India - National Water Management Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No. 16231 IMPLEMENTATION COMPLETION REPORT INDIA NATIONAL WATER MANAGEMENT PROJECT (Credit 1770-IN) JANUARY 10, 1997 Agriculture and Water Operations Division Country Department II South Asia Region This document has a restricted distribution and mav be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World bank authorization.  CURRENCY EQUIVALENTS Currency Unit = Indian Rupees (Rs.) Rs. 13.0( = US$1.00 (Appraisal: 1987) Rs. 19.64 = US$1.00 (Intervening years: 1987-94) Rs. 31.65 - US$I .00 (Completion vear: 1995) WEIGHTS AND MEASURES I meter (m) = 3 28 feet (ft) I kilometcr (kin) 0.62 miles (rni) I hectare (ha) = 2 47 acres (ac) I million cubic meters (Mm3) = 804 acre-feet (ac-ft) I cubic foot per second (cfs or cuscc) 0.0283 cubic meters per second (m3/s) I kilogram (kg) 2.2 pounds (lb) I mctric ton (t) 2.205 pounds (lb) I thousand million cubic feet (TMC) = 28.317 million cubic meters (Mm3) FISCAL YEAR OF THE BORROWER Apnl I to March 31 PRINCIPAL ABBREVIATIONS AND ACRONYMS CCA Cultivable Command Area DCA Development Credit Agreement ERR Economic Rate of Return GOI Government of India ICR Implementation Completion Report ID Irrigation Department IIMI International Irrigation Management Institute IMTI Irrigation Management Training Institute ISTF Institutional Strengthening and Training Fund M&E Monitoring and Evaluation MOWR Ministry of Water Resources NWMP National Water Management Project O&M Operation and Maintenance PIM Participatory Irrigation Management RDS Rajolibunda Diversion Scheme SAR Staff Appraisal Report VVS Vani Vilas Sagar WMC Water Management Cell WRCP Water Resources Consolidation Projects Vicc President D. Joseph Wood Director : R. Drysdalc Division Chief/Manager S. Barghouti Staff Mcmber : K. Oblitas. Senior Economist  FOR OFFICIAL USE ONLY IMPLEMENTATION COMPLETION REPORT INDIA NATIONAL WATER MANAGEMENT PROJECT (Credit 1770-IN) Table of Contents P R E F A C E ..............................................................i EVALUATION SUMMARY ............................................................. ii PART I: PROJECT IMPLEMENTATION ASSESSMENT .......................................1 A. Project Objectives I B. Achievement of Project Objectives 3 C. Implementation Record and Major Factors Affecting the Project .6 D. Project Sustainability 8 E. Bank Performance .8 F. Borrower Performance .10 G. Assessment of Outcome .14 H. Future Operation .14 I. Key Lessons Learned .14 PART II: STATISTICAL TABLES ............................................................. 17 Table 1. Summary of Assessments ........................................................... 17 Table 2. Related Bank Loans/Credits ....................................................... 19 Table 3. Project Timetable ............................................................. 19 Table 4. Loan/Credit Disbursements: Cumulative Estimated and Actual ... 20 Table 5. Key Indicators for Project Implementation .................................. 20 Table 6. Key Indicators for Project Operation .......................................... 21 Table 7a. Training and Studies - Andhra Pradesh ......... ...... ....... 21 Table 7b. Training and Studies - Karnataka ............................................... 22 Table 7c. Training and Studies - Kerala ..................................................... 25 Table 7d. Training and Studies - Madhya Pradesh ..................................... 25 Table 7e. Training and Studies - Tamil Nadu ............................................. 26 Table 8A. Project Costs ..... ................. .......... ............. 28 Table 8B. Project Financing ............................................................. 28 Table 9. Economic Costs and Benefits .................................................... 29 Table 10. Status of Legal Covenants ......................................................... 30 Table 11. Compliance with Operational Manual Statements ...................... 33 Table 12. Bank Resources: Staff Inputs ..................................................... 33 Table 13. Bank Resources: Missions ......................................................... 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 wiLhout World Bank authorization.  APPENDICES IA Aide Memoire ................................................... 36 1B Supplementary Mission - Aide-Memoire ................................................... 55 2 Economic and Financial Analysis ................................................... 60 3 Observations of Ministry of Water Resources on the Draft ICR ................ 193 4 Excerpts-MOWR's Completion Report Lessons Drawn from NWMP I .... 202  i IMPLEMENTATION COMPLETION REPORT INDIA NATIONAL WATER MANAGEMENT PROJECT (Credit 1770-IN) Preface 1. This is the Implementation Completion Report (ICR) for the National Water Management Project (NWMP) in India for which Credit 1770-IN in the amount of SDR93.2 million (US$114 million equivalent) was approved on March 24, 1987 and declared effective on August 10, 1987. 2. The Credit was closed on March 31, 1995, one year behind schedule. Final disbursement took place on June 20, 1995 and the Credit was fully disbursed. 3. This report was prepared by an FAO/CP mission which visited India in February- March 1995 and October-November 1995. It was finalized by Mr. Keith Oblitas and Ms. Teresita Estoque, Agriculture and Water Operations, Country Department II, South Asia Regional Office, and reviewed by Shawki Barghouti, Division Chief, and Ms. Kazuko Uchimura, Project Adviser, SA2DR. Preparation of the ICR was initiated during the Bank's final supervision mission in February-March 1995. It is based on a review of the Staff Appraisal Report and legal documents, supervision reports and project files, as well as field investigations and discussions with the Bank staff and officials of the Irrigation Departments of the ten States participating in the project. The mission also visited selected sub-projects in six States (Andhra Pradesh, Haryana, Karnataka, Kerala, Madhya Pradesh and Tamil Nadu) and held discussions with project officers and participating farmers 4. The Borrower's comments on the ICR's February/March 1995 Mission's Aide Memoire and on the draft ICR are in Appendix 3 to this report. A Project Completion Report, prepared by the Ministry of Water Resources, is available in the Asia Information Center.   ii IMPLEMENTATION COMPLETION REPORT INDIA NATIONAL WATER MANAGEMENT PROJECT (Credit 1770-IN) Evaluation Summary Introduction Agriculture continues to be the dominant sector of the Indian economy, contributing about 36 percent of the GNP and directly employing over 70 percent of the total labour force. Foodgrain production accelerated substantially during the 1960s with the advent of the green revolution and the introduction of high-yielding varieties of wheat and rice. The development of irrigation has been a major factor in improving agricultural performance. Irrigation has contributed directly by increasing the area cropped, reducing crop losses due to deficient rainfall, and by encouraging diversified cropping and the use of high-yielding varieties of seeds and other improved inputs. Despite the undoubted achievement of irrigation development, yields in many irrigated areas have been and continue to be well below potential because irrigation supplies are unreliable; in addition, the distribution of water is often inequitable, leading to overuse in some areas and shortages elsewhere. These problems have resulted from inadequacies in both the management of irrigation schemes and the physical infrastructure. The National Water Management Project (NWMP) was, therefore, formulated mainly to improve the perforrnance of existing irrigation systems. Project Objectives The purpose of the project was to increase agricultural productivity and farm incomes through the provision of a more reliable, predictable and equitable irrigation service. To achieve these objectives and facilitate extension of similar activities throughout the country, the project was formulated to help participating States and GOI to develop an institutional capacity to plan, implement and monitor improved O&M practices and to provide for low-cost infrastructural improvements designed to support an improved operational plan for selected schemes. In each scheme, the most appropriate technique of water supply was to be adopted. The most important element in scheme improvement was to be the preparation of an operational plan. On the basis of water availability, system characteristics, and agricultural options, the plan was expected to define how the system would be operated with respect to the timings and quantities of water deliveries and the responsibilities of those involved.  iii The principal components of the main part of the project (improved water management) were: (a) Scheme Investments required to implement the operational plan - these included repairs and renovations to existing structures; provision of new control structures; field channels and drains; measuring devices; and investments in upgraded O&M facilities and equipment (including housing for staff and transport); and communications equipment, as required to execute the operational plan; (b) Engineering and Administration required to supervise the effective implementation of the scheme investments; (c) Incremental Recurrent O&M Expenditures to support the improved operational plan and maintain it after the completion of the scheme investments; (d) Institutional Strengthening of the State Irrigation Departments, with particular emphasis on helping them to develop O&M planning and support units in the States and strengthening of the Water Management Cell (WMC) of the Ministry of Water Resources (MOWR) of the Government of India (GOI) so that it could provide policy direction, act as a catalyst for extending the approach across State boundaries and administer and coordinate the NWMP as a whole; (e) Training Programs for technical and orientation training for scheme staff and farmers, State O&M staff and officers of the WMC and study tours within India and limited overseas visits along with support for the infrastructure development of the Irrigation Management Training Institute (IMTI) in Tamil Nadu; and (f) Special Studies and Programs to strengthen specific aspects of water management (e.g. monitoring and evaluation, computer applications, communications, dam safety etc.). The project also supported two other components related to the general objectives of improved water management in India - (a) The Maharashtra Induction Training Program designed to support initial vocational training for engineers recruited to the Maharashtra Irrigation Department; and (b) An Institutional Strengthening and Training Fund (ISTF) to be managed and implemented by GOI and state agencies. Loan Covenants. Special loan covenants that were expected to promote achievement of project objectives and sustained operation were: (a) select, plan, design and seek approval of each scheme to be implemented to the satisfaction of IDA; (b) under TOR satisfactory to IDA carry out detailed review of O&M of NWMP schemes and prepare an action plan for improving O&M in the scheme and the State and implement the plan; (c) prepare a monitoring and evaluation program and implement it. Each scheme under implementation has been approved by IDA. O&M review has been carried out in some States but no action plan has been prepared. M&E units have been set up as planned, but operationally have not generated the required data for assessment of project benefits and to correct shortcomings in operation. Evaluation of Project Objectives. The project concept of instituting water management as a significant input into India's Irrigated Agriculture Sector was sound, relevant, appropriate and in line with the GOI's National Water Policy. Water management has been the focus in several other projects financed by IDA in the water sector in recent years. The Staff Appraisal Report (SAR) was clear in defining the project objectives and in setting out the implementation measures to be adopted for achieving these objectives. The implementation period of seven years was appropriate and the project components included for financing were in conformity with the project objectives. However, there was inadequate appreciation of the need to strengthen the state irrigation departments and to involve farmers in scheme designs, operations, maintenance, and cost recovery.  iv Agro-climatic and socio-economic conditions in India vary widely depending upon geographic location and irrigation systems have evolved reflecting this diversity. Major types of irrigation systems found in India (Localization, Warabundi, Shejpali) have been discussed in the SAR. The three original participating States in this project were from the south - Andhra Pradesh, Kamataka, and Tamil Nadu where localization, a demand-based system is practised. In retrospect, sufficient attention was not given at appraisal to the socio-economic aspects of converting from a demand-based system to a supply-based system. The structured system which was developed under the project to ensure equity in distribution of water supply to the beneficiary farmers located in various parts of the command can be considered to be an innovative idea that could be and should be improved upon in follow-up projects. SAR was unrealistic to expect that in the preparation of operational plans, paddy growing areas could be reduced and substituted by irrigated dry crops covering a larger area of the command. For instance, in the operational plan presented for the Rajolibunda Diversion Scheme in Andhra Pradesh, it was proposed to preclude paddy in middle and lower reaches in kharif and throughout the command in rabi by introducing systematic intermittent irrigation. Farmers obtained a court injunction against this proposal. A major deficiency in project concept was its failure to specify how farmers would be involved in the design and implementation. Even though mention is made in the SAR about farmer participation, no workable solution was given as to how this was to be achieved. Implementation Experience and Results The project was approved in March 1987 and started off very slowly in the three original participating States. The Induction Training Program in Maharashtra started at the same time as well. At appraisal, the project was expected to cover about 580,000 ha. During implementation, a number of additional schemes were included and the number of participating States increased from three to nine. The additional States were: Bihar, Gujarat, Haryana, Kerala, Madhya Pradesh, Orissa, and Uttar Pradesh. The total number of schemes, including those covered at appraisal, was 96 covering a command area of about 2.18 million ha. Bridge financing was provided for lining main canal and water courses in Haryana for a period of two years to allow that State to continue work on these aspects until the approval of the Haryana Water Resources Consolidation Project in 1994. At Credit Closure (March 31, 1995) it is estimated that about 32 schemes covering a command area of 337,331 ha would be completed. The remaining 64 schemes are in varying degrees of completion. The actual benefiting area is far less than the command area covered. For instance, in the schemes visited and included in the economic re-evaluation, the actual benefiting area is estimated to be 56,913 ha as compared to their total command area of 241,576 ha included under the project. The ERRs computed for nine schemes are: Thandava -3 percent; Malampuzha 9 percent; Ratapani -16 percent; Thambaraparani 4 percent; RDS -6 percent; Vani Vilas Sagar 9 percent; Sathanur 6 percent; Peechi -6 percent; and Haryana 17 percent. These values are fairly sensitive to the benefits and less sensitive to the costs.  v Project Sustainability. The ERRs are rather low to be consistent with sustained economic performance. The operational plans prepared for each scheme should be monitored and modified as required to ensure beneficiary acceptance of such plans. This would contribute to a large measure in increasing the benefits of the scheme. Adequate funds must be provided by the State governments to the implementing agencies to maintain the schemes in good condition and operate them. Parts of the irrigation schemes may have to be handed over to the farmers for operation and maintenance. Water users' associations would have to be created as a first step. In addition, an O&M culture has to be inculcated in the operating level staff of the irrigation departments, as well as appropriate modifications in institutional structure to enable sustainable O&M capabilities. Actual Cost, Financing and Implementation Timetable. The project costs including physical and price contingencies were estimated to be US$157.0 million (Rs.2,041.1 million). An IDA credit of SDR93.2 million (US$114.0 million equivalent) was to finance about 73 percent of the total costs and the remaining expenditures were to be mnet from the development budgets of the participating States. At Credit closure, the actual project expenditure in Rupee terms was Rs.5041.1 million representing a cost overrun of 148 percent in nominal rupee terns. Expenditure in US$ terms was US$176.1 million representing a cost overrun of 11 percent. Out of this arnount, US$ 125.7 million was disbursed by IDA. The Credit was fully disbursed and the final disbursement was made on June 20, 1995. The appraisal schedule of a seven-year implementation period was extended by one year. The exact number of schemes, as well as detailed costs, were not exactly known at the time of appraisal except for the three schemes - Vani Vilas Sagar, Sathanur and RDS. Comparison of "actuals" with appraisal estimates are therefore not very meaningful. Individual scheme costs and timetables have therefore to be analyzed. Key Factors Relating to Achievement of Project Objectives. The most important deficiencies which hampered project implementation can be classified into: a) Factors not generaly subject to Government control. This included lack of acceptance by beneficiaries of changes in cropping patterns. Achievement of project objective of increasing irrigated areas by using lower water consuming crops has been hampered by this. b) Factors generally subject to Government control. These included delays in timely release of annual funds for the execution of project works, frequent turnover of key professional staff of the implementing agencies and the limited use of training modules prepared by the Indo-Dutch team. These had substantially affected implementation performance in some participating States and contributed to the lack of understanding of the basic concepts of the NWMP by the operating level professional staff of the implementing agencies. c) Factors generaly subject to implementing agency control. These included insufficient diagnostic work to identify system constraints and to prioritize  vi expenditures and ineffective monitoring and evaluation of scheme performance. This has resulted in inappropriate and impractical solutions being implemented to achieve better water use efficiency in the schemes. In addition, monitoring and evaluation units set up in the different Irrigation Departments have, in general, been ineffective in collecting data required for the evaluation of agricultural benefits and scheme performance in a systematic manner. A pervasive major factor affecting the project was its top-down approach and lack of consultation with farmers. Performance of Bank and Borrower. The performance of the Bank and the Borrower has been mixed. The project involved implementing sound new concepts which required adequate and effective supervision and guidance to be provided by the Bank. During project preparation prior to and in the early stages immediately following appraisal, such supervision and guidance were indeed provided by the Bank. However, beyond FY89, the supervision by the Bank was inadequate and ineffective. The Borrower did not follow the detailed guidelines provided in the SAR and in manuals prepared later by the Bank in the preparation of additional schemes. Schemes were prepared without adequate diagnosis of the physical infrastructure, seasonal operational plans and prioritization of investments. In many instances appropriate construction schedules were not prepared with the result about two-thirds of the schemes have not been completed. The performance of the completed schemes is less than satisfactory. In FY89 this was the only project in the agricultural sector in India without any disbursement at all. At this stage, both the Bank and the Borrower were anxious to increase disbursements which appears to have become the over-riding factor. As a result, quality of works suffered. A number of new schemes were initiated covering an area of about 2.18 million ha and this number was quoted in all the reports giving a misleading impression of excellent project performance. At the end of the project, work had been completed covering only about 15% of this area. Project Outcome Assessment. The project outcome is rated as unsatisfactory since it has achieved only partial progress in physical targets (32 schemes covering 337,331 ha completed out of 96 schemes cleared for implementation covering 2.18 million ha and compared to appraisal target of about 500,000 ha) and in the completed schemes, the provision of reliable, predictable and equitable supply has not been achieved as expected at appraisal. Summary of Findings, Future Operations and Key Lessons Learned Findings. Incremental irrigated area fell short of appraisal estimates, reducing incremental agricultural production to much lower than that estimated at appraisal. The ERRs are therefore lower and below opportunity cost of capital except in Haryana. The reasons for the low rates of return are: (a) lower than assumed incremental benefiting areas in each scheme; (b) lower increases in cropping intensity; (c) excessive cost overrun per ha in real terms; and (d) in some cases non-adoption of crops other than paddy which was assumed at appraisal. It is possible to improve the economic performance of other non-completed schemes through changes in operational plans and related rehabilitation/modernization works. A limited awareness has been created in the decision-making levels  vii of the State governments for the need for better water management in existing projects thus exploiting the already created potential to the fullest. The borrower had conducted several short-term training courses and had undertaken a number of studies but there was little impact of these on project design or implementation. Plans for Future Operation. The following actions would be required to ensure completion of remaining schemes and their smooth operation: (a) revision of operational plans taking into account realistic agricultural data on cropping pattern and operational hydrology; (b) completion of ongoing schemes based on proper diagnostic analysis of infrastructure improvements required to execute the operational plan mentioned above on the basis of an appropriate prioritised investment program for the implementation of which continued financial assistance possibly under a follow-up phase would be necessary under stringent and well-defined institutional improvements, (c) training of 1ID staff, relevant agricultural department staff and farmers in the operation of the schemes; (d) training of scheme managers including inculcation in O&M culture; and (e) provision of adequate funds for sustained O&M taking into account farmer turnover concept; (f) revision of approach to include farmers in assessment of needs, planning, design, construction, O&M, and cost recovery; and (g) institutional modifications and strengthening at the level of irrigation departments to create organizational capabilities in water management and farmer organization. GOI has advised the Planning Commission to release adequate funds to the concerned State Governments to complete the ongoing schemes. There would still be a need for external assistance to complete the schemes and consolidate the gains from this valuable program. Lessons for Future Projects in the Sub-sector. The main lessons learned from this project implementation can be grouped under Concept, Design, Implementation, and Follow-up, as follows: Concept a) Extensive equitable irrigation and self-financing irrigation schemes are two objectives which may not be necessarily compatible. Equity should therefore be achieved through the introduction of volumetric water supply, within the limitations of water availability and the effective recovery of appropriate water charges, rather than through the delivery of the same amount of water per unit of area to each farmer; b) Projects of this type which involve not only technical changes to the status quo, but also have significant social aspects, require a high level of farmer participation in irrigation management (P1M) to be successful. Introduction of such concepts will require, in turn, the use of more advanced technologies in irrigation management which are currently available in India and abroad. These would include provision of improved structured system to allow flexible water  viii distribution to water user associations, and, as needed, canal automation for controlling main canal flows. c) Improving the productivity of irrigation systems linked with farmer participation requires strengthening of the capabilities of the irrigation departments and some institutional restructuring to provide emphasis and staff resources to these activities. In the Water Resources Consolidation Projects (WRCPs) (which are the new generation irrigation projects supported by the Bank), a major thrust is on improving the productivity of irrigation systems through modernization linked with farmer involvement. The "systems improvement and farmer turnover" components of the WRCPs integrally link farmer organization, operations and maintenance, and cost recovery with the engineering improvements. Additionally, each project has involved an institutional reorganization and subsequent strengthening program to build up the capabilities of the irrigation departments in scheme modernization, water management, andfarmer organization. Volumetric supplies are also being attempted by water users associations and this experiment should be carefully monitored and nurtured Canal automation is being experimented in Tungabadra and Majalgaon commands. Design a) A baseline survey of the condition of the system infrastructure and cropping practices is a prerequisite to prepare a detailed scheme report for rehabilitation/modernization. b) Detailed diagnostic work would be undertaken and realistic alternate scenarios should be explored in computer simulation models taking into account farmer crop preferences in preparing operational plans. These plans should be prepared with the assistance of specialists in operational hydrology (water availability and water balance studies under the revised scenarios), irrigation agronomy who are familiar with the local conditions, sociologists, representatives of the farming community, and irrigation engineers. The operational plans should be flexible to handle vagaries of nature and forces of the market place (for instance, if a sugar mill is located near a scheme, water-savings cannot be achieved by switching to a less water consuming crop -- innovative irrigation technology has to be adopted with required capital investments). c) An irrigation system should be considered as a whole water management unit. Works should not be carried out in individual distributaries without any possibility of water management in the system in its totality.  