Page 1 Document of The World Bank Report No: 25701 IMPLEMENTATION COMPLETION REPORT ( IDA-26160 ) ON A CREDIT IN THE AMOUNT OF SDR 106.3 MILLION (US$ 150 MILLION EQUIVALENT) TO THE PEOPLE'S REPUBLIC OF CHINA FOR A LOESS PLATEAU WATERSHED REHABILITATION PROJECT April 29, 2003 Rural Development and Natural Resources Sector Unit East Asia and Pacific Region Page 2 CURRENCY EQUIVALENTS (Exchange Rate Effective Sept. 2002 ) Currency Unit = Renminbi (RMB) Yuan (Y) Y 1.0 = US$ 0.12 US$ 1.0 = Y 8.3 FISCAL YEAR January 1 December 31 ABBREVIATIONS AND ACRONYMS ERR Economic Rate of Return FAO/CP FAO and World Bank Cooperative Program GIS Geographical Information System ICR Implementation Completion Report ICB International Competitive Bidding ha Hectare hp Horse power IDA International Development Association km Kilometer LIB Limited International Bidding m Meter MIS Management Information System MWR Ministry of Water Resources N.B.F. Non-Bank Financed NCB National Competitive Bidding NGO Non-Governmental Organization O&M Operation & Maintenance PMO Project Management Office PSR Project Status Report QAG Quality Assurance Group R&D Research & Development SAR Staff Appraisal Report SDR Special Drawing Right sq Square YRCC Yellow River Conservancy Commission Vice President: Jemal-ud-din Kassum, EAPVP Country Manager/Director: Yukon Huang, EACCF Sector Manager/Director: Mark D. Wilson, EASRD Task Team Leader/Task Manager: J. M. Voegele, EASRD Page 3 CHINA LOESS PLATEAU WATERSHED REHABILITATION PROJECT 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 12 9. Partner Comments 13 10. Additional Information 14 Annex 1. Key Performance Indicators/Log Frame Matrix 17 Annex 2. Project Costs and Financing 18 Annex 3. Economic Costs and Benefits 20 Annex 4. Bank Inputs 25 Annex 5. Ratings for Achievement of Objectives/Outputs of Components 27 Annex 6. Ratings of Bank and Borrower Performance 28 Annex 7. List of Supporting Documents 29 Annex 8. Pictorial Overview of Loess Plateau and Project Achievements 30 Page 4 Project ID: P003540 Project Name: Loess Plateau Watershed Rehabilitation Project Team Leader: Juergen Voegele TL Unit: EASRD ICR Type: Core ICR Report Date: April 28, 2003 1. Project Data Name: Loess Plateau Watershed Rehabilitation Project L/C/TF Number: IDA-26160 Country / Department : CHINA Region: East Asia and Pacific Region Sector/subsector : Central Government Administration 5%, General Water 50%, General Agr.45% KEY DATES Original Revised/Actual PCD: 07/05/1991 Effective: 10/03/1994 10/03/1994 Appraisal: 10/06/1993 MTR: 11/22/1997 11/22/1997 Approval: 05/26/1994 Closing: 12/31/2002 12/31/2002 Borrower/Implementing Agency : PRC/MININSTRY OF WATER RESOURCES Other Partners: Shanxi, Shaanxi and Gansu Province, and Inner Mongolia Autonomous Region STAFF Current At Appraisal Vice President: Jemal-ud-din Kassum Gautam S. Kaji Country Manager : Yukon Huang Nicholas Hope Sector Manager: Mark D. Wilson Joseph R. Goldberg Team Leader at ICR: Juergen Voegele Juergen Voegele ICR Primary Author: Xueming Liu; FAO 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: HS Sustainability: HL Institutional Development Impact: H Bank Performance: HS Borrower Performance: HS QAG (if available) ICR Quality at Entry: S Project at Risk at Any Time: No Page 5 3. Assessment of Development Objective and Design, and of Quality at Entry 3.1 Original Objective: 3.1.1 The primary objective of the project was to increase agricultural production and incomes on 15,600 sq. kilometers of land in nine tributary watersheds of the Yellow River in the Loess Plateau. A secondary objective was to reduce sediment inflows to the Yellow River by locating the project areas in parts of the watersheds that had suffered severe soil erosion. The project was designed to: (i) create sustainable crop production on high-yielding level farmland thereby replacing the areas devoted to crops on sloping lands; (ii) plant sloping lands to a range of trees, shrub and grasses for land stabilization and production of fuel-wood, timber for construction and fodder; and (iii) substantially reduce sediment runoff from slope lands and gullies. 3.1.2 The project design attempted to break the cycle of poverty and environmental degradation by supporting sustainable agricultural practices and wealth creation. The objectives were clear, well designed, consistent with the Borrower's and World Bank Group's overall strategies at the time of preparation. The problem of erosion in the Loess Plateau was well recognized and there was a strong consensus among all stakeholders about the project objectives (see Section 3.5). 3.2 Revised Objective: 3.2.1 The original project objectives remained unchanged. 3.3 Original Components: 3.3.1 The project has two main components: (1) Land Development and Erosion Control Works; and (2) Institutional Development. The first component was further divided into the following sub-components: (a) Terracing, (b) Afforestation, (c) Orchards, (d) Grasslands, (e) Sediment Control Dams, (f) Warping Land, (g) Irrigation; and (h) Gully control; while the second component was divided into (a) Training, (b) Research, and (c) Technology Transfer. 