China - Second Irrigated Agriculture Intensification Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No: 33801-CN IMPLEMENTATION COMPLETION REPORT (SCL-43540 TF-25509) ON A LOAN IN THE AMOUNT OF US$300 MILLION TO THE PEOPLE'S REPUBLIC OF CHINA FOR THE IRRIGATED AGRICULTURE INTENSIFICATION II PROJECT December 21, 2005 Rural Development and Natural Resources Sector Unit East Asia and Pacific Region This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  CURRENCY EQUIVALENTS (Exchange Rate Effective June 30, 2005) Currency Unit = Yuan Y1.0 = US$ 0.12 US$ 1.0 = Y8.11 FISCAL YEAR January 1 - December 31 ABBREVIATIONS AND ACRONYMS ADP Agricultural Development Project CAD Comprehensive Agriculture Development CAS Country Assistance Strategy CATEC County Agrotechnical Extension Center COCAD County Offices of Comprehensive Agricultural Development ERR Economic Rate of Return ICB International Competitive Bidding ICR Implementation Completion Report IPM Integrated Pest Management MIS Management Information System MTR Mid-Term Review MOF Ministry of Finance MWR Ministry of Water Resources NCB National Competitive Bidding NDRC National Development and Reform Commission NPV Net Present Value O&M Operation & Maintenance PAD Project Appraisal Document PIP Project Implementation Plan PLG Project Leading Group PMO Project Management Office POCAD Provincial Office of Comprehensive Agricultural Development PPMO Provincial Project Management Office QAG Quality Assurance Group RMB Renminbi (Yuan) TM Task Manager SIDD Self-financing Irrigation and Drainage District SOCAD State Office of Comprehensive Agricultural Development WSC Water Supply Company WUA Water Users' Association Vice President: Mr. Jeffrey Gutman (Acting), EAPVP Country Director Mr. David Dollar, EACCF Sector Director Mr. Mark Wilson, EASRD Task Team Leader/Task Manager: Ms. Qun Li, EASRD  CHINA IRRIGATED AGRICULTURE INTENSIFICATION II PROJECT IMPLEMENTATION COMPLETION REPORT CONTENTS Page No. 1. Project Data 1 2. Principal Performance Ratings 1 3. Assessment of Development Objective and Design, and of Quality at Entry 2 4. Achievement of Objective and Outputs 3 5. Major Factors Affecting Implementation and Outcome 8 6. Sustainability 9 7. Bank and Borrower Performance 10 8. Lessons Learned 15 9. Partner Comments 16 10. Additional Information 17 Annex 1. Key Performance Indicators/Log Frame Matrix 18 Annex 2. Project Costs and Financing 22 Annex 3. Economic Costs and Benefits 25 Annex 4. Bank Inputs 31 Annex 5. Ratings for Achievement of Objectives/Outputs of Components 33 Annex 6. Ratings of Bank and Borrower Performance 34 Annex 7. List of Supporting Documents 35 Annex 8. Environmental Assessment 36 Annex 9. Borrower's Completion Report 38  Project ID: P049700 Project Name: Irrigated Agriculture Intensification II Project Team Leader: Qun Li TL Unit: EASRD ICR Type: Core ICR Report Date: December 21, 2005 1. Project Data Name: Irrigated Agriculture Intensification II Project L/C/TF Number: SCL-43540; TF-25509 Country/Department: CHINA Region: East Asia and Pacific Region Sector/subsector: Irrigation and drainage (40%); General agriculture, fishing and forestry sector (32%); Law and justice (14%); Roads and highways (7%); Power (7%) Theme: Other rural development (P); Water resource management (P); Other environment and natural resources management (P); Land administration and management (P); Rural services and infrastructure (S) KEY DATES Original Revised/Actual PCD: 08/19/1997 Effective: 09/15/1998 12/22/1998 Appraisal: 02/13/1998 MTR: 11/01/2000 01/01/2001 Approval: 06/18/1998 Closing: 12/31/2003 06/30/2005 Borrower/Implementing Agency: People's Republic of China/State Office of Comprehensive Agricultural Development Other Partners: STAFF Current At Appraisal Vice President: Jeffrey S. Gutman (Acting) Jean-Michel Severino Country Director: David R. Dollar Yukon Huang Sector Manager: Mark D. Wilson Geoffrey Fox Team Leader at ICR: Qun Li Richard Reidinger ICR Primary Author: Qun Li; Lang S. Tay 2. Principal Performance Ratings (HS=Highly Satisfactory, S=Satisfactory, U=Unsatisfactory, HL=Highly Likely, L=Likely, UN=Unlikely, HUN=Highly Unlikely, HU=Highly Unsatisfactory, H=High, SU=Substantial, M=Modest, N=Negligible) Outcome: S Sustainability: L Institutional Development Impact: SU Bank Performance: S Borrower Performance: S QAG (if available) ICR Quality at Entry: S Project at Risk at Any Time: No  3. Assessment of Development Objective and Design, and of Quality at Entry 3.1 Original Objective: 3.1.1 The objectives of the project were: (a) to increase agricultural production; (b) to increase farmers' incomes; and (c) to establish institutional mechanisms for the sustainable use, development and management of water and land resources in the Huang-Huai-Hai Plain. 3.1.2 The objectives were relevant and in line with the Bank's Country Assistance Strategy (CAS) for China (Report No. 16321-CHA) in supporting: (a) the continued development of sustainable techniques to upgrade marginal land; (b) rational and sustainable food security programs; (c) the reduction of poverty in some areas; and (d) the expansion, improvement and rehabilitation of existing irrigation and drainage facilities to increase agricultural production. The objectives were also consistent with China's Ninth Five Year Plan (1996-2001) to raise domestic grain output while also increasing cotton, oilseed and other crop production to meet the demand from an increasing population and higher per capita consumption. The project was also intended to contribute substantially to meeting the production targets of the Ninth Plan. 3.2 Revised Objective: 3.2.1 There was no revision of the project objectives throughout the project implementation period. 