Ix Implementation a) Prioritization of expenditures necessary for implementing the selected operational plan should be prepared, together with detailed construction sequences, which take into account the financial capability of the implementing agency, canal closure periods, and number of days available to carry out the civil works in each year. b) Quality of construction in small and scattered works requires particular attention. High quality technical assistance and third party supervision should be considered in future project. c) Priority should be given to training a maximum number of professional staff of the implementing agencies in water management principles, and some of the promising staff should be trained intensively in computer and remote sensing technology to be water managers. Turnover rate of key senior-level staff must be mninimized in projects involving new concepts. Otherwise, the benefits of training and efficiency in implementation are lost. Follow-up a) Simple key parameters should be established for monitoring and evaluating the performance and impact of water management from scheme start-up. An effective monitoring and evaluation unit should be created in each participating State with a multi-disciplinary core team. Monitoring the impact of scheme interventions on agricultural production should be carried out continuously starting from the baseline survey and using the same parameters so that operational plans can be modified if required and also a final evaluation of the scheme performance would be feasible. b) O&M of the scheme should be jointly carried out by department officers and farmer beneficiaries. Adequate funds should be made available for such O&M.  1 IMPLEMENTATION COMPLETION REPORT INDIA NATIONAL WATER MANAGEMENT PROJECT (NWMP) (Credit 1770-IN) PART I: PROJECT IMPLEMENTATION ASSESSMENT A. PROJECT OBJECTIVES 1. The purpose of the project was to increase agricultural productivity and farm incomes through a more reliable, predictable and equitable irrigation service. To achieve these objectives and facilitate extension of similar activities throughout the country, the project was formulated to help participating States and the Government of India (GOI) to develop an institutional capacity to plan, implement and monitor improved operation and maintenance (O&M) practices and to provide for low- cost infrastructural improvements designed to support an improved operational plan for selected schemes. In each scheme, the most appropriate technique of water supply was to be adopted. The most important element in scheme improvement was to be the preparation of an operational plan. On the basis of water availability, system characteristics, and agricultural options, the plan was expected to define how the system would be operated with respect to the timings and quantities of water deliveries and the responsibilities of those involved. 2. The principal components of the main part of the project (improved water management) were: (a) Scheme Investments required to implement the operational plan - these included repairs and renovations to existing structures; provision of new control structures; field channels and drains; measuring devices; and investments in upgraded O&M facilities and equipment (including housing for staff and transport); and communications equipment, as required, to execute the operational plan; (b) Engineering and Administration required to supervise the effective implementation of the scheme investments; (c) Incremental Recurrent O&M Expenditures to support the improved operational plan and maintain it after the completion of the scheme investments; (d) Institutional Strengthening of the State Irrigation Departments, with particular emphasis on helping them to develop O&M planning and support units in the States. In addition, the Water Management Cell (WMC) of the Ministry of Water Resources (MOWR) was to be strengthened so that it could provide policy direction, act as a catalyst for extending the approach across State boundaries and administer and coordinate the NWMP as a whole; (e) Training Programs for technical and orientation training for scheme staff and farmers, State O&M staff, and officers of the WMC, and study tours within India, and limited overseas visits, along with support for the infrastructure development of the Irrigation Management Training Institute (IMTI) in Tamil Nadu; and (f) Special Studies and Programs to strengthen specific aspects of water management (e.g. monitoring and evaluation, computer applications, communications, dam safety etc.). The project also supported two other components related to the general objectives of improved water management in India - (a) The Maharashtra Induction Training Program designed to support initial vocational training for engineers recruited to the Maharashtra Irrigation Department; and (b) An Institutional Strengthening and Training Fund (ISTF) to be managed and implemented by GOI and state agencies.  2 3. The project was to be implemented over a seven-year period and the project costs including physical and price contingencies were estimated to be US$157.0 million (Rs.2,041.1 million) at 1986 prices. An TDA Credit of SDR93.2 million (US$114.0 million equivalent) was to finance about 73 percent of total project cost and the remaining expenditures were to be met from the development budgets of participating States. 4. Special legal covenants that were expected to promote achievement of project objectives and sustained operation were: (i) select, plan, design and seek approval of each scheme to be implemented to the satisfaction of IDA; (ii) under TOR satisfactory to IDA carry out detailed review of O&M of NWMP schemes and prepare an action plan for improving O&M in the scheme and the State and implement the plan; and (iii) prepare a monitoring and evaluation program and implement it. The schemes were selected, planned, designed and submitted to IDA after clearance by GOI. However, the quality of scheme preparations varies and their selection was not subjected to detailed technical scrutiny by IDA. O&M review has been carried out in some States but no action plan for improvement has been prepared. M&E units have been set up as required but operationally have not generated the required data for assessment of project benefits and to correct shortcomings in project operation. In effect, while there was compliance with special legal covenants, the requisite follow-up was inadequate and therefore the intended objectives for their compliance were not achieved. 5. The project concept of instituting water management as a significant input into India's Irrigated Agriculture Sector was sound, relevant and appropriate and in line with the GOI's National Water Policy. Water management has been the focus in several other projects financed by IDA in the water sector in recent years. The Staff Appraisal Report (SAR) was clear in defining the project objectives and in setting out the implementation measures to be adopted for achieving these objectives. The implementation period of seven years was appropriate and the project components included for financing were in general realistic. 6. Agro-climatic and socio-economic conditions in India vary widely depending upon geographic location, and irrigation systems have evolved reflecting this diversity. Major types of irrigation systems found in India (Localization, Warabundi, Shejpali) have been discussed in the SAR. The three original participating States in this project were in the south - Andhra Pradesh, Kamataka, and Tamil Nadu - where localization, a demand-based system, is practiced. In retrospect, sufficient attention was not given at appraisal to the socio-economic aspects of converting from a demand-based system to a supply-based system. The structured irrigation system which was developed under this project to ensure equity in distribution of water supply to the beneficiary farmers located in various parts of the command can be considered to be an innovative idea which can be further improved in follow-up projects. However, where field to field irrigation is practiced (as in most paddy irrigation in the southern states) this system may not be best suited (para 42 below refers). 7. The SAR was unrealistic to expect that in the preparation of operational plans, paddy growing areas could be reduced and substituted by irrigated dry crops covering a larger area of the command. For instance, in the operational plan presented for the Rajolibunda Diversion Scheme (RDS) in Andhra Pradesh, it was proposed to preclude paddy in middle and lower reaches in kharif and  3 throughout the command in rabi by introducing systematic intermittent irrigation. Farmers obtained a court injunction against this proposal. 8. The project could be considered as a pilot in setting up water management strategies and in improving the performance of inigation systems throughout India. The project was initially concentrated in three southern states with a provision for expansion to other States if found to be attractive. The concept of reliable, equitable and predictable water supply has attracted the attention of the Irrigation Departments (IDs) of other States which applied later to participate in the project. Hence, the project can be considered to have had a positive impact on the evolution of sectoral policy of improved water management. 9. The concepts and the implementation procedures were not complex from an engineering viewpoint although they called for strengthening the institutions involved in irrigation management. The setting up of a Water Management Cell and the provision of the ISTF has helped in the implementation of the project. However, at state irrigation department levels, water management cells, if created, were "add-ons" to existing institutional structures, and did not fundamentally improve the capabilities of the departments in water management. Also, the complexities of water management in a large community of farmers was not dealt with adequately during the appraisal of the project. A major deficiency in project concept was its failure to include farmers in the discussion of how they would be involved in the design and implementation. It is stated in the SAR that once an operational plan has been established, a program must be prepared to implement the plan which should cover farmer consultation and involvement in approval and implementation. However no workable solution was given as to how such involvement of farmers is to be obtained. The major risks associated with the project were related to the sustainability of the operational programs to be supported. The risk that the project might be unable to sustain programs involving changes in water allocation to farmers, in the face of vested interests and other pressures that affect some scheme operations was recognized at appraisal. Such risks still exist but with more substantive participation of farmers in the preparation, implementation and operation of schemes, coupled with turnover of parts of the system, these could be mtinimized in follow-up projects. B. ACHIEVEMENT OF PROJECT OBJECTIVES 10. Overall Achievement. The project was approved in March 19871 and the Credit became effective in August 1987. Originally three States were to participate in the project: Andhra Pradesh, Kamataka and Tamil Nadu. A total of sixteen schemes had been identified covering a cultivable conmmand area (CCA) of about 460,076 ha. Breakdown by States is provided in Appendix 3. In addition, anticipating that more schemes would be added in the three participating States and that other States might be interested in participating in such a project, an allowance was made to cover an additional CCA of about 123,029 ha. The Induction Training Program in Maharashtra started at the same time as well. During the implementation of the project, several additional schemes were included and the number of participating States increased from three to nine. The additional States were Bihar, Gujarat, Kerala, Madhya Pradesh, Orissa, and Uttar Pradesh. The total number of schemes cleared for implementation increased from sixteen, identified at appraisal, to 96 covering a command area of 2.18 1 Board approval date was March 24, 1987 and the Agreement was signed on May 12, 1987  4 million ha. In addition, bridge financing was provided for lining of main canal and water courses in Haryana for a period of two years to allow that State to continue work on these aspects until the approval of the Haryana Water Resources Consolidation Project in 1994. In Haryana, project investments were mainly utilized for lining of main canals and watercourses. No operational plans were prepared and no new devices were installed. Thus, in the strictest terms, it can be said that NWMP principles were not followed. It would have been appropriate to have amnended the Development Credit Agreement (DCA) prior to proceeding with bridge financing. The original Credit closing date of March 31, 1994 was extended by one year to March 31, 1995 at which time the Credit was expected to be completely disbursed. At the time of Credit closure, it is estimated that about 32 schemes covering a CCA of about 337,131 ha have been completed, about 15 percent of projected command. Some of these schemes are classified as completed by the various State implementing agencies although anticipated physical works have not yet been completed. Details of schemes initiated under the project, their command areas and their completion status are given in Appendix 3. The primary project objective of providing a reliable, predictable and equitable water supply is reported to have been achieved in some of the completed schemes. An independent review of two projects (Sathanur in Tamil Nadu and Bhadra in Karnataka) carried out by the International Irrigation Management Institute (IMI) indicates that this objective has been achieved primarily along the main canal and in some of the distributaries. In some of the other schemes investigated some of the tail-end farmers have received water as a result of project interventions. 11. Objectives of Sector Policies. The project has created a limited awareness in the decision-making levels of the various participating States for the need for better water management in existing projects thus exploiting the already created potential to the fullest. This awareness has to be translated into concrete action in the future. In this project only limited success has been achieved so far. 12. Institutional Development Objectives. Clear understanding of the concept of water management and inculcation of O&M culture in the lower ranking engineers of the Irrigation Departments of the various States have, in general, not been achieved. Staff were not clear about the objectives of the project suggesting that a longer period is required to reap the benefits of institutional development. This is in part due to the high turnover rate of staff and those that were trained in the early periods of the project were no longer associated with the project. Moreover, the objectives and concepts which are very valid, have yet to be reflected in the regular work program of the Irrigation Departments as a whole. In this connection, it should be noted that the project's institutional arrangements were not conducive to capacity building in the Irrigation Department as a whole. Unlike the new generation of water resources consolidation projects each involving significant institutional reorganization and strengthening, under NWMP various implementation cells were created within the department which were appendages rather than integral to the operational functions of the parent department. The Irrigation Departments could have absorbed the NWMP concepts in their regular program as well, at least in the areas of preparation of operational plans, ensuring efficient delivery of water. On a more positive note, in Maharashtra, despite initial difficulties in getting started with the necessary experts, the induction training programn is reported to have performed well based on information provided by the Water and Land Management Institute which conducted the program in association with the Indian Institute of Technology, Bombay, and it may be advisable to explore the possibility of extending such programs to other States.  5 13. Economic Rates of Return (ERR). The re-estimated ERRs for the 9 schemes evaluated at completion consist of two sets - the first set reflecting those calculated at appraisal, and the second set reflecting the others. The current economic rates of return re-estimated at completion for the first set are 9 percent for Vani Vilas Sagar (VVS), 6 percent for Sathanur, and -6 percent for RDS, compared to 45 percent, 43 percent and 29 percent, respectively estimated at appraisal. No economic rates of return were calculated for the others at scheme preparation. The recalculated ERRs for these schemes at completion are: -6 percent (Peechi), 17 percent (Haryana), 4 percent (Thambaraparani), 9 percent (Malampuzha), -3 percent (Thandava) and -16 percent (Ratapani). The recalculated ERRs are based on actual costs incurred up to the Credit closing date plus costs to complete remaining works until the end of FY96, and the actual and projected stream of agricultural benefits. Underlying assumptions about costs and benefits and other information supporting the analysis are shown in Table 9 of Part II and Appendix 2. The implementation difficulties as evidenced by excessive cost overruns per ha of benefiting area, sharply reduced benefiting area, and funding constraints for maintenance are consistent with diminished project performance. Consequently, the reduction in ERR estimates is not unexpected despite positive impact on high-value cash crops such as coconut and improved crop yields on benefiting area. 14. There are various factors which have contributed to the significantly lower as well as negative ERRs. These include a markedly reduced benefiting area at scheme completion compared to appraisal projections, excessive cost per ha in real terms ranging from Rs.23,655/ha (Haryana) to Rs. 104,020/ha (Peechi) and less than anticipated changes in cropping patterns than was foreseen at appraisal or scheme preparation. 15. Sensitivity Analysis. The potential economic benefits are not being fully achieved in most of the schemes. The three factors which have influenced ERRs are: (a) sharp variations in incremental benefiting area; (b) changes in O&M funding; and (c) the inability of state Irrigation Departments to operate, monitor and manage the different schemes satisfactorily. The sensitivity analysis indicates that an increase in benefiting area by 50 percent will raise ERR for VVS to 12 percent; a reduction in 50 percent in O&M funding (by assuming that future benefits would be reduced by 50 percent) would lower the ERR for Sathanur to about -4 percent, and the improved operation and management would raise ERR of Peechi to nearly -4 percent, assuming that the ineffective operation of the Irrigation Department would depress project benefits by 20 percent. 16. Farm Incomes. As at appraisal, farm incomes at completion were ascertained, based on representative farm models. According to appraisal estimates, farm income should have increased on average throughout the command area of the schemes by 31 percent to 66 percent. Farm incomes in the three schemes evaluated at appraisal (VVS, Sathanur and RDS) have achieved lower rates of growth - 8 percent, 31 percent and 20 percent, compared to appraisal estimates of 30 percent, 31 percent and 66 percent, respectively. Increases in farm incomes in the nine schemes at completion as a direct result of the project have ranged from 8 percent to 89 percent. The highest of 50 percent and 89  6 percent are in the tail end reaches of the Malampuzha and Thandava Schemes which were essentially rainfed prior to NWMP.' Details are given in Appendix 2. C. IMPLEMENTATION RECORD AND MAJOR FACTORS AFFECTING THE PROJECT 17. Implementation Record. The implementation record is mixed. While only 32 out of the 96 schemes initiated under the project have been completed, a limited awareness for water management and improving irrigation efficiencies has been created in the implementing agencies of some States. However, the management systems required to ensure sustainable operation of completed schemes admittedly requiring a longer time horizon is not in place yet. It is not clear whether the State governments with unfinished schemes would have the financial resources and institutional capacity to complete the schemes and generate benefits as anticipated at appraisal. 18. Factors Not Generally Subject to Government Control. This included lack of acceptance by beneficiaries of changes in cropping patterns. The assumption that farmers could be induced to change from paddy (for which their land had been localized earlier) to irrigated dry crops was unrealistic and, in the case of Andhra Pradesh, the farmers of the RDS scheme went to court and obtained an injunction against the Government. In Thandava, the revised cropping pattem has never been put into practice. In Karnataka, coconut in VVS and paddy areas in the Bhadra scheme have increased. In Tamil Nadu, groundnut area has increased at the expense of paddy, millets and sugarcane. Achievement of project objective of increasing irrigated areas by using lower water consumption crops has occurred only to a limited extent. 19. Factors Generally Subject to Government Control. These included delays in timely release of annual funds for the execution of project works, frequent turnover of key professional staff of the implementing agencies and the limited use of training modules prepared by the Indo-Dutch team. These had substantially affected implementation performance in some participating States and contributed to the lack of understanding of the basic concepts of the NWM\P by the operating level professional staff of the implementing agencies. Funding constraints are reflected in the non- completion of any scheme in Andhra Pradesh and partial completion (10 percent) in Bihar and Uttar Pradesh. In Tamil Nadu, thirteen Chief Engineers headed the NWMP operation over the seven-year period whilst in Andhra Pradesh thirteen executive engineers and five superintending engineers handled the RDS project alone over the seven-year period. Such frequent changes in staff are not desirable, particularly in a project introducing new concepts in engineering (structured network) and water management. Dutch technical assistance was provided to prepare training modules on the concept, design and implementation of the NWMP schemes. The use of the modules prepared by this Indo- I Borrower concurs that the project schemes have been able to raise the productivity of land thereby increasing the income of farmers in the command area (refer Appendix 3, para 3.9). The Borrower also states that the lower rates of return are due to (a) reduction in benefited areas; (b) increase in cost of rehabilitation; and (c) less than anticipated changes in cropping patterns. This situation arose because of the fact that the rehabilitation and modernization works were scattered and started at a later date, and therefore some of the important works were still incomplete by the closing date. The farmers were also unwilling to change the cropping pattern as planned in the SAR. Moreover, there is considerable time-lag between the scheme completion and accrual of the full benefits (Appendix 3, para 3.5).  7 Dutch team varied amongst the participating States. Engineers from Madhya Pradesh Water and Land Management Institute were aware of the significance of these modules. Wider use of these modules could have been made by other State govemments. 20. Factors Generally Subject to Implementing Agency Control. These included insufficient diagnostic work to identify system constraints and to prioritize expenditures and ineffective monitoring and evaluation of scheme performance. In FY88, Bank/FAOCP missions assisted the three original States in the preparation of detailed scheme reports. Beyond that time the implementing agencies were to prepare scheme summaries using their funds and then get them cleared by WMC. Once the summaries were cleared, the implementing agencies were allowed to use project funds to prepare detailed scheme reports with the assistance of consultants, if required. But the quality of scheme reports prepared after the involvement of Bank/FAOCP has not been good. In the preparation of revised operating plans, actual cropping pattem data could have been collected from the agriculture department. This has resulted in inappropriate and impractical solutions being implemented to achieve better water use efficiency in the schemes. Monitoring and evaluation units were to be created in the implementing agencies and evaluate the impact of the projects on the agricultural production in the commands of the rehabilitated schemes. This aspect has not been taken up seriously in any of the schemes. Even though such M&E units have been created in various implementing agencies, they have not been effective in collecting baseline and follow-up data relevant for the evaluation of agricultural benefits and scheme performance. Another major factor affecting the project was its top-down approach and the lack of consultation with farmers. 21. Implementation Delays. A number of schemes which were expected to be completed over a period of four years have not been completed even after seven years. In many of the schemes, construction schedules did not take into account canal closure and heavy rainfall seasons properly which led to implementation delays. In Andhra Pradesh, engineers' estimates for construction were quite low leading to poor response from contractors to tender or delay in processing of tenders. 22. Actual Project Cost. The actual project expenditure is given in Table 8A of Part II. In rupee terms, the total cost would be Rs.5061.3 million in current prices representing a 148 percent cost overrun in nominal rupee terms. Similarly, the total expenditure in US dollars up to Credit closing date is estimated to be US$176.1 million representing a cost overrun of 11 percent. The smaller cost overrun in US$ terms is due to the exchange rate movement of the rupee against US$. However, the exact number of schemes as well as detailed costs were not known exactly at the time of appraisal (except the three schemes evaluated at appraisal). Comparison of actuals with appraisal estimates for the whole project are therefore not very meaningful. 23. Project Financing. The IDA Credit is fully disbursed (SDR93.2 million). The remaining expenditure has been met by the various State Governments involved and the GOI. Details are given in Table 8B of Part II.  8 D. PROJECT SUSTAINABILITY 24. Project sustainability hinges primarily on adequate provision of maintenance funds, increased involvement of farmers in O&M and cost recovery, inculcation of O&M culture to the Irrigation Department staff to carry out the O&M activities, and enhancement of institutional capabilities for O&M.. Availability of funds alone is not a guarantee for sustained project operation. Planning for O&M has to be given priority and until this is achieved, project sustainability cannot be ensured. Based on past experience, it would seem unlikely that adequate finds would be spared for maintenance. Even in places where adequate maintenance funds have been released by the State Govemments, the maintenance of structures constructed under NWMP was unsatisfactory. If the operational plans devised under the project are not kept up, it would lead to the deterioration of the systems as well. 25. As costs for O&M increase and the State governments find it difficult to meet these costs, it is necessary to tum over parts of the system to the beneficiaries in addition to increasing the water charges and improving collection rates. Since farmers have not been involved in the preparation and execution of the schemes under this project, it is necessary to take urgent steps to include them in the operational phase. Water Users Associations at different levels in the system would have to be created as a first step. E. BANK PERFORMANCE 26. Identification and Preparation. The project was identified around 1985 and prepared over a two-year period by the Bank/FAO CP with the assistance of national consultants and staff of the Irrigation Departments of the participating States. A number of missions were mounted to prepare detailed studies for three schemes: one each in the three participating States. The defect in identification and preparation which continued through appraisal was the skill mix of the teams which comprised mainly of irrigation engineers, economists and system analysts. There was limited agricultural specialist input in identification, preparation and appraisal. At the time of project preparation and appraisal, beneficiary participation was just beginning to receive attention, however no sociological input was provided either. This might have led to the suggestions for changes in cropping patterns (replacing high water consuming paddy crop with less water consuming crop to increase area irrigated) which were subsequently rejected by the farmers (para 36 and 39). Barring these two drawbacks, the engineering aspects of the three schemes were well prepared. Operational hydrological analyses were carried out through simple simulation models which were appropriate. At appraisal, even though twelve schemes had been identified for inclusion in the project, only three had been prepared in detail. In the internal memoranda discussing the Yellow Cover Report, it was observed that an unusually high Bank/FAOCP input will be required in the post-appraisal/early-project implementation period. This turned out to be true and the Bank/FAOCP did provide the requisite input. However, a similar amount of input required for the other 77 schemes cleared through the implementation period was not practical.  9 27. Design. The Bank engineers, in association with resident mission staff, prepared the structured network concept which can be considered innovative despite some of its constraints. This system delivers water below the structured level proportional to the area served. To respond to this operation, it is necessary to take account of other related factors such as soil moisture holding capacity and variability in the irrigability of lands and associated crops best suited for these lands. 