India - Himalayan Watershed Management Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No. 15637 PERFORMANCE AUDIT REPORT INDIA HIMALAYAN WATERSHED MANAGEMENT PROJECT (LOAN 2295-IN) May 22, 1996 Operations Evaluation Department This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. Currency Equivalents (annual averages) Currency Unit = Indian Rupee (Rs) 1983 US$1.00 Rs 9.50 1988 US$1.00 Rs 14.48 1992 US$1.00 Rs 28.00 Abbreviations and Acronyms ERR Economic Rate of Return FAO/CP Food and Agricultural Organization/World Bank Cooperative Program GOI Government of India GOUP Government of Uttar Pradesh HADD Hill Area Development Department HWMP Himalayan Watershed Management Project M&E Monitoring and Evaluation MTR Mid-Term Review NBC Natural Breeding Centres NGO Non-Government Organization OED Operations Evaluation Department PCR Project Completion Report PR President's Report SAR Staff Appraisal Report T&V Training and Visit Fiscal Year Government: April 1 - March 31 FOR OFFICIAL USE ONLY The World Bank Washington, D.C. 20433 U.S.A. Office of the Director-General Operations Evaluation May 22, 1996 MEMORANDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT SUBJECT: Performance Audit Report on India Himalayan Watershed Management Project (Loan 2295-IN) The Himalayan Watershed Management project (Loan 2295-IN) for US$46.2 million, was approved in FY83. The project was closed in FY92, two years later than planned, and a total of US$17.9 million was cancelled. Attached is the Performance Audit Report (PAR) prepared by the Operations Evaluation Department (OED). The primary objective of the project was to minimize further deterioration of the Himalayan ecosystem in the upper reaches of the Ganges River in the State of Uttar Pradesh (UP). Specifically the project was to arrest land degradation and rehabilitate two badly eroded watersheds covering about 300,000 ha, improve animal husbandry and agricultural practices, meet local needs for fuel and fodder, and develop an approach to watershed rehabilitation which could be replicated on a larger scale. The principal elements of the project (comprising 60 percent of expenditure) were the planting of trees on reserved and communal land, and the provision of seedlings for planting on private land, and the development of conservation measures on these lands, with the aim of improving ground cover and reducing runoff. The program proposed was equivalent to afforesting 50 percent of the project area in the seven year implementation period. Other components included improvement of crop and livestock extension services, development of small scale irrigation, and a program to exchange local cattle for buffaloes from improved milking strains. Field implementation was carried out by individual line departments, coordinated by the Hill Area Development Department (HADD) of the Government of UP. Initial progress was slow and there were major coordination problems. It proved difficult to persuade the line departments to give enough priority to the project elements, for which they were responsible and to second experienced staff to them. A Mid Term Review was held in 1988. As a result two Project Directors (reporting to HADD), one for each major watershed, were appointed and made responsible for implementation. Line department staff were seconded to the project and the physical objectives of the project were changed, including scaling back the forestry programs. Following the MTR, overall progress improved and the revised targets were largely met by the time the project closed. There was great pressure in the initial stages of the project to achieve the forestry targets. This pressure reduced the extent to which participatory methods of implementation were used. Despite the improved coordination after the MTR, this initial handicap was never really overcome and has resulted in poor post-project maintenance of assets developed during the project. Subsequent projects, in UP and other states, have built upon this experience and adopted much more participatory approaches to resource management with greater success. This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. 2 The revised forestry program resulted in the development of 40 percent of the original target, sufficient to meet the fuelwood needs of the area. However, there is little evidence of the impact of the operations on stream flow and, if the original planting program had been adhered to, it is not clear that the additional fuelwood would have been needed. The widespread adoption of new varieties of wheat and rice, and increased milk supply, have improved food supplies in the area. However, project implementation was, and subsequent development has been hampered, by the relative remoteness of the area and the migration of its young men into government or military service elsewhere in India. These two factors have combined to diminish the development of local entrepreneurial capacity and market oriented economic activity. The per hectare cost of the approach to watershed development used by the project is high and its cost effectiveness doubtful. The audit estimates the ERR at II percent, taking into account he overhead costs of the project. The PCR rated this project as satisfactory. However, the audit downgrades this to marginally unsatisfactory, given the high cost and limited replicability of the approach. Institutional development is rated as modest and sustainability as uncertain, confirming the ratings in the PCR. Lessons which can be learned from this project include (i) where a project is to assist communities to rehabilitate resources for which they will subsequently be responsible, it is essential that they be involved from the outset in the definition, design and implementation of these activities; and (ii) participatory approaches will not be effective unless the institutional structure is appropriately open and receptive. Robert Picciotto by Ulrich Thumm Attachment FOR OFFICIAL USE ONLY Contents Preface..... .................................................... 3 Basic Data Sheet................................................... 4 Evaluation Summary ............................................... 7 1. Introduction ................................................... 13 Background ......................................... ........ 13 2. Project Design .......................................... ........ 15 Preparation and Objectives ....................5..... ..............15 Project Description ..................................... ........ 17 Project Organization ............................................... 18 3. Project Implementation............................................20 Mid-Term Review ..............................................21 Project Results ....................................................21 Forestry ................................................21 Conservation Structures.......................................23 Agricultural Production.......................................24 Sustainability ...............................................26 Institutional Development-An Evolution in Approach.....................28 Economic Rate of Return................................. ..............29 Bank Performance.............................................31 Borrower Performance..........................................31 Compliance with Covenants....................................... 31 4. Issues and Lessons.............................................. 32 Ratings.................................................... 32 Quality At Entry.............................................. 32 Social Setting................................................ 33 Conservation Structures......................................... 35 Replicability................................................. 36 Problems of Monitoring......................................... 36 Handover of Assets............................................ 37 Lessons.................................................... 37 Annex A. Borrower Comments on the Project Completion Report ................. ...... 39 Maps IBRD 16811 RI IBRD 16813 This report was prepared by John English (Task Manager) who audited the project in September, 1995. Constance L. Frye and Megan Kimball provided administrative support. i document has a restricted distribution and may be used by recipients only in the performance of their offiilduties. Its contents may not otherwise be disclosed wii.hout World Bank authorization.  3 Preface This is the Performance Audit Report (PAR) of the Himalayan Watershed Management Project for which Loan 2295-IN for US$46.2 million was approved on May 31, 1983. The Closing Date was September 30, 1992, two years later than planned. A total of US$16.0 million was canceled in November 1989 and December 1991. Final disbursement from the loan was made on February 18, 1993, and a balance of US$1.9 million was canceled. The PAR is based on the Project Completion Report (PCR) prepared by the FAO/World Bank Cooperative Program for the Agricultural Operations Division, India Department, South Asia Regional Office, on the Staff Appraisal Report (SAR), President's Report (PR), and the legal documents, on study of the project files, supervision reports and project documents, discussions with involved Bank staff, and on the discussions and findings of an OED mission which visited India in September, 1995, and met with officials of the Governments of India and the State of Uttar Pradesh and its implementing agency, and with farmers affected by the project. The collaboration and assistance from these officials and farmers is gratefully acknowledged. Following standard OED procedures, copies of the draft report were sent for comments to the Government on March 29, 1996. However, no comments have been received. 4 Basic Data Sheet HIMALAYAN WATERSHED MANAGEMENT PROJECT (LOAN 2295-IN) Key Project Data (amounts in USS million) Appraisal Actual Actual as % of estimate appraisal estimate Total project costs 69.1 44.2 64 Loan amount 46.2 - Cancellation - 17.9 Date physical components completed 09/90 09/92 Economic rate of return (%) 24 11 Cumulative Estimated and Actual Disbursements Fiscal Year Appraisal Actual Actual as Estimate (USSM) (USSM) % of appraisal 84 0.22 0.12 55 85 1.22 0.85 70 86 3.42 2.40 70 87 7.52 5.76 77 88 14.82 8.00 54 89 27.82 12.81 46 90 40.07 17.50 44 91 46.20 21.89 47 92 - 26.72 58 93 28.30 61 Date of final disbursement: February 18, 1993. Project Dates Original Actual Identification - 11/79 Preparation - 11/80 Pre-appraisal - 11/81 Negotiations - 04/27/83 Board approval - 05/31/83 Signing - 06/08/83 Effectiveness 08/19/83 09/09/83 Closing date 09/30/90 02/18/93 5 Staff Inputs (staff weeks) Fiscal Identification/ Appraisal Negotiations Supervision Total Year Preparation Pre-81 69.1 - - - 69.1 81 15.8 - - - 15.8 82 73 1.4 - - 74.4 83 - 38.1 2.5 - 40.6 84 - - - 28.4 28.4 85 - - - 60 60 86 - - - 23 23 87 - - - 30.7 30.7 88 - - - 12.7 12.7 89 - - - II 11 90 - - - 10.3 10.3 91 - - - 5.5 5.5 92 - - - 7.4 7.4 93 - - - 4.2 4.2 Total 157.9 39.5 2.5 193.2 393.1 Mission Data Date No. of Staff days Specializations Performance Rating Types of (month/year) persons infield represented a rating b trend problems c Pre-appraisal 12/81 6 26 FE,WM,LU,A - - Preparation 03/82 4 15 FE,A,LO,S - - Commissioning 10/83 5 13 FE,A,A,LO,A - - Supervision 1 05/84 3 5 A,A,E 2 - M,T Supervision 2 11/84 6 10 S,A,A,A,F,H 2 - M,T Supervision 3 06/85 4 6 WM,H,A,F 2/3 - M Seasonal Op. 11/85 4 11 WM,F,H,A - - Supervision 4 01/86 3 11 WM,H,A 3 - M,T Supervision 5 06/86 3 12 WM,H,A 3 - M,T Extension Org. 08/86 3 5 A - - Hort. Dev. 09/86 I 4 H - - Review 10/86 1 5 - - - 6 Date No. of Staff days Specializations Performance Rating Types of (month/year) persons infield represented a rating b trend problems c Research 11/86 2 4 - - - Supervision 6 11/86 3 5 WM,F,SS 3 - M.T Supervision 7 11/87 3 5 WM,H,SS 3 - M Review 05/88 3 14 F,SS,EX - - Supervision 8 08/88 3 5 H,SS 3 - M,T Supervision 9 11/88 3 6 WM,EX 2,3 - M,I Supervision 10 04/89 3 6 WM,H,I 2,3 - M,I Supervision 11 12/89 3 5 A,H,I 2,2 - M,I Supervision 12 06/90 3 5 A,A,H 3,2 - M,I Supervision 13 12/90 3 5 A,A,W 3,2/3 - M,I Supervision 14 07/91 4 5 A,A,F,H 2,2 - M,I Supervision 15 01/92 4 5 A,A,F,W 2,2 - M,I Supervision 16 07/92 3 5 A,A,F 3,2 - M,I a. FE: Forest Economist; WM: Watershed Management; LU: Land Use; A: Agriculturalist; LO: Loan Officer; S: Senior Staff; E: Economist; F: Forester; H: Horticuluralist; SS: Soil Scientist; EX: Extension Specialist; 1: Institutions Specialist; W: Women in Development. b. I = Problem-free; 2 = Moderate problems; 3 = Major problems. c. M = Managerial; T = Technical; I = Impact. Other Project Data Borrower: Government of India; Executing Agency: Government of Uttar Pradesh FOLLOW-ON OPERA TIONS Operation Credit! Amount Board date Loan no. (US$ million) Integrated Watershed Development (Plains) 2131/3197 62.0 05/90 Integrated Watershed Development (Hills) 2100/3175 88.0 03/90 7 Evaluation Summary Introduction 1. The immediate catalyst for this project was the increased level of flooding experienced in the Ganges valley during the late 1970s. After the worst ever floods in northern India in 1978 the Government of India appointed a Working group to formulate an action plan for flood mitigation and control. Increased use of the Himalayan foothills for grazing and cultivation, leading to reduction of soil cover, and increased run-off and erosion, was seen as a major contributing factor to the flooding problem. 2. The primary objective of the project was aimed at minimizing further deterioration of the Himalayan ecosystem in selected areas covering just over 300,000 ha. More specifically the project was to: arrest the land degradation occurring in the project area and rehabilitate badly eroded watersheds; improve animal husbandry and agricultural practices; meet local needs for fuel and fodder; and develop an approach to watershed rehabilitation which could be replicated on a larger scale. 3. The principal elements of the project, approved in 1983, were to support the planting of trees on reserved and communal land and the provision of seedlings for planting on private land; development of conservation measures on these lands, with structures in gullies to reduce stream flow and erosion; improvement of livestock and crop extension services to disseminate improved technologies; a program of exchange of local cattle for buffaloes from improved milking strains; development of small irrigation; and project administration, including the establishment of a National Watershed Development Policy Committee. Implementation Experience 4. Overall responsibility for coordination and execution of the project rested with the Union Ministry of Agriculture. The Secretary, Hill Area Development Department (HADD), Government of Uttar Pradesh (GOUP) was responsible for project implementation at the state level. However, field level implementation was to be carried out by individual line departments, coordinated by HADD. 5. Initial progress was slow and there were major problems in coordination of the overall effort. As had been feared by preparation and appraisal missions, the key problem was that, for almost all the line departments in the UP Government, the hill districts represented a minor part of their overall program, and were seen as low priority. 6. The Project Agreement called for a Mid-Term Review (MTR) to be held after the third year of the seven year implementation period. A week long joint GOI-GOUP-Bank meeting of all the relevant officials was held in early 1988. The meeting agreed major changes to project structure, abandoning implementation through line departments. Two Project Directors (reporting to HADD/GOUP), one for each of the major watersheds, were appointed and made responsible for implementation. Line department staff were seconded to the project. 8 7. Significant changes were also made in the physical objectives of the project, based on the initial experience. The forestry targets were scaled back, while those for conservation measures were expanded. The cattle exchange program was terminated and a breeding program substituted, and greater emphasis was placed on general agricultural improvement. Project Results 8. Following the MTR, overall progress was improved and the revised targets were largely met by the time the project closed in 1992, after two extensions. However, at the outset, the assumption on which the project was based, namely that action was needed to reduce land degradation and flooding, meant that there was great pressure to attain forestry and related targets. This, together with the initial focus on line department implementation, reduced the extent to which there was an effectively participatory basis of operation. Despite the improved coordination achieved after the MTR, the initial handicap was never really overcome and has resulted in post-project problems related to continued maintenance of assets developed during the project. 