatively small (US$200/ha) but the improvements in manageability of the system and the reliability of water supply were spectacular. With RWS design, chaks were to be between 20 and 30 ha, as opposed to the 40 ha approximation determined at appraisal. Blocke I and II will be converted to the RWS system as part of the Stage II works. 21. Water Users' Groups (WUGs). WUGs play an important role in dissemination of appropriate farming techniques, as well as canal maintenance and other water management issues. Without effective and functional WUGs, the RW'S system will not be able to achieve design efficiency. The project had a target of forming 915 WUGs; 802 WUGs, or 88%, were achieved by project closing (Annex A, III). The shortfall is mainly due to delays in land settlement and title assignment (para. 17). 22. During appraisal it was agreed that water would only be regularly available to a tertiary unit after: (a) the WUG had been formed for the unit; (b) field channels had been constructed; and (c) satisfactory distribution routines had been established. Implementation proved this condition to be inappropriate, as it is difficult to form WUGs prior to the availability of reliable water deliveries at the farm level. Once water was available on a  -e - reliable basis, however, WUGs become important narticipants in the irrigation system. MIP Stage II has taken account of this._' 23. Cost Recovery. The DCA required HMGN to recover from beneficiaries: (a) the full cost of operation and maintenance; and (b) in progressive steps over a reasonable period of time, the investment cost of the project. At appraisal, annual charges were levied on each crop sown (NRs 60/crop/ha). In 1986, this system was changed to a flat rate (NRs 100/ha/year). The rate was insufficient to allow recovery of maintenance costs. At IDA urging, it was raised to NRs 200/ha/year for 1987 collections, which is still insufficient to cover operation and maintenance cost. Actual irrigation operation and maintenance cost was estimated at NRs 280 per ha per year. 24. Water charge collection efficiency for the MIP is the highest of all HMGN-managed surface irrigation projects. For fiscal year 1987/88, a collection rate of 78Z was realized (Table 14). This performance is particularly impressive given the fact that cost recovery under the project was impeded by: (a) the rate of settlement and land title assignment; and (b) weak linkages in the assessment/collection system. Measures were taken to correct both problems. Supervision missions emphasized the importance of timely settlement and title assignment. By project closing all land was settled, although land titling in approximately 30? of project area land was outstanding. As collection procedures are based on lists of currently registered land owners, untitled and newly settled lands are not captured in the water charge process. Project Results 25. There are four primary results from the project: (a) increased cropping intensities; (b) increased yields; (c) increases in incomes as well as in training and employment opportunities; and (d) increased water management efficiency. Increased efficiency of water management will be important over the long run, as efficient uw - of water resources is crucial to HMGN's policy of agricultural intensifica on. 26. Agricultural Production. An agricultural impact study carried out in 1989 revealed that production in the project area was 89? higher than SAR projections of production without the project (Table 15). Both cropping intensity and average crop yields rose in the project area. Average cropping intensity increased by 40Z, and average yields of padly (improved), wheat and oilseeds rose by 48?, 64? and 402, respectively (Table 16). These increases are attributable inter alia to successful implementation of the farmer training program at the pilot farm assisted by the agricultural consultant on irrigated farming and on-farm water management (paras. 13, 14 and 19). Estimates of future production in the project area (C992/93) are based on yield levels already obtained by many ordinary farmers in the project area. 27. Employment and Farm Incomes. The project created construction and agricultural employment. Short-term employment of 1,050 laborers during construction generated current benefits, but the agricultural employment will have longer-term impacts. The agricultural labor requirement in the command 3/For example, intensive training will be focused on WUG leaders who will teach/demonstrate water management techniques to other farmers in their respective groups.  area in 1988/89 was 76Z, or 322,100 man-days higher than in 1982/83 (Table 18). The most recent estimate of farm family on-farm incomes is US$130 per year for small farms, and an average of US$192 for all three farm sizes (Table 6). This represents a 10OZ and 192Z increase, respectively, over incomes estimated at appraisal. The average increment to net benefits is currently estimated to be NRs 591 per family member per annum (Table 17). 