28. Institutional Development Aspects. Even though the project aimed at establishing a good O&M mechanism in the Irrigation Departments, such institutional developments are yet to take place. The Bank apparently did not pay enough attention to the institutional development aspects. Funds provided for this purpose were not fully utilized and in the last six months before Credit closure were transferred to cover expenditures on works. 29. Farmer Organization Aspects. The appraisal's focus on engineering also resulted in little attention to the role of farmers in the project. At that time, the benefits of farmer involvement in irrigation were not widely appreciated in Asia, but in retrospect, this was a major missed opportunity. 30. Environment and Social Aspects. The project did not involve any direct environmental concerns and it did not involve any persons or families being displaced. 31. Supervision. Supervision by the Bank was inadequate and ineffective. In the early years of project implementation when only the three original States were involved, a considerable number of staff weeks were spent on supervision. Of course, during this time the Bank was also assisting the State implementing agencies in the preparation of detailed feasibility studies for other schemes. At the time of appraisal, only three schemes had been studied in detail: Sathanur in Tamil Nadu, Vani Vilas Sagar in Karnataka and Rajolibunda Diversion Scheme in Andhra Pradesh. From FY87 to FY89, a total of 387.2 staff weeks were spent for preparation assistance and supervision whereas for the remaining five years only 233.5 staff weeks were spent in reviewing new schemes to be included under the project and supervising ongoing schemes. Given this constraint, alternate arrangements should have been made to scrutinize the schemes properly. The Bank should share some of the responsibility for the poor preparation of the schemes which has led to their poor performance. The supervision missions did not provide adequate guidance to the State implementing agencies regarding the principles and objectives of the NWMP. Instances like in Rajolibunda Diversion Scheme where rehabilitation/modernization work has been carried out in seven out of 47 distributaries in a scattered manner, Ratapani where nothing has been done to arrest the 20 percent leakage observed in the first 800 m but unnecessary proportional modules have been built (Appendix 3 refers) could have been avoided through effective supervision. As the number of schemes under implementation increased, the Bank did not equip itself with appropriate supervision mechanisms. Up to and including FY89, only seventeen schemes were either under preparation or implementation whereas beyond that period another 79 schemes were reviewed and cleared out of which 68 schemes have been under implementation. Yearwise breakdown of schemes under implementation is given in Appendix 3. Poor diagnostics and non- prioritization of physical investments could have been avoided if the number of schemes had been kept to a manageable level.  10 32. Eleven supervision mission reports were prepared. In order to cover the large number of States, supervision work was carried out over a four to six-month period and covered in a single report. Supervision effort was also divided amongst Bank staff . It is therefore somewhat surprising that such anomalies in implementation were not spotted and corrected. 33. In the early years of the project, implementation rate was slow and correspondingly disbursements were low. This was the only project in the agriculture and irrigation sector not to register any disbursement at all in entire FY89. In the first three years of the project implementation, progress was very slow which was understandable since new concepts were being introduced to the Irrigation Departments' staff and several new engineering designs had to be prepared using scientific hydraulic engineering principles involving the mechanics of flows. At the end of March 1990, less than half of the predicted disbursements had materialized. Both the Bank and GOI were anxious to increase disbursements and this appears to have become the overriding concern and additional schemes in the three original and another six States were approved. Even though it is clearly specified in SAR para 3.33 that no new schemes would be started after June 1990, about seventy schemes were cleared for implementation after this date although it was evident that they would not be completed before the original Credit closing date of 31 March 1994. The total estimated cost of the schemes cleared by GOI and approved by the Bank was also far in excess of the funding available under the project (Appendix 3 refers). It appears from Bank documents that as early as 1989, when the project performance was behind schedule and the impact of the project could not have been evaluated, plans were under way for a follow-up project and these additional schemes were cleared assuming that there would be continued financing of the schemes under such a follow-up phase.' At the same time, other internal communication indicates that there was considerable concern about non-collection of reliable data on the project's impact on agricultural production. Monitoring and evaluation component was not supervised and guided properly in spite of such observations. F. BORROWER PERFORMANCE 34. Scheme Preparation. The strengthening of the Water Management Cell (WMC) in the MOWYR has been accomplished well and it has been staffed as per SAR requirements except for the post of agronomist. The WMC has been responsible for clearing scheme summaries first and then the detailed scheme reports for further clearance by the Bank. The WMC achieved this through an appraisal committee consisting of senior experts from other relevant agencies of the GOI and chaired by the Additional Secretary of the MOWR. As the number of schemes under implementation increased, the WMC did not have adequate technical lower-level staff to independently check the various scheme designs presented. Accordingly, scheme reports which were not up to SAR standards were cleared for implementation. Alternate checking mechanisms such as the regional offices of the Central Water Commission may have to be investigated for future follow-up projects. The WMC had, unlike the State implementing agencies, continuity of senior staff In fact, the commissioner's post was occupied by the same person until September 1994. I Borrower states in para 3.2 (iii) of their comments (Appendix 3) that ....an impression crept in that the second phase of the NWMP would follow concurrently without any transition period between the two phases." And that .... "GOI/WB cleared several projects for implementation even at a later date fully realising that they would not be completed before the closing date of the project."  I1 35. Scheme preparations are not of a high standard. Design and conceptual weaknesses were evident in some of the schemes visited by the mission. In Tainil Nadu, the designer has taken advantage of existing drainage courses in sub-catchments to convey irrigation flows to tanks but in designing control structures for flow division the natural run-off and floods usually handled by such courses have been ignored. As a result, water has backed up on farmers' fields (inadequacy of the structures) and the annoyed farmers have destroyed the flow division structures. No attempt has been made to review the situation and rebuild the structures. In Andhra Pradesh, small structures built on black cotton soils have failed due to the swelling and shrinking forces. While larger structures have been constructed with appropriate intermediate layers of cohesive non-swelling soils, the same attention has not been given to smaller structures such as proportionate dividers at field outlet levels. Cement plasters have been directly applied to random rubble masonry without any bonding layer resulting in failure of the plasters. Pointing would have been more appropriate in these cases. The quality of construction in small and scattered structures requires particular attention in future projects. In the Ratapani scheme in Madhya Pradesh, no work has been carried out in the first 800 m of the main canal where (on the day of the rnission's visit) 30 percent water losses by leakage was measured, but money has been spent further downstream with no significant benefits. Reduction of leakage losses in this section would have provided better water control and significant additional benefits. The performance of the Vani Vilas Sagar scheme in Karnataka is directly related to the inflows into the reservoir which are inadequate (the reservoir has not filled in recent years). A comprehensive water balance analysis of the total system should have been carried out and operational plans devised accordingly. The performance of the Sathanur scheme in Tamil Nadu is also dependent on reservoir inflows. The scheme experienced successive drought years and in 1991, no releases were made to irrigation command or the system tanks. Still about 30 percent of the land was irrigated which indicates the extent of groundwater usage in the command. Recently several discharge measuring structures have been completed. However, improvements have not been carried out in several distributaries which has led to the malfunctioning of outlets. At Peechi in Kerala, the problem is abundant growth of weeds and bushes in the canals, and canal banks restricting access to the system. A priority investment should have been the purchase of bush, weed and sediment clearing machinery which would have enabled water to reach the farthest points of the command. If the basic concept provided in the SAR of analyzing the performance of the scheme, identifying the constraints, and then adopting appropriate solutions to implement an operational plan had been followed, the maintenance constraint would have been identified as the major problem in water management in this scheme. 36. In resectioning canals, a standard theoretical approach has been taken which is leading to O&M problems. Manning "n" value of 0.025 applicable to straight earth channel in good condition has been used in designs together with trapezoidal sections which the O&M staff are unable to maintain. Instead, it would have been preferable to adopt natural regime sections (in unlined canals) with more realistic values for the roughness coefficients with appropriate backwater computations in the design. It was observed that the field staff were unable to push required discharges after a year of completing some of the resectioning because even a shift in "n" value to 0.030 (due to inadequate maintenance) results in a 20 percent reduction in the carrying capacity of the channel. Bed slopes are, by necessity, quite small due to topographic constraints. It was found that proper maintenance of the canals to maintain design bed slopes is critical for the success of the proportional outlets which are based on hydraulics of flow and maintaining predetermined energy gradients in the channels. In most of the  12 schemes visited (except Thandava in Andhra Pradesh) field-level staff were not quite conversant with the hydraulic principles of NWMP and the importance of proper maintenance in achieving these. 37. Involvement of agricultural department staff and farmers in the preparation of seasonal operational plans was, at best, minimal. Actual prevailing cropping patterns vary considerably from those in the design records. These have evolved over the years and to some extent are contributing to the inequitable and often inefficient irrigation service. It would probably not be feasible to provide adequate and equitable supplies to the entire command in some of the schemes which are water-short due to these cropping patterns. However, simulations could have been carried out using these actual cropping patterns, coupled with direct farmer consultation (para 26 and 39), in order to determine logical alternatives to improve water management. Instead, it appears that theoretical patterns were used in the analysis and the alternatives recommended have turned out to be impractical. 38. Implementation. The implementation performance varies from State to State. The size of the schemes rehabilitated range from medium to large. Kerala, which joined the project later (in 1990) appears to have performed well. In Tamil Nadu and Karnataka, reports prepared by IIMI indicated that some progress has been achieved in pushing water further down the main canal and distributaries. Since all proportional outlets have not yet been constructed, equitable distribution has not yet been achieved in its entirety. In most of the 32 schemes, according to the reports prepared by the State Irrigation Departments, cropped areas have increased, cropping patterns have changed and total production has increased. Of particular relevance is how much of the reported improvements is directly attributable to the project interventions and how much is due to other factors such as increased groundwater use, improved agricultural practices, etc. It was found that in limited areas where reliability of water supply has been assured, use of inputs has gone up and hybrid varieties are replacing local varieties. On the whole, NWMP has somewhat improved the situation, but the original project development objectives have not been achieved. In Kerala, Andhra Pradesh and Madhya Pradesh, there is an overall appreciation for water management at higher levels of Government. 39. Water Management. During field visits it was observed that the concept of water management has not been clearly understood by most of the operating level engineers of the Irrigation Departments. The idea was to prepare seasonal operational plans based on actual and realistic cropping patterns for good management of available scarce water resource and then to determine the interventions required to improve the physical infrastructure to carry out such plans. Obviously, deferred maintenance would be part of such interventions but only to the extent required to manage the water resource of each scheme. However, it was found that deferred maintenance had become the prime objective during project implementation. This is evident from the fact that in some schemes, structures have been renovated, resectioning of canals have been carried out, and new structures have been built in a scattered manner, which will not assist in any rational water management. For instance, in Rajolibunda Diversion Scheme, seven out of 47 distributaries have been repaired but not in any contiguous unit. As a result, in most schemes actual benefiting areas would be much smaller than the Cultivable Command Area (CCA). Similarly, the quality improvement in irrigation service delivered to the farmers would be minimal.  13 40. Beneficiary Involvement. Farmers were not consulted in any systematic manner (as required in the SAR through scheme-level committees, etc.) in preparing alternative operational plans, including changes in cropping patterns, although such plans were to define the rules for their implementation and iiiclude structural modifications at the outlet and distributary levels necessary to achieve a more reliable, predictable and equitable irrigation service directly affecting the farmers. 41. Monitoring and Evaluation (M&E) of key paramneters (physical progress as compared to targets, in cropping patterns, cropped areas, productivity and production data at various locations in a scheme command) has not been carried out in a systematic -manner and such information was not readily available even in those schemes where M&E units were functional. Measuring devices (provided for in the SAR) have not been constructed in most of the schemes visited and where they have been constructed, they are either inoperational or readings are not being taken and analyzed to evaluate the efficiency and adequacy of the operational plan in achieving improved water management.1 Remote sensing methods were employed in selected schemes to determine the effectiveness of the project interventions. However, sufficient attention was not paid to ground truthing and in determining whether improvements identified by remotely sensed data were in fact due to project interventions or due to other factors. For instance, in RDS where project work was undertaken in only seven out of the 47, the remote sensing unit reported increases in area irrigated throughout the command and attributed it to NWMP interventions. In fact there were such increase in irrigated area during the monitored year, but it was due to other factors. In spite of these oversights, the use of remote sensing from satellite imagery should be encouraged in other projects both during preparation and appraisal and later during and after implementation. Excellent facility exists for this purpose in India. 42. Training and Institutional Strengthening, both at the national and at the State level was a key element to project success and this has been stressed very clearly in the SAR. Even though a number of studies were undertaken and short-term training courses both in India and abroad have been carried out there was little impact on scheme designs, preparation of operational plans and implementation. This aspect has not been given the importance it deserves during implementation and the Bank supervision missions failed to follow up the implementation of this most essential component. In fact, the funds provided for this purpose were not fully utilized and in the last six months were transferred to cover expenditures of works. The net result is that after eight years of project implementation, only a limited awareness has been created within the IDs on the importance of water management. In retrospect, better results could have been achieved in converting ID engineers from construction people to water managers despite longer term requirement for management changes, if sufficient attention had been paid to this component 43. The structured network system which was developed in the early years of project implementation was not tested in a pilot project in humid rice-growing areas of eastern and southern India. The application of this system in paddy growing areas where field to field irrigation is practiced should be evaluated further. It is difficult to reach a firm conclusion based on practical experience at Borrower concurs with the ICR observation "....regarding the absence of proper measuring devices and other necessary infrastructure.... However, data on physical progress of works, financial expenditure, cropped area, reservoir levels, gauge, and discharge of important canals etc. were maintained by the project authorities right from the beginning and were reviewed from time to time." Refer para 3.8 Appendix 3.  14 this stage since even in the completed schemes of this type, operations have not yet been carried out. The limitation of the structured system in such an environment is that water is delivered at the outlet proportional to the area served, but farmers in the first fields continue to take a larger share before sending it to the next field and so on. In addition, the system does not have the flexibility to conjunctively use rainfall in a localized area. An improved structured system l is currently under preparation and will be applied in other Bank-supported projects. G. ASSESSMENT OF OUTCOME 44. The project outcome is rated unsatisfactory. At the time of appraisal, only a limited number of schemes had been prepared in detail and subsequent scheme preparations were to be vetted by WMC and cleared by an appraisal committee. Provisions were made under the project to strengthen the WMC and, in fact, additional staff in different disciplines were recruited with the exception of an irrigation agronomist. However, the WMC did not have adequate technical support staff to analyze in detail the various scheme reports prepared by the States. Only 32 out of the 96 schemes initiated under the project have been completed at Credit closure. The detailed scheme reports cleared for implementation were deficient in two major aspects contributing to poor performance: (i) proper diagnostic of irrigation infrastructure was not carried out and prioritization of investments was not made; and (ii) outdated agricultural data on crops and cropping patterns were used in the preparation of the operational plans resulting in impractical water management solutions. The result is that NWMP objectives of equitable, predictable and reliable irrigation supplies to all farms in a scheme command have not been achieved in most of the completed schemes and incremental benefiting areas have remained well below 50 percent of expectations in most schemes. H. FUTURE, OPERATION 45. The following actions would be required to ensure completion of remaining schemes and their smooth operation: (a) revision of operational plans, in consultation with fanners, taking into account realistic agricultural data on cropping pattern and operational hydrology; (b) completion of ongoing schemes based on proper diagnostic analysis of infrastructure improvements required to execute the operational plan mentioned above on the basis of an appropriate prioritized investment program for the implementation of which continued financial assistance from the Bank possibly under a follow-up phase would be necessary under stringent institutional obligations; (c) training of ID staff, relevant agricultural department staff and farmers in the operation of the schemes; (d) training of scheme managers including inculcation in O&M culture; and (e) provision of adequate funds for sustained O&M taking into account farmer takeover concept. During the wrap-up meeting with the ICR mission and in a subsequent letter, GOI has indicated that they have advised the Planning Commission to release adequate funds to the concemed State Governments to complete the ongoing schemes. There would still be a need for external assistance to complete the schemes and consolidate the gains from this valuable program. Improved Structured System in India by J.-P. Baudelaire, Principal Irrigation Engineer, SA2NA refers  15 1. KEY LESSONS LEARNED 46. The main lessons learned from this project implementation can be grouped under Concept, Design, Implementation, and Follow-up, as follows: Concept a) Extensive equitable irrigation and self-financing irrigation schemes are two objectives which may not be necessarily compatible. Equity should therefore be achieved through the introduction of volumetric water supply, within the limitations of water availability and the effective recovery of appropriate water charges, rather than through the delivery of the same amount of water per unit of area to each farmer; b) Projects of this type which involve not only technical changes to the status quo, but also have significant social aspects, require a high level of farmer participation in irrigation management (PIM) to be successful. Introduction of such concepts will require, in turn, the use of more advanced technologies in irrigation management which are currently available in India and abroad. These would include provision of improved structured system to allow flexible water distribution to water user associations, and, as needed, canal automation for controlling main canal flows. c) Improving the productivity of irrigation systems linked with farmer participation requires strengthening of the capabilities of the irrigation departments and some institutional restructuring to provide emphasis and staff resources to these activities. In the Water Resources Consolidation Projects (WRCPs) (which are the new generation irrigation projects supported hy the Bank), a major thrust is on improving the productivity of irrigation systems through modernization linked with farmer involvement. The "systems improvement and farmer turnover" components of the WRCPs integrally link farmer organization, operations and maintenance, and cost recovery with the engineering improvements. Additionally, each project has involved an institutional reorganization and subsequent strengthening program to build up the capabilities of the irrigation departments in scheme modernization, water management, andfarmer organization. Volumetric supplies are also being attempted by water users associations and this experiment should be carefully monitored and nurtured. Canal automation is being experimented in Tungabadra and Majalgaon commands. Design a) A baseline survey of the condition of the system infrastructure and cropping practices is a prerequisite to prepare a detailed scheme report for rehabilitation/modernization. b) Detailed diagnostic work would be undertaken and realistic alternate scenarios should be explored in computer simulation models taking into account farmer crop preferences in preparing operational plans. These plans should be prepared with the assistance of specialists in operational hydrology (water availability and water balance studies under the revised scenarios), irrigation agronomy who are familiar with the local conditions, sociologists, representatives of the farrning  16 community, and irrigation engineers. The operational plans should be flexible to handle vagaries of nature and forces of the market place (for instance, if a sugar mill is located near a scheme, water- savings cannot be achieved by switching to a less water consuming crop -- innovative irrigation technology has to be adopted with required capital investments). c) An irrigation system should be considered as a whole water management unit. Works should not be carried out in individual distributaries without any possibility of water management in the system in its totality. Implementation a) Prioritization of expenditures necessary for implementing the selected operational plan should be prepared, together with detailed construction sequences, which take into account the financial capability of the implementing agency, canal closure periods, and number of days available to carry out the civil works in each year. b) Quality of construction in small and scattered works requires particular attention. High quality technical assistance and third party supervision should be considered in future project. c) Priority should be given to training a maximum number of professional staff of the implementing agencies in water management principles, and some of the promising staff should be trained intensively in computer and remote sensing technology to be water managers. Turnover rate of key senior-level staff must be minimized in projects involving new concepts. Otherwise, the benefits of training and efficiency in implementation are lost. Follow-up a) Simple key parameters should be established for monitoring and evaluating the performance and impact of water management from scheme start-up. An effective monitoring and evaluation unit should be created in each participating State with a multi-disciplinary core team. Monitoring the impact of scheme interventions on agricultural production should be carried out continuously starting from the baseline survey and using the same parameters so that operational plans can be modified if required and also a final evaluation of the scheme performance would be feasible. b) O&M of the scheme should be jointly carried out by department officers and farmer beneficiaries. Adequate funds should be made available for such O&M.  17 PART II. STATISTICAL TABLES Table 1: Summary of Assessments A. Achievement of obiectives Substantial Partial Nealigible Not Applicable (/) (,/) (/) Macro policies l O O EII Sector policies E l1 El O Financial objectives ED El L Institutional development El El El Physical objectives El El El Poverty reduction El E0 E3 Fo Gender issues El El E E] Other social objectives El El El FE Environmental objectives El El [I E Public sector management El El El Wi Private sector development El El El Oi Other (specify) 1/ [E El B. Proiect sustainabiliry Likely Unlikely Uncertain (/) (.f (/) Highlv C. Bank performance satisfactory Satisfactory Deficient GO (V) V) Identification l E l Preparation assistance En El Appraisal El1 E Supervision EE W i 1/ Structured systems and satellite remote sensing applications,  18 Highlv D. Borrower performance satisfactory Satisfactory Deficient (.T (1) (e) Preparation L J E Implementation F F g Covenant compliance D Ei Operation (if applicable) l E E Highlv High&v E. Assessment of outcome satisfactory Satisfactory Unsatisfactorv unsatisfactory C: El E  19 Table 2: Related Bank Loans/Credits Loan/credit title Purpose Year of approval Status Preceding operations 1. UP Second Public Construct improved 1983 Closed. PCR issued 1992 Tubewell Project public tubewell systems, (Cr. 1332-IN) modernize existing US$101 million standard tubewells and connect to exclusive 11 KV feeder lines of existing standard tubewell systems. Following operations 1. Bihar Plateau Help GOB efforts to 1992 On-going. Development Project increase rural incomes (Cr.2439-IN) and reduce the incidence of poverty in the plateau area of Bihar. 