3.3 Original Components: 3.3.1 The components of the project are summarized below; the estimated costs at appraisal included physical and price contingencies. (a) Irrigation and Drainage Component (US$554.4 million). The component consisted of investments in: the construction of farmland irrigation water distribution facilities; the rehabilitation and improvement of irrigation and drainage pumping stations and power-driven tubewells; spreading of water-saving irrigation technologies; increasing the area of irrigated land and enhancing the reliability of irrigation and drainage standards; and the expansion of associated rural road and electric service networks. These improvements were to be implemented over an existing irrigation service area of about 1,534,700 ha. (b) Agricultural Intensification and Support Services Component (US$163.9 million). The component included investments in: the improvement of land and soil through land levelling and deep plowing, including support for increased use of organic and compound fertilizers, over 770,000 ha; the improvement and strengthening of the service delivery system for agricultural technology, agricultural mechanization, compound fertilizers, and improved seeds and crop varieties at the county and township levels; the dissemination and extension of both improved agricultural sciences and technology; training of farmers to enhance the levels of their scientific farming technologies and to promote Integrated Pest Management (IPM); and the supply of modern agricultural production materials, mainly chemical fertilizers, low-toxicity pesticides and plastic membranes for agriculture, as needed to increase agricultural input availability. (c) Afforestation, Environment Protection and Monitoring Component (US$46.9 million). The component consisted of investments in the establishment of forest-shelter networks for farmland to increase forest coverage by about 30 percent, mainly as shelter-belts around fields and along canals; and the establishment of networks of stations to monitor the groundwater, quality of surface water and soil fertility - 2 -  in the project area. (d) Institutional Development and Support Component (US$54.2 million. The component consisted of investments in: the support of institutional development through strengthening and development of a program approach in the State Office of Comprehensive Agricultural Development (SOCAD) and the Provincial Office of Comprehensive Agricultural Development (POCAD); the promotion and establishment of Self-financing Irrigation and Drainage Districts (SIDDs) and Water Users' Associations (WUAs); the development and establishment of a modern computerized management information system (MIS) for the project; and establishment of a national MIS and training center under SOCAD for the Comprehensive Agriculture Development (CAD) program. (e) Survey, Design and Management Component( US$29.9 million). The component was made up of investments in: engineering survey, investigation and designs for project works and facilities; the supervision of works and facilities under construction; buildings, offices, equipment and vehicles for project management. 3.4 Revised Components: 3.4.1 There was no major revision of the project components during project implementation. 3.5 Quality at Entry: 3.5.1 There was no Quality at Entry assessment or Quality Assurance Group (QAG) analysis during the course of the project. However, the quality at entry was considered satisfactory as the project was based on SOCAD's CAD program in the Huang-Huai-Hai Plain and the successful results of past Bank-financed agricultural development projects (i.e., the first Irrigated Agriculture Intensification Project (IAI) as well as the North China Plain Project, the Shandong Agricultural Development Project (ADP), the Hebei ADP, the Henan ADP, and the Anhui Pi-Shi-Hang project). Quality at entry was further ensured with feasibility reports prepared for all sub-project by the County Offices of Comprehensive Agricultural Development (COCADs). These sub-project feasibility reports were reviewed and approved by the POCADs and by the SOCAD. During the project appraisal, the Bank reviewed and spot-checked about twenty percent of the county feasibility reports to ensure soundness and adequacy of preparation. 4. Achievement of Objective and Outputs 4.1 Outcome/achievement of objective: 4.1.1 All project components (except SOCAD's National Training Center) were completed by December 31, 2003 as scheduled, with all the main physical targets having been met. The Loan's Closing Date was extended by eighteen months to enable completion of the Training Center, which was delayed for reasons explained below (see paragraph 4.2.5). The primary objective of increasing agricultural production and incomes through more efficient and sustainable use of the land and water resources exceeded the expectations stated in the Project Appraisal Document (PAD). Irrigated farmland improved under the project totaled some 1.53 million ha (100 percent of the PAD estimates), of which some 672,100 ha (137 percent of the PAD estimates) were provided with water-saving irrigation facilities. Although the surface water irrigation area was reduced by about 3,900 ha, groundwater irrigation increased by about 41,000 ha (from 360,520 ha to 401,550 ha -111 percent of PAD estimates) based on a more efficient use of available water resources. Overall irrigation system efficiency improved from 0.68 to 0.74 (108 percent of PAD projections). Construction of canal linings, better canal control structures and the use of sub-surface PVC - 3 -  pipes for irrigation were completed to a high quality and have contributed to this good performance. 