3.3.2 These components were well designed to achieve the project objectives. They took into account lessons learned from other related projects, integrated local experience, and were based on extensive trials. A participatory approach combining outside expertise and experiences with local knowledge was an integral element in the project design (see Section 3.5). 3.4 Revised Components: 3.4.1 The original components remained largely unchanged throughout the implementation period. However, the Staff Appraisal Report (SAR) targets of dam construction, shrub plantation and grassland development were adjusted and complementary livestock development activities were added at the mid-term review to reflect evolving priorities and better use of funds (see Section 4.2). 3.5 Quality at Entry: 3.5.1 Quality at Entry is rated satisfactory. The project objectives were consistent with the Bank’s CAS in China, which calls for sustainable agricultural development in impoverished and environmentally vulnerable areas and for diversified agricultural production; and with the Government's long term policies which focus on increased production, income generation and employment creation in rural areas. Project design incorporated valuable experiences and lessons learned from other Bank-financed watershed development projects and previous experience in the Loess Plateau area. The integrated watershed planning approach adopted by the project provided a sound basis for the project design and implementation—it combined a well designed participatory planning and implementation methodology with strong technical support and a sound monitoring and evaluation system. It also fully addressed the Bank's safeguard policies in project design. - 2 - Page 6 3.5.2 However, the project could have, in hindsight, placed more emphasis on allocating costs to households, local governments and the central government for the sediment retention structures. While the SAR clearly stated the significant downstream benefits and benefits for the society as a whole, the Central Government did not agree to bear a greater share of the costs for that sub-component of the Project. 4. Achievement of Objective and Outputs 4.1 Outcome/achievement of objective: 4.1.1 The outcome of the project is rated highly satisfactory. 4.1.2 The project was completed slightly ahead of the schedule envisaged in the SAR with all the main physical targets having been met. Large-scale investments into terracing and irrigation, afforestation, grasslands, livestock and animal husbandry, and dams and silt retention structure were completed to a high quality. The primary objective of increasing agricultural production and incomes through more efficient and sustainable use of the land and water resources exceeded SAR expectations. Likewise, the secondary objective of reducing sediment inflows into the Yellow River was substantially met (see Annex 1). At completion, the project showed itself to be one of the largest and most successful erosion control programs in the world, bringing unprecedented benefits to households in the project area and for those downstream. The project has developed and demonstrated an economically viable and environmentally sustainable integrated watershed management model (see Sections 4.3 and 6.1), which is already being replicated throughout the Loess Plateau. 4.1.3 Over one million farmers in the project area have directly benefited from the project with raised annual grain output from 427,000 tons to 698,600 tons, and fruit production from 80,000 tons to 345,000 tons, and farmers’ incomes per capita increased from Y 360 to Y 1,263. The annual sediment inflow to the Yellow River has been reduced by 57 million tons. In addition, production on terraces has significantly reduced labor inputs and together with road access to the terraced fields has allowed for mechanization. These labor savings have permitted many farmers to pursue new income earning activities such as off-farm jobs and livestock production. Increased income and the prospect of sustainable agriculture have allowed farmers to invest in a wide range of enterprises and social programs, which have benefited the communities as a whole. 4.1.4 The project has benefited from the favorable environment facing agriculture and contributed to the profound changes that have recently taken place in agriculture throughout China. In particular, land tenure reforms created strong incentives for investment in land improvement; while improved communications have removed regional marketing constraints and provided farmers with countrywide market channels. The availability of new technology and management practices has further expanded production options and raised yields. 