2. A.P. Cyclone Reconstruction and 1990 Closed. ICR issued 1995. Emergency Recon- rehabilitation of struction Project infrastructure damaged by (APCERP) the cyclone of May 1990. (Cr.2179-IN) and Ln.3260-IN) Table 3: Project Timetable Steps in project cycle | Date planned [ Date actual/ l l latest estimate Identification (Executive Project Summary) over a 2 year period 1984-1986 Preparation November 1985 First IDA mission Appraisal May 1986 Negotiations January 19-23, 1987 Letter of development policy (if applicable) n.a. n.a. Board presentation March 24, 1987 Signing - May 12, 1987 Effectiveness June 15, 1987 August 10, 1987 First tranche release (if applicable) n.a. n.a. Midterm review (if applicable) n.a. n.a. Second (and third) tranche release (if applicable) n.a. n.a. Project completion March 31, 1994 March 31, 1995 Credit closing March 31, 1994 March 31, 1995  20 Table 4: Loan/Credit Disbursements: Cumulative Estimated and Actual (US$ million) FY87 FY88 FY89 FY90 FY91 FY92 FY93 FY94 FY95 Appraisal estimate 1.0 11.20 27.00 47.40 71.10 91.40 106.1 114.0 114.00 Actual - 8.43 8.43 15.66 26.13 37.94 67.5 106.9 129.97 I Actual as % of 0 75 47 33 37 42 64 94 110 estimate Date of final June 20, 1995 disbursement ' Represents depreciation of US dollar against SDR. A total of SDR93.2 million was disbursed, which is equivalent to US$129.97 million. The higher US$ value than at appraisal reflects the exchange rate movements of the US$ vis-a-vis the SDR. Table 5: Key Indicators for Project Implementation A number of schemes covering a total of 2.18 million ha have been initiated in a number of States compared to the twelve schemes identified at appraisal covering 459,976 ha and unidentified schemes covering an area of about 123,029 ha. State-wise details are given in Appendix 3.  21 Table 6: Key Indicators for Project Operation (NOT IDENTIFIED AT APPRAISAL) Table 7a: Training and Studies - Andhra Pradesh Si. Name of Sub- Name of Study Status of Study Main Findings of the Study No. Scheme I 2 3(a) 3(b) 4 1. Rajolibanda Monitoring & Evaluation Completed. Analysis of satellite data covering kharif season of Diversion Scheme of RDS Command Area 1991 and 1992 and rabi seasons of 92-92, 92-93, (RDS) by NRSA 93-94 indicate significant increase in irrigated area even exceeding the service area. Water in both kharif and rabi seasons is now reaching even the tail end of most distributaries, with substantial irrigated areas in this part of command area. Irrigation can be further extended to larger than recommended irrigated area under paddy crop. The condition of paddy under NWMP is showing improvements in both kharif and rabi seasons. Similar improvements in the condition of semi-dry crops is noticed during the rabi season. While more equitable condition in paddy is noticed across distributaries during the kharif season, the condition of semi-dry crops decreases towards the tail end distributaries. The gap in head reaches and tail reach is more in semi-dry crops than paddy. Focus should be on equitable water supply across the command area (Executive Summary is appended). 2. NWMP Phase I Evaluation Study of On- Draft report is The report is under scrutiny. Going Sub-Projects of received. NWMP Phase I  22 Table 7b: Training and Studies - Karnataka STUDIES .... Technical Assistance/Studies Under N WMP . SI. Name of Sub-Schemes Main Findings for the Study No. Name of Study Status of Study (1) 1 (2) (3a) (3b) (4) r, Tungabhadra Sub-Project Modification of canal Study completed. Modified As per the study, the canal Distributary No. 76 of section and distributary design suggested. section is modified to TBLBC No. 76 accommodate required flow of 8.50 n3/sec. 2. Bhadra and Tungabhadra Studies of structures Study completed. Modified An operational plan dividing the Sub-Project irrigation networks structures suggested. distributary into two reaches is recommended to be adopted. Accordingly structures have been modified. 3. Bhadra Sub-Project Monitoring of irrigation Study completed for rabi of Interim report was fumished for water management in 93-94. 1993-94. Rabi season final Bhadra command area report is awaited from NRSA. taken through satellite RSA TRAINING |SI. |Course Title |No.|-l Training During the Year 1990-19911" 1. 2nd Training course on "Monitoring of Irrigation System" for Engineers of Irrigation Projects. 2. Training Course on "Water & Land Management Practices under New Operational Plan for Dharma Scheme" for Sowdies and Work-Inspectors at Hangal. 3. One day for farmers training course on 'Water & Land Management Practices under Operational Plan for Dharma Scheme" for farmers at Hangal. 4. One day for farmers training course on "Water & Land Management Practices under New Operational Plan for Dharma Scheme" for farmers at Hangal. 5. One day for farmers training course on "Water & Land Management Practices under New Operational Plan for Dharma Scheme" for farmers at Hangal. 6. 2nd training course on Evaluation Performance of Irrigation Systems for Engineers of Irrigation Schemes at Hangal. 7. 2nd training course on "Main Canal Operation" for Engineers at Dharwad. 8. Training course on "Water & Land Management Technology under New Operational Plan" for WI & Sowdies of Marconahalli command area at Marconahalli. 9. One day farmers training course on "Water & Land Management Practices under New Operational Plan for farmers at Marconahalli. 10. One day farmers training course on "Water & Land Management Practices under New Operational Plan for farmers at Marconahalli.  23 TRAINING Si. Course Title No.I 11. One day farmers training course on "Water & Land Management Practices under New Operational Plan" for farmers at Marconahalli. 12. 7th short duration training course on "Planning & Design of New Operational Plans" for engineers at Dharwad. 13. Training course on "Water & Land Management Practices under New Operational Plan" for WI & Sowdies for Chandramapalli command at Chandramapalli. 14. One day farmers training course on "Water & Land Management Practices under New Operational Plan" for Chandramapalli for command at Chandramapalli. 15. One day farmners training course on "Water & Land Management Practices under New Operational Plan" for Chandramapalli command area at Chandramapalli. 16. One day farmers training course on "Water & Land Management Practices under New Operational Plan" for Chandramapalli command area at Chandramapalli. Training During the Year 1991-922/ 1. One day refresher course for field level functionaries of Bhadra and Vanivilas Sagar Projects. 2. - do - 3. - do - 4. - do - 5. Field Level Functionaries training course of Work-Inspectors and Sowdies of Areshankar and Ramanahalli Projects under NWMP at Bijapur from 25 to 27 February 1992. 6. One day farmers training course for Areshankar Project under NWMP at Jeeralbhavi on 28-2- 1992. 7. Field Level Functionaries training course of Work-Inspectors and Sowdies of Tungabhadra Project at Munirabad under NWMP from 3 to 5 March 1992. Training During the Year 1992-933' 1. Training course on "New Operation Plans and their implementation" for Engineers of NWMP Schemes. 2. Training course on "Water & Land Management Techniques under NWMP of Kanva & Maidala Schemes for Field Level Functionaries. 3. Training course on "Monitoring of Implementation of NOP of NWMP schemes" for Engineers of NWMP Schemes. 4. Training course on "Evaluation of performance of NWMP Schemes" for Engineers of NWMP Schemes. 5. Training course on "Water & Land Management Techniques under NWMP Schemes" for Field Level Functionaries. 6. Training course on "Main Canal Operation Unit (MCOU) of NWMP Schemes" for Engineers of NWMP Schemes. 7. Training course on "Engineering Design of Canal and Structures of NWMP Schemes" for Engineers of NWMP Schemes.  24 TRAINING Si. jCourse Title No.| 8. Training course on "Water and Land Management Techniques under NWMP Schemes" for Field Level Functionaries of NWMP Schemes. Training During the Year 1993-9441 1. "Water & Land Management Techniques under NWMP Scheme" for Field Level Functionaries of Dharma Naregal Channels and Madag Masur NWMP Schemes. 2. "New Operation Plans and their Implementation" for Engineers of NWMP Schemes. 3. "Water and Land Management Techniques under NWMP Schemes for Field Level Functionaries of Chandramapalli Project. 4. "Monitoring of Implementation of NOP of NWMP Schemes" for Engineers of NWMP Schemes. 5. "Evaluation of Performance of NWMP Scheme" for Engineers of NWMP Schemes. 6. "Water and Land Management Techniques under NWMP Schemes" for Field Level Functionaries. 7. "Main Canal Operation Unit (MCOU) of NWMP Scheme" for Engineers of NWMP Schemes. 8. "Engineering Design of NWMP Scheme Works" for Engineers of NWMP Schemes. 9. "New Operation Plans and Improved Water Management Practices" for Agricultural Personnel of NWMP Schemes. " Number of persons required to be trained under NWMP: 970 Number of persons actually trained: 673 2 Number of persons required to be trained under NWMP: 913 Number of persons actually trained: 377 31 Number of persons required to be trained under NWMP: 264 Number of persons actually trained: 166 4/ Number of persons required to be trained under NWMP: 174 Number of persons actually trained: 280  25 Table 7c: Training and Studies - Kerala Si. Name of Sub- No. of Persons No. of Persons ActuaBy Remarks (Main items No. Schemes under Required To Be Trained Under NWMP covered under training etc. NWMP Trained may be indicated) (1) (2) (3a) (3b) (4) 1. Malampuzha 466 257 Management Improvements ________________ Training was impaned. 2. Mangalam 72 43 3. Pothundy 78 49 4. Peechi 232 127 5. Vazhani 82 53 Table 7d: Training and Studies - Madhya Pradesh TRAINING . Training Details in India/Abroad Name of Name of Persons to be No. of Persons ActuaUy Remarks (Main items SI. No. Sub-Schemes under Trained Trained under NVWM covered on Training etc. Farmers Officers Farmers Officers may be indicated) FARMERS TRAINING 1. Billa project, Shahagarh. 60 20 75 19 Farmers participation and Distt. Sagar Institutional organisation for development of WUOs 2. Tawa project 60 20 98 41 -do - Powerkhed, H'bad 3. Wainganga project 60 20 77 55 - do - Kewalari, Distt. Seoni 4. Chandrakesar project 60 20 97 23 - do - Dewas 5. Harsi project Chinnor, 60 20 84 32 - do - Gwalior 6. SAS, Sachi, Distt. Raisen, Vidisha 7. Avadha Tank Sheopurkala, 60 20 69 16 - do - Morena STUDIES Si. Name of Sub-Schemes Name of Study Status Study Main Findings for No. the Study 1. Bama Project Development of WOUs with To be started in Can be given alter a ultimate aim to turnover the January 1995. completion study. system to WUOs 2. SAS Project (Haliali) - Proposal is - do - - do - - do - under consideration  26 Table 7e: Training and Studies - Tamil Nadu TRAINING .... Training Details in India/Abroad .... Si. Name of Sub- No. Schemes under No. of Persons No. of Persons Remarks (Main items covered under NWMP Required to be Actually Trained training etc. may be indicated) Trained under NWMP && S. && 1. Sathanur 120 + 750 = 870 82 + 448 = 530 Officials include officers from Public Works Department, Agriculture Engineering Department and are trained 2. Kodayar 124 + 1250 = 1374 102 + 2223 = 2325 for: 1) new operational plans 3. Thambaraparaniand 125 + 1100 = 1225 101 + 11150 = 11251 4. Chirtar 2) their implementadon adopting modem techniques for optimisation of water 5. Amaravathy 120 + 500 = 620 104 + 333 = 437 use; 3) to create an awareness among farmers 6. Cumbam Valley of their responsibilities in 7. Marudhanadhi and 120 + 150 = 270 71 + 2010 = 2081 implementing various activities under 8. Manjalar the new operation plans successfulty; 9. Sethiathope and 143 + 1250 = 1393 135 + 489 = 624 4) micro computer applications. 10. Tholudur && = No. of departmental staff. ** = No. of users/farmers. IMTI: Irrigation Management Training Institute at Tiruchirapalli. STUDIES Si. . . . Technical Association/Studies under NWM,MP No. Name of Sub-Schemes Main Findings of the Name of Study Status of Study Study I 1. Sathanur Feasibility survey for Studies completed. Finalising a design for 2. Kodayar establishing VHF dependable, reliable and 3. Thambraparani communications system for economical statewide 4. Amaravathy National Water Management message. Data transmission 5. Sethiathope Project Schemes facility by establishing VHF communication for NWMP l ________________________ sub-projects. l 11 1. Thambraparani Computer Aided Simulation Studies completed. Developing a computer studies for Thambraparani based simulation model for system Thambraparani basin with sufficient detauils to study the major operations for water management and to prepare revised rules of regulation and also for providing implementtion programme for real time l________________________ operation.  27 STUDIES Si. ... Technical Association/Studies under NWMP No. Name of Sub-Schemes Main Findings of the Name of Study Status of Study Study III I. Sathanur Probable maximum flood Studies completed. To design a dependable and 2. Thambraparani (PMF) studies for 11 dams reliable maximum flood for 3. Kodayar in Tambraparani, Kodayar, eleven dams in Ponniar, 4. Amaravathy Sathanur & Amaravathy Thambraparani, Kodayar systems. Conducted under and Amaravathy systems. Dam safety cell IV 1. Sathanur Pilot study on inventorying Studies completed. Identifying the major crops 2. Thambraparani and monitoring Irrigated in the command areas and Agriculture in estimating the cropped area. Thambraparani and Sathanur Command Area V 1. Tholudur Computer Aided Rainfall Studies completed. The arrived rainfall analysis in the Command probabilities will be helpful area of Wellington Reservoir to plan irrigation releases of Vellar Basin for the command area of reservoir. VI I. Sathanur Computer Aided Studies completed. To study in detail the Hydrological study on Ponnaiyar Basin and to characteristics of inflows evolve the scientific reason into Sathanur Reservoir for the decline of inflows in to Sathanur Reservoir in the past years. The impact of rainfall pattem in the past five years is also found out and to prepare the software package on the above aspects.  28 Table 8A: Project Costs Appraisal Estimate Actual" Estimate Item Rs US$ Rs US$ ................................ m illion INVESTMENT COSTS - Scheme Investments 958.6 73.7 4,648.2 158.2 - Engineering & Administration 86.3 6.6 2/ 2/ - Institutional Strengthening 111.3 8.5 357.2 15.7 - Training 32.7 2.5 31 3/ - Special Studies and Programmes 34.5 2.7 31 31 - Maharashtra Induction Training 33.6 2.6 39.8 1.6 - Institutional Support & Training Fund 130.0 10.0 16.1 0.6 Total Investment Costs 1,387.0 106.6 5,061.3 176.1 RECURRENT COSTS - Incremental O&M 109.5 8.5 31 31 Total Baseline Costs 1,496.5 115.1 5,061.3 176.1 Physical Contingencies 125.1 9.6 | Price Contingencies 419.5 32.3 | TOTAL PROJECT COSTS 2,041.1 157.0 5,061.3 176.1 The exact number of schemes as well as detailed costs were not known at the time of appraisal, except for the three schemes - Sathanar, Vani Vilas Sagar and RDS. Comparison of actuals with appraisal estimates for the project as a whole are therefore not meaningful. 2/ Disaggregated data are not available; state-wise expenditures include engineering and administration and O&M along with Scheme Investments. 3/ In the absence of disaggregated data, actuals for training, studies and support are shown within institutional strengthening. Table 8B: Project Financing Source Appraisal Estimate Actual/Latest Estimate (US$ M) (US$ M) IDA 114.0 130.0 GOI 43.0 46.0 TOTAL 157.0 176.0  29 Table 9: Economic Costs and Benefits ._____ ____ _ Appraisal Estimate ICR Estimate Culturable Command Area Under Rehabilitation (ha) 583,105 2,180,000 Increased Production (tons/year) - Three schemes combined (SAR) - Paddy - 7,700 - 1,650 - Groundnuts + 19,980 + 4,510 - Sorghum + 2,870 - Millet + 6,220 - 2,900 - Sugarcane + 20,510 - 38,240 - Cotton + 1,980 + 30 - Coconut + 2,660 + 1,220 Increased Farm Income (per ha) RDS 3,260 3,534 Vani Vilas Sagar 5,220 2,790 Sathanur 1,720 4,848 Thandava 3,412 Malampuzha 8,885 Ratapani 2,665 Thambaraparani 5,034 Peechi 732 Haryana 3,705 Net present value (Rs M) RDS + 108 - 28.8 Vani Vilas Sagar + 74 - 13.7 Sathanur + 78 -13.8 Thandava - 28.0 Malampuzha - 7.0 Ratapani - 7.0 Thambaraparani - 49.0 Peechi - 25.5 Haryana 322.0 Economic Rates of Return (%) - RDS 29 - 6 - Vani Vilas Sagar 45 9 - Sathanur 43 6 - Thanadava - 3 - Malampuzha + 9 - Ratapani - 16 - Thambaraparani + 4 - Peechi - 6 - Haryana + 17 Underlying Assumptions Project life 25 years 25 years Standard Conversion Factor 0.80 0.90 Construction Conversion Factor 0.75 0.90 Unskilled labour valued at: two-thirds 0.90 Financial wage rate (Rs/man-day) 10.0 27-85  30 Table 10: Status of Legal Covenants India National Water Management Project Original Revised Covenant Present fulfilment fulfilment Description of Agreement Section type status date date covenant Comments 3.01a,PA I C ll 11 Maintain separate records and accounts. 3.01b,PA 1 CP ll ll Furnish to IDA certified accounts as audited no later than 6 months after the end of each fiscal year. 3.02,PA I CP 11 ll Maintain accounts of expenditures for which withdrawals are to be on basis of SOEs; retain at least 1 year after audit all records for such expenditures; and include in audited accounts certification that SOEs can be relied upon to support withdrawals. 3.03,PA 11 NC 11 ll Provide adequate O&M funds budget for O&M of being provided each sub-project. by the States are inadequate. SCH2,03,PA 10 C ll 11 Select, plan, design and seek approval of each scheme to satisfaction of IDA. SCH2,04,PA 5 CP 11 11 Establish and maintain PPM and O&M Cells pursuant to satisfactory program. SCH2,05,PA 5 NC 11 11 For each scheme, establish a Committee one year prior to completing investment and thereafter maintain it. SC1H2,06a,PA 9 CP ll lI Under TOR satis- Study carried factory to IDA, carry out in a few out detailed O&M participating Review of NWMP States. schemes.  31 Original Revised Covenant Present fulfilment fulfilment Description of Agreement Section type status date date covenant Comments SCH2,06b,PA 9 NC 12/31/91 11 Review results of Review reports studies with IDA. yet to be received from the States. SCH2,06c,PA 10 NC 12/31/92 11 Prepare an Action Depended on Plan for improving compliance of O&M in the State, SCH2,06b. then implement the Plan. SCH2,07,PA 9 CP 12/31/87 11 Prepare M&E Program and there- after implement said program. SCH2,08,PA 9 CD 06/30/88 11 Furnish progress reports within 3 months of end of reporting period (Apr.-Sept. & Oct.- March: the first report due 6/30/88). SCH2,09,PA 10 CP 12/31/87 11 Prepare Training Training has Program and there- been carried out after implement said in most of the program. States. SCH2,10,PA 11 CD 11 11 Furnish to IDA by end-November (beginning 11/30/87) proposed ftnancial requirements for following fiscal year. SCH2,11,PA 5 CP 12/31/87 11 Establish a Dam DSP esta- Safety Panel with blished in most TOR satisfactory to of the States. IDA and promptly carry out any needed repair work. G014.01a,DCA I C 11 /1 GOI to maintain records and accounts adequate to reflect resources and expenditures for Part A(iii) of the project. GO14.01b,DCA I CD 11 11 Have the records and accounts audited for each FY by independent auditors acceptable to the Association and furnish to the Association.  32 Original Revised Covenant Present fulfilment fulfilment Description of Agreement Section type status date date covenant Comments G014.03,DCA 1 CP ll ll GOI shall have the special account for each FY audited and furnish to the Association not later than 6 months after the end of each year report of such audit. 2,06,PA,MAH 5 C 12/31/87 11 Establish state level committee to coor- dinate Maharashtra component. 3.01a,PA,MAH I C ll ll Maintain records and accounts adequate to reflect resources & expenditures in respect of Part B of the project of departments or agencies responsible for Part B. 3.Olb,PA,MAH I C ll ll Have records for Part B of the project audited and furnish Audit Report to the Association not later than 9 months after end of such year. 3.02,PA,MAH 1 C ll ll For withdrawals on SOEs, maintain separate records and accounts and ensure that these are included in the annual audit under Section 3.01. Status: Covenant Type: C - Complied with 1 - Accounts/audit CD - Compliance after delay 2 - Financial performance CP - Complied with partially 3 - Flow and utilization of project funds NC - Not complied with 4 - Counterpart funding NYD - Not yet due 5 - Management aspects of the project SOON - Compliance expected in reasonably short time 6 - Environmental covenants 7 - Involuntary resettlement 8 - Indigenous people 9 - Monitoring, review and reporting 10 - Implementation 11 - Resource allocation 12 - Regulatory/Institutional action 13 - Other  33 Table 11: Compliance with Operational Manual Statements Statement number and title Describe and comment on lack of compliance 1. OMS 2.25 Cost Recovery Policy The OMS emphasizes the importance of recovering the full cost of operation and maintenance of irrigation schemes with a reasonable proportion of investment costs, subject to repayment capacity of the beneficiaries. None of the participating States are in compliance. Table 12: Bank Resources: Staff Inputs Planned Revised Actual Stage of project cycle _ Weeks US$ Weeks US$ Weeks US$000 Through Appraisal n.a. n.a. n.a. n.a. 154.4 305.2 Appraisal-Board n.a. n.a. n.a. n.a. 164.9 297.4 Board-Effectiveness n.a. n.a. n.a. n.a. 5.5 11.5 Supervision n.a. n.a. n.a. n.a. 495.9 623.9 Completion n.a. n.a. n.a. n.a. 40.0 14.4 TOTAL 860.7 1,252.4  Table 13: Bank Resources: Missions Types of Performance rating Problems Number Specialized Implementation Development CLC PMP FA Stage of Month/ of Days in staff skills status objectives project cycle year persons field represented Through appraisal Appraisal through Numerous missions were mounted by Bank/FAO CP together with National Consultants, amounting to a large number of man-months Board approval to assist the States to prepare the project. This assistance continued in the early years of the project implementation period. Board Approval through effectiveness Supervision I June-July 1987 2 7 E,IE 2 2 2 2 Supervision 2 May 1988 3 - IE(3) 2 2 2 2 Supervision 3 March 1989 12 at different IE(7),E,Ag(2),SA,PE 2 2 2 2 times over a period of six months Supervision 4 August 1989 4 Series of IE(2),E,SA 2 2 2 2 2 missions between May 8 and June 28 Supervision 5 December 89 to July 6 IE(4),E,SA 2 2 2 2 2 1990 Supervision 6 June 1990 6 IE(4),E,SA 2 2 2 2 2 December 1990 8 IE(3),E,FA.Ag(2),PE 2 2 2 f' 2 2 Supervision 7 September 1991 3 IE,E 2 2 2 2 2 Supervision 8 Jan. to Jun/July 92 3 IE,E 2 2 2 2 3 Supervision 9 Jan. to June 1993 3 (issued Oct. 93) IE,E 2 1 2 3 (issued Jul. 93) IE,E 2 2 2 2 3 Supervision 10 Dec. 93 - June 94 2 IE(2),E S S 1 _ 2 (issued July 94) Supervision 11 July 94 to March 95 1 IE(2) ICR Mission I Feb-Mar 95 3 21 IE,E,Ag l ICR Supplementary Nov 95 3 21 IE.E,Ag Mission  35 CLC - Compliance with Legal Covenants. PMP - Project Management Performance. FA - Fund Availability. FA(2) Moderate problems of availability of funds. FA(3) Major problems of availability of funds, which are being addressed adequately. IE - Irrigation Engineer; E - Economist; SA - Systems Analyst. FA - Financial Analyst; Ag - Agriculturalist; PE - Procurement Engineer. Status 3 in Tamil Nadu for Audits and Andhra Pradesh for Overall Status. Missions were undertaken at different periods to different States and a summary supervision report has been prepared.  36 IMPLEMENTATION COMPLETE REPORT INDIA NATIONAL WATER MANAGEMENT PROJECT (Credit 1770-IN) Appendix lA Aide Memoire A. Introduction 1. An FAO/CP mission" visited India between 18 February and 20 March 1995 to review the implementation performance of the National Water Management Project (NWMP) and to collect relevant information and data to prepare the Implementation Completion Report (ICR). As required by Bank regulations, the FAO/CP mission coincided with the last supervision mission for the project carried out by Mr. N.K.Bandyopadhyay"' from the World Bank Resident Mission in New Delhi (SA2NA). Mr. B.Foley"' (Operations and Maintenance (O&M) Specialist) participated in the FAO/CP mission under separate Terms of Reference to evaluate the O&M capabilities of the various State Irrigation Departments (IDs) and to identify the requirements for setting up O&M units (as appropriate) and strengthening existing units in a possible follow-up project. 2. After initial meetings in New Delhi with the officers of the Water Management Cell (WMC) of the Ministry of Water Resources (MOWR) of the Government of India (GOI), the mission undertook field visits to selected subprojects in four States (Tamil Nadu, Kerala, Andhra Pradesh and Madhya Pradesh)'. ,fn Tamil Nadu, Andhra Pradesh and Madhya Pradesh the mission held wrap-up meetings chaired by the Secretary of the ID of each State. In Andhra Pradesh, the mission also visited the National Remote Sensing Agency (NRSA) at Hyderabad and held detailed discussions about the applications of satellite remote sensing technology in irrigation water management with the scientists of the Water Resources Group led by Dr.Thiruvengadachari. On returning to New Delhi, the mission met with the officers of the Irrigation departments of Bihar, Gujarat, Haryana, Karnataka, Orissa and Uttar Pradesh at the World Bank Resident Mission Office. The Director of the Water and Land Management Institute (WALMI) of Maharashtra briefed the mission on the implementation progress of the Maharashtra Induction Training Program financed under NWMP in that State. In New Delhi, .the mission also actively participated in the workshop organized to discuss the findings-of the independent evaluation studies of two of the subprojects covered under NWMP (Sathanur subproject in Tamil Nadu and Bhadra subproject in Kamataka). The evaluation studies were carried out by the International Irrigation Management Institute (IIMI) headquartered in Colombo. The workshop was attended by representatives from the "Messrs.S.Rajagopal (Water Resources Specialist, Mission Leader), Mr. W.l.Sorrenson (Economist, TCII) and Mr. S.B.Hukkeri (Agronomist, National Consultant). 21Task Manager. " Mr.Foley was financed by the Govemment of Australia and was recruited by SA2AG. Mr. Mathur, Senior Joint Commissioner, WMC (MOWR) accompanied the mission during its field visits to Tamil Nadu and Kerala. Mr. Parashar, Commissioner, WMC (MOWR) participated in the wrap-up meeting held in Hyderabad covering the Andhra Pradesh component.  37 Irrigation Departments of the States participating in NWMP, MOWR,GOI and the World Bank Resident Mission. 3. The mission wishes to thank all concerned staff of the Irrigation Departments of the States visited and the WMC of MOWR for the assistance and hospitality extended to the mission members. B. Project Objectives 4. The purpose of the project was to increase agricultural productivity and farm incomes through a more reliable, predictable and equitable irrigation service. To achieve these objectives and facilitate extension of similar activities throughout the country, the project was formulated to help participating States and GOI to develop an institutional capacity to plan, irnplement and monitor improved O&M practices and to provide for low-cost infrastructural improvements designed to support an improved operational plan for selected schemes. In each scheme, the most appropriate technique of water supply was to be adopted. The most important element in scheme improvement was to be the preparation of an operational plan. On the basis of water availability, system characteristics, and agricultural options, the plan was expected to define how the system would be operated with respect to the timings and quantities of water deliveries and the responsibilities of those involved. C. Project Components 51. The principal components of the main part of the project (improved water management) were: (a) Scheme Investments required to implement the operational plan - these included repairs and renovations to existing structures; provision of new control structures; field chanmels and drains; measuring devices; and investments in upgraded O&M facilities and equipment (including housing for staff and transport); and commrunications equipment, as required to execute the operational plan; (b) Engineering and Administration required to supervise the effective implementation of the scheme investments; (c) Incremental Recurrent O&M Expenditures to support the improved operational plan and maintain it after the completion of the scheme investments; -(d) Institutional Strengthening of the State IDs, with particular emphasis on helping them to develop O&M planning and support units in the States. In addition, the WMC of MOWR was to be strengthened so that it could provide policy direction, act as a catalyst for extending the approach across State boundaries and administer and coordinate the NWMP as a whole. In each instance, the project provided for staffing costs, buildings, equipment, vehicles and associated recurrent costs; (e) Training Programs for technical and orientation training for scheme staff and farmners, State O&M staff and officers of the WMC and study tours within India and limited overseas visits along with support for the infrastructure development of the Irrigation Management Training Institute (IMTI) in Tamil Nadu; (f) Special Studies and Programs to strengthen specific aspects of water management (e.g., monitoring and evaluation, computer applications, communications, dam safety etc.). The project also supported two other components related to the general objectives of improved water management in India - (a) The Maharashtra Induction Training Program designed to support initial vocational training for engineers recruited  38 to the Maharashtra ID; and (b) An Instirutional Strengthening and Training Fund (ISTF) to be managed and implemented by GOI and state agencies. D. Project Costs and Financing 6. The project costs including physical and price contingencies were estimated to be US$157.0 million (Rs2,041.1 million)". An IDA Credit of SDR93.2 million (US$114.0 million equivalent) was to finance about 73% of total project costl and the remaining expenditures were to be met from the development budgets of participating states. The project was to be implemented over a seven year period. E. Evaluation of Project Objectives 7. The project concept of instituting water management as a significant input into India's Irrigated Agriculture Sector was most appropriate and in full agreement with the GOI's National Water Policy. Water management has been the focus in several other projects financed by IDA in the water sector in recent years. The Staff Appraisal Report (SAR) was clear in defining the project objectives and in settina out the implementation measures to be adopted for achieving these objectives. The implementation period of seven years was appropriate and the project components included for financing were in general realistic. 