9. Project Benefits. The widespread adoption of the new varieties of wheat and rice (both of which are largely used for subsistence) and increased milk supply, have improved food supplies. Production for the market remains limited, and so the project is not likely to have significantly improved cash incomes. Reflecting the project's origins, approximately 60 percent of project expenditure was for the forestry and related components. The PCR estimated an ERR for this component of 8.5 percent compared to the appraisal estimate of 24 percent. The lower PCR figure reflects sharply lower yield assumptions. Data from measured plots in similar conditions indicate that the amounts of fuel and fodder which might reasonably be extracted from the plantings is significantly higher. Based on these figures, and taking into account the amounts likely to be utilized per household, the audit has re-estimated the ERR as 14 percent (or 11 percent if the overhead is applied pro-rata). 10. However, these results should be treated with caution. The amounts of fuelwood available are estimated to be close to the appraisal estimates of requirements on a per household basis. Thus, if the initial, much larger, planting goals had been achieved, there is doubt whether the additional fuelwood would be utilized, because of the limited local demand. In that case the re-estimated ERR would be significantly lower. Thus, the project may have, rather fortuitously, undertaken planting at an economically viable level. However, this level of planting left it far short of its resource management objective of increasing ground cover. Sustainability I. The PCR rated the sustainability of the project as uncertain. Concerns primarily relate to the maintenance of the assets created under the project, project performance in relation to the goals for reduction of land degradation, and continued support from public agencies. These concerns are also overlain by the difficulties resulting from the non-participatory nature of the project as initially designed and implemented and also, perhaps deeper problems, related to the social structure and characteristics of the region. 12. Project Assets. Because of the limited participatory nature of the project/beneficiary relationship (para. 8), the work undertaken did not always reflect the priorities of the community. 9 In addition, in part to offset this problem, the activities were undertaken on terms favorable to the communities, i.e. there was no charge for the services of the bulls, or for the irrigation facilities, and villagers were paid to carry out plantings on community lands. This further blurred the question of responsibility for the assets. In addition, termination of project funding was rather abrupt. As a result, the future responsibility of the local communities for the maintenance of assets created under the project (especially NBCs) was not adequately clarified and agreed before the withdrawal of support, and it is clear that many of these assets are not now being adequately maintained. 13. In order to address this issue, the audit recommends that a follow-up survey be carried out to assess whether community or communal management of these assets is satisfactory; to emphasize that there is a responsibility on the part of the community to adequately maintain them; and to determine, where there is significant incidence ofpoor management, its causes and whether there are steps which could be taken to address it. It would seem useful for GOI to set a precedent in this area. Conclusions 14. Overall Assessment. The PCR rated this project as satisfactory. However, the audit downgrades this rating to marginally unsatisfactory. On the positive side, the audit reaches a more favorable assessment of the outcome of conservation activities, and the re-estimated ERR (based solely on the return to forestry activities) is satisfactory. However, this ERR is not robust and the analysis suggests that if the original forestry target had been achieved (designed to meet resource management goals) the ERR would be significantly lower. The project did not achieve other major objectives. The replicability of its approach to watershed rehabilitation is doubtful. The sustainability of project achievements is uncertain and institutional development modest. Bank and Borrower performance are both rated as satisfactory. The Bank's role in pressing for a mid-term review at the outset and carrying it out was crucial in achieving the required change in approach to implementation. While the Borrower initially resisted change, it has moved substantially since the MTR to incorporate the initial lessons into this and other operations. Overall, although this project did not meet its expectations, it played an important role in the evolution of the approach to watershed management in India. 15. Quality at Entry. Project design was heavily influenced by the perception of the time that mismanagement of land in the hills was a principal cause of flooding in the Ganges Valley and that a massive effort was needed to overcome this situation. Thus, the project aimed to establish, or replant, perennials (mainly trees), on 167,000 ha of the 300,000 ha in the project area (i.e. more than half). This was unrealistic, especially if there was to be significant community involvement in local operational design and implementation. Within this total it was planned that the project would supply seedlings for planting by villagers equivalent to 87,000 ha (farm forestry). This amounted to over I ha per household and was unrealistic with many men working outside the region and limited local fuelwood markets. The pressure of these targets, and weak coordination, led to the top-down approach to implementation, the effects of which could not be overcome, even after organizational changes were made at the Mid-Term Review. 16. Institutional Development. The project played a significant role in the evolution of thinking in India on the approach to resource management oriented programs. Early Bank preparation reports proposed the creation of a separate directorate to manage the project with seconded staff, but this was not accepted. The initial experience of the project made clear the 10 deficiencies of the approach through line departments and the inability to achieve effective coordination. The MTR was a turning point in this regard. Although the impact on the HWMP was limited (para. 8) this approach was adopted in subsequent projects, leading currently to a directly participatory approach, using multi-disciplinary teams and participatory appraisal techniques to determine the needs and follow-up for individual communities. Thus, although the initial design was flawed, steps have been taken during and after the project period to incorporate the lessons from HWMP into subsequent operations across the whole of the government National Watershed Program. However, within the project institutional development is rated as modest. 17. Social Setting. The project area has a number of characteristics which have impacted upon the outcome of the project and are influencing subsequent developments in the area. Outmigration of men seeking employment has been a feature for several generations. The striking feature of the project area is the extent to which the migrants move towards the public sector, particularly the military, but also teaching and general public service. The result is to inculcate bureaucratic rather than entrepreneurial skills. There is only limited evidence of remittances being used for investment purposes, and few people see the area as providing income earning opportunities. Thus, if the overall Indian economy progresses in the next few years, outmigration of families may become more common, which would help reduce the resource pressure. 18. This lack of entrepreneurial/managerial skills among men also suggests that the communities may have to look more to women as a source of these skills. However, the project made little progress in this area and the extension staff were overwhelmingly male. 19. Conservation Structures. There is uncertainty over the effectiveness of these structures. The PCR judged that these are not effective, while Government and the Project Agency disagree. Definitive data is not available, but the audit concluded that, in general, the structures appear to be achieving their objectives. However, given that the same methods are being used in current projects in the hill region, a more definitive answer to this question is desirable. The audit recommends that a follow-up survey be carried out on a systematic basis to assess the situation, and to recommend changes, either in the initial measures or subsequent management, if these appear appropriate. 20. Replicability. One of the objectives of the project was to develop a replicable approach to watershed rehabilitation. The results of the project indicate that this was not achieved. The line agency based structure initially adopted led to a poorly coordinated and top-down approach. This was modified at the MTR, but the result was not fully satisfactory, as the further changes which have been introduced into subsequent projects testify. The results of the audit also cast doubt on the replicability of the approach to watershed rehabilitation based on large scale tree planting, mainly as fuelwood plantations. This approach is costly and of doubtful economic viability (para. 10). More cost effective means of improving ground cover, should be tested, e.g. greater use of natural regeneration. 21. Problems of Monitoring. The disagreement over the effectiveness of the conservation measures is typical of a broader problem of how to monitor progress in such resource management areas. The issue is whether the conservation measures have fostered increased vegetative cover and reduced run-off and erosion. There is considerable disagreement over appropriate indicators for monitoring progress in this area, particularly at the micro-level. In II order to met this need, there would seem to be a role, as a part of local monitoring, for the use of photographs to record progress. Such visual 'data' are also likely to be more meaningful to local populations than numbers, largely related to characteristics which are not of great priority to them. Lessons 22. Where a project is to assist communities develop assets or rehabilitate resources, for which they are subsequently to be responsible, it is essential that they be involved from the very beginning in the design and implementation of these activities. 23. In areas such as that covered by the project, characterizing natural resource management problems as emergencies, requiring crash programs to cope with them, is likely to be counterproductive. 24. Participatory approaches will not be effective unless the institutional structure is appropriate, and appropriate skills applied. 25. A Mid-Term review can be of great value in requiring all parties to meet and review in depth problems which have arisen, and reach decisions on modifications in the project's objectives and its structure to respond to them. 26. There is a need for a broader range of indicators to monitor and assess progress in achieving resource management objectives. Particularly where operations involve the support of small scale, village level activities, the indicators have to be meaningful to local participants.  13 1. Introduction Background 1.1 The Himalayan and Shiwalik regions of the two states of Uttar Pradesh (UP) and Himachal Pradesh (HP) extend for about 600 kms along the south/southwest slopes of the Himalayan Mountain chain, with elevations of between 300 and 3,000 m above sea level. Two- thirds of the area is classified as forest, although actual forest area is much less. The area has a very fragile ecosystem, highly erodible soils, and very steep slopes in the upper catchments which, combined with intense rain, result in a high natural rate of erosion. High densities of people and livestock result in firewood and fodder extraction which can exceed the annual biomass production, thereby causing accelerated denudation. Because of increasing population pressure, crop cultivation extended to unsuitably steep unprotected slopes, thus contributing to erosion. 1.2 The immediate catalyst for the project was the increased level of flooding experienced in the Indo-Gangetic Plain during the 1970s. After the worst ever floods hit Northern India in August 1978, the Government of India (GOI) appointed a Working Group to formulate an action plan for flood control in the basin. It was envisaged that the program would integrate downstream engineering works with upstream conservation, reforestation and other watershed treatment measures on the flood prone rivers and their tributaries. As part of the proposed program, Bank assistance was sought for actions in the two states of Himachal Pradesh (HP) and Uttar Pradesh (UP). 1.3 It was quickly seen that there would be advantages to uncoupling the upstream and downstream elements of the proposed program, as the design and planning frame for major engineering works would be extensive, and there was no overriding reason for moving ahead as a single operation. It was also agreed that the initial priority should be on 'critically eroded watersheds', and that this would require detailed local level work. This initial approach was influenced by the preparation experience of the Kandi Watershed Protection Project (Ln. 1897) in Punjab, which had been appraised in 1978. This was a pilot effort to tackle the problem of degradation in the Punjab section of the Shiwalik range, and which proposed to undertake a series of linked actions in each of five watersheds: (a) rehabilitation in the upper catchment areas through reforestation, pasture development and soil conservation works; (b) construction of storage or retention dams in the middle catchment to moderate floods; (c) flood protection, reclamation and irrigation of farmland in the lower parts of the watershed. 1. The original concept for a two state project covering selected watersheds in HP and UP was carried through to appraisal. However, shortly before negotiations, Himachal Pradesh withdrew because of disagreement with GOI over onlending terms for the credit. In this report numbers relate to UP only, unless otherwise indicated. 14 1.4 The project became effective in 1980 and had only just got into it's stride when the Himalayan project was appraised and negotiated in 1982-83. And, the early results from the Kandi project may have been misleading rather than helpful to the design of the Himalayan project. The PCR and the audit for the Kandi project note that although the efforts of the individual departments, livestock, forestry, horticulture etc. were generally technically sound and they met their targets, the whole did not, in the end, equal the sum of the parts. There was a tendency to over-engineer some works, e.g. in soil conservation the emphasis was placed on bench terracing rather than less expensive, more flexible and less capital intensive vegetative methods. More importantly, coordination was weak and the impact on the land use systems as a whole fell short of expectations. For example, a cattle program made improved dairy stock available, and rehabilitation efforts in upper watersheds generated increased forage production. However, the latter was disposed of by auction and the livestock farmers in the lower parts of the catchments were outbid. They had continued difficulties in increasing quality and quantity of fodder, and many of the improved stock were sold as poor nutrition reduced milk yields. Initially the Kandi project may have appeared to be making good progress, but by the end of the implementation period it was clear that the ultimate impact of the efforts was modest at best. 15 2. Project Design Preparation and Objectives 2.1 The project was to focus on watersheds in the outer Himalayas and Shiwaliks. This area forms a belt adjacent to the plains which varies in width from 30 to 80 km, and has altitudes up to 2,200 meters. These were the areas where the most serious denudation and soil erosion was taking place. It was decided that the less seriously affected middle and inner Himalayas would not be included in a first project. Also, flood control based on extensive construction would not be included as such programs took much longer to plan, design and engineer than upper catchment works, and were the responsibility of the states' irrigation departments. 