4.2 Outputs by components: 4.2.1 Land Development Component ( US$212.51 million). The component as a whole is rated highly satisfactory with individual sub-components evaluated below: Terracing (US$96.75 million). Terracing was the cornerstone of the project and was a precondition for sustainable farming practices. The conversion of slopelands to level fields virtually eliminated erosion and helped maintain precious moisture in the soils, allowing higher crop yields and more diversified cropping patterns. Some 90,446 ha of terraces were constructed, realizing 108 percent of the planned - 3 - Page 7 target. Most of the terraces were constructed by local bulldozer operators, with some manual labour for shaping the riser and terrace lips. Roads were built by the bulldozer operators to reach the areas terraced. The standard of terrace construction was very high. The benefits of the terraces were highlighted in a dramatic way during the droughts which affected the area in 1999, 2000 and the first half of 2001. On treated slopelands farmers obtained good yields while non-treated lands suffered complete crop failures (See Additional Annex 8 Pictures 5 to 12). Afforestation and Orchards (US$72.21 million) . This component has established some 290,000 ha of arbor, shrub and economic trees. So called ‘arbor trees’, such as black locust, Chinese pine, spruce and poplar, have been established on wastelands. Land preparation was usually carried out by farm families employed and supervised by specialized teams, organized at township work stations, and trained by the county forestry specialists. In most areas, specialized teams did the planting. Severe droughts have caused low survival rates requiring replanting, in some cases several times. The arrangements for taking up contracts on the arbor tree plantations by households varied from county to county. In some counties the land was contracted before plantation and the farmers were involved in tree establishment. In others the PMOs undertook planting and also kept control for a few years. Households have often been motivated to take up contracts for tree plantations from a sense of pride of ownership and participation in what they see as sound environmental practice, even though the timber bearing trees will yield no financial return for many years. At appraisal it was planned that wastelands would be planted with shrubs, mainly for erosion control. However, most farmers preferred trees and there was a trend toward planting trees on the steeper slopes, sometimes together with shrubs in alternating rows. Over the project implementation the counties saw less justification to plant shrubs for erosion control (except in desert areas) because of the success of grazing bans, which generated in many areas a dense natural vegetation cover at minimal costs (see below). The planting of desert willow has proved particularly successful in Inner Mongolia for stabilizing sand dunes, as well as providing an alternative source of income for farmers, through the sale of the woody stems for the production of particle board. In the northern part of Shanxi, and in Inner Mongolia, areas of seabuckthorn were dying due to a widespread infection by the seabuckthorn-moth, combined with the severe stress from the prevailing drought. Despite these setbacks, the planting has proved successful with the fruit being sold for the production of juice, providing an additional income source for farmers. The target in the SAR for Economic Trees and Orchards was 27,000 ha. However, interest in these trees was far higher than anticipated, around 57,000 ha were planted by the completion date of the project. Some 27,000 ha of kernel apricot, chestnuts, walnuts, and Chinese date trees were planted on sloping wastelands, and more than 30,000 ha of orchards, mainly apples and some pears, were planted on terraces. The land was contracted to farmers, mostly before the trees were planted. The common practice was for farmers to interplant with vegetable crops while the trees were young. Standards of development were high and the farmers have had access to privately produced good quality seedlings. Good market prospects exist for high quality fruit. Grasslands (US$16.84 million). The project has planned to provide funds for establishment of some 155,000 ha of artificial grasses mainly astragalus and alfalfa. The main purpose of this component was to produce fodder for pen-fed animals in order to reduce the pressure from unsustainable grazing on slopes. Progress of this component has been mixed. Before grazing restrictions were widely imposed - 4 - Page 8 plantation of grasses was only attractive in some areas. After the widespread adoption of grazing bans the growing of artificial grasses is successfully practiced on large scale in most project areas. The attractiveness of fodder production is very high and some farmers also use terraces for planting grasses. At the time of the ICR some 100,400 ha, representing 65% of the estimated target, of astragalus and alfalfa have been established mainly on flat or gently sloping wasteland, but much more areas are planted by farmers using their own funds. During project identification and appraisal, free grazing was found to be a major cause of soil erosion in the Loess Plateau. The Project Management Offices (PMOs) persistently worked towards grazing restrictions in the project and adjoining areas. As a result a number of project counties have now adopted strictly enforced grazing bans and the provincial authorities are confident that all counties will soon adopt this practice. An important impetus came through the arbor tree and shrub planting