4.1.2 The outcome of the agricultural intensification and support services has been fully achieved. Increased production of quality seeds, together with land levelling, farm mechanization and input supply, has increased total crop production from about 16.3 to 22.4 million tons (137.4 percent of the PAD target), despite a decline in total food grain production of about 11 percent. This decline was caused by falling grain prices after the government withdrew the grain subsidy and freed the grain market, on accession to the World Trade Organization (WTO), and farmers shifted to higher-value crops. However, food grain production was sustained and increased from 9.3 to 11.2 million tons through higher yields achieved by the increased use of quality seeds, farm mechanization and production inputs when cropping patterns were appropriately adjusted to reflect market demand. Commodity cash crop planting was increased from 557,200 ha to 908,600 ha with total production increasing from 3.8 to about 11.2 million tons (295 percent of PAD target). 4.1.3 The net financial return to beneficiaries was higher than that estimated at appraisal. As a result of the improved irrigation facilities and agricultural production conditions and services, farmers' incomes in the project area have increased substantially. The average per capita annual income has increased from Y714 to Y1,171 for low-income households (115 percent of the PAD estimate and 168 percent of the "without-project" scenario), from Y1,017 to Y1,497 for medium-income households (111 percent of the PAD estimate, and 147 percent of the "without-project" scenario), and from Y1,237 to Y1,769 for high-income households (109 percent of the PAD estimate and 138 percent of the "without-project" scenario). Farmers' incomes are expected to increase further as the project reaches full development. 4.1.4 Although poverty reduction was not one of the project's stated objectives, the project has reduced poverty levels in some 19 State-designated poverty counties (with average per capita incomes of less than Y637) and 17 province-designated poverty counties in the project area. Located in Hebei, Henan and Anhui, a total of Y1,278 million was invested in these poverty counties in irrigation and drainage facilities, land and soil improvement, agricultural production technologies and services, quality seed production, rural roads and electricity, and SIDD/WUA development. As a result, these poor counties benefited from increased agricultural productivity and higher standards of living. Compared to pre-project levels, the average net per capita income of project-area farmers increased substantially (in Anhui by 60 percent, in Hebei by 92 percent , in Henan by 61 percent, and in Jiangsu by 103 percent) and some formerly poor households have reached medium income levels. 4.1.5 Institutional strengthening and development has achieved substantial results. The project staff under SOCAD, POCADs and COCADs in a total of 133 counties gained considerable experience in project preparation, implementation and management, especially on supervision, procurement and the use of computerized MIS for project accounting and financial reporting, and monitoring and evaluation. A major objective of SOCAD in undertaking the project was to improve its internal management capabilities, and the experience under the project will certainly further strengthen SOCAD's national CAD program in the five provinces and also throughout the country. Development of SIDDs (both for Water Supply Companies (WSCs) and WUAs), introduced on a pilot basis under the project, substantially exceeded the physical targets; eleven WSCs were established (compared to eight planned) and 516 WUAs were established (compared to 181 planned). The completed SIDD/WUA program covers about nine percent (136,500 ha) of the total project area of 1.53 million ha, 365,550 households and a farming population of 1.43 million. The pilot SIDD program under the project has demonstrated substantial benefits, such as improved irrigation performance, better maintenance of irrigation facilities, water saving, reduced water conflict, and improved payment of water charges, and it has provided valuable experience and lessons for expanding and promoting similar development in the project area, in future similar comprehensive - 4 -  agricultural development projects and in the national CAD program. 4.1.6 The overall Economic Rate of Return (ERR) for the project at completion is estimated at 30 percent and is highly satisfactory compared to 24 percent estimated at appraisal (Section 4.3 and Annex 3). Overall, the project achieved its main objectives, reached or exceeded its expected outputs and outcomes, and developed and demonstrated an economically-viable and environmentally-sustainable integrated comprehensive irrigated agricultural development model (see section 4.3 and 10) which is already being replicated through the new project (Irrigated Agriculture Intensification III) and the national CAD program. The achievement of objectives is therefore rated as satisfactory. 