2.2 Objectives were outlined in the project brief as: (a) "to arrest, and in some places to reverse, the manmade ecological destruction, which has been going on in the outer Himalayas and Shiwaliks for over a century, by introducing systems aimed at ensuring that the resource base is able to sustain the predicted growth in the rural population; (b) to rehabilitate badly eroded watersheds and thereby mitigate the damage done to downstream engineering works and farmlands by flood waters containing high levels of sediment; (c) to meet the rural population's requirements for fuel, fodder and small timber; (d) to improve animal husbandry, agricultural and horticultural practices and productivity; (e) to provide enhanced off-farm employment opportunities through project works and selected cottage industries; and (f) to develop an approach to watershed rehabilitation that can be replicated on a large scale and form the basis for sound investment in the future."2 2.3 It was recognized that the achievement of these objectives would require significant changes in the land use and agricultural practices of the rural population. If that was to be achieved it would be essential that the activities supported should find easy acceptance among the population. Some would be less acceptable and would have to be combined with more 'popular' measures which would counteract possible unfavorable reaction to the project as a whole. Care would also have to be taken to ensure local participation in planning whatever was done, through the panchayats and village meetings. 2.4 The major disagreements between the Bank and the Borrower during preparation and appraisal centered around these issues. An early preparation report discusses the need for and implications of a participatory approach at some length, " It is also an axiom (to which much lip- 2. Himalayan Integrated Watershed Development Project: Project Brief. June 17, 1980, para. 18. 16 service is paid) that no plan or project could succeed without the cooperation of the proposed beneficiaries-residents within the micro-watershed. This cooperation cannot mean the creation of plans that would provide benefits in the abstract for the villagers and the involvement of the villagers only in their implementation. Every watershed plan must be made with the villagers of the concerned watershed."3 This same report discusses the operational and organizational implications of a 'participatory' approach at some length (see paras 2.03 - 2.14), including the desirability of establishing a separate directorate in the state Hill Development Department to manage the effort, using staff seconded from other line agencies However, the national and state governments were of the opinion that the necessary expertise could be effectively provided through the existing line department structure, and the Bank eventually accepted that approach. 2.5 "However, if these objectives are to be achieved, if watershed rehabilitation measures are to be successful, project design will have to ensure that the great majority of persons living in the micro-watershed perceive themselves as beneficiaries. Furthermore, some of the benefits will have to be immediate (i.e. village water supplies) if villager support is to be assured. In other words once micro-watersheds with critical erosion problems have been selected for rehabilitation under the project, the next objective (after dealing with the conservation aspects) would be to improve productivity per unit area, income and employment. Only in this way will it be possible to secure local participation."4 2.6 In this same vein, it was indicated that "an essential prerequisite for project success is participation by residents in the planning and implementation process. This could be achieved, under existing panchayat legislation, through the formation of "joint committees" [quotation marks in original] whose membership would include a selected number of the concerned panchayats, local leaders, and representatives of the implementing departments. The functions of the joint committee, set up for each micro-watershed, would include "coordination, communication, and assistance with the planning and implementation of project works." 2.7 It is clear that the notion of participation implicit in this approach was 'facilitation of the achievement of government objectives' rather than development of dialogue with the aim of maximizing joint goals. Implicitly, it seems to have been assumed that the sustainability of agriculture and the maintenance of the land resource base were not local priorities. This approach had a distinct top-down bias and previewed some of the major implementation difficulties. 2.8 The key element in the thinking behind the project was the need to establish sustainable agricultural systems in order to avoid resource degradation in the form of erosion and its down stream costs. "It is absolutely clear that an integrated approach must be taken involving reforestation, soil conservation, pasture/animal husbandry development, agricultural (rainfed and irrigated) development and the expansion of off-farm activities, otherwise the objectives would not be achieved."5 3. See "Integrated Watershed Development in the Himalayas: Socio-Economic perspective - UP," Annex to Identification Report, para. 2.02, January 1980. 4. Project Brief, June, 1980, para. 19. 5. Project Brief. June, 1980, para. 21. 17 2.9 This thinking led to an emphasis in project preparation and implementation on detailed planning for individual watersheds, based on their constituent micro-watersheds. Interdisciplinary teams were established in each state to oversee the preparation of plans for a sample of micro-watersheds. The initial criteria for selection of watersheds were: * "overall condition of the watershed (degree of past erosion/denudation and susceptibility to land slides etc.); * sediment yield at micro-watersheds final outlet; * micro-watersheds natural recuperative powers and prospects of reducing erosion/sediment yield using man made measures; * existing or planned engineering structures downstream; * road access, proximity to administrative centers and demonstrative effect; * people's willingness to cooperate (as evidenced by previous self help achievements); * strength of local organizations; and * good potential for minor irrigation, water harvesting and agriculture."6 Project Description 2.10 The primary objective of the project was aimed at minimizing, in selected areas covering just over 300,000 ha, further deterioration of the Himalayan ecosystem, "caused by depletion of forest cover, overgrazing, bad land use and careless road construction, all of which had resulted in erosion in these areas." In addition efforts were made to improve agricultural productivity and enhance rural incomes. 2.11 The project consisted of the following components: (a) establishment of 59,000 ha of mixed species plantations (for fuel and timber), 27,800 ha of fodder tree plantations on government and panchayat owned land, and fuel and fodder trees on some 81,000 ha of privately owned 'wastelands', (poor and degraded land not regularly cultivated) and on field boundaries; (b) construction (by the Forest Department) of soil conservation structures designed to protect existing infrastructure such as roads and villages, including some 1,650 brushwood/stone checkdams, 1,300 cratewire dams and simple drop structures. Construction (by the Agriculture Department) of terraces on 6,000 ha of arable farming land; (c) livestock development, including a cattle exchange program in which 10,500 buffaloes would be exchanged for local cattle, and provision of better livestock services through improvements to 40 existing livestock dispensaries; (d) improvement of agricultural extension services through provision of staff, equipment, housing and staff training; (e) horticultural development, including multiple grafting of wild-growing trees with improved material in 3,900 ha of existing, but wild, orchards, and establishment of 1,500 ha of new fruit orchards; 6. Project Brief, June, 1980, para.. 20. 18 (f) irrigation development (by Soil Conservation Department), including improvement, construction and lining of 250 km of small channels and construction of 650 small water tanks; (g) development of research and training activities connected with watershed development; (h) project administration at State, Watershed and Sub-Watershed level; and (i) establishment of a National Watershed Development Policy Committee, assisted by a Watershed Development Council for appraisal of sub-projects, and monitoring and evaluation (M&E) of project progress. 2.12 The total project cost, over a period of seven years, was estimated at appraisal to be about US$69.12 million (Rs 656 million), of which only US$0.80 million (Rs 7.3 million) would be foreign expenditures. Approximately 70 percent of the project cost net of taxes and duties (US$66 million) was to be financed by a Bank loan of US$46.2 million. Project Organization 2.13 Overall responsibility for coordination and execution of the project rested with the Union Ministry of Agriculture. The Secretary, Hill Area Development Department (HADD), Government of Uttar Pradesh (GOUP) was responsible for project implementation in the state. 2.14 Comments in the review process focused on the economic analysis and the issue of participation. There were concerns as to whether the cost-benefit analysis double counted benefits, particularly production of fodder and milk. Concerns were also raised that the appraisal report did not adequately address how adequate participation was to be . This was a valid concern, as later experience showed. 2.15 The major issues raised at the Decision Meeting, in addition to participation, were project management and the duration of financing. Management concerns revolved around the problem of coordinating multi-disciplinary and multi-agency activities. Questions were also raised over the numbers of committees envisaged, and the cumbersome organizational structure. It was noted, however, that a single department was to have prime responsibility for implementation, and that it was anticipated that this would minimize the problem. However, the way in which the issue was posed is itself illustrative of the problem. More clarity on committee structures etc. would not have eliminated future difficulties, which really resulted from the basic approach being adopted. One reviewer was closer to the mark when he asked for more information on what would 'happen to those farmers who are asked to give up their land to enclosure.' 2.16 It was recognized that the type of program being considered would require an extended implementation period. The initial proposal had been for the Bank to finance the first five years of the expected nine year implementation period. However, supporting the program for only five years would not be sufficient to assess the development impact of some project elements, and could leave the state government with a significant financing burden after the fifth year which, if follow-on support did not materialize, would give rise to a number of uncompleted works. It was 19 finally proposed that Bank support be for seven years, with a provision for a mid-term review of the progress of the project after three years. 2.17 No ma or issues arose during negotiations and the project was approved by the Board on May 31, 1983. 7. Except that just before negotiations Himachal Pradesh withdrew from the project and the operation was scaled back. 20 3. Project Implementation 3.1 The project became effective in September, 1983, but initial progress was slow and the predicted difficulties of coordination were soon validated. The initial rate of uptake of the loan was expected to be slow. The estimate for disbursement by June, 1986 (almost 3 years into implementation) was only US$3.4 million (8 percent of the total). However, actual disbursement by that time was only $1.8 million. 3.2 An indication of the initial problems of the project is given by the following passages from the supervision report of July, 1985:8 "Detailed subwatershed 'plans' have been submitted and the supervision mission visited the areas where these were being implemented only to find minimal impact in the field. Components had been covered on a target basis only, (e.g. forest nurseries had been planted with the required number of pots but the seedlings were dead or had not germinated due to lack of water and the clear plastic pots had perished; soil conservation components had been poorly constructed and in inappropriate areas; other components had been constructed with absolutely no knowledge of what impact they would have on the people or the watershed, or how they interrelated with each other). In short, there has been much 'planning' without any understanding of what needs to be done. As there is no lack of enthusiasm, and both the staff and the GOUP are eager to know what they can do, it is felt that the project can be 'turned around' towards a more successful conclusion, with the planned field implementation assistance this planting season." 3.3 "The problem over the past 18 months has been a breakdown in communications between GOUP, Project staff, the GBPUAT (GB Pant University of Agricultural Technology) and the Line Departments responsible for implementation, but the mission felt that this had been brought about by the total lack of knowledge of all parties, on just what was exactly to be implemented in a watershed project, apart from the normal T&V components. What were the new technologies for stabilizing the watershed and increasing agricultural production? From the supervision review, it was obvious that there was no coherent plan, everybody was working in isolation of everybody else and the project became 'target oriented' at the expense of quality or rationality."9 3.4 Over the next two years little improvement occurred. "The project has now completed four years; it has so far only achieved 20 percent of its program which has had little impact on the project's main objectives. ... As stated by previous missions, the project is being severely constrained by poor management, which is too dispersed and decentralized to handle the day-to- day problems associated with running a development project scattered over 312,000 ha in parts of three districts. There is no focus on the important component needs; the project's main objectives are not being addressed; and there is little cooperation between line departments and project management which has led to increasingly low morale among staff and field managers and confusion as to what needs to be implemented." 8. Himalayan Watershed Management Project. July 10, 1985, Supervision Summary, Section 6: "Summary of Project Status". 9. Op. cit. Annex 2, para. 6. 21 "It is now clear that the project as it is now designed cannot be replicated over the large areas involved, neither financially nor in terms of administrative requirements. Policy changes are also needed to initiate attitudinal changes by the line departments towards more appropriate implementation of their programs." 0 3.5 As had been feared by preparation and appraisal missions, the key problem was that, for almost all of the line departments in the UP Government, the hill districts represented a minor part of their overall program. There was overwhelming political pressure for improved agricultural performance in the main populated areas of the Ganges plains. Thus, the coordination task facing the management of this project was almost impossible. This was the major issue for the Mid-Term Review. Mid-Term Review 3.6 Some difficulty was initially faced in getting GOI-GOUP to prepare a document which could be the focus of the review. However, once the initial reluctance was overcome a useful document was prepared, which formed the basis for a week long joint GO]-GOUP meeting of all the relevant officials. The review was held in early 1988. This meeting enabled many of those involved in the project to discuss, for the first time, the basic objectives and operations of the project at length and allowed the air to be cleared. As a result a series of decisions were made, the most important of which were to abandon implementation through the line departments, and to revise the operational targets. 3.7 It was agreed that two Project Directors (reporting to GOUP), one in each of the major watersheds covered, would be appointed and be wholly responsible for the implementation of the program in their areas. Each Director would receive a budget direct from the state Finance Department and would allot budgets to the line program heads (now redesignated as Deputy Project Directors). Line department staff were seconded to the project. It was also agreed that two "Centres of Excellence" would be established, one in each region, as a means of providing a demonstration of the approaches which could be adopted. 