8. Agroclimatic and socio-economic conditions in India vary widely depending upon geographic location and irrigation systems have evolved reflecting this diversity. Major types of irrigation systems found in India (localization, warabundi, shejpali) have been discussed in the SAR. However, the bias shown in the SAR towards the warabundi system was, in the mission's opinion, unrealistic. Distinction should have been made between the productive irrigation systems (found in the southern States) and the protective irrigation systems (found in the north and northwest). The three original participating States in this project were from the south - Andhra Pradesh, Karnataka, and Tamnil Nadu. The structured irrigation system which evolved in the early years of project implementation has apparently experienced difficulties in these states (para 22 refers). 9. SAR was also unrealistic to expect that in the preparation of operational plans, paddy growing areas could be reduced and substituted by irrigated dry crops covering a larger area of the command. For instance, in the operational plan presented for the Rajolibunda Diversion Scheme in Andhra Pradesh, it was proposed to preclude paddy in middle and lower reaches in khaMif and throughout the conmmand in rabi by introducing systematic intermnittent irrigation. E. Irnplementation Performance "Cost estimates were based on price levels at the end of 1986 and included about USS9.1 million as foreign exchange cost (6% of the total project cost). vEquivalent to about 70% of the water management components and 100% of the special training and technical assistance activities.  39 10. The project was approved in May 1987 and started off very slowly in the three original participating States. The Induction Training Program in Maharashtra started at the same time as well. At the time of appraisal, the project was expected to cover about 580,000 ha. During the implementation of the project several additional schemes were included and the number of participating States increased from three to nine". In addition bridge financing was provided for lining of main canal and water courses in Haryana for a period of two years to allow that State to continue work on these aspects until the approval of the Haryana Water Resources Consolidation Project in 1994. The original Credit closing date of March 31, 1994 was extended by one year to March 31, 1995 at which time the Credit was expected to be completely disbursed. At the time of Credit closure, it is estimated that about 35 schemes covering a command area of about 345,676 ha would be completed. Since most of these 35 schemes are just getting completed, it is difficult to evaluate their impact on agricultural production and income precisely. 11. ;I appears that the Maharashtra Induction Training Program was well designed and a number of engineers have been given intensive training in hydraulics, hydrology and irrigation related subjects. 12. The implementation perfoimance varies from State to State. The sizes of the schemes rehabilitated range from medium to large. Kerala, which joined the NWMP later (in 1990) appears to have perforrned well. In Tamil Nadu and Karnataka, reports prepared by IIMI indicated that some progress has been achieved in pushing water further down the main canal and distributaries. Since all proportional outlets have not yet been constructed, equitable distribution has not been achieved in its entirety. In most of the 35 schemes, according to the reports prepared by the State irrigation departments, cropped areas have increased, cropping patterns have changed and total production has increased. The mission has collected data from the aaricultural departments and from its own farmer interviews in the schemes visited. These would be compared with the data provided in the reports mentioned above and the impact on agricultural production and on farmers' incomes in these schemes would be evaluated further in the ICR. Of particular relevance is how much of the reported imnprovements is directly attributable to the project interventions and how much is due to other factors such as increased groundwater use, improved agricultural practices, etc. The mission has found that in limited areas where reliability of water supply has been assured, use of inputs has gone up and hybrid varieties are replacing local varieties. On the whole, NWMP has somewhat improved the situation, but the project development objectives have not been achieved. In Kerala, Andhra Pradesh and Madhya Pradesh, there is an overall appreciation for water management at higher levels of Government. 13. It appears, however, that the concept of water management has not been clearly understood by the operating level engineers of the IDs. The idea was to prepare seasonal operational plans for good management of available scarce water resource and then to determine the interventions required to improve the physical infrastructure to carry out such plans. Obviously, deferred maintenance would be part of such interventions but "The additional States were Bihar, Gujarat, Kerala, Mvladhya Pradesh, Orissa, and Uttar Pradesh. The total number of schemes including those included at appraisal was 96 covering a command area of 2.13 million ha.  40 only to the extent required to manage the water resource of each scheme. But, the mission found that deferred maintenance has become the prime objective under the project during implementation. This is evident from the fact that in the schemes visited by the mission structures have been renovated, resectionina of canals have been carried out and new structures have been built in a scattered manner, which will not assist in any rational water management. For instance, in Rajolibunda Diversion Scheme (para. 9), seven out of 47 distributaries have been repaired but not in any contiguous unit. As a result in most schemes actual benefitting areas would be much smaller than the Culturable Command Area (CCA). Similarly, the quality improvement in irrigation service delivered to the farmers would be minimal. 14. Design and conceptual weaknesses were observed in some of the schemes visited by the mission. In Tamil Nadu, the designer has taken advantage of existing drainage courses in sub-catchments to convey irrigation flows to tanks but in designing control structures for flow division the natural run-off and floods usually handled by such courses appear to have been ignored. As a result water has backed up on farmers' fields (inadequacy of the structures) and the annoyed farmers have destroyed the flow division structures. No attempt has been made to review the situation and rebuild the structures. In Andhra Pradesh, small structures built on black cotton soils have failed due to the swelling and shrinking forces. While larger structures have been constructed with appropriate intermediate layers of Cohesive Non-Swelling soils, the same attention has not been given to smaller structures such as proportionate dividers at field outlet levels. Cement plasters have been directly applied to random rubble masonry without any bonding layer resulting in failure of the plasters. Pointing would have been more appropriate in these cases. In the Ratapani scheme in Madhya Pradesh, no work has been carried out in the first 800 m of the main canal where (on the day of the mission's visit) 30% water losses by leakage was measured, but money has been spent further downstream with no significant benefits. Reduction of leakage losses in this section would have provided better water control and perhaps, significant additional benefits. 15. In resectioning canals a standard theoretical approach has been taken which is leading to, in the mission's opinion, O&M problems. Manning "n" value of 0.025 applicable to straight earth channel in good condition has been used in designs together with trapezoidal sections which the O&M staff are unable to maintain. Instead, it would have been preferable to adopt natural regime sections (in unlined canals) with more realistic values for the roughness coefficients with appropriate backwater computations in the design. The mission observed that the field staff were unable to push required discharges after a year of completing some of the resectioning because even a shift in "n" value to 0.030 (due to inadequate maintenance) results in a 20% reduction in the carrying capacity of the channel. Bed slopes are by necessity quite small due to topographic constraints. The mission found that proper maintenance of the canals to maintain design bed slopes is critical for the success of the proportional outlets which are based on hydraulics of flow and maintaining predetermined energy gradients in the channels. In most of the schemes visited (except Thandava in Andhra Pradesh) it appeared that field-level staff were not fully conversant with the hydraulic principles of NWMP and the importance of proper maintenance in achieving these. 16. Involvement of agricultural department staff and farmers in the preparation of seasonal operational plans was, at best, minimal. Actual prevailing cropping patterns  41 vary considerably from those in the design records. These have evolved over the years and to some extent are contributing to the inequitable and often inefficient irrigation service. It would probably not be feasible to provide adequate and equitable supplies to the entire command in some of the schemes which are water-short due to these cropping patterns. However, simulations could have been carried out using these actual cropping patterns in order to determine logical alternatives to improve water management. Instead, it appears that theoretical patterns were used in the analysis and the alternatives reconmmended appear to have tumed out to be less than satisfactory. 17. Farmers do not appear to have been consulted in any systematic manner (as required in the SAR through scheme-level committees, etc.) in preparing alternative operational plans although such plans were to define the rules for their implementation and include structural modifications at the outlet and distributary levels necessary to achieve a more reliable, predictable and equitable irrigation service. 18. Monitoring and evaluation of key parameters" has not been carried out in a systematic manner and such information was not readily available even in those schemes where M&E units were functional. Measurina devices (provided for in the SAR) have not been constructed in most of the schemes visited and where they have been constructed, they are either inoperational or readings are not being taken and analysed to evaluate the efficiency and adequacy of the operational plan in achieving improved water management. F. Factors Affecting Implementation Performance 19. At the time of appraisal, only a limited number of schemes had been prepared in detail and subsequent scheme preparations were to be vetted by WMC of MOWR and cleared by an appraisal committee chaired by the Additional Secretary, MOWR. Provisions were made under the project to strengthen the WMC and in fact additional staff in different disciplines were recruited with the exception of an irrigation agronomist2". However, the WMC did not have adequate technical support staff to analyse in detail the various scheme reports prepared by the States. In the mission's opinion the scheme reports were deficient in two major aspects contributing to poor performance: (i) proper diagnostic of irrigation infrastructure was not carried out arid prioritization of investments was not made; and (ii) outdated agricultural data on crops and cropping patterns were used in the preparation of the operational plans resulting in less than optimal water mangement solutions. 20. The original project scope was to cover a limited number of schemes in the three participating States and to carry out pilot schemes in other interested States. It was stipulated in the SAR that no new schemes would be started after June 1, 1990 (paragraph 3.33 refers). In the first three years of the project implementation, progress was very slow which was understandable since new concepts were beina introduced to the ID staff and several new engineering designs had to be prepared using scientific "Physical progress as compared to targets, crops, cropped areas, productivity and production data at various locations in a scheme command. 'This post was never filled and was subsequently cancelled.  42 hydraulic engineering principles involving the mechanics of flows. At the end of March 1990 less than half of the predicted disbursements had materialised. It appears that at this point, the focus on the development objectives was lost. Both the Bank and GOI were anxious to increase disbursements and this appears to have become the overridifig concern and additional schemes in the three original and another six States were approved beyond the date specified in the SAR". In fact all new schemes were approved only beyond this cut-off date although it was evident that they would not be completed before the original Credit closing date of 31 March 1994. The total estimated cost of the schemes cleared by GOI and approved by the Bank was also far in excess of the funding available under the project2". As the number of schemes to be cleared was quite substantial, the WMC was not. in a position to evaluate these aspects in detail, nor was the Bank's supervision mechanism adequate to effectively cover the large number of schemes. Poor diagnostics and non-prioritization of investments in physical infrastructure could have been avoided if the number of schemes had been kept to a manageable level. 21. Training and Institutional Strengthening both at the GOI level and at the level of the participating States was a key element to project success and this was stressed very clearly in the SAR. This aspect appears to have been ignored and the Bank supervision missions did not also follow up the imnplementation of this most important component. In fact, the funds provided for this purpose were not utilised and in the last six months were transferred to the Works category to cover expenditures. The net result is that after eight years of project implementation, only a limited awareness has been created within the IDs on the importance of water management. In retrospect, the mission is of the opinion that better results could have been achieved in converting ID engineers from construction people to water managers if sufficient attention had been paid to this component. 22. The structured network system which evolved in the early years of project implementation as a result of the bias shown in the SAR towards the warabundi system, was not tested in a pilot project. The mission agronomist is sceptical about the success of such systems particularly where field-to-field irrigation is practiced in paddy growing areas. However, it is difficult to reach a firn conclusion based on practical experience at this stage since even in the so-called completed schemes of this type, operations have not yet been carried out. G. Project Sustainability 23. Project sustainability depends on inculcating O&M culture to the irrigation department staff besides providing adequate funds to carry out the O&M activities. Availability of funds alone is not a guarantee for sustained project operation. Planning for O&M has to be given priority and until this is achieved, project sustainability cannot be ensured. The mission noted that even in places where adequate maintenance funds "Table I page 10 refers J It appears that these additional schemes were cleared assuming that there would be continued financing of the schemes under a follow-up phase.  43 have been released by the State Governments, the maintenance of structures constructed under NWMP was unsatisfactory. If the operational plans devised under the project are not kept up, it would lead to the deterioration of the systems as well. 24. As costs for O&M increase, it is necessary to turn over parts of the system to the beneficiaries. Since farmers have not been involved in the preparation and execution of the schemes under this project, it is necessary to take urgent steps to include them in the operational phase. Water Users Associations at different levels in the system would have to be created as a first step. H. Future Operation / Action Plan 25. The following actions would be required to ensure completion of remaining schemes and their smooth operation: (i) revision of operational plans taking into account realistic agricultural data on cropping pattern and operational hydrology; (ii) completion of ongoing schemes based on proper diagnostic analysis of infrastructure improvements required to execute the operational plan mentioned above on the basis of an appropriate prioritised investment program for the irnplementation of which continued financial assistance from the Bank possibly under a follow-up phase would be necessary ; (iii) training of ID staff, relevant agricultural department staff and farmers in the operation of the schemes; (iv) training of scheme managers including inculcation in O&M culture; and (v) provision of adequate funds for sustained O&M taking into account farmer takeover concept. In addition, it would seem appropriate to provide required technical and financial assistance to the States for preparing a limited number of schemes for inclusion under a possible follow-up project. I. Lessons Learned 26. The main lessons learnt from this project implementation are - A baseline survey of tih& condition of the system infrastructure and cropping practices is a must to prepare a scheme sunmmary (rapid rural appraisal would be required). - Detailed diagnostic work should be undertaken and realistic alternate scenarios should be explored in computer simulation models taking into account farmer crop preferences in preparing operational plans. These plans should be prepared with the assistance of specialists in operational hydrology (water availability and water balance studies under the revised scenarios) and input from a specialist in irrigation agronomy who is familiar with the local conditions. The operational plans should be flexible to handle vagaries of nature and forces of the market place. - Prioritization of expenditures necessary for implementing the selected operational plan should be prepared, together with detailed construction sequences, which take into account the financial capability of the implementing agency, canal closure periods and number of days available to carry out the civil works in each year.  44 While a ceiling on the expenditure to be incurred is desirable to avoid too much money being spent on one scheme, it is necessary to review this concept. The clearing agencies should ensure that the ceiling amount does not become an impediment to execute even priority works and is not used by those preparing scheme reports as a target amount to be spent. Simple key parameters should be established for monitoring and evaluating a scheme from scheme start-up. A monitoring and evaluation unit should be created in each State with a multidisciplinary core team prior to credit effectiveness of any follow-up project. An O&M agency should be established at the State and scheme level (see separate report of B. Foley). An irrigation system should be considered as a whole water management unit. Schemes should not be carried out in individual distributaries without any possibility of water management in the system in its totality. Priority should be given to training a maximum number of ID staff in water management principles and some of the promising people should be trained intensively in computer and remote sensing technology to be water managers. Bank-supported projects in India should introduce advanced technologies in irrigation management currently available and adopt them in demonstration schemes. J. Follow-up 27. Relevant parts of a draft ICR would be prepared and transmitted to the Bank after internal reviews by mid-MIay 1995.  45 Table 1. Scheme Initiation Schedule 1987 1988 1989 1990 [ 1991 1992 [ 1993 1994 Tamil Nadu 3 1 2 2 I 1 1 (10 subprojects) Kerala 1 4 (5 subprojects) _ Karnataka 9 1 7 2 11 (30 subprojects) Andhra Pradesh I 1 5 3 (10 subprojects) Madhya Pradesh 2 12 4 (19 subprojects) Orissa 8 (8 subprojects) Gujarat 2 2 (2 subprojects) Uttar Pradesh 10 (10 subprojects) Bihar 2 (2 subprojects) Maharashtra Induction Training Program Haryana Bridge Financingc Total 10 5 1 17 4 30 28 (96 subprojects) _ _ ] _ _  46 Attachment I Observations of MOWR on the Draft Comments of ICR MIission Aide-Memoire of ICR Mission on NWMP-I l. The draft aide-memoire The appraisal of this project wvas carried submitted by the ICR Mission, after their out with detailed investigation of three month long stay in India, during Feb- schemes. Given the large number of March, 1995 deals primarily with their schemes under execution, it was decided to findings on the performance of four visit and collect relevant information from NWVMP sub-projects visited by them selected schemes. The mission excluded during the above period. Even though the schemes already under evaluation by IIMI Mission had detailed discussions with the since their results could be usedfor ICR project authorities from States other than work as well. The discussions wvith ID where the Mission made personal visit, the officers held at NAew Delhi revealed that aide-memoire does not touch upon the the schemes in their State were just getting outcome of such discussions. This may started Hence they were not discussed in have to be added in the aide-memorre. the Aide-M1emoire. However, information collectedfrom these officers will be reported in detail in the ICR. 2. It would have been more appropriate to have the comrments of the concerned State Governments of the findings of the ICR iMission as brought out in their aide-memoire. However, since referring the matter to the State Governments would have taken more time and as the World Bank have requested for a quick look at the draft before it is finalised by the ICR Mission, only broad observations of the MOWR are being communicated in the following paragraphs.  47 3. Ninety-eight sub-projects were The information given in paragraph 10 of undertaken under the NWMIP programme the RAAMI regarding scheme completions. is in 11 participating States. Of these, 44 based on data collected from all the IDs. sub-projects are either completed or are in The mission has kept the Hat-yana scheme a very advance stage of completion. In as Bridge Financing scheme. This is view of the fact that in the initial 3-4 years shown in Table 1 of the RAAMf. The ICR of NWMP not much progress could be mission is encouraged to learn that GOI made under the project, completion of has requested the Planning Commission to almost half of the schemes undertaken for provide adequate funds for the timely implementation can be considered as a completion of unfinished schemes. good achievement. The remaining schemes are also in various stages of implementation and the Government of India has advised the Planning Commission to provide adequate funds to the State Governments so that the remaining works are completed early.  48 4. The Ministry of Water This comment has been taken into account Resources have separately forwarded to the in paragraph 12 of RA1. The concept part World Bank a copy of the Completion has been discussed at length during the Report prepared by the Government of wrap-up meetings held in Tamil Nadu, India on the NWMP, which deals at length Andhra Pradesh, MlGadhya Pradesh and with various aspects of the programmes New Delhi. This aspect wvould be discussed implementation, including the performance further in the ICR. The mission disagrees of individual states on the various with GOI comment on this subject. components of the projects. It would be seen that water management and maintenance have appreciably improved in the schemes completed under NWMP. The schemes which have already been completed or are nearing completion have shown increase in productivity ranging between 15% and 67% per hectare of land, through the additional irrigated area and by better water use efficiency. The, project interventions have also created conditions for equitable distribution of irrigation water. The tail commands of completed NWMP schemes have started getting irrigation water, some for the first time after the inception. It is not clear what made the ICR team to infer that "the concept of water management has not been in general clearly understood by ID staff'. In fact the above improved situation has been possible only because the concerned irrigation staff followed the principles of NWMP (para F, sub-para 2).  49 5. The schemes included in the The concept of the project and its NWMP programme definitely lacked understanding by various officers of the - maintenance in the past. To bring the implementing agencies including the WMC canal network to proper shape was, have been discussed at length. This therefore, one of the important features of comment itself reflects the situation. This the infrastructural intervention proposed in aspect has been discussed clearly in the project implementation. Unless the paragraph 13 of the RAM. canals were made capable of running adequate waters, as provided in the initial designs, it was not possible to ensure feeding to the tail reaches, as also distributing the irrigation water reliably and equitably to the beneficiaries. Comments of the Mission that "deferred maintenance had become the prime objective under the project during implementation" is, therefore, not an accurate reflection of the performance (para F, sub-para 2). 6. It is not clear how without The sentence referred to in the comment proper investigations by a competent has been deleted from the RAM There construction engineer, the Mission have were two engineers with construction drawn conclusions adversely on the quality experience who accompanied the mission. of construction. Moreover, merely by However, paragraph 14 of the RAM has paying a short visit to one project in been revised suitably. Andhra Pradesh and one in Madhya Pradesh, the mission has expressed the view that the "cement content of plasters in Andhra Pradesh and Madhya Pradesh appears to be quite low". Similarly, quality of construction in some of the schemes in Tamil Nadu has been stated to be poor. It would have been more appropriate if such comments on the quality of construction were made after proper investigations by a construction specialist (para F, sub-para 3).  50 7. Likewise, without going The mission disagrees with this through the detailed designs of the observation. There was plenty of structures, the Mission has drawn opportunity for Tamil iVadui officers to conclusions with regards to flaws in their bring oiut the designs and discuss them. designs (para F, sub-para 3). Such deficiencies wvere noted by the ICR mission leader during his involvement in the preparation and appraisal of the Tamil iVadu WRCP. 8. Instead of banking on the The mission has collected detailed scheme statements of the field level staff, who are reports from the schemes visited and the in most of the cases not fully conversant values for the parameters were taken from with the parameters adopted in the designs these reports. of the structures, it would have been more appropriate if the Mission had discussed such aspects with the responsible technical experts associated with the design of structures (para F, sub-para 4).  