4.2 Outputs by components: 4.2.1 The key outputs by components are described below. Outputs achieved are compared to the physical targets as revised during the mid-term review (MTR) and, where appropriate, to appraisal targets. Revised targets at MTR reflected changes mainly in investment policy for the irrigation component (e.g., in favor of water savings, as well as adaptation to local conditions and requirements). It should be noted that the project was designed as a "program approach" project and such changes in components were anticipated in the project design because of the nature of the project, with many small-scale works and activities scattered across some 1,100 sites in 133 counties. Flexibility to adapt to changing or unexpected conditions was built into the project design. Detailed outputs by each component are presented in Annex 1. Irrigation and Drainage Component: 4.2.2 Overall, the component achieved 105 percent of the revised physical targets agreed at the MTR. Although a few categories of work were slightly less than 100 percent (e.g., inverted siphons at 93 percent), all other categories of work exceeded their MTR targets, with tubewells at 115 percent mainly because of the increase in sub-surface pipeline length for tubewells which was expanded as a water-saving measure. For all categories of work under this component, expenditures greatly exceeded 100 percent of the appraisal targets. Comparison with appraisal targets is not meaningful in terms of project achievements because a number of the categories were vastly expanded as a result of policy change to emphasize water savings (e.g., the category of sub-surface pipeline was expanded by about 46 percent, from 16,623 km at appraisal to 24,333 km at completion). Agriculture Intensification and Supporting Services Component: 4.2.3 Overall, the component achieved more than 100 percent of the revised physical targets agreed at the MTR. On quality seeds, production achieved 103 percent (219,000 tons compared to MTR target of 212,016 tons) and quality seed coverage has reached 97.2 percent of the project area. Production of blended fertilizer from new plants reached 394,270 tons (105 percent of the MTR target). On agricultural mechanization, 1094 agri-machinery stations were established (140 percent of the MTR target) and the overall degree of farm mechanization reached 72.7 percent compared to 69.3 percent projected at the MTR. Pest control and forecast now cover 96 percent of the project area (235,930 ha). On extension, 709 township agri-technology extension stations were established or improved with 39,272 new extension staff added. This enabled the training of technical farmers for about 54,000 man-months and of farmers for 100,665 man-months, and establishment of 205,100 ha of demonstration/pilot areas. Afforestation, Environmental Protection and Monitoring Component: 4.2.4 Overall, the component achieved 106 percent of the revised physical MTR targets and 143 percent - 5 -  of the appraisal targets. Key outputs were the establishment of 3,873 ha of new nurseries and 87,718 ha of forest shelter belts. As a result, the vegetation cover in the project area increased from 11.6 percent at appraisal to 17.2 percent at completion, contributing to a more favorable ecological environment. Institutional Strengthening and Development Component: 4.2.5 Overall the component achieved 110 percent of the MTR and 141 percent of the appraisal targets. All components were close to or exceeded the MTR targets except the SOCAD National Training Center. It was delayed due to the design changes for unforeseen works for bank protection during the construction of foundations, the adverse weather conditions and the consequent need for additional work. Compared to the expected targets at the MTR, the lowest target achieved was that of staff study tours (85.4 percent) reflecting SOCAD's strict control to ensure that international study tours were most useful, and research and studies (94 percent) for which a number of research activities were considered duplicative. All studies and technical assistance provided under the project were fully completed, achieving good results in strengthening the management and technical capabilities of SOCAD and lower-level CAD offices and providing technical guidance to field staff and farmers. 4.2.6 The main output was in the establishment of WUAs which achieved 221 percent of the MTR target and 285 percent of the appraisal targets (increasing from 181 WUAs planned at appraisal to 233 planned at the MTR to 516 actually achieved); pilot WSCs achieved 100 percent of the MTR target. The increase in the number of WUAs was the result of greater emphasis on water savings, to which improved on-farm management by WUAs contributed substantially, and on increasing the capability and confidence of the various CAD offices in WUA development, which was introduced for the first time in the CAD system under the project. The pilot SIDDs (WSCs and WUAs together) now cover about nine percent (136,500 ha) of the total project area and serve 365,550 households with a total population of 1.43 million. The completed SIDDs and WUAs will serve as useful models to expand WUA development and better, more participatory water management in the project area as well in the national CAD program and future similar projects. 