3.8 The review also agreed significant changes in the physical objectives of the project, based on the initial experience. The forestry targets were scaled back, while those for conservation structures were expanded. The cattle exchange program was terminated and a breeding program substituted, and greater emphasis was placed on general agricultural improvement. These changes are discussed below in the context of the sectorial programs. Project Results Forestry 3.9 The project envisaged operations on three different types of land. Individually owned land, on which farmers would be encouraged to plant seedlings made available to them, community land (primarily used for fodder or fuel supplies) and reserved forest from which local 10. Himalayan Watershed Management Project. Supervision Report, December 4, 1987. Annex 1, paras. 2-3, 5. 22 communities had rights to extract fuelwood and fodder. Generally, community land is closer to the village than the reserved land on the higher slopes. 3.10 The initial emphasis was on planting in reserved areas and provision of seedlings to farmers, since communities were unclear about the benefits which they would obtain from fodder or fuelwood plantings on community land, especially since they had to agree to stop grazing on this land for several years until the plantings were fully established. Local people thought that permitting a government department to plant trees would result in further diminution of the community's control over the area. 3.11 By the MTR the project had planted over 16,000 ha of reserved forest land with tree seedlings (a considerable achievement). However, in the short period involved, there was little visible impact on stabilizing the degrading slopes. Also, enough fodder tree seedlings had been distributed under the farm forestry program to plant, in theory, almost 18,000 ha. However, the modest interest of farmers in the latter activity was revealed by low survival rates, itself reflecting a number of factors. Low on-farm prices for fuelwood and construction poles, the absence of men (discussed below), a high aversion to risk, and the continuing availability of forest resources to many villages, together with the prospect of future supplies from village and/or departmental plantings, seems to have inhibited individual plantings on cropland. 3.12 However, the original targets for these activities for the whole project period amounted to 167,000 ha. This amounts to just over half the total area in the project watersheds and would have been a formidable undertaking using direct, force account methods. This was recognized at the MTR and the targets were, therefore reduced from 59,000 ha to about 32,000 ha for fuelwood plantations, from 28,000 to 4,000 for fodder plantations, and from 81,000 to 18,000 ha for private farm forestry. At completion the targets for fodder and farm forestry were reckoned to have been met, while that for fuelwood was exceeded by 30 percent. As the project went on, interest among the villagers increased as the benefits in terms of increased supplies of fuel and fodder became apparent. In some cases, additional indirect benefits were noticed, such as the increased flow of water from springs as a result of increased infiltration of water and reduced run-off. 3.13 The audit mission noted that there appears to be widespread adherence to the management practices recommended by the Forestry Department, that is reduced grazing, helped by the program of cattle upgrading (discussed below) which has encouraged greater stall feeding. While the concept of fodder plantations was sharply cut back at the MTR, the project subsequently placed greater emphasis on improvement of pasture areas. This included the exclusion of cattle from grazing areas and greater use of cut forage. This practice increased the growth of indigenous grass species and improved ground cover and is estimated by the PCR to have increased grass yields by 30-40 percent. V-ditches were also cut on the contour and vegetative barriers planted with mixed success: vetiver generally performed poorly, but the overall effect was to increase infiltration and increase available moisture for plant growth. Plantings of Love Grass and clovers have also given generally good results. 3.14 The PCR estimated that the fuel and fodder yields from the plantings carried out by the project are low, despite the higher yields of the improved pastures. For example, the PCR estimates a total fuelwood yield over the life of the plantings of almost 19 tons per hectare, only I1. In this context fodder includes both grass and green material from trees, cut and carried to livestock. 23 half of the appraisal estimate. Such estimates are hampered by the lack of any direct measurement, but the PCR estimates are also based on species composition assumptions which are disputed by the implementing agency. The PCR assumption, that a high proportion of the trees planted were of slow growing species, meant that it was assumed that little lopping for fodder etc. would be feasible during the first ten years of tree life. In practice, it is clear that this is not the case, and fodder is being gathered earlier than assumed by the PCR. 3.15 The PCR also claimed that the nurseries established under the project had been closed. In fact they were turned over to the state Forestry Department. Because of the termination of the project's planting program, the overall demand for seedlings has fallen, and some nurseries have been closed, but several remain open, and are supplying seedlings to farmers. Conservation Structures 3.16 The project placed considerable emphasis on building conservation structures in gullies, especially in areas covered by fuelwood or fodder plantations. The PCR concluded that these structures were providing little direct benefit and that many were poorly constructed and not working properly. Observation at a sample of sites visited by the audit did not confirm this finding. 3.17 The PCR mission was carried out in late March and early April, before the start of the monsoon and at the end of the driest period of the year. In contrast, the audit visited in late September, a month after the end of the monsoon. As a result, vegetative cover in treated gullies is likely to have been much more extensive at the time of the audit mission, and may explain the differing conclusions. The true situation remains unclear and is discussed further below. 3.18 It is also clear that the key to reducing gully erosion in this region, lies in increasing infiltration of water when (or soon after) it hits the ground, and not in diverting or slowing runoff. In the absence of improved plant cover, (preferably combined with measures to slow run off, e.g. contour planting of suitable species to create vegetative barriers and/or contour ditches), structures in gullies will achieve little. Once a large volume of water is moving over the surface or in channels, slopes as steep as those in the project area, (most 30 degrees or more), it will do so quickly and cause damage. However, structures in the gullies, combined with the other measures, will improve the moisture retention near the edges of the gullies, hold soil and encourage further plant growth increasing the overall infiltration of rainfall. This was clearly happening at several of the sites observed by the audit. 3.19 The results of these activities are illustrated on p. 25. The top right picture shows well established fuelwood plantations. The area in the right foreground was planted 6-7 years ago, that in the right background, about 10 years ago. Both are solid stands, and contrast with the left foreground, which illustrates an area with traditional management. However, in this case, the benefit from the planting on the right is less than it would have been, as no work was done in the gully and water continues to move down it at speed. 3.20 The two pictures on the left are of the same plantings. In the lower picture, the foreground left is of a 7-8 year old fuelwood planting and to the right is a 'fodder' plantation of similar vintage. In the middle distance a heavily grazed area typical of the whole slope at the beginning of the project can be seen. In the upper left picture (which looks up-slope) the man is standing next to a small checkdam, almost hidden within the vegetation which has encroached 24 into the earlier gully. That the combination of check dams and dense vegetation on the slopes has greatly reduced the impact of water moving down the gully, is clearly shown by the amount of vegetation in the gully in September, shortly after the end of the monsoon. Data on which to assess the effectiveness of the structures developed does not at present exist. It is recommended that a sample survey be carried out for this purpose (paras. 4.08-4.11). Agricultural Production 3.21 Improved Grains. As discussed in the PCR, the project initiated a number of steps to increase agricultural production, which have had a significant impact. For example, new varieties of wheat and rice are estimated by the PCR to have been adopted by about 80 percent of households, a significant performance. Given the variability of micro-climates in the project area, yield increases are also very variable. Both crops are primarily used for subsistence, frequently a barrier to adoption of new varieties because of taste preferences and variations in cooking quality. However, a large proportion of the men in this region spend much of their lives working in the plains and have developed a taste for the types of rice grown there and generally eaten in India. Since the new varieties introduced by the project had characteristics closer to this preference than the traditional ones, they met ready acceptance. The wheats were also preferred because they are awned12 and, as such, have been subject to greatly reduced losses from birds and bears (which are a nuisance in parts of the project area). As a result, harvested yields are estimated to be 50 percent or more greater than from traditional varieties. 3.22 Livestock. The initial cattle exchange program, in which milk cows were exchanged for high yielding buffalo cows was not successful. This was largely because rural households were reluctant to give up their cows, to which they were closely attached, to an uncertain fate under state protection. After two years the project attempted to establish an artificial insemination (AI) service to upgrade the offspring of the existing herd but, for largely logistical reasons, conception rates were low, and this was also abandoned. After the MTR, the project concentrated on developing Natural Breeding Centers (NBCs) at which one or more bulls of improved strains (largely Jersey or Jersey crosses with local stock) were maintained, and local cows brought to them for service. This program worked well and the offspring are reported to be producing 4 to 5 kg of milk per day compared to only I to 2 kg for the traditional stock. 3.23 However, this program is now floundering because of problems following project completion. During the project about one quarter of the NBCs were directly managed by the Livestock Department and the remainder by farmers' groups. In both cases, the operation was subsidized by the project, including free bull service. After completion, responsibility for all centers was turned over to the Livestock Department, which was unable to continue the subsidies. Payment for the services of a privately owned bull is accepted practice in the project area. However, villagers are reluctant to accept that this should apply under what had been a government program. In consequence, the Livestock Department has followed through on its threats to close the centers if the costs were not covered locally. Subsequent projects have levied charges from the start and included arrangements for a revolving fund for bull replacement, but many of the NBCs operated under the HWMP are now unused and empty. This situation is part of the wider issue of the responsibility for assets created by the project and is discussed further below. 12. The grain has a course bearded shell and is unattractive to birds and other grain caters. bid~ ho( lefi: fen vear old plantation on right contrasts sharply with unprotected grazing on the left and an unprotected gully running . down the center. A bove: Foreground: eight year old fuehwood plantation. In the center of the picture a protected fodder plantation on the right contrasts with heavily grazed area ,n the left. with a stone wall between them. Le/ hlie rock cheekdam is barely \isibie in the center, but (together with plantings and natural growth on the slopes) has reduced strearn flow and permitted extcnsive growth of vegetation in the gully. 26 3.24 Horticulture. The project also aimed to stimulate horticultural production, to increase the marketable surplus. This objective included the supply of fruit tree seedlings, subsidies for the development of orchards, and the grafting of improved material onto existing stocks (topworking - mainly of apples and pears) to upgrade existing orchards. While grafting met limited interest, and only one quarter of the area projected at appraisal was actually improved, the project claims to have exceeded targets for new planting. 3.25 However, this development is not very apparent in the project area, reflecting the audit's overall impression of the project area, as one lacks commercial orientation and entrepreneurial dynamism. There is very little evidence of spontaneous investment either in improvement of housing or of agricultural assets, or of development of commercial facilities. As noted, the majority of the younger men work away from the area, and this has led to the phenomena of unused or abandoned terraced land. The project attempted to interest villagers in improving these areas with little response, reportedly because households with unused land were reluctant to make it available to others who might be interested in renting it, because of concerns about their future ability to reclaim the land when the male members of the household returned. The result is a general air of stagnation in the area with the villages being seen as places from which families would move if they had a real prospect of achieving improved living conditions (especially housing) elsewhere. Sustainability 3.26 The PCR rated the sustainability of the project as uncertain, and the audit maintains the same rating. Concerns on this front primarily relate to the maintenance of the assets created under the project, project performance in relation to the goals for reduction of land degradation, and continued support from public agencies. These concerns are also overlain by the difficulties resulting from the non-participatory nature of the project as initially designed and implemented and also, perhaps, to deeper problems in the social structure and characteristics of the region. 3.27 Project Assets. The problem of asset maintenance is seen most clearly in relation to the NBCs (see para 3.23), but also occurs with respect to small irrigation systems built with project funds, and to management of community lands which have been improved through project fuelwood or fodder plantations. Anecdotal evidence suggests that irrigation systems, which have been turned over to the communities, are not always being effectively maintained, and the differences between communities in the quality of management of community lands are considerable. The audit recommends that a follow-up survey be carried out with three broad purposes: (a) to assess the extent to which community or departmental management of these assets is satisfactory; (b) to determine, where it is significant the course of poor management and steps needed to rectify the situation, and. (c) to emphasize that there is a responsibility on the part of the community (or other agency) to adequately maintain assets entrusted to them, and that Government will check on this. 27 Since this project is one of the first of its kind, it would seem useful for Government to set a precedent in this area. 3.28 Recall that initially, the project was designed and implemented in a top-down manner. This partly reflected the approach of the time, but was also partly determined by the design of the project, which called for it to be implemented through individual line departments with limited coordination through the Forestry Department. This design, with different agencies working in the same communities, greatly reduces the likelihood that a participatory approach will be adopted, as each agency does not have the resources (or skills) to adopt such an approach, and even if they did it would be duplicative and counterproductive. 3.29 As a result, the work undertaken often did not reflect the priorities of the community. This problem was recognized and, as a partial offset, project activities were subsidized, i.e. there was no charge for the services of the bulls, or for the irrigation facilities, and villagers were paid to carry out plantings on community lands. This further blurred ownership and responsibility for the assets. 