51 9. The project authorities Refernce should be made to SAR formulated revised operational plans in paragraph 3.03 page 14 line 6: and most of the completed or nearly completed would provide for selected schemes for schemes and water delivery practice had low-cost infrastructural improvements accordingly commenced under NWNIP. designed to support an improved However, the progress in adoption of the operational plan: Also paragraph 3.08 of operational plans and the riules had not the SAR refers. been studies. In most of the cases the operational plans were either notional or still in draft stage to be provided by the State Governments. It is to be appreciated that the full tempo of the NWMP was attained only in the fourth year of the project. In most of the schemes which have been completed or are nearing completion the operational plan as per the concept of NWMP could be applied only after the various structural interventi.ns were completed. During their field visit to the Thambraparni Project in Tamil Nadu the Mission was shown by the project authorities the operational plans followed by them, both before the commencement of the NWIvIP programme and that adopted during the implementation stage. It, therefore, does not seem to be correct to cormment that no operational plans were being prepared. It was explained by the project authorities to the ICR team during their field visits to Kerala and Tamil Nadu that farmers were associated by them while finalising the irrigation schemes before commencement of every irrigation season. Agriculture and revenue authorities were also invited in the meetings held for the purpose. The operation of the irrigation systems were made strictly as per the schedules finalised under such meetings. (para F, sub-para 6).  52 10. The land holdings in India are This is not what is stated in the aide- in general very small with the result that memoire. Reference should be made to number of beneficiaries under each paragraph 17 of the R4A. irrigation system is very large. It is therefore, not physically possible to deal with individual farmers with regards to running of the irrigation systems. Likewise, it is not possible for the project authorities to inform individual farmers of the structural modifications carried out by them in the conveyance system. It is generally considered adequate to bring the changes carried out by them to the notice of the farmers representatives in the meetings held with them from time to time (para F, sub-para 7). 11. More than eighty-five percent This aspect will be discussed further in the of the funds provided under the NWMP ICR. had been for structural interventions. It was considered necessary to bring the systems to their design conditions first, so that the concepts provided under the NWMP could be made workable. It is, therefore, not justified to say that major stress of the project was on construction of structures and not on management of the system (para F, sub-para 10). 12. The Water Management Cell, This aspect will be furher explored in the in the Union Ministry of Water Resources, ICR with relevant supporting though in itself has been a small unit, but documentation. it has all round technical support from other specialised organisations like CWC, ICAR, Ministry of Agriculture, Planning Commission etc. Even in the Project Appraisal Committee high level representatives from the concerned specialist organisations participate. It is, therefore, not correct to say that WMC did not have technical support staff. Moreover, all the schemes cleared by the GOI were also examined by the World Bank before they could be taken up for implementation (para G, sub-para 3).  53 13. For implementation of the Reference to thinly spreading resources is various sub-projects undertaken under the not with reference to the fund constraint. NWMP programme, initially the funds Paragraph 20 of the RAM refers. were provided by the concerned State Governments out of their own State plans. Only after the works were executed, requests were made by them for reimbursement. At no point of time throughout the duration of the NWMIP programme, fund constraints were experienced by any of the project authorities. The observations of the ICR Mission that the programme suffered because a very large number of projects were undertaken under the programnme which resulted in thinly spreading of the resources is, therefore, not correct (para G, sub-para 2). 14. The remarks of the ICR Mission findings will be presented with Mission with regard to the execution of supporting documentation in the ICR. the NWM`P programmes are contrary to Reports of supervision missions will be the findings of the supervisory Missions of critically reviewed under the category the World Bank which visited the various "Bank Performance". projects from time to time during the life of the NWMP programme. Each of the scheme had adequate engineering and supporting staff provided by the State Irrigation Department. At no point of time the implementation of the NWMP programme suffered on account of shortage of manpower. The implementation of the schemes were also carried out in accordance with a well planned schedule (para G, sub-para 3).  54 15. The available technology with Discussions were held wvith NRSA as the National Remote Sensing Agency reported in paragraph 2 of the RAM. In (NRSA) is still not capable of undertaking fact the ICR mission would recommend - the detailed monitoring as is required in increased use of satellite imagery but with the NWMIP. The development of cloud good ground truthing in any possible penetration technology by NRSA is still in follow-up project right from the start. the infancy stage and they are not able to There are other international agencies monitor at present the coverage under with expertise in this area who could kharif crops, which is predominantly a collaborate wvith NVRSA water resources monsoon period crop. A good amount of group. benefits under the NWMP project are, therefore, left unmonitored in using the remote sensing technique. Moreover, with the available capability they are able to identify the Rabi crops only in three categories, namely (i) paddy, (ii) sugarcane, and (iii) other crops. Break up of crop area under crops other than paddy and sugarcane is still not possible. lDue to this deficiency, gross value of produce of different crops cannot be correctly worked out. The available technology with NRSA is also not capable of interpreting accurately the degraded lands like area under waterlogging, salinity etc. Due to the above reasons, only limited use of remote sensing technology was made for monitoring purposes (para G, sub-para 3).  55 Implementation Completion Report India National Water Management Project (Credit 1770-IN) Appendix LB Supplementary Mission Aide Memoire A. Introduction 1. During February and March 1995, an FAO/CP mission was mounted to prepare the Implementation Completion Report (ICR) for the above-captioned project. The draft ICR was presented to the World Bank in June. The project had been consistently rated as having performed well by the Bank supervision missions. The ICR mission's findings were contradictory to these ratings. Lively discussions followed the submission of the draft ICR. Though the Bank reviewers agreed, in general. with the findings of the ICR mission, some concerns were raised, regarding the lack of information and economic analysis to compare the performance of the three schemes upon which project appraisal was based (RDS, Sathanur and VVS). The Bank, GOI and CP agreed that a Supplementary mission should be mounted to: (i) visit and collect information for economic and other analysis for the SAR schemes not visited earlier and to evaluate their performance; (ii) visit one more scheme in agreement with the Water Management Cell (WMC) of the Ministry of Water Resources (MOWR) to enlarge the sample size; (iii) visit Haryana to assess the implementation of NWMP principles for this component which represented about 25% of NWMP expenditures; and (iv) confirm or amend the earlier ICR mission findings as required. 2. In accordance with the above, an FAO/CP mission visited India between 30 October and 22 November 1995 to undertake the necessary follow-up activities to update and finalize the draft ICR prepared in June 1995. After initial discussions with Messrs. Ridley Nelson (Chief, SA2NA), J.P. Baudelaire (Principal Engineer, SA2NA) and P.C. Mathur (Commissioner, WMC, MOWR, GOI), the mission undertook field visits to Vani Vilas Sagar scheme in Tamil Nadu and Peechi scheme in Kerala selected by the WMC. In addition, the mission visited Haryana to determine to what extent the activities undertaken in this State, involving significant expenditures, followed NWMP principles. In each of these four States, the mission held wrap-up meetings chaired by the Secretary of the Irrigation Department of each State. 3. The mission wishes to thank all concerned staff of the Irrigation Departments of the States visited and the WMC of MOWR for the assistance and hospitality extended to the mission. 4. This aide-memoire supplements the earlier one presented in final form in April 1995, a copy of which is enclosed for ready reference. B. Mission Findings 5. The findings and conclusions of the earlier ICR mission were confirmed during the present mission. Data collected in respect of the four schemes will be used in updating and finalizing the ICR including the required additional economic and financial analysis. The specific findings related to the individual schemes are given below.  56 INDIA: National Water Management Project (Cr. 1770-IN) Implementation Completion Report Supplementary Mission Appendix IB Supplementary Mission - Aide-Memoire 6. Vani Vilas Sagar Scheme (CCA 12,250 ha). At appraisal, it was expected that the project works comprising rehabilitation of canals and structures, provision of new control structures and field channels/drains in the extended command and lining of vulnerable reaches of the main and distributary canals would lead to improvements in water use efficiency, increase in high-value garden crops, mainly coconuts and assurance of rabi cropping. In this scheme, construction consisted of installing proportional outlets, duckbill weirs and tail clusters with limited lining of main canals (LBC, RBC and HLBC). The construction work is, in general, good. Mission observed breakage of some of the outlets by beneficiaries. Unfortunately, re-sectioning of distributaries which is necessary to maintain the full supply depths (which, in turn, are necessary for the proper functioning of the APMs) has not been carried out. The performance of the VVS scheme is directly related to the inflows into the reservoir which are inadequate (the reservoir has not filled in recent years). The mission was advised that minor irrigation works are being constructed in the catchment of the reservoir resulting in further reduction of inflows. A comprehensive water balance analysis of the total system should be carried out and operational plans should be revised accordingly. At the same time, the adjoining non-project area has also experienced expansion of coconut plantings at a higher rate than anticipated at appraisal. In addition, due to inadequate storage in the reservoir, khariff irrigation has not been provided in recent years. As a result, the net benefits are likely to be much lower than levels projected at appraisal. The water balance study is, therefore, essential in determining a sustainable operation for the scheme. Sustainability also depends on adequate maintenance funds being provided to the scheme operators. At the present time the funds provided are inadequate. They only just about cover establishment charges with minimal amounts left over for scheduled maintenance. The operation of the scheme with NWMP principles requires more intensive training for the engineers and the operating level staff of the Irrigation Department. 7. Sathanur Scheme (CCA 18,210 ha). Several measures (improved system control, rehabilitation of canals, tanks and structures, modification of canal capacities, extension of the distribution system as required and provision of direct supply to the tanks) were planned at appraisal to improve water use efficiency and planting of ID crops in both the LBC and RBC. At the time of appraisal, LBC was localized for paddy. A number of proportional outlets, duckbill weirs and by-pass channels to existing system tanks have been constructed. Works have been carried out mainly in the LBC system. The duckbill weirs are as expected, serving their purpose by maintaining water levels in the main canals. The construction quality appears to be good. In Sathanur, as in VWS, improvements have not been carried out in the distributaries and this has led to the malfunctioning of the outlets. Farmers interviewed by the mission complained that water levels are fluctuating and not being maintained properly during the rotation cycle resulting in variable discharges at the outlets. They acknowledged damaging these outlets due to this. Operational level staff of the Water Resources Organization of GOTN should be given proper training in the operation of these structured networks. Additional work on the distributaries is necessary if these outlets are to perform as designed. It may be advisable to construct a full hydraulic model using the  57 INDIA: National Water Management Project (Cr. 1770-IN) Implementation Completion Report Supplementary Mission Appendix lB Supplementarv Mission - Aide-Memoire Froude Number criterion at Sathanur as a training tool to show the operational staff and the farmers how the system is supposed to work. The performance of Sathanur scheme is also dependant on reservoir inflows. The scheme experienced successive drought years and in 1991, no releases were made to irrigation command or the system tanks. Still about 30% of the land was irrigated which indicates the extent of groundwater usage in the command. This would be taken into account in analyzing the impact of the scheme investments. Several flow measurement structures have been established but no measurements have been taken so far. It is recommended that attention be paid to this important aspect which would enable the operators of the scheme to maximize the water use. The scheme did not experience a decline in paddy area as expected at appraisal while the output of groundnuts has also fallen short of appraisal expectations. Sugarcane production has also not reached appraisal projections. 8. Peechi Scheme (CCA 16,000 ha). The physical investments included selective lining, repairs to head regulators, replacement of damaged pipes, sluicegates and provision of measuring flumes. In this scheme, all the outlets have been modified to act as proportional outlets and gates have been installed as well to operate on a fully-open or fully-closed position as required. The construction quality is good. Some measuring flumes have been constructed but in most of them no measurements are being taken. Where measurements are being made, rating curves have not been established. The project engineers informed the mission that after completing the structured outlets, they were able to push the irrigation flows further downstream in each distributary. They have not been able to do this in the current year. The problem in Peechi is the abundant growth of weeds and bushes in the canals, and canal banks restricting access to the system. In addition irrigation is not well developed because of the high rainfall (annual rain around 3,000 mm). Labour is expensive and maintenance funds are inadequate to clear the weeds and bushes as often as required. In fact, at Peechi, priority investment should have been the purchase of bush, weed and sediment clearing machinery which would enable water to reach the farthest portions of the command. If the basic concept provided in the SAR, of analyzing the performance of the scheme, identifying the constraints and then adopting appropriate solutions to implement an operational plan had been followed, the maintenance constraint would have been identified as the major problem in water management in this scheme. At the time of scheme preparation and clearance, it was anticipated that with the physical improvements in the infrastructure, full irrigation would be provided to the second and third paddy crop. This has not been achieved. Throughout the State, the paddy area has been declining due to high cost of inputs, particularly labour and better profitability of cash crops, such as coconut, banana and rubber. Because of these changes, the incremental benefits would tend to be smaller than expectations, i.e. even without the project, increased cash cropping would have continued. 9. Haryana (CCA 1,184,500 ha). Bridge financing was provided to Haryana to continue work on the Bhakra and Yamuna Canal Systems as a follow-up to the Haryana II project financed by the Bank until the WRCP project came on line. Project investments were mainly utilized for lining of main canals and watercourses. No operational plans were  58 INDIA: National Water Management Project (Cr. 1770-IN) Implementation Completion Report Supplementary Mission Appendix IB Supplementary Mission - Aide-Memoire prepared since the rotational system was already in place. No new devices were installed. Thus in the strictest terms, it can be said that NWMP principles were not followed. It would have been appropriate to have amended the DCA prior to proceeding with bridge financing. However, due to lining, some water savings can be expected. Due to the scarcity of measurements, it is possible to derive benefits based on theoretical savings. On this basis, there should be an increase of 38,200 ha in area supplied. There is no change in cropping pattern but increased cropping intensity is a major benefit from lining of canals and watercourses. The quality of lining work is good. C. Impact Evaluation 10. A comparative statement of appraisal estimates and final costs are given in the table below: Appraisal Actual Up 95 Scheme Estimate Expenditure Cost to Complete Final Cost Cost Overrun (Rs million) (Rs million) (Rs million) (Rs million) (Underrun) Vani Vilas Sagar 30.0 52.9 1.4 54.3 81% (Kamataka) Sathanur 46.8 61.9 4.5 66.1 42% (Taniil Nadu) Peechi 25.5 42.0 0 42.0 65% (Kerala) I Haryana 1,220 1' 949.4 0 949.4 (12%) ' Original estimate of Rsl.743 million over dte entire command of 184,600 ha revised to Rsl,220 million to keep expenditre within budget ceiling. 11. Schemes had varying cost overturns with Vani Vilas Sagar at 81%, followed by Peechi (6.5%), and Sathanur (42%). However, Haryana had a cost underrun of 12%. In terms of physical achievement, average lining cost has gone up by 25 % (from Rs 16 per sq. ft. to RsI9.8 per sq.ft.). Cost overrun was caused mainly by price escalation and partly by quantity changes. The relative share of these two elements will be elaborated in the ICR. 12. At this stage, it is evident from available data, supplemented by interviews with farmers, that the actual incremental benefits are much lower than projected at appraisal and scheme preparation. Overall changes in area benefitted, cropping intensity, crop yields and diversification into high-value crops with supporting data collected by the mission will be reviewed and the effect of these changes will be assessed and the results will be presented in the ICR. Preliminary indications are that with the cost overrun, implementation delays and reduced benefits, the ERRs will be substantially lower than those projected at appraisal.  59 INDIA: National Water Management Project (Cr. 1770-IN) Implementation Completion Report Supplementary Mission Appendix IB Supplementary Mission - Aide-Memoire D. Follow-Up 13. The draft ICR prepared in June 1995, will be updated and finalized taking into account the comments received from various reviewers and the findings of the supplementary mission and transmitted to the Bank by end-December 1995.  60 IMPLEMENTATION COMPLETION REPORT INDIA NATIONAL WATER MANAGEMENT PROJECT (Credit 1770-IN) APPENDIX 2 ECONOMIC AND FINANCIAL ANALYSIS A. INTRODUCTION General 1. The NWMP was a multi-state project with different interventions of widely varying impact. Project benefits were expected to arise from management changes, which were assumed to involve "little economic cost." Accordingly, the upper limit for scheme investments was set at Rs2,500/ha, in order that the approach could be replicated and that attention would be concentrated on the highest priority investments. The primary achievement was expected to be the establishment of a systematic and replicable approach to improving the utilization of previously created assets. It was also foreseen that relatively minor increases in agricultural value added would more than justify the direct costs involved and result in a high rate of return. Some conceptual issues relevant to the project merit attention in carrying out re-evaluation at project completion. 2. Future Without-Project. There could be some doubt about the level of incremental agricultural benefits which would have occurred on the assumption that, without project, canals and watercourses would have been neglected - resulting in a falling level of production. The pertinent question to ask to clarify the without-project situation is whether 0 & M works would not eventually be implemented in the place of NWMP if the project were not implemented. The related question is whether 0 & M works with state funds are a genuine alternative to the NWMP. Past records, however, show that funds were allocated for emergency repairs and works, although state budgets for regular 0 & M absorbed staff salaries to a large extent. This is evidenced by the structures which have not gone out of use and the vocally active farmers demanding corrective action which would prevent total breakdown of the system. It is therefore assumed, conservatively, that without-project production will not fall materially. 3. Cost Data. The nominal cost figures of the selected schemes have little relation to the real costs because of adding rupees of quite different purchasing power. To make cost figures more representative of the real resources being used in the schemes, each year's rupee cost was converted into rupees of a constant value before summing them. A Wholesale Price Index (WPI) has been chosen to adjust all rupee costs to the input purchasing power of rupees in FY95.  61 INDIA: National Water Management Project Implementation Completion Report Appendix 2: Economic and Financial Analysis 4. Accrual of Benefits. Foregone benefits on land converted to other crops have been adjusted to derive the incremental net benefits. A one-year time lag is assumed for the accrual of benefits as the farmers receiving water anew or a larger supply than before need time to take many steps towards a more effective use of water for irrigated farming. 5. Comparison of 'Actuals' with Appraisal Estimates. The exact number of schemes as well as detailed costs were not known exactly at the time of appraisal, except for the three schemes - Vani Vilas Sagar, Sathanur and Rajolibanda. Comparison of actuals with appraisal estimates for the project as a whole are therefore not very meaningful. 6. The evaluation of investments shows that none of the scheme interventions, except in Haryana, were able to generate an acceptable rate of return. The calculation does not however include benefits from the utility to households of drinking water, reduction in travel time from head to tail and saving of land width after lining of canals. These have not been quantified for analysis in the absence of reliable monitoring data. The principal factors contributing to the diminished economic performance were much reduced agricultural benefits from a far less benefiting area than was projected at appraisal, excessive cost overruns and the less than anticipated changes in cropping patterns (which at appraisal were expected towards irrigated dry crops requiring less water), than was foreseen at appraisal. 7. This appendix details the economic and financial re-evaluation of NWMP carried out by the ICR mission. It compares the analysis and results of the re-evaluation to the ex-ante economic and financial assessments of NWMP detailed in the SAR. 8. The economic analysis in the SAR was based on detailed analysis of 3 schemes viz. Rajolibanda Diversion Scheme (RDS) in Andhra Pradesh, Vani Vilas Sagar Scheme in Karnataka and the Sathanur Scheme in Tamil Nadu. Project costs were extrapolated to involve some 625,000 hectares. In the SAR an upper limit on scheme investments was set of Rs2,500/ha (1986 prices inclusive of physical contingencies). 9. The ICR mission visited 9 schemes, viz. the Thambaraparani and Sathanur Schemes in Tamil Nadu, the Malampuzha and Peechi Schemes in Kerala, the RDS and Thandava Schemes in Andhra Pradesh, the Ratapani Scheme in Madhya Pradesh, the Vani Vilas Sagar Scheme in Karnataka and the Haryana Scheme. Of the 9 schemes visited by the ICR mission, only Vani Vilas Sagar, Sathanur and the RDS Schemes were analyzed in detail in the SAR. Only a small portion of the proposed physical works of the RDS scheme had been completed by March 19951/, and farmers have taken court action in protest of the Irrigation Department's proposed changes in cropping patterns. The economic re-evaluation reported in this appendix is based on the 9 schemes visited by the ICR mission. Assumptions and Results of the SAR Economic Analysis 10. The major assumptions of the SAR economic analysis were: I/ Out of 47 distributaries only 7 have been completed up to March 1995.  62 INDIA: National Water Management Project Implementation Completion Report Appendix 2: Economic and Financial Analysis (a) economic prices of the tradable crops (rice, cotton, sugarcane, copra and groundnuts) and fertilizers were derived from World Bank commodity price projections; (b) prices of non-traded crops and inputs were derived from prevailing and projected domestic market prices and converted to economic prices by applying a Standard Conversion Factor of 0.8; (c) hired farm labour was shadow priced at 0.67 of the market rate; (d) a specific conversion factor of 0.75 was applied to the civil works; (e) investment costs were assumed to occur in the first 3 years phased at the ratio of 20:50:30; (f) except for the Vani Vilas Sagar scheme, increases in agricultural output due to changing cropping patterns, irrigation intensities and yields would increase in four equal steps from years 2 to 5; (g) the cost and benefit streams were discounted over 25 years and the opportunity cost of capital was taken to be 12%; (h) the total command area of each scheme was assumed to benefit from the project. 11. ERRs were calculated for the three schemes analyzed in detail. In addition an ERR was calculated for the three schemes taken together. Results are tabulated below: ERR RDS Scheme 29% Vani Vilas Sagar Scheme 45% Sathanur Scheme 43% Three Schemes (combined) 37% 12. It was concluded in the SAR that high economic returns could be expected even with the "limited agricultural impact assumed in the analysis." The SAR recognized that the major risks of NWMP would relate to the sustainability of the operational plans to be supported, in the "face of vested interests and the many other factors that affect scheme operations. " 13. It was concluded from sensitivity analysis of variations in crop yields and increases in irrigated areas, that although high economic returns could be expected even with the "limited" agricultural impact assumed under the project, small changes in the assumptions would significantly  63 INDIA: National Water Management Project Implementation Completion Report Appendix 2: Economic and Financial Analysis affect the outcome. By contrast changes in project costs and agricultural prices, were anticipated to have a much smaller impact on the predicted economic outcome. Assumptions and Results of the ICR Economic Re-evaluation 14. The basic assumptions of the re-evaluation are: (a) Following appraisal methodology the project has been re-evaluated over 25 years. (b) The re-evaluation of each scheme has been carried out in early 1995 constant prices. Historical investment and O&M costs for each scheme evaluated have been inflated to 1995 prices by applying the National Wholesale Price Index (WPI) - see Table 1. (c) Project investment and O&M costs have been restated in economic terms by applying the Standard Conversion Factor (SCF) of 0.9 in the absence of disaggregated cost data. (d) Water charges have been excluded from the economic re-evaluation (as they are transfer charges). (e) Border prices have been used for the tradable commodities - rice, groundnuts, wheat, sorghum, soybean, cotton, sugarcane, copra, urea, DAP and MOP, based on World Bank forecasts dated 11 January 1995 for the year 2000 - refer to Table 2. (f) For the non-traded farm commodities and inputs, including hired labour, the current (early 1995) financial prices at each scheme location have been converted to economic prices by applying the SCF of 0.9. (g) As in the SAR the opportunity cost of capital is assumed to be 12%. (h) Like the SAR, benefits of the project have been quantified in terms of increased agricultural production attributable to the scheme interventions. In contrast to the SAR (see para 10, point (h)) only the areas which have actually benefited from the NWMP works have been included in the economic re-evaluation. 