4.2.7 Training of project staff focused on local training and study tours to meet the urgent needs of project staff. At completion, training and study tours for staff totaled 40,717 man-months of domestic training, 8 man-months of international training, 4,915 man-months of domestic study tours and 123 man-months of overseas study tours covering 33 topics and in 22 countries (reflecting the technical breadth and size of the project). Technical assistance totaled 197 man-months mainly for domestic consultancy services. Some 120 applied research activities and studies were carried out by qualified national and provincial institutions in the five project provinces and by SOCAD. The applied research and studies covered irrigation and drainage (35 topics), agriculture (37), forestry (14), 34 others covering environment, SIDD and project management. The research and study results are relevant and have been disseminated in the agricultural production areas. 4.2.8 A major output of the project for SOCAD was to upgrade its own programs and staff capabilities, and the staff training, technical assistance and research/studies have been fully used and achieved good results. The overall impact of project institutional strengthening and development component is rated as substantial. Survey, Design and Management Component: 4.2.9 The component achieved 98 percent of its MTR and 106 percent of its appraisal targets. At the MTR, the number of vehicles was substantially reduced and the number of computers for the MIS was - 6 -  increased. The computerized MIS with all sites linked through the internet played an important part in improving overall management of the project as well as at individual sites and was an important modern management tool introduced into SOCAD under the project. 4.2.10 Overall, the main outputs of project components were completed satisfactorily by the original December 31, 2003 Closing Date, except for the SOCAD National MIS/Training Center, delays in the construction of which necessitated the extension of the Closing Date to June 30, 2005. Similarly, safety and remedial measures and up-grading for three dams (Fuzhiling and Longhekou in Anhui Province, and Xiaotashan in Jiangsu Province) which were required by the legal covenants but not financed under the project, were delayed but were eventually completed by June 30, 2005. The delay in completion of the center and the upgrading of the dams did not affect project performance in achieving agricultural production and farmer income objectives. 4.3 Net Present Value/Economic rate of return: 4.3.1 Following the methodology used during the appraisal, the net present value (NPV) and the ERR were re-estimated for each sub- project area in Hebei, Jiangsu, Anhui, Shandong and Henan province, and for the project as a whole. The ERR for the project is estimated at 30 percent (NPV: Y7,953 million) compared to 23.9 percent estimated at appraisal. The principal reasons for this better-than-expected performance include: (a) an increase in irrigated cropping areas by about 9 percent; (b) the farmers' adjustment of their cropping patterns enabled by project investments and consequent increase in the total cash-crop area by 151 percent (from 603,442 ha estimated at appraisal to 908,603 ha at ICR); and (c) substantial increases in crop yields and output (e.g., the average yield of food field crops - including rice, wheat and corn - increased by about 28 percent from 4,655 kg/ha to 5,976 kg/ha in most sub-project areas) and the output value of the main cash crops increased by three times compared to the appraisal values. However, the actual project investment costs were very close to the estimated costs at appraisal (US$849.29 million at appraisal compared to US$867.3 million at ICR). Returns for all components have therefore far exceeded appraisal expectations (see Annex 3 for more details by province). 4.4 Financial rate of return: 4.4.1 Farm-level analysis was carried out to indicate the project's impact on beneficiary farmers' incomes. The farm model analysis shows a substantial increase in beneficiary farmers' incomes as a result of the comprehensive and integrated improvement in irrigation and drainage facilities, crop production and agricultural support services. In line with the appraisal approach, representative farm households were selected in five subproject areas for the ICR. The following assumptions have been considered to calculate and assess the financial impacts of the project on household incomes: (a) average farm holding size in each subproject area, combined with the actual cropping intensity in each area due to the change from rainfed and partially irrigated land to fully irrigated land; (b) the average 2003 financial farm gate prices prevailing in the project area were used to value outputs and inputs; and (c) from 1998 to 2004 the actual cropping pattern, intensity and the per ha financial crop budgets (for "with-project" and "without-project" analysis) were applied for annual and perennial crop production in each sub-project area, with increased crop yields derived from the development of irrigation and drainage systems and relevant agricultural technology. More details are included in Annex 3. 