components, where grazing bans were first introduced to protect the plantations. As livestock was banned from the planted areas, natural regeneration occurred, providing a visual demonstration of the dramatic effect that the bans could have. More and more areas, where grazing bans were put in place, showed a dramatic increase in vegetative cover even in drought-affected areas. Despite the droughts, natural shrubs, grasses and trees are now well established on steep slopes that had previously been grazed bare. The erosion control benefits from these measures are substantial and unprecedented in this area (See Additional Annex 8 Pictures 16 to 18). Livestock production did not suffer from the various grazing bans and other conservation measures. On the contrary, productivity and incomes rose sharply as farmers moved to more intensive production system. Farmers invested in animal sheds and pen construction, fodder processing equipment, and animals suitable for pen feeding (such as cows and small tail sheep). Informal credit and some project loans were the main sources of financing. Under the project, some 116,000 square meters of sheds were constructed, 10,800 heads of livestock procured and 125 small feed processing units set up. Significantly, over the course of the project farmers have benefited from higher yields, twining rates and wool quality. Grazing bans imposed by counties and townships have full support from farmers. Sediment Control Dams, Gully Control and Irrigation (US$26.7 million). Studies have shown that over 50% of the sediment runoff come from the sides of the deeply incised gullies that are typical of the Loess Plateau. The aim of sediment control dams is to intercept the sediment at its source. The gullies are uninhabited and provide excellent sites for earth dams. There are three types of sediment control dams. Key Dams are the largest dams and they control catchment areas of 10-15 sq. km. They are homogeneous earthfill dams from 15-30 m in height. Warping Dams are about 5 meters high and are built in the lower reaches of a gully. They intercept sediment and thereby create arable land. They are sited several hundred meters apart in the wider parts of a gully downstream of a key dam. Check Dams are small dams, 1-2 meters high built of rock or brushwood. Their purpose is to slow the flow of water in the steep tributary gullies and prevent undercutting of the gully sides. A notable feature of the key dams is that they are usually built without spillways; at most sites it is cheaper to size the dam and the reservoir to store the entire volume of the design flood than to provide a spillway. The height of the dams can also be readily increased in the future to restore any loss in storage capacity due to sediment. In the tributaries and the main stream of the Yellow River, sediment inflow causes the riverbeds to rise and flood embankments have to be periodically raised. Also, the storage capacities of the reservoirs are depleted by sediment. Therefore, sediment control has considerable downstream benefits. Local benefits in the case of key dams are water supply for irrigation and villages, and both key dams and warping dams create new arable land on the sediment deposits. Under the Project, 149 key dams have been built or upgraded, and 1,956 check dams and 1,140 - 5 - Page 9 warping dams have been built, representing 63, 57 and 52 percent of the planned SAR implementation targets respectively. Sediment reduction capacity by the various project works is estimated to be about 57 million tons per year. The number of key dams envisaged in the SAR was reduced at the mid-term review. The village households did not see the local benefits to be commensurate with the construction cost and maintained that the Central Government should bear most of the cost of the key dams in view of the significant downstream benefits. The cost savings were mainly used for terrace construction, afforestation and livestock development. Some 7,100 ha of irrigated lands have been developed, equivalent to 111 percent of the planned SAR target. The development of irrigation systems has resulted in crop diversification and increased yields. Investments in irrigation were mainly in Inner Mongolia where extremely low rainfall means that terracing alone would not sufficiently raise yields to bring about a sustainable agricultural system. Of note is an award winning system developed in Inner Mongolia consisting of ‘one dam and one pond’. Water is retained behind a dam, from where it drains into the sandy soil and subsequently into a downstream pond. A pipeline system is used to pump the water to nearby terraced fields. Fields irrigated in such a way produce maize yields of some seven to eight tons per ha. The high benefit from irrigation was made visually obvious during the drought years, when neighboring non-irrigated fields failed to produce a yielding crop. 4.2.2 Institutional Development Component ( US$39.54 million) is rated highly satisfactory. Under this component, the project financed: (i) the construction and equipping of training centres at provincial, prefecture and county levels; (ii) vehicles and equipment for extension