3.30 The situation was further complicated by the rather abrupt termination of the project. Project staff claim that they had been led to believe in 1991 that there would be a third, one year, extension of project closure and had planned a work program on this basis. The Bank indicated a willingness to extend the Closure date, on condition that Government submit detailed proposals for the transfer and future management of the assets created under the project, including the areas reforested. These were not received and, in consequence, the final decision to terminate was made at the last minute. As a result, programs were halted as funding was terminated and staff were transferred to other projects. Thus project staff had little time to discuss with villagers what would happen when the project ended, and to try and at least get the community to take some steps to prepare for the increased responsibilities it was now to shoulder. 3.31 In consequence, at least some villages were left uncertain as to what their responsibilities were to be and, given the modus operandi of the project, tended to assume that another agency of government (e.g. normally the relevant line department), would take over responsibility for the assets, and that project arrangements would continue. This largely happened in the case of the NBCs (para 3.23). 3.32 Institutional Support. At project closure, staff were withdrawn from the project area somewhat abruptly, and reassigned to other projects being implemented by the UP Watershed Management Directorate. People assisted by the project then had to revert to the relevant line departments of the UP State Government. Those with whom the audit mission spoke did not indicate that any specific measures were taken by the line departments to initiate any follow-up contact. All three of the farmer groups met by the audit mission indicated some degree of dissatisfaction with the current situation and described areas in which they were seeking advice or assistance and had not been able to obtain it, e.g. disease problems on vegetables, replacement seed, operational problems with the irrigation system constructed by the project. These problems seem to reflect social problems seen in the area (lack of young men and the limited entrepreneurial spirit) which is discussed further below. Nevertheless, there appears to have been inadequate follow-up by the line departments, who are also limited by their own budget. 28 Institutional Development-An Evolution in Approach 3.33 Preparation of the project began in 1979. It was recognized that components selected for inclusion under the project should be acceptable to the local population, and be designed so that the majority of the people in the project areas would perceive themselves to be beneficiaries of project activities, such as fuelwood plantations and improved pastures. Early in preparation a Bank report discussed the implications of a participatory approach at some length (para. 2.4). 3.34 This same report recognized that adoption of a participatory approach to the identification, preparation and implementation of local level activities would have organizational implications. There would need to be a coordinated approach to village communities and that could best be ensured if the overall program were managed by a separate directorate in the Hill Development Department, with staff from the relevant line agencies being seconded to the directorate. However, the central and state governments were cool to this idea. 3.35 The impetus for the program was the perceived need to improve management of land, primarily by increasing plant cover. This implied the need for a push to achieve appropriate action. Therefore, multi-disciplinary teams were established to prepare detailed plans for a sample of micro-watersheds. At the same time committees were established with members from concerned panchayats, local leaders and representatives of the line departments to facilitate planning and implementation. 3.36 Recall that, by the early 80's, the Bank was beginning to question the wisdom of using separate project implementation units for rural development projects, based on experience with such arrangements in the late 1970s. This meant that, unless the preparation/appraisal team could come up with very strong reasons for a single project agency, the Bank was unlikely to make it a "make-or-break" issue, and the Government view prevailed. 3.37 However, as discussed in paras. 3.3-3.6, this structure did not work well. People's participation was seen as something to be achieved through social planning. This approach was reinforced by designing programs with sector specific funding arrangements and with minimal funding for cross-cutting social development activities. The aim was to encourage people to self manage their resources with the collaboration of the line agencies. Community participation did not have a central role in planning or management. Unfortunately, sectoral line agencies lacked funds to adequately fill these roles after project funding ceased (para. 3.23). 3.38 The Mid-Term Review in 1988 was a turning point, not only for the HWMP, but also for the approach to this type of project. For the first time senior staff from both the union and state governments and the Bank talked through the objectives of the project at length and the problems being faced in implementation. As a result, organizational changes were made with staff being seconded from the line departments to a Watershed Directorate, located within the HADD, which took overall responsibility for implementation as had been proposed during preparation. This greatly facilitated operations and implementation proceeded much more effectively afterwards. 3.39 At the same time, changes were made in the field level approach. Multi-disciplinary teams were used as before to prepare village plans, but now these were discussed with a village consultative committee, comprising local resource persons. Modifications were incorporated, and the plan presented for final approval by the committee. Only then were works carried out. 29 In reality, given the short time horizon of the project at that stage, this village planning approach proved to be too time consuming. . While there was a distinct improvement in performance, as measured both in physical terms, and in the relationship with the communities, the initial problems resulting from the general lack of consultation and participation, could never be fully overcome. Project staff attempted to ensure local participation through monthly micro- watershed meetings and entrusted some activities to local women's and youth organizations. But, overall the top down approach persisted and few communities were fully involved 3.40 This 'semi-participatory' approach was incorporated into two, parallel, EEC funded projects which were initiated in 1989 and 1990 and implemented by the UP Watershed Directorate (Bhagirathi II and Bhimtal), and has subsequently evolved into a more directly participatory approach being used in the Doon Valley Project, also EEC funded. In that project participatory Rural Appraisal techniques are being used to better understand community dynamics, resources, aspirations and needs and in generating realistic plans. The project was initiated with a substantial training program as staff were used to working in a hierarchical structure that militated against effective, joint work with beneficiary communities. An NGO was engaged to organize this training and subsequently links were developed with more than 10 NGOs which are working in partnership with the project. Beneficiary an user groups are being formed including representatives of all community groups, including women. Locally elected motivators are identified to interact with these groups.13 It is too early to assess the success of this new approach in the hills of UP. 3.41 While in the HWMP the initial approach incorporated is generally accepted to have been flawed, steps were taken during the life of the project to modify it when problems became clear and the process of identification and incorporation of lessons learned during the project period has continued after its closure. Because the project was large and visible enough to command the attention of central and state governments it had an impact which went beyond it's own immediate operations and affected the overall approach to land resource management problems not only in the lesser Himalaya but more broadly across the National Watershed Program funded by the Bank and other agencies. The MTR decisions enabled the subsequent EEC funded projects to begin with a more appropriate institutional structure and move to a more participatory approach. Economic Rate of Return 3.42 At appraisal separate estimates were made of the economic rates of return (ERR) to forestry activities, horticulture, irrigation, and the whole project. The PCR estimate of the ERR was based on the forestry components only (which, however, amounted to 60 percent of project field expenditures). It considered that available data on the production of grains, fruits and vegetables and milk in the project area were insufficient to enable any estimate to be made for those components. However, since forestry represented over half of the field expenditure of the project, it does provide an initial indication of overall performance. 3.43 The PCR estimate for the ERR of forestry related activity was 8.5 percent, which was reduced to 5 percent if a pro rata share of project overheads was added. The audit has prepared a 13. See: Thapiyal, et. al. "Participatory Watershed Management in the Lesser Himalayas: Experiences of the Doon Valley Project", Delhi, 1994. 30 revised estimate of the ERR for forestry activities based on additional data. This indicates a return of 14 percent; 11 percent if overheads are added. Nevertheless, these estimates are not very robust as prices are only guestimates-there is no active real market in the relevant commodities. If the value of the fuelwood and fodder is reduced by 25 percent the latter rate declines to eight percent and to three percent with 50 percent lower values. To extend this partial estimate of the project ERR for the whole project is not possible owing to inadequate data. However, for the ERR for the entire project to be 11 percent requires that the ERR for the non-forestry components also be at least 11 percent. This is unlikely since, although some yields have been increased significantly, the extent of the total increase in production appears limited. 3.44 The forestry operations had three elements-fuelwood plantations, fodder plantations, and farm forestry. "Fuelwood plantations" involved tree planting with a mix of species some of which have edible leaves and can be lopped for fodder and fuel, as well as fuelwood species which can be lopped or pollarded during the 20 year period covered in the analysis. Some of the species are of lower value for fuel and fodder but can be maintained for later felling for timber. "Fodder plantations" comprised reduced tree planting of species with fodder and/or fuel value, and improved seeding or planting of grasses. Grazing in the area, thus improved, was to be restricted to increase ground cover, although grass could be cut and carried to livestock. "Farm forestry" primarily involved the provision of seedlings to farmers for them to plant on their own land. 3.45 The PCR considered that the fuelwood and fodder production projections at appraisal were too optimistic. They prepared a sharply lower estimate. In turn, the implementing agency disputes the principal basis for the PCR estimate, i.e. the low assumed share of species suitable for use as fodder (see Borrower comments to PCR, attached as Annex A). Also data from trial measurements indicate that considerably greater quantities can be "harvested" per tree than assumed by the PCR. The PCR assumed that material would be removed from only 150 trees per ha, compared with a typical fuelwood planting of 1,500 trees per hectare, of which up to 50 percent typically were fuelwood species. Therefore, significantly greater output is likely. 3.46 Based on the experimental results noted above, the audit re-estimated production from the plantings. This revised estimate amounts to an output of about 10 kg of fuelwood and 6 kg of fodder per day, for each household in the project area. This is in line with the requirements estimated at appraisal, and also with what it is reasonable to expect a household to collect on a regular basis. The plantings undertaken under the project amount to about I ha per village household in the area. These areas do not constitute the only fuel supplies available to the populations as other wooded areas exist, in addition to trees around villages and in cultivated areas. Thus, as project plantings mature, fuelwood supplies in the area should be adequate. Fuelwood is being supplied to larger settlements in the area but, given the high cost of transport, it is unlikely that it will be shipped out of the project area to the population centers in the plains. In other words, local demand is likely to constrain the value of the product, not supply. Thus, the project may have inadvertently achieved a level of planting which was near the optimum. 3.47 However, the primary objective of the project was to improve ground cover in the project areas to reduce runoff by lessening peak river flows and seasonal variations in flow, thereby diminishing the sediment load in the rivers leaving the region and thus reducing erosion. Although no data are available on the impact of project actions on river flow at the discharge points of the watershtds, there remain significant areas on which ground cover is modest, as shown in the photographs on page 39. As appraised, the project aimed to plant, in one way or 31 another, half the project area. In the event, about 20 percent was achieved. If the original target had been met, at the cost levels achieved in the project, the decline in value of fuelwood etc. would have significantly lowered the return to the forestry component. Bank Performance 3.48 Bank performance was mixed. Although the major issue of the need for a participatory approach and its organizational implications were flagged by Bank staff early in preparation, this was not followed up strongly, and Government's views prevailed. However, implementation experience validated the initial view and the organizational structure adopted at the MTR was close to that proposed at the outset. The Bank also accepted unrealistic targets in some components of the project, particularly in forestry. There was some initial reluctance on the part of the Borrower to prepare a background paper and schedule the MTR, but the Bank pressed for it and prevailed. 3.49 The Bank also committed heavy inputs to supervision in the early stages but this emphasized a limited range of technical issues. Following the MTR inputs diminished significantly, leaving the impression on the Borrower side that the Bank had somewhat lost interest. The unfortunate consequences of the abrupt termination have been discussed above (para. 3.29). While the Bank was technically correct in that the Borrower had not met the agreed conditions for a further extension, the nature of this condition meant that an abrupt cut-off was always a distinct possibility, and a modest extension, to allow orderly wind down, would have been preferable. Borrower Performance 3.50 Borrower performance mirrored that of the Bank. Although initial performance was poor, the extent to which the lessons of the MTR were learned by the UP Government and the Hill Development Department in particular and incorporated into practice in operations funded both by the Bank and others was commendable. However, the present uncertainty about the status of the assets created under the projects, does not reflect well on the relevant line departments. Compliance with Covenants 3.51 Overall compliance with covenants was good. There was some slippage in meeting the planned date for submission of the Mid-Term Review, as discussed above (para 3.6), but other covenanted actions were almost completely satisfied. 