4.4.2 Increased crop outputs and changed cropping patterns have greatly increased farmers' incomes, especially for low-income farmers. Project monitoring indicators show that the per capita income of low-income, medium-income and high-income farm households in the project areas reached Y1,170.98, Y1,497.29 and Y1,769.54, respectively, in 2003. This represented an increase of 68 percent for - 7 -  low-income farmers, 47 percent for medium-income farmers, and 39 perent for high-income farmers, compared to appraisal estimates. 4.5 Institutional development impact: 4.5.1 Institutional strengthening and development included in the project have achieved good results and impact, in terms of both project implementation and strengthening of CAD capabilities, and especially through the introduction and demonstration of farmer-managed WUAs (see also Section 4.2). Exposure and participation of project staff with no prior experience with Bank-financed projects provided the rare opportunity of acquiring valuable project implementation and management skills and capabilities building. Project staff at all levels benefited from training and study tours in various aspects of project management, including planning, supervision, monitoring and evaluation, financial management, and computerized applications, as well as exposure to innovations including WUAs and farmer associations and cooperatives as applied in other countries. Through contact with the Bank Group experts and consultants and Bank project management, SOCAD and lower-level CAD staff were able to acquire new methodologies and technologies to enhance their technical capability, especially with the computerized MIS. The SIDD program, introduced as a pilot project for the first time under SOCAD's CAD program, was successful and laid the foundation for democratic water management and for farmers eventually to take over the operation and maintenance of the tertiary irrigation facilities themselves through the WUAs established. Overall, the institutional development impact of the project is rated as substantial. 5. Major Factors Affecting Implementation and Outcome 5.1 Factors outside the control of government or implementing agency: 5.1.1 Only one major factor outside the control of the government or implementation agency affected project implementation. Bad weather affected project implementation to some extent. Severe droughts occurred in Henan and Shandong, and the Huai River floods in Anhui in 2003 slowed down project implementation as local governments had to divert financial resources to deal with the impact of natural disasters. In addition, the heavy rains and floods also affected the construction works for the SOCAD National MIS/Training Center and delayed its completion for about six months. 5.2 Factors generally subject to government control: 5.2.1 One of the most positive factors in project success was the firm commitment by central, provincial and local governments, including the city and county governments, to implement the project. Participation of the local government leadership and staff was active and effective, paving the way for smooth project implementation. At start-up, government establishment of the Project Leading Groups (PLGs) and Project Management Offices (PMOs) at all levels was rapid and well coordinated. However, government policies to free grain markets and withdraw the production subsidies had some negative impacts on the project as falling grain prices caused the grain cultivated area to decline and adversely affected farmers' incomes. Other negative factors affecting project implementation included the delay in reviewing and approving detailed individual project feasibility reports, and reduced budget allocations and funding constraints faced by some project counties. 5.3 Factors generally subject to implementing agency control: 5.3.1 The project was fortunate to have an effective SOCAD and a POCAD in each of the five provinces in leading project implementation. SOCAD and POCADs, with their existing staff, were able to launch the - 8 -  project immediately after appraisal by organizing PMOs at the city and county levels. Many of the city and county PMOs were new to Bank-financed projects and required substantial training on project management and procedures, especially for procurement and reimbursement. The POCADs organized special training for them in procurement, disbursement, project supervision and financial management. Written manuals were also issued to project staff for their guidance. After an initial learning process, all city and county PMOs were able to operate efficiently, and project supervision and management at the county level was progressively improved. 5.3.2 SOCAD, located in Beijing, was able to keep frequent touch with the Task Manager (TM) who for most of the project implementation period was located in the Bank's Beijing office. This close coordination and continuous dialogue had a substantial positive impact on the project, helping to resolve implementation problems and issues at all levels rapidly, and also enabled the Bank to monitor closely the project progress and problems before they became severe. With 131 project counties and many hundreds of sub-projects, difficulties in maintaining supervision and construction quality and engineering standards were encountered, especially for works carried out by township and village construction teams, but these were resolved rapidly with the proximity of the Bank's TM to the field. Project expenditures using the MIS were prepared diligently by each Provincial PMO and sent to SOCAD for processing before being submitted to the Bank for disbursement. The timely disbursement of project funds greatly helped the cash flow of the project entities which had made payments in advance for works completed; project costs and expenditures were closely monitored and rigorously controlled by SOCAD and POCADs resulting in very low cost overruns. 