28. Economic Rate of Return (ERR) . The appraisal estimate of the ERR from this project was 142, assuming five years for command area build-up, and seven years for yield build-up. Although initially there were concerns that a reduced-scope project would not be economically viable, the 1984 revised estimate was a 17X ERR, or 13Z assuming ten years (instead of five) for command area build-up, and seven years (instead of five) for yield build-up. The project ERR at PCR preparation time was estimated to be 20X, higher than projected under the most optimistic scenario at reassessment (Table 6(b)). This rate of return is particularly impressive in light of the reduced project scope, and is largely attributable to the high levels of incremental agricultural production achieved in the project area. The increase in crop production is mainly attributable to a reliable irrigation water supply. This created confidence and motivation for farmers to participate in farmer training conducted at the pilot farm, and to follow guidance provided by extension workers. The project management throughout the project period, conscientiously supported all aspects of agricultural activities. Project Sustainability 29. The project had a significant positive impact on agricultural development and farm incomes. The institutional strengthening that took place during implementation (paras. 13 and 18), particularly the establishment of effective WUGs, facilitated sustained project operations under public sector management with increased users involvement. Strengthening bo.h the pilot farm and the Project Authority under the Stage II project will ensure that local institutions are in place to maintain the original project momentum. The high level of water charge collection efficiency (para. 24) should also aid the Stage II project. Bank Performance 30. HMGN started preparing the project as early as 1977, with technical assistance from expatriate consultants. The project was not prepared in sufficient detail, which resulted in implementation delays (paras. 9-10). The SAR identified the key project components required to overcome deficiencies in the existing systems and confirmed the need for the installation of sufficient control structures to ensure efficient water delivery. The SAR also outlined a well-balanced development plan for overcoming constraints that could otherwise impede improvement of the agricultural performance. Eighteen supervision missions visited the project at an average interval of six months. Consistent Bank staff attention to the proiect was apparent in the supervision reports, and in the recommendations contained therein. Bank staff concentrated not only on assuring compliance with Credit covenants, but also on ensuring the quality of project implementation. The latter is evidenced in recommendations such as a shift to RWS and preparation of a crop manual for the command area. The following lessons may be learned from the project:  - 8 - (a) Appropriate farmer motivation and incentives are crucial to project success. Formation of WUGs is not possible until reliable delivery of irrigation water has occurred. It is important to form effective WUGs in time for farmers to be involved in planning and design at the tertiary level. This early involvement results in a greater sense of ownership, and a higher level of participation by farmers. (b) The following weaknesses encountered under the project: (i) deficiencies in project preparation as evidenced by the mapping error, which resulted in project revision (para. 10); (ii) overoptimistic implementation plan which resulted in extension of the closing date by three years; (iii) inadequate assessment of the relative feasibility of LCB and ICB civil works contracts (para. 12). The latter two can, however, be related to the short experience, at appraisal time, with implementation of large civil works in Nepal. Borrower's Performance 31. Except for the initial delays due to changes in the project scope and the mode of procurement for the main civil works contract, the project authority has been successful in implementing the project works. Table 7 contains the review of performance of the Borrower. Except for delays in submitting audited accounts and audited reports, the Borrower's compliance with the Credit covenants was satisfactory. During project preparation and appraisal, the project authority provided all the support needed. After project revisions in late 1984, project organization was considerably strengthened. The revised organization played a major role in assuring the project's successful implementation of the project. Project Relationships 32. Bank relationships with HMGN, local government and the Project Authority have been good. Consulting Services 33. The project consultants employed by the Mahakali Irrigation Development Board (MIDB) provided in-service training, advice and guidance to project engineering staff in all cycles of the project implementation, as well as preparation of bid documents and evaluation. Their performance was exceptionally good, particularly for engineering supervision, developing an irrigation system operation manual for the command area, publishing a crop manual for the farmers, and guiding the field staff in conducting the impact study of the project. It is noteworthy that the consultants have maintained a good working relationship with the Project Authority and project management, and also worked closely with the beneficiaries. The stu.ccess of the project implementation could not have been attained without the quality of consultant inputs that were provided during the project.  - 9.- Project Documentation and Data 34. Although there were some weaknesses in preparation (para. 30), preparation and appraisal concentrated on the appropriate issues. The SAR provided a useful framework for the Project Authority to implement the project, and for Bank review of project progress. As stated earlier, the submission of the project accounts and audits within six months after the end of each fiscal year (para. 16), and the formation of WUGs and construction of field channels prior to release of irrigation water (para. 30a) stipulated in the DCA were actually difficult for HMGN to comply with. The data relevant to the preparation of the PCR were readily available.  - 10 - PART II. PROJECT REVIEW FROM BORROWER'S PERSPECTIVE Evaluation of Performance Bank's Performance 35. Consistent IDA attention to the project contributed to its success. While the supervision mission persoiinel were highly qualified in their respective disciplines, absence of an agriculturalist on early missions was not optimum from a project perspective. Despite initial delays in implementing civil works, the project area already under cultivation could well have benefited from agricultural expertise. This was rectified in later IDA missions, after 1984. The project appreciated the prompt response and quick action of IDA in helping solve problems encountered during the implementation, and providing suggestions for improvement of the system. Without this level of support the project would not have been able to achieve its revised targets. Borrower's Own Performance 36. Overall performance of institutions participating in the project was satisfactory. During project implementation, the project organization developed into a strong and competent institution. Coordination of project management with other institutions was good. To achieve the revised targets, project staff, consultant and main contractor coordinated their efforts t) work as a single team. This enabled the civil works to be completed six months ahead of schedule, to international quality standards. To achieve the level of collection of water tax, the project staff have coordinated their efforts with the Land Revenue Department (LRD), and have convinced farmers to pay the charges. A major factor in convincing the farmers has been the reliable supply of irrigation water. The agricultural impact to date has been ahead of the targets, as detailed in the Agricultural Impact Study Report. This is due to the level of agricultural training, extension and demonstrations at the pilot farm. Production of vetiver grass is underway, as suggested by IDA: vetiver will help in developing low cost, appropriate protection for drains and canal embankments. Lessons Learned 37. In Nepal, any sizeable construction contract should be let under ICB procedures to ensure quality, price and timely completion. However, to assist the development of the local contracting industry, maximum use of local firms as sub-contractors should be encouraged so that local contractors will be able to rapidly gain the necessary experience. 38. Participation of the farmers to the extent possible in the planning, design and implementation of the project is vital to the achievement of the objectives. In Stage I, farmer participation was initially minimal, and certain problems arose, which would not have arisen if farmers had actively been involved at all stages. Farmer participation in planning, design and implementation is an integral part of Stage II. 39. Reporting requirements should be reasonable within the domestic administrative context; requirements which conflict with local practices and timetables should be avoided if possible. Agreement to unrealistic covenants  - 11. - in the DCA for the submission of project accounts and audits to IDA can distract supervision and project management attention from more urgent issues directly related to project implementation.  - 1.2 - PART III. SUMMARY OF STATISTICAL DATA Table 1: Related IDA Credits Credit Number/ Year of Purpose of Project Title Approval Project Status 373-NEP 1973 To complete and upgrade 28,700 ha Completed. Followed by Birganj Irrigation of the Nepal Eastern Canal Stage II (Narayani Zone) command area and to provide tubewells north of the canal 617-NEP 1976 One component provided for Completed First Rural improvements to small-scale Development irrigation schemes in the Hills 654-NEP 1976 Development of 64 public tube- Completed. Followed by Bhairawa-Lumbini well systems and strengthening of Stage II Stage I agricultural support services in the western Terai 812-NEP 1978 Improvement of Chatra Canal Completed. Followed by Sunsari-Morang command area, including drainage, Stage II Stage I river training and sediment control works (12,000 ha) 856-NEP 1978 Follow-on to Birganj Project, Completed. Followedby Narayani Stage II completed deferred work Stage III 939-NEP 1979 Included a component for Completed Second Rural improving hill irrigation in Development the Mahakali Zone 1101-NEP 1981 Improvement of small farmer In progress. To be Hill Food managed schemes totalling completed in 1990 Production about 3,000 ha 1316-NEP 1983 Completion of Stage I works Nearing completion. To Bhairawa-Lumbini and introduction of improved be followed by Stage II designs for construction of Stage III 38 new tubewell systems 1715-NEP 1986 Rehabilitation of early blocks In progress. To be Narayani Stage III for improved (i) water management completed in 1994 capacity and (ii) dry season agricultural production  - 13 - Credit Number/ Year of Purpose of Project Title Approval Project Status 1727-NEP 1986 Includes construction of 32 and In progress. To be Third Rural rehabilitation of 18 existing completed in 1993 Development schemes in the Hills 1814-NEP 1987 Follow-on to Stage I, continuing In progress. Will be Sunsari-Morang work on 23,000 ha completed in 1994 Stage II 1924-NEP 1988 Follow-on to Mahakali I, to In progress. To be Mahakali Stage II extend command area completed in 1995. Table 2: Project Timetable Date Date Date Item Planned Revised Actual Identification (Executive Project Summary) 02/75 02/75 Preparation 03/77 03/77-11/78 Appraisal Mission 01/80 01/08-28/80 Credit Negotiations 05/80 05/20-23/80 Board Approval 06/09/80 06/09/80 Credit Signature 09/29/80 09/29/80 Credit Effectiveness 12/31/80 03/31/81 02/27/81 Credit Closing 12/31/85 12/31/88 12/31/88 Credit Completion 06/30/86 06/30/89 06/30/89 Table 3: Cumulative Disbursements FY81 FY82 FY83 FY84 FY85 FY86 FY87 FY88 FY89 Appraisal Estimate (US$M) 0.3 3.8 7.7 11.1 14.6 16.0 - - - Actual (US$ M) 0.1 0.5 1.2 1.7 3.2 5.2 11.5 13.7 16.0 Actual as % of SAR estimate 33 13 16 15 22 33 72 86 100  - 14 - Table 4: Physical Achievements Item Appraisal Revised Actual Achievement as % of Target Target Achieved Appraisal Revised IMPROVEMENTS TO EXISTING SYSTEM: delivery (ha) 3,400.0 2,700.0 2,700.0 79 100 canals (km) tertiary - 35.8 35.8 - 100 distributary - 37.0 37.0 - 100 drainage 200.0 219.0 219.0 110 100 EXTENSION OF SYSTEM: delivery (ha) 3,200.0 1,900.0 2,100.0 66 ill canals (km) tertiary 115.0 170.0 170.0 148 100 distributary 30.0 14.6 14.6 49 100 ROADS: public (km) 40.0 15.0 15.0 38 100 service (km) 65.0 56.0 56.0 86 100 Table 5: Proiect Costs and Financing (a) Proiect Costs (US$ thousand) ADpraisal Estimate Revised Estimate Actual Category Local Foreign Total Local Foreign Total Total Civil Works Vehicles & Equipment 5,421 4,003 9,424 2,662 7,832 10,494 12,872 Project Administration 1,136 159 1,295 720 180 900 68 Technical Services 300 1,218 1,518 865 1,635 2,500 3,033 Fellowships 50 934 984 0 0 0 0 Contingencies 3.140 3.140 6.280 893 2.085 2.978 0 TOTAL 10,047 9,454 19,501 5,140 11,732 16,872 15,973 Total Disbursement 15,973 Credit Amount 16,000 Amount Cancelled 27  - 15 - (b) Project Financing (US$ million) Source Agreement Actual Actual as % of Agreement IDA Expenditure Categories 16.0 16.0 100 Domestic 3.5 3.4 97 TOTAL 19.5 19.4 99.5 Table 6: Project Results (a) Direct Benefits Benefit Actual Estimated at Benefit Item Appraisal Achieved Farm families to benefit: 2,600-2,800 3,100 -/ Changes in irrigated area: (a) improvement to existing system 3,400 ha 2,700 ha (b) increase in command area +3,200 ha +2,100 ha Increase in production: (a) foodgrains +120% +181% (+7,500 tons) (+10,040 tons) (b) all crops +173% +167% Incremental farm and self employment: +53% +76% Average increase in per capita income: Est. from US$65 to US$118 per year +82% Achieved from US$65 to: US$ 130 per year for small farms +100% US$ 190 per year average all farm sizes +192% a/ Agricultural impact study (1989) reveals that there are more smaller farm size than assumed in SAR.  (b) Economic Cost nd anoM tit 5tt*fh (NRa 000) mItts YOAR 2248 2""6 6 164 6146to 266 161641 in?" 166666 I"M66 m0o2 l wa` I 1maa* ISt4 wit PP*JGCt 64.f012 MM Ism o 17426 2146sU 44"6 460t6 4522 u7u 70154 62 228' 284672 *1 s P*j0ct Woltsa 37182 M19 M278 2181 1Lon 1782 2i786 I6S 21021 21324 26128 2136 t._tt~.2.1 AfficwIt'wa 6*1f82 s .1026 .446 14 162 2tM "Its2 216t0 Into 46027 60901 16021 208S64 RTs Cael"I