32 4. Issues and Lessons Ratings 4.1 Overall Assessment. The PCR rated this project as satisfactory. However, the audit downgrades this rating to marginally unsatisfactory. On the positive side, the audit reaches a more favorable assessment of the outcome of conservation activities, and the re-estimated ERR (based only on the return to forestry activities) is satisfactory. However, this ERR is not robust. The principal objective of the project was to improve groundcover and reduce seasonal variations of downstream flow ( and floods). However, the project only reached 40 percent of the original target, and the analysis suggests that if this target had been achieved the local population would not have fully utilized the available fuelwood and fodder and the ERR would be significantly lower. The project did not achieve other major objectives. The replicability of its approach to watershed rehabilitation is doubtful. The sustainability of project achievements is uncertain and institutional development modest. Bank and Borrower performance are both rated as satisfactory. The Bank's role in pressing for a mid-term review at the outset and later undertaking it was crucial in achieving the required change in approach to implementation. While the Borrower initially resisted change, it has moved substantially since the MTR to incorporate the initial lessons into this and other operations. Quality at Entry 4.2 The principal reason for the project outcome was that the initial design was defective. This primarily resulted from the perception of the time that the major cause of the downstream flooding in the Ganges Valley was mismanagement of land in the hills, and that a massive effort to overcome was required to rectify the situation. In a sense the project, although large, was seen as a pilot effort covering only 300,000 ha.14 In the intervening 15 years views on the cause of the downstream problems have changed and the contribution of the population in the hills is now seen as less significant. 4.3 The perceived need to move ahead was a factor in the Bank's agreeing with the Borrower that the objectives could best be achieved by limited coordination of line agencies, rather than by establishing a new department or office to undertake the program. As discussed earlier (paras. 3.33-3.4 1) experience during the project period has shown the necessity of a participatory approach to land resource management, where the local population is primarily responsible for land management, especially in fragile environments, and the importance of a single, multi- disciplinary point of contact with the community. This approach is now being used elsewhere in the hills of UP (para. 3.40) and in other regions, but once the HWMP had been initiated on another basis, it was not possible to return and re-establish it on a more participatory basis. As a result, the capacity left by the project for action at the community level for natural resource management is limited, as exemplified by the remaining problems related to the management of the assets created by the project (paras. 3.27-3.31). 14. The proposed Himachal Pradesh component was withdrawn just before negotiations. 33 4.4 The pressure to rapidly improve ground cover, and the fact that the lead in preparation of the project was taken by the Forestry Department, meant that the prime means selected to achieve this objective was the development of plantations (fodder and fuel) and the support of "farm forestry" (the planting of trees by farmers). The total area planned for planting during the project period was 167,000 ha, or more than half of the project area. This physical target was clearly unrealistic and there was certainly no way in which it could have been achieved in seven years using participatory rather than direct methods. The sub-target for "farm forestry" was to plant the equivalent of 87,000 ha of trees on individual holdings. This would amount to more than I ha per household, and would also have been an unattainable target under the conditions (with many of the able bodied men away), even if there had been a potential market for the product, which there was not. 4.5 The cost of the forestry related activities which were undertaken was high (approximately US$400 per ha). Data to show whether the increased biomass had any impact on stream flows and silt load are not available. It is also clear that improved ground cover could be obtained more cheaply by encouraging the growth of shrubs and grasses, than by planting trees, and that the indirect methods of intensifying livestock production, especially upgrading cows, has an important role to play here in encouraging more sustainable methods of fodder production and use. 4.6 Thus, although the project has achieved many of its subsidiary objectives, and lessons have been learned and applied, it was fatally handicapped by the initial perception of the problem it was to tackle and the impact of that perception on project design. Thus, it is difficult to judge the outcome as other than marginally unsatisfactory. And, for the same reasons, the audit judges Bank and Borrower performance to have been only marginally satisfactory. Social Setting 4.7 It was also, perhaps, unfortunate that the UP hills were the site of the first major project aimed to tackle land resource management problems. The project area has a number of particular social characteristics which have affected the outcome of the project and are influencing subsequent developments in the area. Like many mountainous areas, the UP hills are a region of outmigration. Although road links with the remainder of India are adequate, routes are tortuous and transport costs relatively high. Given the steep slopes and, therefore, the relative difficulty of agricultural production, the potential for increasing incomes and accommodating increased population in the area is limited. As a result there has been a long tradition of outmigration, particularly in the form of migration by males for work. Thus, there is a marked absence of younger males, in the 20 to 50 year age group. Communities are primarily made up of women, children and elderly men. 4.8 There is a particular tradition in the region that the migration is to public sector employment. In the villages visited by the mission, the overwhelming majority of the absent men were in the military, teaching, or other public service. This means that when they return they have a pension, even if small, which provides some support. More importantly, the skills developed during this employment, and which are brought back to the villages, are not entrepreneurial but bureaucratic. There was a marked tendency in these villages, when discussing issues with villagers for them to view the problem from a bureaucratic rather than a managerial/entrepreneurial perspective. That is, the response was to consider "which agency can 34 we approach to assist us with this problem, and what do we need to do to obtain assistance?," rather than to focus on what the group itself could do to tackle the problem. 4.9 A striking feature of the project area is the apparent lack of private developmental activity. The villages generally comprise close groupings of houses, surrounded by terraces. They show little signs of improvement i.e. few new roofs or indications of houses being extended, and there are few signs of new shops or related investments at the crossroads. Agriculture is primarily subsistence oriented, focusing on rice and millet, with a few vegetables and fruits. There is little evidence of production for sale, or goods waiting for market. When husbands working elsewhere can obtain suitable long term accommodation, families leave. There was discussion throughout the audit of the problem of 'abandoned terraces', belonging to owners now unable to use them but unwilling or unable to sell the land or allow others to use it. However, the project failed to tackle this problem. In short, it is not an area in which the inhabitants see a future. 4.10 This situation has a number implications for the assessment of the HWMP and the follow-up to it. These relate to the roles of women and of its implications for line departments; the future of the area; and possible modifications of the land management practices now being advocated by the project agencies. 4.11 Role of Women. The migration of younger men and the effects of the type of employment most of them pursue, mean that men in general do not provide entrepreneurial leadership in the villages. Thus it must come from the women, at least in the shorter term. However, there is little tradition of women engaging in business activities in this culture and also, in the project area, many of them are heading what are effectively 'single parent households' and, therefore, have onerous day-to-day responsibilities which limit their ability to undertake new activities. 4.12 At the same time the project made little progress in addressing women. At the beginning it recruited some women motivators, but these were part time and only received a minimal stipend. It is reported that little was achieved. In any event, the project soon switched to a more intensive extension input along T&V lines, and women motivators were not considered necessary. Later in the project life it became apparent that women's needs were not effectively being met and some women extension agents were recruited. This was too late to have any real impact in the HWMP project area, but their numbers are being increased in successor projects and there is some hope that, when they are a feature of a project from the outset, rather than an add-on, more effective relationships may be formed with women's groups and women's priorities will be more effectively addressed. 4.13 Land Management Practices. Under the HWMP great stress was placed on the need to reduce grazing pressure on the forest and pasture lands and villages were required to close areas improved under the project and to cut and carry forage to their animals. In addition to increased milk production, the program for upgrading of cattle was linked to stall feeding, with the same objective in mind. Limited emphasis was also placed on fodder production on crop land. Limitation of grazing does result in improved ground cover and infiltration of rainfall, and also protects young trees from damage by animals. However, manual cutting and carrying of forage from pasture areas is a very time consuming exercise. By necessity it is mainly undertaken by the women, who have other claims on their time. For reasons noted above and especially if recent macro-economic policies increase economic growth in other regions of the country, 35 pressure on land in this region may ease. In this situation, if the project actions improve pasture productivity, the women may not cut fodder in the more inaccessible or steep areas. This can pose a fire hazard, and some have already occurred. Thus, it appears advisable to change the approach to management of the fuel and fodder plantations to a!low limited, short duration, rotational grazing of these areas by dry cows and young stock. This would reduce the demand on women's time, and mean that some of the vegetation which otherwise might not be cut at all, would be grazed down. Conservation Structures 4.14 As noted (paras 3.16-3.18), there is uncertainty over the effectiveness of the work carried out under the project in gullies. This involved the development of vegetative barriers in the upper stretches of existing gullies, and physical structures (generally rock filled gabions) in the lower stretches. In general, these structures have been undertaken where tree planting, and/or improvement of grasses, was also undertaken. This often included the planting of vegetative barriers along the contour, or digging furrows, so as to increase infiltration of rainfall. The PCR judged that these efforts were not effective, while Government and the project agency disagree. Definitive data is not available, but the audit concludes that, in general, the structures appears to be achieving their objectives.15 4.15 However, given that the methods used in this project are still being used in subsequent operations in the UP Hills, and that a major objective to reduce erosion and related land degradation, this is a serious issue. Even if a small percentage of the work is ineffective, this should give rise to concern, and changes in methods. In addition, by the time a survey could be carried out it will be at least 4 years after project completion, and 7 or 8 years after most work was done, it may be appropriate in any case to review the current state of these areas. 4.16 The audit, therefore, recommends that a follow-up survey be carried out to assess the current situation on a systematic basis. This could review a sample of the areas on which work was carried out under the project, but should be large enough (e.g. 25 -33 percent) and geographically spread, to obtain a reasonable coverage of the range of conditions within the project area. This should agree on a set of criteria on which the gullies are to be judged, and should be carried out over a relatively short period of time, e.g. three to four weeks at the end of the monsoon season, when stream flow may still be moderate. It should also obtain data from local communities on their perception of any related variables, such as spring and stream flow, water quality, and the growth of vegetation in the area. Photographic recording of each area might also be useful, to ensure that all sites are actually visited and to enable visual comparisons to be made between sites on characteristics not easily quantified in this type of survey, e.g. extent of vegetation cover. 4.17 The study should attempt to assess the proportion of gully works which are not performing satisfactorily and to attempt to determine the factors related to this poor performance, such as current management practices of the local community(ies), agro- ecological/climatic conditions, slope, or soil conditions. This should enable appropriate 15. The PCR and audit were carried out at different times of the year when the state of growth of vegetation in gully areas is likely to be significantly different (see para 3.17). 36 measures to be taken to adjust ongoing programs, and to determine possible remedial measures and their feasibility. Replicability 4.18 One of the objectives of the project was to develop replicable approach to the rehabitation of watersheds (para. 2.2). The achievement of this objective can be assessed in institutional and economic terms. In both respects, the findings of the audit suggest that the project failed to meet this objective. 4.19 The project relied heavily on the development of plantations as the principal means of increasing ground cover. This is a high cost approach. Project expenditure for forestry amounted to about US$400 per ha in current dollar terms. The analysis carried out by the audit suggests that, although the ERR for the rate of planting achieved by the project was satisfactory, if the original planting targets had been achieved at this unit cost level, then the rate of return would have been reduced significantly, because of limited local demand for the fuel and fodder. 4.20 This outcome suggests that greater consideration be given to alternative, lower cost, methods of increasing ground cover. This could be achieved by greater control over grazing, especially in the early part of the season, which would encourage natural regeneration. This could be supplemented with some hard seeding of tree and shrub species, and minimal planting in the more degraded areas. Greater emphasis on livestock improvement through breeding, as undertaken by the project after the Mid-Term Review, linked to some subsidies to cut and carry and to conserve fodder, could increase the acceptability of restrictions on land use. However, this approach, which would result in lower levels of use of available grass, might increase the fire risk. With time this risk might be lowered as grasses are replaced by trees and shrubs. 4.21 On the institutional side, the shortcomings of the project's initial approach have been discussed at length (paras. 3.33-41 and 4.11-12). Significant changes in the approach initially adopted have been made in subsequent operations in India casting considerable doubt on the approach developed by the project, which was heavily constrained by the initial design. Problems of Monitoring 4.22 The disagreements between the PCR mission on the effectiveness of conservation works in gullies and of tree planting have been noted (paras 3.14 and 3.16). At the same time the PCR and others have been critical of the project for being too focused on physical targets at the expense of working more closely with local people and emphasizing more general improvements in the state of land resources. 4.23 Both of these concerns basically relate to the problem of assessing the performance of efforts to improve resource conservation. The World Bank, FAO and others are in the process of attempting to devise indicators which will provide some operational guides in this area. The debate to date has been marked by a wide range of views on where the priorities for measurement should lie, and what is really measurable in any consistent way. The basic problem is caused by the site specificity of conditions and the fact that what is aimed for is an improvement over the pre-existing situation, not necessarily achieving some target level of physical performance. Given the accepted need to increase local participation and 'ownership' in 37 resource management activities, there is also a need to find indicators which are meaningful to local people. 