5.4 Costs and financing: 5.4.1 Total project cost was estimated at US$849.3 million (Y7,083.6 million) at appraisal, US$854.8 million (Y7,083.6 million) at MTR in 2001, and US$867.3 million (Y7,181.7 million) at project completion. The increase of total project cost at completion was about US$18 million or 2.1 percent over the appraisal estimate. This marginal increase was primarily due to currency exchange differences, increases in labor and materials cost, and enhanced work quantities and engineering standards (see Annex 2 for a summary of costs for each component at appraisal, MTR and completion). At completion, Provincial and local governments had financed about 34 percent of total project costs, farmers had contributed 32 percent, and the Bank loan had provided 34 percent. This was only slightly different from appraisal estimates (see Annex 2 for a summary of project financing). 6. Sustainability 6.1 Rationale for sustainability rating: 6.1.1 Sustainability of the project is rated as likely. (a) The project areas in the five provinces support a large rural population and are intensively cultivated for annual grain and commodity crops. Beneficiary farmers contributed labor and funds to construct the facilities, especially the tertiary and on-farm irrigation and drainage facilities upon which their livelihoods depend. They feel "ownership" for the facilities and are likely to take care of them and use them effectively. (b) The existing government agencies involved (agriculture, irrigation and forestry) took over responsibility for project facilities for operation and maintenance as soon as they were completed. Their continued operation and maintenance is assured because the agencies are part of existing local government - 9 -  entities which provide basic services to farmers. Where farmer WUAs were established, they have taken over responsibility for operation and maintenance of the facilities, and in non-WUA areas local governments are providing satisfactory annual budgets to meet the annual operation and maintenance costs. (c) Being a major agricultural production base, the project area has strong policy and financial support from the central government which will ensure that the project can continue to achieve good results and sustain its operations on a long-term basis. (d) SOCAD has previously undertaken successfully major comprehensive agricultural development programs in the five project provinces and will continue to extend its program in a large part of the remaining area under their jurisdiction. SOCAD's commitment and continued presence in the five provinces would help ensure that completed project facilities would continue to be sustainable. 6.2 Transition arrangement to regular operations: 6.2.1 There was no need for transition arrangements to regular operation. Operation and maintenance responsibilities for the project facilities were pre-determined and defined before appraisal. Upon completion, facilities at each sub-project were handed over to the concerned local units at county and township level for operation and maintenance, or to the farmer WUA established under the project, as appropriate. 7. Bank and Borrower Performance Bank 7.1 Lending: 7.1.1 The Bank's performance for lending is rated as satisfactory. The Bank responded promptly to the government's request for assistance. The project was first proposed to the Bank in late 1996 and an identification mission was sent to China in February 1997. The Borrower's needs were assessed, the project scope and content were formulated and agreed with SOCAD, and the project was prepared and processed on a fast-track basis as requested by the government. Identification was followed by the Bank preparation and pre-appraisal missions in September and November 1997; the project was appraised in February 1998 and the loan was approved in June 1998. The whole process, from project identification to approval, took about 16 months. The relatively rapid processing of the loan was facilitated by the Borrower's thorough project preparation, close cooperation with the Bank and timely provision of project data and information, and prompt action on Bank recommendations. In order to facilitate early project start-up during the idle period for farming activities, the Bank made provision for US$15 million for retroactive financing for small works construction through beneficiary/farmer participation during the 1997/98 season. 