15. Full details of the ICR economic re-evaluation for each of the 9 schemes analyzed by the rnission are presented in Section D. A summary of the results is set out below:  64 INDIA: National Water Management Project Implementation Completion Report Appendix 2: Economic and Financial Analysis ERR RDS -6% Vani Vilas Sagar 9% Sathanur 6% Thambaraparani Scheme 4% Malampuzha Scheme 9% Thandava Scheme -3% Ratapani Scheme -16% Peechi -6% Haryana 17% 16. The results of the economic re-evaluation should be considered indicative and not definitive for the following reasons: (a) NWMP works on the 9 schemes re-evaluated have only just been completed. With more time operational plans may be finer-tuned so that the areas benefiting from the NWMP works may possibly increase, although in the opinion of the mission this is an unlikely scenario; (b) only 9 schemes, out of a total of 35 schemes, which will be completed in NWMP, were re-evaluated. It could be argued that the schemes re-evaluated are not representative of the completed schemes. Whilst this is a valid criticism, three of these schemes were, however, chosen by the IDs concerned. It is more than likely therefore that the schemes re-evaluated are representative of the more successful schemes included in NWMP. It should also be pointed out that the remaining schemes which were started under NWMP will in all probability take longer to complete than those which have been included in the re-evaluation, and it is possible that they may never be completed, due to a shortage of funds. Therefore, the economic rates of return from these schemes will most likely be lower than those that have been calculated by the ICR mission in the economic re-evaluation; (c) the agricultural impact of the project at the scheme level has not been closely monitored (refer para 22) necessitating basing the re-evaluation on data and information collected by the mission during the field visits.1' In the mission's opinion, the data collected were of a reasonably high standard and acceptable for re- estimating the ERR of the schemes evaluated.  65 INDIA: National Water Management Project Implementation Completion Report Appendix 2: Economic and Financial Analysis B. AGRICULTUJRAL PRODUCTION IMPACT 17. According to the SAR the project would increase agricultural output primarily through more effective use of canal supplies. Changes in the timing of irrigation, methods of water delivery, efficiency of water use, and the effectiveness of agricultural support services would result in modified cropping patterns, increased irrigation intensity, and higher crop yields and value added. It was recognized in the SAR that the relative importance and effect of the different interventions would vary widely in the schemes that would eventually be included in the project. 18. The three schemes evaluated in detail in the SAR were used "to illustrate the likely agricultural production impact". The SAR states that the project was expected to have the following impacts: (a) a decline in paddy and sugarcane; (b) a marked increase in the output of oilseeds and pulses; (c) a limited impact, either positive or negative, on the output of coarse grains; (d) a positive and substantial impact on high-value cash crops such as cotton and coconut. 19. In the SAR changes in cropping patterns and crop yields were applied to the whole command area of each scheme. The validity of this assumption must be questioned in view of the fact that not all areas within a scheme conmnand could benefit from the NWMP works, since generally adequate irrigation water is supplied throughout the head and middle reaches of irrigation canal schemes; also, it is well known in India that farmers do not readily change their cropping patterns, especially substituting paddy for less water demanding ID crops"' . Despite the fa,t that crop yield increases that were anticipated as a result of NWMP were regarded in the SAR as "modest", the crop yield increases assumed in the SAR (over the entire command areas), were optimistic ranging from an average 15%, 19% and 32% for the Vani Vilas Sagar, Sathanur and RDS schemes respectively. 20. Based on the assumption that the three schemes analyzed in detail would be representative of the NWMP as a whole, the following agricultural impact was estimated in the SAR: 1/ In schemes (RDS, Thandava, Malampuzha, Ratapani and Thambaraparani) visited by the mission, farmers repeatedly said that their first choice of crop to be grown was paddy. The two most important explanative reasons for this behaviour are: their desire to satisfy basic family food requirements first and foremost over and above their desire to earn money from the sale of surplus production; the ease of marketing surplus rice compared to other crops.  66 INDIA: National Water Management Project Implementation Completion Report Appendix 2: Economic and Financial Analysis 3 Schemes NWMP in Total (67,000 ha) (625,000 ha) (tons) (tons) Paddy -7,700 -725,000 Groundnuts +19,980 +186,400 Sorghum +2,870 +26,800 Millet +6,220 +58,000 Sugarcane +20,510 +191,300 Cotton +1,980 +18,500 Coconut (copra equiv) +2,660 +24,700 21. In the SAR it was predicted that, through the introduction of "improved irrigation plans" the distribution of water to farmers was expected to change, leading to the volume of water delivered to the head-reaches decreasing and that to the tail-end increasing. Furthermore, in the SAR it was predicted that cropping pattems would change significantly with ID crops substituting for paddy. 22. Monitoring and Evaluation (M&E) of the schemes visited by the mission has not been sufficiently comprehensive to permit an accurate assessment of the impact on cropping pattems and crop yields of NWMP. Therefore the mission has had to rely on information supplied by Irrigation Department (ID) field staff on the areas within the commands of the schemes which have benefited from the NWMP works. Complementary to this, the mission interviewed farmers in the scheme areas in the tail, middle and head reaches, to establish what changes have occurred, and are likely to occur, pre-NWMP, during NWMP and post-NWMP, with respect to cropping pattems, farming practices, and crop yields. Data on crop areas and yields (the latter from crop cutting trials within the scheme commands) were collected and discussions were held with Agriculture Department field staff" . 23. Assessments of the agricultural impact of NWMP were made by the mission based on information collected from the field visits. Due consideration was also taken of data presented by IDs in their ICRs, as well as the monitoring data provided by the National Remote Sensing Agency (NRSA) I/. 24. Despite difficulties encountered by the mission in obtaining accurate field data on crop areas and yields, and the adequacy of irrigation supply throughout scheme commands, it is nevertheless 1/ Agriculture Department field staff accompanied the mission in the field. 2/ NRSA have monitored 4 NWMP schemes, viz. Malampuzha, RDS, Sathanur and Bhadra. Interpretation of data from the satellite monitoring of these schemes is limited by the following: commonly occurring cloud cover which inhibits the reliability of data interpretation; data analysis of an insufficient time period (at least 5 years are needed) prior to the initiation of NWMP to establish a meaningful pre-NWMP crop area and crop yield base; the lack of rigorous "ground truthing" which is required to reliably correlate satellite data with field level crop yield data, assess the significance of sources of water other than surface irrigation water (viz, groundwater and rainwater) on crop growth, and allow for the impacts of the policing of water delivery on crop growth; inability to differentiate between crops other than paddy and sugarcane.  67 INDIA: National Water Management Project Implementation Completion Report Appendix 2: Economic and Financial Analysis very clear that the agricultural impacts of NWMP predicted by the SAR have not eventuated in the schemes visited by the mission. It is obvious that IDs have spent most, if not all of their time on construction activities, and insufficient emphasis has been allocated to the main project objective of improved water management. Very limited interaction between ID staff and farmers has occurred. The main repercussion of this is that new operational plans devised by the IDs have not been accepted unanimously by farmers. The main reason for non-acceptance of the new operational plans is that farmers place an overwhelrming priority on the growing of paddy. 25. NWMP appears to have had a limited impact on cropping patterns in most of the schemes. The impact of the project in all of the schemes visited by the mission has been limited to certain tailend reaches where water was previously not reaching, or it was reaching but not in sufficient quantity to permit crops to be fully irrigated, so that partial irrigation only was being achieved. 26. No impact of the project was observed in terms of farmer (and ID field staff) consciousness of improved water management. Wasting of water particularly in head reaches continues. Very little if any shift has occurred in the schemes visited from high water demanding crops to lower water demanding more profitable crops, due to the project interventions. 27. Tabulated below are the areas which are benefiting from NWMP in each of the schemes visited by the mission. These areas are compared to the originally planned areas. With the exception of Kerala, the other states visited by the mission have considered the total scheme command areas as having benefited from the NWMP works, which is not consistent with field observations.  68 INDIA: National Water Managenent Project Implementation Completion Report Appendix 2: Economic and Financial Analysis Scheme Areas Benefiting from NWMP (Hectares) Scheme Mission findings Originally Planned Area RDS 1,400 35,410 Vani Vilas Sagar 580 12,250 Sathanur 2,490a' 17,490 Thambaraparani 9,540 45,282 Malampuzha: Tailend 1,108 20,553 Middle 1,890 Thandava 1,040 20,828 Ratapani 350 1,655 Peechi 380 16,000 Haryana 38,165 72,108 " Conversion to groundnut from paddy (853 ha), sugarcane (478 ha) and millet (1,159 ha). 28. The impact of NWMP on agricultural production in the 9 schemes analyzed by the mission is tabulated below. Paddy accounts for significant share of incremental production attributable to NWMP. This is in marked contrast to the agricultural impact predicted on the SAR (para 18) that paddy production would decrease and other crops, particularly groundnuts and sugarcane, would markedly increase.  69 INDIA: National Water Management Project Implementation Com,pletion Report .\Appndix 2: Economic and Financial Analysis Scheme/Crop Incremental Production (tons) RDS Paddy +910 Groundnuts +120 Sunflower +80 Cotton +30 Vani Vilas Sagar Coconut (copra) +1,220 Groundnuts -1,160 Sathanur Paddy -2,560 Sugarcane -38,240 Mllet -2,900 Groundnuts +5,550 Thambaraparani Paddy +4,540 Malampuzha Paddy +4,090 Thandava Paddy +1,090 Groundnuts +190 Sesame -50 Ratapani Wheat +120 Gram +40 Lentil -20 Peechi Paddy +1,050 Haryana Wheat +31,600 Paddy +12,470 Cotton +9,770  70 INDIA: National Water Management Project Implementation Completion Report Appendix 2: Economic and Financial Analysis C. FARM INCOMES SAR Assessment of Farm Income Impacts 29. Representative farm models were compiled for the 3 schemes analyzed in detail in the SAR. The results are summarized below: Increases in Farm Income Estimated in the SAR Scheme Increase in Net Farm Income RDS 66% Vani Vilas Sagar 30% Sathanur 31% 30. Farm net incomes (inclusive of the return to family labour) were estimated in the SAR to increase on average throughout the whole command areas of the schemes by 30% to 66%. ICR Re-evaluation of Farm Incomes 31. In the re-evaluation at completion, representative farm models were prepared for the benefiting areas of each of the 9 schemes visited by the ICR mission. The analysis is elaborated in detail in Section D and the results are summarized below. It must be stressed that these changes in farm incomes refer to fanns in the limited benefiting areas only of the schemes, and not to the all of the farms in the total command areas of the schemes, which the SAR farm income estimates applied to. Comparison with SAR estimates is therefore not very meaningful. NET FARM INCOME (Rs/ha - 1995 prices, per benefiting hectare) Scheme Without Project With Project % Increase RDS 18,020 21,554 20% Vani Vilas Sagar 32,940 35,730 8% Sathanur 15,597 20,445 31%  71 INDIA: National Water Management Project Implementation Completion Report Appendix 2: Economic and Financial Analysis Without Project With Project % Increase Thambaraparani Scheme Manimathur Command 11,683 13,248 13% Drainage Zone 14,545 16,752 15% Thambaraparani Command 16,746 18,008 8% Malampuzha Scheme Tailend 14,179 21,241 50% Middle Reach 21,682 23,505 8% Thandava Scheme Tailend 3,822 7,234 89% Ratapani Scheme Tailend 6,111 7,791 28% Middle Reach 8,142 9,127 12% Peechi 8,020 8,752 9% Haryana 12,550 16,255 30% 32. It is seen that none of the appraised schemes (Vani Vilas Sagar, Sathanur and Rajolibanda) were able to generate as high rate of income growth as expected at appraisal. Increases in farm income have ranged between 8% and 89%/o. The highest of 50% and 89% are in the tailend benefiting areas of the Malamapuzha and Thandava schemes respectively which were essentially rainfed prior to NWMP. In the Thambaraparani and Ratapani schemes NWMP has resulted in a more reliable irrigation supply which has led to: (a) a changed cropping pattern (generally a more intensive use of high water demanding crops - normally paddy) and higher crop yields; or (b) higher crop yields without any impact on cropping patterns. However, a relatively higher farm income in absolute figures in Vani Vilas Sagar reflect the prevalence of high-value coconuts.  72 INDIA: National Water Management Project Implementation Completion Report Appendix 2: Economic and Financial Analysis D. ECONOMIC RE-EVALUATION 33. The benefits and costs assessed for 9 of the schemes visited by the ICR mission, are detailed next. The parity prices used in the re-evaluation of the tradeable commodities are given in Table 2. In Tables 3-8 the financial investment and O&M expenditures (in current Rupees) are given. These 12-monthly expenditure figures were inflated to 1995 prices using the National Wholesale Price Index (WPI). The WPI series used is given in Table 1. RDS Scheme 34. Benefits. No reliable data are available on areas which have benefited from the project. The mission has estimated that some 1,400 ha (about 4% of the command), may be benefiting from additional water. From ID records there has been no marked change in cropping patterns following NWMP. 35. The economic benefit and cost flows of the RDS scheme economic re-evaluation are given in Table 3, and the phasing-in of the benefiting areas are shown in Table 9. The cropping patterns and crop budgets derived to assess the agricultural benefits are presented in Tables 11 and 12 and Tables 13-21 respectively. The annual net production benefits with- and without-project are shown in Table 3. The without-project O&M costs are considered as a benefit (saved cost) in the analysis, since the full O&M costs during and after project implementation have been treated as a cost. 36. Costs. The project economic investment costs used in the re-evaluation are given in Table 3 and are based on the actual expenditure incurred during implementation of the works in financial years 1988/89 to 1995/96. The financial costs have been converted to economic costs by applying the SCF of 0.9 and inflating the costs to 1995 prices using the WPI. O&M costs (expressed in 1995 economic prices) incurred from 1988/89 to 1994/95 and in subsequent years are shown in Table 3. These costs were based on the Government of Andhra Pradesh O&M data which indicates that in 1995 prices approximately Rsl 00/ha was spent on O&M prior to NWMP, compared to Rsl 75/ha at present. 37. Economic Analysis Results. Discounting the net benefit stream (see Table 3) at 12% the NPV is calculated at minus Rs28.843 million and the ERR is minus 6%. 38. Farm Income Results. Farm models were prepared for two distributaries which have been included in NWMP. Distributary 16a is considered as representative of the kharif distributaries, and Distributary 24 representative of the rabi distributaries. Per hectare gross income, costs of purchased inputs and net farm income for Distributaries 16a and 24 are presented in Tables 22 and 23 respectively. The results of the farm model analysis are summarized below:  73 INDIA: National Water Management Project Implementation Completion Report Appendix 2: Economic and Financial Analysis FARM MODEL WITHOUT-PROJECT WITH-PROJECT % NET INCOME NET INCOME CHANGE (Rs/ha) (Rs/ha) Distributary 16a 9,679 11,351 17% Distributary 24 8,341 10,203 22% Net incomes are assessed to have increased by 17% in Distributary 16a and 22% in Distributary 24. Vani Vilas Sagar Scheme 39. Investment and 0 & M Costs. At appraisal, the physical works included rehabilitation of canals and structures, provision of new control structures and field channels! drains in the extended command, and lining of vulnerable reaches of the main and distributary canals. The completion of these works was expected to result in improvements in water use efficiency, increase in high value garden crops, mainly coconut, and assurance of rabi cropping. At project closure, construction consisted of installation of proportional outlets, duckbill weirs and tail clusters with limited lining of main canals - left bank, right bank and high level left bank. Investment costs in financial terms amounted to Rs54.23 million (comprising Rs52.88 million up to March 31, 1995, and Rsl.35 million costs to complete), compared to the appraisal estimate of Rs3O.00 million, giving a cost overrun of about 81% in nominal terms or 25% in real terms. The cost per hectare (coconuts) has gone up by 547% from RslO,OOO/ha (appraisal) to Rs64,655/ha (Credit closure) in real terms. 40. Incremental 0 & M costs consist of actual expenditure claimed from the Bank from FY93 to FY95 (100% reimbursement) and 2% of investment costs (at Rsl.085 million) thereafter. 41. Economic Benefits. Following appraisal methodology, only incremental agricultural production benefits have been quantified for economic analysis. At appraisal, the incremental area assumed to benefit is taken at 3,000 ha for coconuts, compared to a much reduced area (580 ha) at scheme completion. Production benefits from annual crops are projected to accrue with a one-year time lag from the completion of works each year while benefits from new planting (coconut) are expected to begin six years from the date of planting. The cropping pattems and crop budgets required to estimate agricultural benefits are presented in Tables 24-27 The foregone benefits on land converted to coconut with-project have been adjusted to derive the incremental net benefits. The annual net incremental benefit is given in Table 28. 42. Economic Costs. Historical investment and 0 & M costs have been inflated to 1995 prices by applying the wholesale price indices (WPI). These costs, including labour, have been restated in border terms by using the standard conversion factor (SCF) of 0.9. The economic costs thus derived are shown in Table 4.  74 INDIA: National Water Management Project Implementation Completion Report Appendix 2: Economic and Financial Analysis 43. Economic Analysis. The economic analysis has been carried out in 1995 constant prices. Financial prices and economic values of crops and inputs are presented in Tables 2 and 26. Future prices are based on World Bank Primary Commodity Forecasts dated February 1995. Based on the above assumptions, the cropping pattems and per ha crop yield assumptions shown in Tables 24-27 and a 25-year period of analysis, the current estimate of ERR for the scheme is 90/o, 36% lower than the appraisal estimate. At completion, a smaller area of 580 ha has actually benefited from project works, compared to 3,000 ha projected at appraisal, taking only coconut as the benefiting crop. It was assumed that this area of 580 ha is 50% of the increase while the other 50% is irrigated by groundwater, irrespective of project investment. The foregone benefits from this conversion to coconut from annual crops have been adjusted to derive increased incremental benefits as shown in Table 28. The major factors, which contributed to lower ERR include excessive cost overruns per ha and lower production benefits from a much reduced incremental benefiting area (580 ha) compared to appraisal estimate (3,000 ha), which are not fully offset by better prices for scheme production, mainly copra. A sensitivity analysis shows that a 50% increase in area irrigated will bring the ERR to 12%, while a reduction in costs by 50% will raise the ERR to nearly 15%. 44. Farm Incomes. At appraisal, a representative farm model (farm size of 3 ha) had been used to estimate the project impact on farm incomes at the then prevailing financial prices which were used to value outputs and inputs. The results showed an incremental income of Rs5,220, compared to Rs8,385 generated at scheme completion (see Table 29). Farm Income (Farm size - 3 ha) Without Project With Project (Rs) (Rs) Gross value of production 113,407 121,665 Production cost including water charges 14,593 14.466 Net return 98,814 107,199 Incremental = Rs8,385 45. The increase in net farm incomes over appraisal estimates is largely explained by the higher prices of copra.  75 INDIA: National Water Management Project Implementation Completion Report Appendix 2: Economic and Financial Analysis Sathanur Scheme 46. Investment and 0 & M Costs. Several measures were planned at appraisal to improve water use efficiency and planting of irrigated dry (ID) crops in both the left and right bank canals. At project closure, a number of proportional outlets, duckbill weirs and by-pass channels to existing system tanks had been constructed. Works have been carried out mainly in the LBC system, which was localized for paddy at appraisal. The resulting expenditures in financial terms amounted to Rs68.5 million (comprising expenditures of Rs6l.9 million up to March 31, 1995, plus costs to complete of Rs6.571 million), compared to the appraisal estimate of Rs46.8 million, giving a cost overrun of 46% in nominal terms or a cost underrun of 90/0 in real terms. 47. Incremental 0 & M costs consist of actual expenditure claimed from the Bank (100% reimbursement) from FY92 to FY95 and 2% of investment costs (at Rsl .37 million) thereafter. 48. Economic Analysis. At appraisal it was expected that a rainfed dry cropped area of 9,777 ha would be converted into irrigated dry crops. At scheme completion, a much smaller area (2,490 ha) had been converted to groundnuts from paddy (853 ha), sugarcane (478 ha) and millet (1,159 ha). The foregone benefits on land converted to groundnuts with-project have been adjusted to derive incremental net benefits, as shown in Table 30. It is assumed that about 50% of total area of 1,368 ha irrigated by groundwater wells would be from well recharged by canal supplies with-project. 49. Historical investment and incremental 0 & M costs have been inflated to 1995 prices using WPI. These costs including labour have been restated in border terms by applying the SCF of 0.9. The economic costs derived in this manner are given in Table 5. 50. Economic Analysis. The economic analysis has been carried out in 1995 constant prices. Financial prices and economic values of crops and inputs are shown in Tables 2 and 31. Future prices are based on World Bank Primary Commodity Forecasts dated February 1995. Based on the above assumptions, the cropping patterns and per hectare crop yield assumptions shown in Tables 32 and 33 and a 25-year period of analysis, the current estimate of ERR for the scheme is 6% compared to the appraisal projection of 43%. A sensitivity analysis shows that decreasing costs by 50% will improve the ERR to about 21% while improving benefits by 50% will push up the ERR to 14%.  76 INDIA: National Water Management Project Implementation Completion Report Appendix 2: Economic and Financial Analysis 51. Farm Incomes. The result showed an incremental income of Rs4,848 (based on farm size of 1.5 ha) compared to Rs1,720 at appraisal (see Table 34). Farm Income (Farm size - 1.5 ha) Without Project With Project (Rs) (Rs) Gross value of production 26,808 33,231 Production cost including water charges 11.211 12.786 Net return 15,597 20,445 Incremental = Rs4,848 Thambaraparani Scheme 52. Benefits. The mission, in consultation with ID field staff, established that there are three discrete areas benefiting from the NWMP works carried out to date: (i) 740 ha in the Manimuthar command where, because of a more assured and timely supply of irrigation water, the area of the pishanam paddy has increased by approximately 67 ha and the average crop yield has increased about 10% from 4,500 to 5,000 kg/ha. (ii) 4,000 ha has benefited from improved drainage works carried out as part of NWMP. Over this area the pishanam paddy yields have increased on average 15% from 3,500 to 4,000 kg/ha. (iii) 4,800 ha of the Thambaraparani command. Although no change has occurred in cropping patterns throughout this area attributable to the NWMP works, paddy yields for both the pishanam and kar crops have increased approximately 5%. 53. The economic benefit and cost flows of the Thambaraparani economic re-evaluation are given in Table 6 and the phasing-in of the areas benefiting from the project are shown in Table 10. The cropping patterns and crop budgets derived to assess the agricultural benefits are presented in Tables 35-37 and Tables 38-51 respectively. 54. The agricultural benefits have been phased directly in relation to the expenditure profile achieved during implementation. The annual net production benefits with- and without-project are shown in Table 52. The without-project O&M costs, which are based on the 1988/89 actual O&M costs, are treated in the analysis as a benefit (see Table 52).  77 INDIA: National Water Management Project Implementation Completion Report Appendix 2: Economic and Financial Analysis 55. Costs. The project economic investment costs used in the re-evaluation are given in Table 52 and are based on the actual expenditure incurred during implementation of the works in financial years 1988/89 to 1994/95. The financial costs have been converted to economic costs by applying the India SCF of 0.9 and inflating the costs to 1995 prices using the WPI (Table 1). O&M costs (expressed in 1995 economic prices) incurred from 1988/89 to 1994/95 are shown in Table 52. 56. Economic Analysis Results. Discounting the net benefit stream (see Table 52) at 12% the Net Present Value (NPV) is calculated at minus Rs48.658 million and the Economic Rate of Retum (ERR) is 4%. 57. Farm Income Results. Three farm models representative of the areas benefiting from the NWMP works are shown in Tables 53, 54 and 55, where the per hectare gross income, costs of purchased inputs and net farm income are presented. The crop budgets used to prepare the farm models, calculated in constant 1995 financial prices are detailed in Tables 53, 54 and 55. The results of the farm model analysis are summarized below: FARM MODEL WITHOUT-PROJECT WITH-PROJECT % NET INCOME NET INCOME CHANGE (Rs/ha) (Rs/ha) Manimuthar Command 11,683 13,248 13% Drainage Affected Area 14,545 16,752 15% Thambaraparani Command 16,746 18,008 8% 58. Net incomes are assessed to have increased by 13% and 8% in two of the benefiting areas (Manimuthar and Thambaraparani commands respectively) and 15% in the zone that is affected by poor drainage. 59. Retums to family labour were also calculated and results are tabulated below: FARM MODEL WITHOUT-PROJECT WITH-PROJECT % LABOUR RETURN LABOUR RETURN CHANGE (Rs/day) (Rs/day) Manimuthar Command 328 361 10% Drainage Affected Area 306 353 15% Thambaraparani Command 294 316 7%  78 INDIA: National Water Management Project Implementation Completion Report Appendix 2: Economic and Financial Analysis 60. Retums to family labour are calculated to have increased 10% and 7% respectively in the Manimuthar and Thambaraparani Commands, and 15% in the improved drainage zone. Malampuzha Scheme 61. Benefits. In the Malampuzha scheme, irrigation is provided for the second crop of paddy while the first paddy crop is rainfed. In consultation with ID field staff, the mission established that 1,108 hectares have benefited in tailend portions of the project, and a further 1,890 hectares in the middle reaches. The tailend areas were rainfed prior to NWMP and are now irrigated. Average yields have increased from about 1,800 kg/ha to 4,300 kg/ha. The middle reach areas were partially irrigated prior to NWMP and are now receiving adequate irrigation water. Average yields have increased from 3,800 kg/ha to 4,500 kg/ha. 62. The economic benefit and cost flows of the Malampuzha economic re-evaluation are given in Table 56 and the phasing-in of the areas benefiting from the project are shown in Table 57. The cropping patterns and crop budgets derived to assess the agricultural benefits are presented in Tables 58 and 59 and Tables 60-65 respectively. 