4.24 The problem is illustrated in the photographs reproduced on page 39. Aerial photography can be used to estimate the average biomass over an area. However, such figures are not very meaningful on small areas such as those involved in the HWMP. Relatively dense trees may be a less effective ground cover than the vegetation shown in the left hand photos with a mixture of trees, shrubs and grass. Crude stream flow data--e.g. extent of dry periods, presence of silt, etc. may give an indication but are site specific, and difficult to aggregate. What is most important is perceived local benefits. 4.25 In order to meet these varied needs, there would seem to be a role, as a part of local monitoring, for the use of photographs to record progress. Where as is now frequently the case, communities draw up outline plans with project agencies, which are then incorporated in a small written report, with the appropriate plans, photographs of slopes or gullies prior to improvement, and then follow-up photographs at agreed intervals at specific times of the year, would provide a valuable record of change. Such visual 'data' are likely to be much more meaningful to the population than numbers related to characteristics which may not relate well to their preoccupations. They would also enable assessments, such as that at completion, to obtain a broader view than is possible in a brief field visit, and reduce the potential for the rather unproductive interchange which occurred in the PCR of the HWMP. Handover of Assets 4.26 Post-project management of the assets created by the project is a matter of concern. However, evidence on the situation is anecdotal rather than systematic. It is recommended that a survey be carried out to assess the extent of the problem and to determine its causes. These might stem from the initial design of the activity or from the post project actions of those now responsible for the assets. The survey would also help: (a) demonstrate that Government remains interested in the assets created using its funds; (b) emphasize that local communities are considered accountable for them; and (c) assess what might be done to remedy the situation and to modify the design of ongoing and future operations, so as to minimize the problem in future. Since larger scale operations of the same type are now underway, it would be valuable to establish these principles now in one of the first large operations which has been completed and handed over. Lessons 4.27 Where a project is to assist communities develop assets or rehabilitate resources for which they are subsequently expected to be responsible, it is essential that the communities be involved from the very beginning in the selection, design and implementation of these activities. This lesson has been well accepted and has been incorporated in subsequent natural resource management projects in UP and in India in general. In fact the initial difficulties of the project were a catalyst in bringing this issue to the governments' attention and in beginning the process of change to incorporate this lesson. 38 4.28 Treating natural resource management problems, in areas such as that covered by the project, as emergencies, requiring crash programs to cope with them, is likely to be counterproductive. Such crash programs, in which specific and short term targets are envisaged, are not likely to be compatible with participatory approaches. The only way in which the huge targets envisaged in this project for forest related activities (over half the area to be planted) was by direct action by the responsible line agencies, whatever might have been said about participation, and meeting the needs of the villagers. 4.29 Participatory approaches will not be effective unless the institutional structure is appropriate, and the project deploys staff with appropriate skills. Where more than one line department is likely to be involved in the implementation of activities on the ground in the same communities, they cannot effectively engage in initial participatory activities. Some other entity, which may not be governmental, is needed to take the lead in problem identification and in helping communities select activities for which line department support is subsequently provided. Under the HWMP the initial institutional structure, with identification and implementation being carried out by line agencies directly, could not have led to a satisfactorily participative approach, even if the agencies had been so inclined. 4.30 A Mid-Term Review can be of great value in requiring all parties to meet and review in depth problems which have arisen, and reaching decisions on modifications to respond to them. Without a review in 1988, it is unlikely that this project would have been redesigned. After two more years of frustration it would probably have been terminated at the original closing date in 1990 and declared to have been unsuccessful. Quite possibly, this would have delayed by several years the development of effective approaches to land management programs. 4.3 1 There is a need for a broader range of indicators to monitor and assess progress in achieving resource management objectives. Particularly where operations involve the support of small scale, village level activities, the indicators need to be meaningful to the local participants. Some of these indicators will have to be qualitative in nature. Photographic records, either from aerial or ground views, can be valuable for this purpose. 39 Annex A Borrower Comments on the Project Completion Report The Borrower provided comments on the PCR. Unfortunately, the material which was attached to the final version of the PCR was missing several pages. Most crucially these missing pages contained the comments on the yield of fuelwood and fodder from the fuelwood plantations established under the project. An additional copy of these comments was supplied to the audit mission. However, because of repeated copying of a carbon copy, the quality is now poor, and it was decided to reproduce the text. The retyped version of the text is attached. Annex A 40 Comments on PCR (Part I and III) Subtitle Page Para. Comments Evaluation Summary Implementation Experience (ii) 3 The shortfall in planting of fuelwood and fodder tree species was against SAR targets. Since SAR targets were revised during MTR, the SAR targets ceased to be effective. There was no shortfall against the MTR targets. In fact these were exceeded. Results (iii) I As mentioned above, the SAR targets became redundant after MTR. Hence, the observation that the project fell far short of attaining its targets in the forestry and soil conservation components, and the inference drawn from this observation is misleading. In fact, the MTR targets of these components were not only fully attained, they were exceeded. Sustainability (iii) 3 &4 Comments on these observations have been given in the relevant paragraphs in the following pages. Part I Project Design and Organization 4 4.5 Although at the time of the launching of the project, the social issues were not emphasized enough, this aspect was taken into consideration during MTR. Himalayan Watershed Management was a pilot project. As such, the problems experience during the course of project implementation could have not been foreseen. However, a "Project Launch Workshop" would have been highly beneficial in proper orientation of the project staff. Project Design and Orientation 4 4.7 It is agreed that setting up of an M & E unit would have helped. However, basic information such as the number of beneficiary families and the area and average yields of improved cropping were properly documented by the project. Project Implementation 5 5.4 PCR para. 5.4 mentions that SAR plantation target of 86,000 ha. was based on extrapolation of data from Maachlad SWS, and it was subsequently found that only 55,000 ha. was actually available for afforestation and 41 Annex A Subtitle Page Para. Comments consequently, the MTR reduced the afforestation target to 35,905 ha. Even then, while making observations on forestry component, SAR targets have been quoted and used, which apparently has resulted in aberrated and skewed assessment of the component. 5 5.5 The project activities were carried out by the line departments of the Government of U.P. from the year of start in 1983 till 1988. The recommendations of the World Bank supervision missions were regularly sent to these line departments for follow-up. The line departments took the project work as an additional ity to their normal departmental activities. This was not one of the reasons why the World Bank supervisions missions' recommendations could not be translated into action in the pre-MTR period. There were a number of other drawbacks in the line department system of working which led to major basic changes in the organizational structure, implementation strategy and project philosophy after Mid-Term Review was conducted. As a result the project's focus was shifted to meet local community needs for fuel and fodder. Project Implementation 5 5.6 Forest nurseries in the project areas have not been closed but transferred to the line departments. The observation that there was a tendency to adopt a blue print approach for making contour furrow is not true. There was considerable elasticity in the approach and the methodology was changed where there were abrupt changes in slope angle. Vetiver was not used exclusively for forming vegetative barriers along the contours. Other suitable indigenous species like Dodonea viscosa, love grass, etc. were also planted on the contour furrows. 5.7 Targets for physical gully control structures as mentioned in the SAR or MTR were only indicative because it is impossible to predict the number of structures which would be required to treat a nala without carrying out detailed survey of the area. Hence, under soil conservation and gully control work, a series of small, low-cost check dams were constructed after proper survey of the nalas. Though at a glance, the number of the structure constructed seems quite high, it is not so if one takes into account the quantum of problems of gully erosion in the area. The average cost of these structures comes to Rs.360 per Annex A 42 Subtitle Page Para. Comments brushwood/stone checkdams; Rs.2670 per cratewire dam and Rs.6400 per drop structure. It is apparent that the unit cost per structure has been quite low. There might have been some instances of faulty or inadequate designing of some of the structures but overall quality of the soil conservation work, designs of the structures and their cost-effectiveness has been adequate. In the various World Bank review missions reports too, it has been observed that the technology of construction was found to be appropriate and cost-effective. 6 5.8 The instruments installed by the project to measure run-off and soil loss in some mini-watersheds were supported by a standby instrument which was used in case of any breakdown in the installed one. Hence, the run off and sediment data was regularly collected and properly maintained throughout. 5.10 The observations that most horticultural nurseries within the project area have ceased production is erroneous and misleading. The project did not establish its own fruit-seedling nurseries but obtained most of the horticultural planting material from the line department's nurseries and other sources. Even after the closure of the project, the line department's nurseries are continuing their production. 5.11 The cattle exchange program was dropped after two years and not after three years as mentioned in the report. 5.12 The limitations of "Line Department Concept" were apparent, and it was not in time with the project approach. As a result the project could not get adequate momentum and proper orientation. This crucial aspect was taken into account during MTR. It was only after the MTR in 1988-after a lapse of five years-that "Unified Commands" system was adopted which gave a discernible impetus to the project. The performance of the project-both in qualitative, as well as quantitative terms-improved drastically after the switch over to "Unified Command" approach which it reflected in the World Bank Supervision Mission's reports also. The "Unified Command" approach could not have been made applicable at the start because there was no experience in the department for such approach. It is obvious that the "Unified Command" approach was developed after the results of "Line Department" 43 Annex A Subtitle Page Para. Comments approach were examined. This is a development process which, naturally, takes time to evolve. Project Results 7 6.1 The SAR targets were revised in the MTR and more realistic targets were fixed which were not only fully achieved, but were even exceeded in case of Forestry and Soil Conservation components. This is also reflected in para. 6.2 of the report. Before MTR, when the work was executed by the line departments, it was scattered all over the project area. However, after MTR a multi-disciplinary team concentrated their efforts on some selected microwatersheds in an integrated manner. As a result, the deterioration of the ecosystem in the selected MWS was minimized. 7 6.3 The shortfall in forestry mentioned in this paragraph is with respect to SAR targets. As already stated, the SAR targets had ceased to exist and were replaced by MTR targets. Hence it is erroneous and unreasonable to refer every time to SAR targets for assessment of achievements. The PCR mission has grossly underestimated the fuelwood yields from forestry and fodder plantations as 18.9 and 10.0 t/ha respectively after 20 years. Even the Mission has agreed that these estimates cannot be verified. Central Soil and Water Conservation Research and Training Institute (CSWCRT) have given the following estimates of yield per tree from Grewie oppositifolia in their Fakote ORP. Fakote ORP area is similar to that of Himalayan Watershed Management Project area. Though Grewia oppositifolia is slow growing species, the yield data for this species in the Fakote ORP can be used to make a conservative assessment of the expected yield from the plantations raised by the project. (A) Yield of twigs from lopping of trees (10th year Onward) Frequency of lopping: 4 lopping once a month for 4 months in a year, i.e. maximum 16 loppings/year. (1) Estimated yield in 10th to 15th years: 2.5 kg of twigs per tree per lopping, 16 loppings will give 2.5 x 16 = 40 kg/tree/year Annex A 44 Subtitle Page Para. Comments (2) Estimated yield in 16th year: 50 kg/tree/year (3) Estimated yield in 17th and 18th years: 60 kg/tree/year (4) Estimated yield in 19th and 20th years: 80 kg/tree/year (B) Yield from thinning in the 15th year Minimum Height 3 M Minimum DBH 20 M Minimum yield per tree: 0.1Cum = 0.5 Qt/tree (C) Yield from final felling in the 20th year Minimum height 3.5 M Minimum DBH 22 Cm Minimum yield per tree: 0.14 cum = 0.75 Qt/tree Total yield per tree at the end of 20 years (A) From lopping 10th to 15th years 40x6 = 240 kg. 16th year 50xI = 50 kg. 17th & 18th years 60x2 = 120 kg. 19th & 20th years 80x2 = 160 kg. Total 570 kg. or 5.7 Qtl/tree (B) From 15th year thinning 0.5 Qtl/tree (C) From 20th year final felling 0.75 Qtl/tree If the yield of twigs from lopping is taken as consisting of 25 percent of fuelwood (1.425 Qti per tree) and 75 percent of fodder leaves (4.275 Qtl per tree), the total fuelwood yield at the end of 20 years is estimated to be 1.425 + 0.5+ 0.75 = 2.675 Qtl/tree as per the CSMCRT data for Fakote ORP. It means that 38.9 t/ha fuelwood yield will accrue at the end of 20 years from a 45 Annex A Subtitle Page Para. Comments plantation which has 145 stem per ha. Similarly 38.9 t/ha of fuelwood will be available from lopping only in a five-year-old plantation with 273 stems per ha. In the post-MIR period the number of trees planted in the three plantation models was 1600, 2500, and 1000 per ha. Hence the average number of seedlings planted per ha may safely be taken as 1700 which is more than six times that of 273 stems and more than II times that of 145 stems needed to produce 38.9 t/ha fuelwood as per the calculations given above. Estimated yield of fodder from plantations As mentioned above, the total yield of twigs from lopping (10th year onwards) comes to 570 kg/tree, and if 75 percent of the lopped material constitutes fodder, the yield of fodder comes to 4.275 Qtl/tree over a period of ten years (i.e. 10th year to 20th year). Thus the average yield of tree fodder from the plantation only comes to 0.43 Qtl/treelyear. If we take the average number of trees planted in different plantation models as 1700 per ha, and out of these, on an average only 200 trees per ha of fodder yielding species., the yield of fodder from plantations only comes to 0.43 x 200 = 86 Qtl/ha or 8.6 t/ha/year which is more than four times that of the assessment made by the PCR mission. The yield of grass from the plantations and the pasture development areas will be in addition to the above. The observation that the stands included about 40 percent of species (Quercus, cedrus, cupressus) that would not be harvested in the first 20 years does not reflect the ground truth. In fact, the species mentioned above can be grown only in areas with elevation more than 500 M.. The major species planted in such areas are Robinia, Acacia, mollissima, Quercus Serrate, etc. which are fast-growing species. All these species give adequate intermittent yields through lopping after six to ten years. Cedrus and cupressus form a very low percentage of the total plants. The MTR estimate of fuelwood has been taken as 60 percent of the SAR estimate which again is a wrong interpretation of the details given in MTR. In MTR report, it has been mentioned that the physical achievement in forestry sector up to March 1988 ranges Annex A 46 Subtitle Page Para. Comments from 46 to 60 percent of the SAR physical targets for the same period. It does not mean that in MTR the i yields are estimated to be only 60 percent of the original SAR targets. The approach of the project towards meeting the fuelwood demands in the area has been two-pronged: 1. By enhancing the availability of fuelwood through different types of plantation models. 