7.2 Supervision: 7.2.1 The Bank's performance for supervision is also rated as satisfactory. The project was supervised at roughly six-months intervals and a total of ten times over the implementation period. Bank supervision missions usually lasted two to three weeks each, covered several project sites in each of the five provinces, and generally was preceded by a SOCAD/POCAD internal supervision mission to each province whose findings were provided to the Bank for information. Appropriate technical experts were included in each Bank mission (see Annex 4), and Bank's supervision mission findings and recommendations were highly appreciated by SOCAD and the POCADs, especially on procurement, disbursement, project monitoring - 10 -  and specific technical issues. Mission Aide-Memoires were provided to SOCAD after each mission and were disseminated promptly by SOCAD to the POCADs for follow-up action. During 2000/2001, Bank missions assisted SOCAD in carrying out the MTR of overall project progress, expenditures and costs and in deciding on the adjustments needed. The review resulted in some adjustments in costs between the categories of works and of goods, an increased disbursement percentage for works under Category 1 (a) from 20 percent to 45 percent, and the revision of targets for a number of sub-components based on project experience to that date. 7.3 Overall Bank performance: 7.3.1 The Bank's overall performance is rated as satisfactory. The Bank responded promptly to the Borrower's requests and adequately prepared and appraised the project on a fast-track basis. It provided close and effective supervision and maintained close contact and interaction with the Borrower. Taking into consideration the various implementation constraints, the Bank approved revisions promptly and provided the appropriate extensions of the Loan Closing Date so that the project could be completed as planned. Borrower 7.4 Preparation: 7.4.1 The Borrower's performance in project preparation is rated as satisfactory. SOCAD has gained long experience in project preparation under its CAD programs throughout China. After agreeing with the Bank on the project's scope and content, and on the five provinces that will participate in the project, SOCAD together with the five POCADs prepared the project feasibility reports in adequate detail for review by Bank missions. SOCAD and the POCADs also jointly prepared the Project Implementation Plan (PIP) and the various specific subject reports according to Bank recommendations and requirements. Preparation by the Borrower was considered adequate and of satisfactory quality. 7.5 Government implementation performance: 7.5.1 The Government implementation performance is rated as satisfactory. Following project appraisal, each province established Project Leading Groups (PLG) at province, city and county levels. The Provincial PLG was headed by a Vice-Governor. Project implementation responsibility was assigned to the respective COCADs. The key role played by the provincial governments was in providing policy and technical guidance to the implementing units, effective inter-agency coordination, adequate and timely counterpart fund allocations, and prompt approval of staff for the PMOs. The central government implementation role was mainly delegated to SOCAD to monitor and supervise project implementation by POCADs and local governments, operate the loan Special Account for the project, handle international procurement and provide various required reports to the Bank. 7.6 Implementing Agency: 7.6.1 Project Organization and Management. PMOs were rapidly established at central, provincial, city and county levels. Project staff, mostly transferred or seconded from existing government offices and agencies, were effective and capable after the initial orientation and training. These PMOs were adequately maintained throughout project implementation. Other line agencies, especially the Water Resources, Agriculture and Forestry Bureaus, provided the necessary technical support, especially at the county level. The provincial, city and county PLGs were established accordingly, but met on an ad-hoc basis. These PLGs also provided some coordination but in some cases could have played a stronger leadership role and - 11 -  performed better. 7.6.2 Physical Construction. Physical construction of small-scale sub-project works and facilities, mainly tertiary canals and drains, land levelling, and tree planting, was mainly carried out by project beneficiaries through community participation during the idle season for farming activities (November to March). Larger works, such as canal ancillary structures, tubewells, rural roads, bridges and culverts, and small buildings were carried out by village and township contractors through price quotations or National Competitive Bidding (NCB) procedures managed by the respective county PMOs. Construction of sub-projects started during the first or the second year of project implementation, and over fifty percent of the sub-projects were substantially completed by the end of third year; Jiangsu and Shandong Provinces in particular progressed faster, particularly on the irrigation and drainage component. Construction of facilities for agricultural support services was comparatively slower because of the geographical dispersion and small size of many of the facilities. Project construction works were virtually completed by the end of the fifth year as scheduled, except for the SOCAD National MIS/Training Center. Construction quality has generally been satisfactory, but varied among the five provinces and also among the counties in each province. Bank and SOCAD/POCAD supervision missions helped to ensure good construction quality and standards as well as progress. 7.6.3 Procurement and Disbursements. Procurement under the project was in accordance with Bank guidelines and procedures and was carried out under three categories