63. From interviews with local farmers and Agriculture Department field staff, it was assumed that tailend farmers change from rainfed to irrigated farming practices over a three year period after receiving irrigation water. In the analysis it was assumed that middle reach farmers would respond immediately to fully irrigated practices following the arrival of additional water. The annual net production benefits with- and without-project are shown in Table 56. The without-project O&M costs, which are based on the 1990/91 actual O&M costs, are considered as a benefit (saved cost) in the analysis and are shown in Table 56. 64. Costs. The project economic investment costs used in the re-evaluation are given in Table 56 and are based on the actual expenditure incurred during implementation of the works in financial years 1990/91 to 1994/95. The financial costs have been converted to economic costs by applying the India SCF of 0.9 and inflating the costs to 1995 prices using the WPI (Table 1). O&M costs (expressed in 1995 economic prices) incurred from 1990/91 to 1994/95 are shown in Table 56. 65. Economic Analysis Results. Discounting the net benefit stream at 12% the NPV is calculated at minus Rs6.779 million and the ERR is 9%. 66. Farm Income Results. Two farm models were prepared representative of the tailend and middle reach benefiting areas of the scheme (see Tables 58 and 59). Per hectare gross income, costs of purchased inputs and net farm income are presented. The crop budgets used to prepare the farm models, calculated in constant 1995 financial prices, are detailed in Tables 60-62. The results of the farm model analysis are summarized below:  79 INDIA: National Water Management Project Implementation Completion Report Appendix 2: Economic and Financial Analysis FARM MODEL WITHOUT-PROJECT WITH-PROJECT % NET INCOME NET INCOME CHANGE (Rs/ha) (Rs/ha) Tailend 14,179 21,241 50% Middle Reaches 21,682 23,505 8% 67. Net incomes are assessed to have increased by 50% in the tailend areas row filly irrigated having previously grown rainfed paddy only, and 8% in the middle reach areas which have been fMlly irrigated following NWMP. 68. Returns to family labour were also calculated and results are tabulated below: FARM MODEL WITHOUT-PROJECT WITH-PROJECT % LABOUR RETURN LABOUR RETURN CHANGE (Rs/day) (Rs/day) Tailend Benefiting Areas 284 386 36% Middle Reach Benefiting 394 427 8% 69. Returns to family labour are calculated to have increased by 36% and 8% on the tailend and middle reach benefiting farms respectively. Thandava Scheme 70. Benefits. According to ID field staff an additional 1,040 ha in tailend reaches are now receiving adequate irrigation water, when previously most of these areas were rainfed. There has not been any marked change in cropping patterns following NWMP, except that sesame grown under rainfed conditions has been replaced by more profitable ID crops where water has been assured. The areas of paddy which were previously grown under rainfed conditions are now irrigated in the tailend benefiting areas. 71. The economic benefit and cost flows of the Thandava scheme economic re-evaluation are given in Table 66 and the phasing-in of the areas benefiting from the project are shown in Table 67.  80 INDIA: National Water Management Project Implementation Completion Report Appendix 2: Economic and Financial Analysis The cropping pattems and crop budgets derived to assess the agricultural benefits are presented in Tables 68 and 69 and Tables 70-73 respectively. In the analysis it is assumed, based on discussions held with farmers, that benefits have been realized immediately additional water has been supplied. The annual net production benefits with- and without-project are shown in Table 66. The without-project O&M costs, which are based on the 1988/89 actual O&M costs, are considered as a benefit (saved cost) in the analysis since the fulil O&M costs during project implementation have been treated as a cost. 72. Costs. The project economic investment costs used in the re-evaluation are given in Table 66 and are based on the actual expenditure incurred during implementation of the works in financial years 1988/89 to 1995/96. The financial costs have been converted to econornic costs by applying the India SCF of 0.9 and inflating the costs to 1995 prices using the WPI (Table 1). O&M costs (expressed in 1995 economic prices) incurred from 1988/89 to 1994/95 and in subsequent years are shown in Table 66. These costs were based on the Government of Andhra Pradesh O&M data which indicates that in 1995 prices approximately RslOO/ha was spent on O&M prior to NWMP, compared to RsI75/ha at present. 73. Economic Analysis Results. Discounting the net benefit stream at 12% the NPV is calculated at minus Rs28.43 million and the ERR is minus 3%. 74. Farm Income Results. A farm model was prepared to represent the tailend benefiting areas of the scheme (see Tables 74 and 75). Per hectare gross income, costs of purchased inputs and net farm income are presented. The crop budgets used to prepare the farm model, calculated in constant 1995 financial prices are detailed in Tables 76-79. The results of the farm model analysis are summarized below: FARM MODEL WITHOUT-PROJECT WITH-PROJECT % NET INCOME NET INCOME CHANGE (Rs/ha) (Rs/ha) Tailend 3,822 7,234 89% Net incomes are assessed to have increased by 89%/o in the tailend NWMVIP works benefiting areas. 75. Retums to family labour were also calculated and results are tabulated below: FARM MODEL WITHOUT-PROJECT WITH-PROJECT % RETURN LABOUR RETURN CHANGE (Rs/day) (Rs/day) Tailend 233 249 7%  81 INDIA: National Water Management Project Implementation Completion Report Appendix 2: Economic and Financial Analysis 76. Family labour input is assessed to have increased from 66 man-days to 84 man-days, an increase of 27% on the tailend benefiting farms, while returns to family labour are calculated to have increased 7%. Ratapani Scheme 77. Benefits. In consultation with ID field staff the mission established that 160 hectares have benefited at the tailend of scheme main canal, and 190 hectares in the middle reaches. The tailend benefiting areas were receiving only one watering in the rabi prior to NWMP and are now receiving two waterings. Farmers have continued to plant drought tolerant local varieties of wheat in these areas. The middle reach areas are now receiving adequate irrigation water and farmers are planting HYV of wheat where they were previously planting local wheat varieties. 78. The economnic benefit and cost flows of the Ratapani scheme economic re-evaluation are given in Table 80 and the phasing-in of the areas benefiting from the project are shown in Table 81. The cropping patterns and crop budgets derived to assess the agricultural benefits are presented in Tables 82 and 83 and Tables 84-87 respectively. The analysis assumes, based on mission discussions with farners, that benefits have been realized immediately additional water has been supplied. The annual net production benefits with- and without-project are shown in Table 80. The without-project O&M costs, which are based on the 1991/92 actual O&M costs, are considered as a benefit (saved cost) in the analysis since the full O&M costs during project implementation have been treated as a cost. 79. Costs. The project economic investment costs used in the re-evaluation are given in Table 11 and are based on the actual expenditure incurred during implementation of the works in financial years 1991/92 to 1994/95. The financial costs have been converted to economic costs by applying the India SCF of 0.9 and inflating the costs to 1995 prices using the WPI (Table 1). O&M costs (based on data provided by the Government of Madhya Pradesh and transformed into 1995 economic prices) incurred from 1991/92 to 1994/95 and in subsequent years are shown in Table 80. 80. Economic Analysis Results. Discounting the net benefit stream at 12% the NPV is calculated at minus Rs6.79 million and the ERR is minus 16%. Should the remaining 180 hectares at the tailend also receive an additional irrigation, assuming the same investment costs, the ERR increases to 12%. 81. Farm Income Results. Two farm models were prepared representative of the tailend and middle reach benefiting areas of the scheme (see Tables 82 and 83). Per hectare gross income, costs of purchased inputs and net farm income are presented. The crop budgets used to prepare the farm models, calculated in constant 1995 financial prices, are detailed in Tables 88-91. The results of the farm model analysis are summarized below:  82 INDIA: National Water Management Project Implementation Completion Report Appendix 2: Economic and Financial Analysis FARM MODEL WITHOUT-PROJECT WITH-PROJECT % NET INCOME NET INCOME CHANGE (Rs/ha) (Rs/ha) Tailend 6,111 7,791 28% Middle Reaches 8,142 9,127 12% 82. Net incomes are assessed to have increased by 28% in the tailend areas and 12% in the middle reach areas benefiting from NWMP works. 83. Retums to family labour were also calculated and results are tabulated below: FARM MODEL WITHOUT-PROJECT WITH-PROJECT % LABOUR RETURN LABOUR RETURN CHANGE (Rs/day) (Rs/day) Tailend Benefiting Areas 226 279 24% Middle Reach Benefiting 275 296 8% 84. Returns to family labour are calculated to have increased by 24% and 8% on the tailend and middle reach benefiting farms respectively. Peechi Scheme 85. Investment and 0 & M Costs. The physical investments at appraisal, included selective lining, repairs to head regulators, replacement of damaged pipes, sluice gates and provision of measuring flumes. At project closure, outlets in the scheme have been modified to act as proportional outlets and gates have been installed as well to operate on a fully-opened or fuilly-closed position, as required. These works required a total expenditure of Rs42.03 million, compared to an estimate of Rs54.34 million made at scheme preparation, giving a cost underrun of 23% in nominal terms or 33% in real terms. The cost per hectare has, however, gone up by 140% from Rs43,370/ha (scheme preparation) to Rsl 04,020/ha (Credit closure) in real terms. 86. The incremental 0 & M costs consist of actual expenditure claimed from the Bank (100% reimbursement) and 2% of investment costs thereafter. 87. Economic Benefits. Only agricultural production benefits have been quantified for economic analysis. At appraisal, the incremental benefiting area was assumed at 1,253 ha, compared with a much reduced area of 350 ha at scheme completion. The foregone benefits on 350 ha converted  83 INDIA: National Water Management Project Implementation Completion Report Appendix 2: Economic and Financial Analysis to irrigated second crop of paddy are assumed negligible as no crop could be grown without irrigation. Production benefits from annual crops are projected to accrue with a one-year time lag from completion of works each year. The cropping pattems and crop budgets required to estirnate agricultural benefits are presented in Tables 92 and 93. The annual net incremental benefit is given in Table 94. 88. Economic Costs. Historical investment and incremental 0 & M costs have been inflated to 1995 prices by applying WPI. The resulting costs, including labour, have been restated in border terms by using an SCF of 0.9. The economic costs derived in this way are shown in Table 7. 89. Economic Analysis. The economic analysis has been carried out in 1995 constant prices. Financial prices and economic values of crops and inputs are presented in Tables 2 and 94. Future prices are based on World Bank Primary Commodity Forecasts dated February 1995. Based on the above assumptions, the cropping patterns and per hectare crop yield assumptions shown in Tables 88 and 89 and a 25-year period of analysis, the current estimate of ERR for the scheme is -6% (see Table 95). The negative ERR is attributed to a substantially reduced benefiting area, 350 ha compared with 1,253 ha at scheme preparation and excessive real cost overrun per ha. No ERR was calculated at scheme preparation. A sensitivity analysis shows that a 50% increase in area irrigated with jack up the ERR to only -1%, while a 50% reduction in costs will raise the ERR to 2%. 90. Farm Incomes. ICR estimates an incremental farm income of about Rs450/ha based on an average farm size of 0.25 ha (Table 96) as follows: Farm Income (Farm size - 0.25 ha) Without Project With Project (Rs) (Rs) Gross value of production 6,760 7,209 Production cost including water charges 4 755 5,021 Net return 2,005 2,188 Incremental Rsl 13/0.25 ha or Rs450/ha  84 INDIA: National Water Management Project Implementation Completion Report Appendix 2: Economic and Financial Analysis Haryana Scheme 91. Investment and 0 & M Costs. Project investments were mainly utilized for the lining of main canals and watercourses. These works were a continuation of the lining financed under the Haryana Irrigation II Project. The resulting expenditures in financial terms amounted to Rs949.41 million, 22% lower than the revised estimates at scheme preparation. The original estimate of Rs1,743 million was revised to Rs1,220 million to keep expenditure within a budget ceiling. The cost underrun in nominal terms was 22% and in real terms 26%. The cost per hectare has, however, gone up by 14% from Rsl6,920/ha (scheme preparation) to Rs23,655/ha (Credit closure) in real terms. 92. Incremental 0 & M costs were estimated at 2% of total investment in lining of watercourses at completion, which is about Rs5.29 million, assumed to occur from FY96 onwards, as indicated in Table 8. It is assumed that no incremental maintenance cost on canals would be incurred as maintenance costs on lined canals would not differ materially from those on unlined canals when maintenance is adequate and regular. 93. Economic Analysis. The economic analysis has been carried out in 1995 constant prices. Financial prices and economic values of crops and inputs are shown in Tables 2 and 97. Future prices are based on World Bank Commodity Forecasts dated February 1995. Historical investment costs have been inflated to 1995 prices using WPI. The resulting costs, including labour, have been restated in border terms by applying an SCF of 0.9. The economic costs derived in this manner are shown in Table 8. 94. The lining of watercourses generates savings in water which is available only to farmers within the command of the outlet. In the case of canal lining, such savings become available for use anywhere in the system. Both these savings will allow irrigation to cover an increased area. Crop yield increases are likely due to larger depth of application to the crops. There is not enough data to disaggregate these effects. Increased area has therefore been evaluated, albeit conservatively, as the principal impact of the project. It is difficult to isolate the direct impact accurately because of a number of parameters involved - rainfall, river flows, cropping pattern and farmers' choice in allocation of water. Moreover, lack of measurements under the project allows only theoretical savings to be considered. 95. On this basis, there should be an increase of 38,165 ha (11,350 ha saved by watercourses and 26,815 ha saved by canals) in the area supplied. There has been no change in cropping pattern but increased cropping intensity is a major benefit from the lining of canals and watercourses. Overall benefits have been adjusted for reduced output from the adjacent areas using wells which will not be recharged by seepage from the canals, after lining. 96. Based on the above assumptions, the cropping patterns and per hectare crop yield assumptions shown in Tables 98 and 99, and a 25-year period of analysis, the current estimate of ERR is 17% (see Table 100). A sensitivity analysis was done in view of the fact that the actual increase in irrigated area ascribed to the project cannot be accurately known. The results show that a 20%  85 INDIA: National Water Management Project Implementation Completion Report Appendix 2: Economic and Financial Analysis decrease in area irrigated will bring the ERR down to 13%, still an attractive investment, while an increase in costs by 20% will lower the ERR to 14%. 97. Farm Incomes. No estimate was made at scheme preparation. ICR estimates an incremental farm income of about Rs14,800 based on a representative farm model (farm size - 4 ha) (see Table 101): Farm Income (Farm size - 4 ha) Without Project With Project (Rs) (Rs) Gross value of production 92,350 121,350 Production cost including water charges 42.141 56,330 Net return 50,209 65,020 Incremental = Rs14,81 1  86 INDIA: National Water Management Project Implementation Completion Report Appendix 2: Economic and Financial Analysis Table 1. Exchange Rates and Wholesale Price Indices (WPI) Used Financial Year Project Exchange Rate WPI (81/82=100) Inflator 1986/87 12.79 132.7 2.07 1987/88 12.97 143.6 1.91 1988/89 14.48 154.3 1.78 1989/90 16.66 165.7 1.66 1990/91 17.95 182.7 1.50 1991/92 24.52 207.8 1.32 1992/93 26.41 228.7 1.20 1993/94 31.36 247.8 1.11 1994/95 31.65 274.5 1.00 Source: World Bank - India Country Economic Memorandum, May 95.  INDIA - Natonal Water Management Project I - ICR Appendx2 :Economic and Financial Analysis TatUe 2: DertvatlonofParltyPrices PAFI TY (1) 1 DI DI'][' [E] [E] VI] DI 0] [1 UNITS RCE G NUTS WHEAT SORGHUM SOYBEAN COTTON(5) SUGAF COPRA UREA DAP MOP CANE PROJECTED WORLD MARKET PRICE (YEAR 2oo) In i99o constant dollars (2) US$1MT 271.00 568.00 162.00 96.00 233.00 1500.00 242.00 420.00 142.00 162.00 104.00 Adjfactor to 1995 dolars 3) 1.08 1.08 1.08 1.08 1.08 1.08 1.08 1.08 1.08 1.06 1.08 In constant 1995 dollars US$/MT 292.68 635.04 174.96 103.68 251.64 1620.00 261.36 454.00 153.36 174.96 112.32 QUALITY ADJUSTMENT % 80.00 100.00 80.00 90.00 90.00 90.00 97.00 95.00 100.00 100.00 100.00 WOFLD MARKET EQUIVALENT US$/MT 234.14 635.04 139.97 93.31 226.48 1458.00 253.52 431.00 153.36 174.96 112.32 International shippng USt/MT 50.00 100.00 60.00 60.00 70,00 -250.00 -70.00 120.00 50.00 50.00 50.00 CIF INDIAN PORT US$/MT 284.14 735.04 199.97 153.31 296.48 1208.00 183.52 551.00 203.36 224.96 162.32 EXCHANGE RATE (4) PS/USI 31.85 31.85 31.85 31.85 31.85 31.85 31.85 31.85 31.85 31.85 31.85 CIF INDIAN PORT Rs/MT 9049.99 23411.02 6368.98 4882.99 9442.76 38474.80 5845.09 17550.00 6477.02 7164.98 5169.89 Port charges Rs/MT 226.25 585.28 159.22 122.07 236.07 -961.87 -146.13 400.00 161.93 179.12 129.25 0 Internal transport Ps/Mr 150.00 200.00 200.00 200.00 200.00 -40.00 -20.00 200.00 200.00 200.00 200.00 Internalhandling Rs/MF 120.00 120.00 120.00 120.00 120.00 -200.00 -150.00 100.00 120.00 120.00 120.00 PROCESSEDVALUE Rs/MT 9546.24 24316.30 6848.21 5325.06 9998.83 37272.93 5528.96 18250.00 6958.94 7664.10 5619.14 Processingrato % 6200 3500 9000 90.00 90.00 3500 10.00 100.00 100.00 100.00 100.00 Processing costs Rs/MT -150.00 -200.00 -150.00 -150.00 -150.00 -500.00 -150.00 -100.00 0.00 0.00 0.0D Value of cotton seed Rs/MT 2500.00 WHOLESALE VALUE Rs/MT 5768.67 8310.70 6013.38 4642.56 8848.95 15045.53 402.90 18150.00 6958.94 7664.10 5619.14 Local marketingflransport Rs/MT -200.00 -200.00 -200.00 -200.00 -200.00 -300.00 -60.00 250.00 200.00 200.00 200.00 FARMGATE PRICE Rs/MT 5568.67 8110.70 5813.38 4442.56 8648.95 14745.53 342.90 17900.00 7158.94 7864.10 5819.14 ECONOMIC FARMGATE PRICE Rs/KG 5.57 8.11 5.81 4.44 8.65 14.75 0.34 17.90 7.16 7.86 5.82 FINANCIAL FARMGATE PRCE (Andca Pradesh) Rs/KG 3.80 8.50 4.50 3.0D 9.00 2(100 0.55 30.00 ) 3.00 8.40 3.80 (1) Parity: V] Import (E] Export (2) World Barnk pice forecasts dated February 1995 for the year 2000 a Inflated by MLA Index (average Ibr 1994 and 1995 bEkse 1990) (4) Based on World Bark Office Memorandum dated 11 January 1995 re Exchange Pates and Price ContIngencIes for Pro ject Analysis In Indba verage value for 1994 and 1995) (5) Wholesale value refers to seed cotton 6 Price at KarnaiatQ. File: PARITY.WK3  INDIA: Natbnal WaterManagement Pmject I - ICR Appendix 2: EaDnomic and FnancialAnalysis ROS Sub-pojoect Table 3: Economic Costs and Benefits (FRs million, 1995 Dontint pices) YEAR 1 2 3 4 5 6 7 8 9 10 t1 12 PROJECT COSTS Investment Costs (1) 0.020 2.130 3.632 3.950 11.739 12.053 39.142 WitM-pmjectO&M Costs (2) 3.187 3.585 3.984 4.382 4.780 5.179 5.577 5.577 5.577 5.577 5.577 TOTAL COSTS 3.207 5.715 7.615 8.332 16.520 17.231 44.719 5.577 5.577 5.577 5.577 PROJECT BENEFITS Without-project Net Pioducton Benefits 0.036 0.541 1.403 2.340 5.126 7.986 17.274 17.274 17.274 17.274 With-pmject Net Productin Benefits 0.045 0.686 1.780 2.969 6.503 10.131 21.914 21.914 21.914 21.914 Incremental Net Benefits 0.010 0.145 0.377 0.629 1.377 2.145 4.639 4.639 4.639 4.639 Without-piojectOiM (Saved) Costs (3) 3.187 3.187 3.187 3.187 3.187 3.187 3.187 3.187 3.187 3.187 3.187 00 TOTAL BENEFITS 3.187 3.196 3.332 3.564 3.815 4.564 5.332 7.826 7.826 7.826 7.826 NET BENEFITS -0.020 -2.519 -4.283 -4.769 -12.704 -12.668 -39.387 2.249 2.249 2.249 2.249 NPV (12%, 25 years) - 28.843 ERR (25 years) - 6.21 % NOTES: (1) Taken hom GOA P actual expenditure records. Financial costs have been converted ID economic costs multiplying by the SCF of 0.9. (2) Assu med b be Rs175Aha CCA (35.410 ha) hom 1994/95 increased linerally from Rs100ha from commencment of tlhe sch eme works. Economic costs have been estimated by applyng the SCF of 0.9 (3) Taken at RslOOAiaof CCA (35,410 ha), and oDnverted in lo economic terms by rumltiplying by the SCF of 0.9. File: RDSEA.WK3  INDIA: National Water Management Project (Cr.1770-IN) Imptementation Completion Report Appendix 2: Economic and Financiat Analysis Table 4. Financial and Economic Costs - Vani Vilas Sagar Scheme (Rs mittion) FY87 FY88 FY89 FY90 FY91 FY92 FY93 FY94 FY95 FY96 FY97 onwards Financial - Investment costs 0.194 2.061 6.638 9.023 6.284 12.766 4.660 7.277 3.973 1.350 - - In FY95 prices" 0.402 3.937 11.816 14.978 9.426 16.851 5.592 8.077 3.973 1.350 - CO - Incrementat 0 & M - - - - - 0.405 0.123 0.391 1.0853 1.0853o - In FY95 prices" - - - - - 0.486 0.137 0.391 1.0853' 1.085 ' Economic - Investment costs2' 0.362 3.543 10.634 13.480 8.483 15.166 5.033 7.269 3.576 1.215 - Incremental 0 & M2/ - - - - - - 0.437 0.123 0.352 0.976 0.976 Using inflators shown in Table 1. 2 Adjusted by using SCF of 0.9. 3 Assumed at 2X of totat investment costs after compLetion of works.  INDIA: National Water Management Project (Cr.1770-IN) IrWpLementation Completion Report Appendix 2: Economic and FinanciaL Analysis Table 5. Financial and Economic Costs - Sathanur Scheme (Rs million) FY87 FY88 FY89 FY90 FY91 FY92 FY93 FY94 FY95 FY96 FY97 orwards Financial - Investment costs - 0.750 2.163 7.169 12.324 13.436 8.160 1.871 16.053 6.5711 - - In FY95 prices2' - 1.433 3.850 11.901 18.486 17.736 9.792 2.077 16.053 6.571 - 0 - Incremental 0 & M -- - - 0.301 0.285 0.497 1.368 1 .3703 1.37031 - In FY95 prices2' - - - - 0.397 0.342 0.552 1.368 1.3703' 1.3703 Economic - Investment costs4' 1.290 3.465 10.711 16.637 15.962 8.813 1.869 14.448 5.914 - - IncrementaL 0 & M4 - - - - 0.357 0.308 0.497 1.231 1.233 1.233 ' IncLudes Rs2.071 million for terminated works to be carried out. Using inftators shown in Table 1. 3, Assumed at 2% of total investment costs after compLetion of works. V Adjusted by using SCF of 0.9.  91 INDIA: National Water Management Project Implementation Completion Report Appendix 2: Economic and Financial Analysis Table 6. Annual Scheme Investment and O&M Expenditures (Current Rupees Million) SCHEME 1986/87 1987/188 1988/89 1989/90 1990/91 1991/92 1992/93 1993/94 19949 THAMA8ARAPARANI Investment 0.745 6 539 23.132 27.413 46.893 15.872 66.619 O&M 6.925 8.612 9.717 13.082 12.751 19.500 21.000 MALAMPUZHA Investment 1761 6.867 20.368 29.386 21.584 OSM 2.100(1) 6.500(1) THANtAVA Investment 0.030 1 200 4.130 11.250 15.600 19.600 14.S44 0&M (2) (2X3) RATAPANI Investment 0.980 1.920 3.100 9.500 OSM 0.520 0.800 NOTES: (1) 1994 prices. (2) In 1995 prces OCM expenditure was RslO0/ha in 188/89 and Rs175Sha in 1994/95. (3) Anticipated NVvMP investment expenditure in 1995/96 = Rs3 856 million.  INDIA: National Water Management Project (Cr.1770-IN) Implenentation Completion Report Appendix 2: Economic and Financial Anatysis Table 7. Financial and Economic Costs - Peechi Scheme (Rs miLLion) FY87 FY88 FY89 FY90 FY91 FY92 FY93 FY94 FY95 FY96 FY97 orwards Financial - Investment costs - - - - 0.005 16.065 25.957 - In FY95 prices" - - - - - - 0.006 17.832 25.957 - - - IncrementaL 0 L M - - - - - 0.841" 0.8412/ - In FY95 prices - - - - - - - 0.8412 0.8412/ Economic Investment costs3 . - - - - - 0.005 16.049 23.361 - - - IncrementaL 0 & M 3/ - - - 0.757 0.757 "' Using inflators shown in TabLe 1. 2/ Assumed at 2% of total investment costs after compLetion of works. 3 Adjusted by using SCF of 0.9.  INDIA: National Water Management Project (Cr.1770-IN) Impiementation Comptetion Report Appendix 2: Economic and Financial Analysis Table 8. Financial and Economic Costs - Haryana Scheme (Rs million) FY87 FY88 FY89 | FY90 FY91 FY92 FY93 FY94 FY95 FY96 FY97 onwards Financial - Investment costs - - - - - 386.708 510.148 52.553 - - - In FY95 prices' - 464.050 566.269 52.553 - - Incremental 0 & M - - - - - - - - - 5.88 5.88 - In FY95 prices" - - - - - - - - - 5.88 5.88 Economic - Investment costs3 - - - - - 417.645 509.642 47.298 - - - Incremental 0 & M3' - - - 5.29 5.29 Using inflators shown in TabLe 1. Assumed at 2X of total investment costs on lining of watercourses after completion of works. 3i Adjusted by using SCF of 0.9.  INDIA: National Water Management Project I - ICR Appendix z: Economic and F lnanuaA Anatysis RDS Sub-project Table 9: RDS Sub-project Benefitting heas (Hectares) YEAR 1 2 3 4 5 6 7 8 9 10. 25 KharlfT Distributary Tailend (epresented by DIstributary 16&) - Incremenbl area benefittng 1 21 35 38 113 116 377 'I -accumulatedtoblarea benefitting 1 22 57 95 208 324 701 700 700 4> Babi Distribuiry Tailend (represented by Distributary 24) - incremental arem benefitting 1 21 35 38 113 116 377 -accumulatedtoblarea beneftting 1 22 57 95 208 324 701 700 700 NOTES: 1. This sub-project has notbeeneffectively monitored Theareas tht wil beneft from 995/96wereassessed by the missionat 4% of the toal command area (1,416 ha). Theareas beneftting inany particubr year during the project ImplementaBon priod have been calculated In drect proportlon to the Investment costs ti the preceding year. a itisassumedtriatlarmerswll benefitrromtneadditionalwatersupplyimmediately. File: RDSEA.WK3  INDIA: Nabonal Water WMnagement Project I - ICR AppenOx)2: Economicanc Flnancial AnalysIs Thambaraparani Sub-project Table lo: Thambaraparani Sub- p-olect Beneriting Areas YEAR 1 2 3 4 5 6 7 8 9 10 .25 Manimuthar Command - incremental area benefltfng 4 36 116 121 185 58 221 - accumulated total area benefitting 4 40 156 277 461 519 740 740 740 Thambaraparani Draiage AffectedArea U - tncrenentat area benetAtUn 23 192 626 653 999 3t4 At92 - accumulated total area benefitting 23 216 842 1495 2494 2808 4000 4000 4000 Thambaraparani Command - mcremental area benentting 28 231 751 784 1199 377 1431 - accumulated total area benemitbng 28 259 1010 1794 2993 3369 4800 4800 4800 NOTES: 1. This sub-project rias not been effectively monKtaed. The areas thatwill benefitfrom 19956 were assessed bythe mission together vrth the ield staff. The areas benelltting In any particular year ouring the project Implementation period have been calculated relaive to the Investment costs In the preceding year. 2. It Is assumed that farmers wIl benefit from the adoDtional anrcYa more timely water supply. or Improved Orainage condiibons Im medately. File: THAMBAEA.WK3  INDIA: National Water Management Project I - ICR Appendix 2: Economic and Financial Analysis RDS Sub- project Table 11: 1 Hectare Farm Model - Distributary 16a CROP 1 HECTAFE FARM MODEL WITHOUT PROJECT WITH PROJECT AREA Kharif - paddy 1.00 1.00 Rabi (well irrgation) - groundnut 0.20 0.20 xo - sunflower 0.15 0.15 o GROSS INCOME 24388 30503 COST OF PURCHASED INPUTS (1) 9439 11810 NET INCOME 14949 18693 NOTES: (1) Includes cost of family labour File: RDSEA.WK3  INDIA: National Water Management Project I - ICR Appendx 2: Economic and Financial Analysis RDS Sub- project Table 12: 1 Hectare Farm Model - Distributary 24 CROP 1 FE CTAFE FARM MODEL WITHOUT PROJECT WITH PROJECT AREA Kharit - paddy 0.25 0.25 - sesame 0.20 0.20 - sorghum 0.20 0.20 - sunflower 0.20 0.20 Rabi - paddy 0.25 0.25 - groundnut 0.50 0.50 - cotton 0.10 0.10 - sunflower 0.15 0.15 GROSS INCOME 19690 24620 COST OF PURCHA