2. By reducing the demand of local people for fuelwood through alternate sources of energy (such as biogas) and introduction of energy-efficient, fuel-saving devices (such as smokeless chulhas, improved wood stoves, pressure cookers, etc.) In the former case, the PCR mission has omitted to take into account the introduction of energy plantations over an area of 1085 ha for production of seasonal energy biomass. Such energy plantations are expected to yield 14t of fuel per ha. In the latter case, it is estimated that the demand for fuelwood is reduced by 14 t per year for each biogas plant installed; 0.7 t per year for each smokeless chulha and improved wood stove and 15 t per year for each pressure cooker introduced in the project area. If all the fuelwood saving devices introduced in the project area are taken into account, the net saving of fuelwood amounts to equivalent of 47,346 t of fuelwood produced per year. 8 6.4 The leaf fodder production from fuelwood and fodder plantation has also been grossly under-estimated. As is clear from the details given in the comments against para. 6.3 fodder yields is estimated to be 8.6 metric tonnes/ha/year from lopping of 200 trees only. The project had adopted a multi-pronged approach towards meeting the fodder demands of the area. The supply of fodder was increased through enhanced availability of grass and leaf fodder from different plantation models and pasture development program carried out in the non-arable land. In addition to this, cultivation of fodder crops was encouraged in the farmers' fields by introducing fodder minikits. The 47 . Annex A Subtitle Page Para. Comments fodder minikits distributed during the project period (80,750 minikits) covered an area of 3230 ha. The production from these fodder minikits was estimated to be about 15 t/ha. This additional production from the cultivated fodder has not been taken into account in the PCR mission's report. The other aspect of fodder management was to reduce the wastage of fodder and to ensure maximum utilization of the available fodder and agricultural waste as animal feed. With this object, chaff, cutters, thrashers, feed troughs, grass choppers, etc. were introduced in the area which resulted in reduction of pressure on forests for fodder and grass. The PCR mission has not taken this aspect into account while making their assessment of fodder balance. 6.6 In the pre-MTR period, the World Bank Supervision Missions' recommendations were regularly sent to the line departments for follow-up. However, the line department concept had its limitations. This has been dealt with in the PCR par II prepared by the Project. A brief mention has also been made in the comments given against para. 5.12 above. IT may, however, be mentioned here that in the highly problematic areas of the project, the top soil-which supports the vegetative measures-is either washed away, or is consistently on the move. In such cases construction of low cost engineering control works is essential to arrest the washing down of the top soil. Only then the site conditions are improved enough to support vegetative measures of soil conservation. The assessment of labor costs of check dam does not take into account that construction of brushwood and stone check dams is exclusively labor-oriented. In case of cratewire dams and drop structures the labor component is 50 percent and material component is also 50 percent. As mentioned in para. 5.7 above, the average cost of a cratewire dam and a drop structure comes to Rs.2670 and Rs.6400, respectively. Hence, the approximate labor component for these structures comes to Rs.1335 and Rs.3200 only. Taking the average unit labor cost as Rs. 18 per man day, labor cost for one cratewire dame comes to 74 and for one drop structure comes to 177 man days only. Thus it is obvious that the PCR mission has grossly over-estimated the labor cost of these items. Annex A 48 Subtitle Page Para. Comments There is no denying of the fact that some technical faults in a few soil conservation structures might have crept in, but no generalization can be made from such stray cases. The World Bank Supervision Mission in their field reports had repeatedly observed that the cross section of these structures seen by them during field visits were found to be cost-effective and the technology of construction was appropriate. 9 6.10 The PCR Mission's report states that the cost of establishing the fuelwood and fodder plantations were much higher than expected. The statement is unwarranted and does not represent the true picture. In fact, the details given by PCR Missions in this regard have been quoted out of context from an old report (May-June 86) of World Bank Review Mission. At that time (about seven years back) the line departments were executing the project activities as an additionality to their normal departmental work. The observations of World Bank Review Mission's report (May-June 1986) were later looked into and appropriate corrective measures were adopted in the subsequent year. From 1988 onwards, i.e. after MTR, United Command system was adopted. The quality of work markedly improved after MTR which was also reflected in World Bank Missions field reports. PCR Mission's report also states that the planting rate of five to 12 trees per man day was excessively high. It is not clear how these figures have been arrived at by PCR Mission. In this context it is pertinent to note that all works in the Project are carried out as are the approved schedule of rate. The prevailing wage rate is taken into account while deciding upon the schedule of rate. The present schedule of rate for planting is Rs.0.18 per seedling at a wage rate of Rs.25 per diem. It indicates that the planting rate is 139 seedlings per man day. Even if we add the cost of carriage (head load for average distance 3 km), the planting rate comes to 42 seedlings per man day as per the details given below: Schedule of rate per seedling (a) Carriage of plants (head load) Rs.0.14/km Average distance of carriage 3 km Cost of carriage Rs.0.42/seedling 49 Annex A Subtitle Page Pam. Comments (b) Cost of planting Rs.0.18/seedling Cost of carriage (3 Km) + Planting Rs.0.60/seeding Wage rate Rs.25/day Planting rate, including carriage 3 Km = 42/man day It is obvious from the above that the PCR mission's statement that the planting rate was five to 12 trees per man day is erroneous. The observation that dry stone walls were ineffective in protection of plantations because they were easily breached and seldom maintained is incorrect. Protection of the plantations was mostly done by Mahila Mangal Dais. They usually engaged a chowkidar for this purpose who used to patrol the plantation and maintain the cobble wall wherever there was a breach. It has been stated that although only 37 percent of the SAR targets for plantations were achieved, plantation costs were 73 percent of the SAR estimate. This is gross misrepresentation of the facts and statistics. The statement is totally misleading and presents a completely wrong picture. The physical and financial targets of forestry component as given in SAR are presented below: Item Unit Cost SAR Physical (Rs/ha)' targets (ha) 1. Fuelwood Plantation 59,000 (35.1%) (a) Forest Reserves 2615 (b) Civil Soyam 3623 2. Fodder plantation 27,800 (16.6%) (a) Private Land 108 (b) Civil Soyam 1481 3. Farm Forestry 13 81,000 (48.3%) Totals 218.35 M 1,670,800 (100%) a. 1983 level as per Machlad Report Annex A 50 Subtitle Page Para. Comments (A) Total cost of Farm Forestry over 81,000 ha (48.3 percent of total physical targets) @ Rs .13/ha =Rs.1.05 M (0.5% of total cost) (B) Total cost of fuelwood and fodder plantation over 59,000 + 27,800 = 86,800 ha (5 1.7% of total physical targets) =Rs.217.3M (99.5% of total cost) It is apparent from the details given above that though the physical targets of farm forestry comprised 48.3 percent of total physical targets, the financial outlay under this component constituted a bare 0.5 percent of the total financial targets of forestry component. since farm forestry was dropped after MTR, the physical was achievement under farm forestry did not record progress. However, since farm forestry was a low-cost item, smaller physical achievements could never result in a proportionate reduction in financial costs. Similar was the case with fodder plantations on private land which also had a low unit cost (Rs. 108/ha). this item was dropped after MTR resulting in reduced physical achievement against SAR targets but without a proportionate reduction on the expenditure side. After MTR, the forestry component was drastically revised and the following plantation models were adopted: 1. On degraded forest 1600 plants/ha 2. On blank areas 2500 plants/ha 3. Near villages 1000 plants/ha Technology adopted for these plantation models included digging of contour furrows and raising of vegetative barriers along these contour furrows in addition to planting of seedlings in pits dug along contours. It may be mentioned here that in the pre-MTR period only planting of seedlings was to be carried out. The post-MTR models with improved technological 51 Annex A Subtitle Page Para. Comments package entailed higher unit costs as compared to pre- MTR models. From the details given above, it becomes quite clear that PCR mission's report has not been able to present a straight and truthful record of factual position. The report gives a totally biased assessment of forestry component through erroneously presented statistics insinuating that merely 37 percent of the physical targets were achieved after spending a whopping 73 percent of the estimated costs given in SAR. The cost over run in case of project administration was mainly due to revision of the pay scales of the employees. 6.11 The PCR mission has calculated the Economic Rate of Return (ERR) for forestry on the basis of actual areas planted and anticipated yields. The PCR mission report has attributed the low ERR of forestry to low yields per ha and high labor content. Both these premises are incorrect as is clear from the comments given against paras. 6.3, 6.4, and 6.10. 6.12 Strategy of the project, especially after MTR, has been to work simultaneously on different aspects of integrated watershed management. Fuelwood and fodder plantations were raised to increase their availability and horticultural activities were carried out to bring those cultivated private lands, which are not suitable for cultivation under permanent vegetation. On the other had, efforts were made to reduce pressure on the forests for fuelwood and fodder by providing alternate sources of energy (biogas), energy saving devices, cultivation of energy biomass on farmers' fields, maximum utilization of available fodder and agriculturable waste through supply of implements (like thrashers, chaff cutters, etc.) and encouragement for staff feeding through breed improvement program. The resultant of all these activities put together sold have been taken into account while making assessment of the impact on environment. The PCR mission report has failed to take into account the holistic Annex A 52 Subtitle Page Para. Comments approach of the integrated project. It has commented on the impact on environment by dwelling upon individual components taken as independent entities and on piecemeal basis. Project Sustainability 10 7.1 The plantains raised by the project were formally handed over to Gramsabhas along with a future management plan. 11 7.5 As mentioned earlier, the infrastructure (including nurseries) of the line department (Horticulture Dept.) still exists in the project areas. Hence the sustainability of the project supported horticulture sub-component has not been put in jeopardy. Project Documentation and Data 13 13.2 Monitoring activities included the monitoring and evaluation of physical and financial progress, but were not limited to this aspect only. regular collection of data with respect to crop yields, fodder and grass yields, number of cross-bred progeny produced production of energy biomass, etc. was carried out. 53 Annex A Comments on Part III of PCR Subtitle Page Comments 1 Related Bank Loans and Credits 14 Comments are not required. 2. Project Time Table 14 O.K. 3. Credit Disbursement 14 O.K. 4. Project Implementation 15 O.K. A. Physical Targets O.K. B. Yields (i) Incremental Forestry 15 The per ha yields given under different items have been grossly under-estimated-Details are given in the comments on Part I. (ii) Crops 15 O.K. 5. Project Costs and Financing A. Project Costs 16 The SAR costs as shown in the table are almost as per the SAR document. Actual costs are also shown in MUS$, hence comments cannot be given. Actual expenditure against Agriculture is 213 percent of the SAR estimate due to introduction of T & V. Actual expenditure on administration is 196 percent of the SAR estimate which is mainly due to revision of pay- scales of the employees. B. Project Financing 17 The figures are given in MUS$, hence comments cannot be given. 6. Project Results A. Direct Benefits 18 O.K. B. Economic Impact Economic rate of return for forestry has been estimated at 4.9 percent which is too low. The premises on which the ERR has been calculated are incorrect as mentioned in the comments given against para. 6.12 of Part I. 7. Status of Covenants 19 O.K. 8. Use of Bank Resources A. Staff Inputs (Staff Weeks) 19 No comments are possible. B. Missions 20 O.K. Appendices Appendix I No comments. Annex A 54 Subtitle Page Comments Appendix 2 No comments. Table I to 12 Appendix 3 Under pasture development, work done in 1992-93 (1113 ha) has not been taken into account. Hence, the total of pasture development work should be 3649 ha. Appendix 4 The yield assumptions for different species and other assumptions like percentage of different species are incorrect. The details are given in relevant paras. in Part I Appendix 5 O.K. Appendix 6 No comments are possible as project costs have Tables I and 2 been expressed in US$. Appendix 7 The economic analysis is based on inaccurate and unrealistic premises. Hence the ERR estimation is too low. Appendix 8 No comments needed. Note: Some tables and relevant appendices of Part III need to be revised in view of the details given in the comments of Part I, IBRD 16811R1 07k 810. 82 FEBRUARY 1985 Srinagar I N D I A HIMALAYAN WATERSHED MANAGEMENT PROJECT HIMACHAL PRADESH JAMMU and KASHMIR HIMAC PRADES Himachal Pradesh Physical Relef (meters): Project Sites Jammu . Uttar Pradesh 3000 Proiect Sites 500 32* Arterial Roads 200 0 Selected Towns 100 Batal and Villages A rMain Rivers State Boundaries Amr3 sar International Boundaries Jullundur a 0 100 200 Kilometers CHANDGARt. c N Abala 30' Saharanpur Mu ffarnagar - r7 NAR 3 HARYANA D Hi New D lhi hazia Pokhiro U T T A R Aligarh Shahiohanpur Kathrn n u S karMallhura P R A D E H JaprEtawah Lucknaw tihari RAJASTHANanr Gwalior 26°- -26° rra Kota ÄaramArn 780 800820284 -24- M A D H Y A P R A-'D E S H Sagar rh,s .ap h.s b.e PreP.red by The Kb,id Benk s st.# ~ vmu..y for the conwrw ~c or the readers and os excksi~l for the nternal us of The ~be Bank and the 0~5n ~1na Bho pal blF~nace Corporaion The denor-a~on usöd and IOm bo~nae ~hw an ts rmap do not Y jiJoa lpur Iply o the part of The ~ rd Ban and thö eh~nbl F~nce Corort.on an j~ @ n on the kef sttu of ayw ter~or or &ny ~ndrsmr accept~nc Or Such boundn~ 76° 78° 80° 82° 84°  긷：   IMAGING Report No: 15637 Type: PPAR