services and project management; (iii) hardware, software and related training for GIS and project information system management; (iv) local and overseas training and study tours; and research and the services of external experts. The Bank team helped to introduce and maintain a strict project management system and supervision in the field with random spot-checking at all levels. The rules were rigorously enforced at the beginning to establish a precedent, and were maintained throughout project implementation. This component has substantially improved and developed the Research & Development (R&D) capacity of the existing research institutions in the areas of crop protection, balanced fertilisation, orchard and livestock management (detailed research and extension topics are listed in the Additional Information). The component has also achieved far-reaching impacts on institutional development and R&D capacity (see Section 4.5 on Institutional Development Impact). 4.3 Net Present Value/Economic rate of return: 4.3.1 A recalculation of the economic returns of the project investment shows an Economic Rate of Return (ERR) of the project as high as 29%, if environmental and downstream benefits are included and 19%, if only on-site benefits are considered. Returns for most components met or exceeded SAR estimates, and in particular returns from the livestock and animal husbandry sector have far exceeded SAR expectations. Most project investments were largely local labour based and a significant under-employment in the rural project areas suggests that the opportunity cost for this labour was very low. The major differences in the ICR estimation as compared to the SAR estimation include: (i) significant additional benefits of the project from the livestock sector, which turned out to become an important determinant of the project economic success, while the SAR calculated returns from shrub and grass plantation only, (ii) an up-dated value for the sediment reduction benefits from Y1-Y2/ton, - 6 - Page 10 (iii) the inclusion of the carbon sequestration benefits of the project, (iv) lower than expected direct benefits from afforestation due to lower than expected growth and survival rates of arbor trees, and (v) lower economic labour costs due to the prevailing underemployment in the project areas. ERR Comparison SAR ICR without sediment reduction and global benefits with sediment reduction benefits With sediment reduction and carbon sequestration benefits without sediment reduction and global benefits with sediment reduction benefits with sediment reduction and carbon sequestration benefits Terracing 17 18 N/A 23 26 26 Dam/Warping Land 4 17 N/A 5 26 26 Irrigation 17 18 N/A 32 32 32 Afforestation 18 19 N/A 10 13 17 Orchards 26 26 N/A 25 26 26 Economic Trees N/A N/A N/A 22 22 22 Livestock N/A N/A N/A 33 36 36 Total Project 17 19 N/A 19 22 a) 29 a) a) Includes benefits from grazing ban not covered by individual components . 4.4 Financial rate of return: 4.4.1 As per the SAR approach, the calculation of a financial rate of return for the project is not considered applicable. The financial impact on individual farms and households was directly measured in field surveys showing a significant increase in per capita income for one million farmers. (from Y 360 at baseline to current Y 1,263 against SAR target of Y 900, see Annex 1). 4.5 Institutional development impact: 4.5.1 The project has taken the approach of working with and developing the existing institutions in the Chinese Government and public administration. This not only ensured sustainability, but also created models, which could be spread beyond project areas. The institutional development impact of the project is substantial and far-reaching, as demonstrated in the following: (i) The project management system not only called for an unprecedented level of accountability and discipline in the use of project funds and recording of project activities and achievements, but also allowed evaluation of public sector institutions, PMOs and individuals according to their performance and efficiency. This has been practiced widely and has introduced a completely new level of screening and control of government and public administration. Offices or individuals performing well have been awarded or promoted, others have been faced with stricter control or in few cases even been excluded from the project. In most project areas the political leaders at various levels have taken personal responsibility for the project exercising a tight system of self-control and self-evaluation. The fact that a number of project management practices have been widely adopted in other national programs indicates how strongly those methods have been internalized in the Government system already; (ii) the integrated approach for watershed development taken by the project required multi-disciplinary planning and working teams under one PMO. In many cases and often for the first time the different departments and - 7 - Page 11 bureaus such as soil and water conservation, agriculture, livestock and animal husbandry, and forestry have made joint efforts in planning and implementing integrated watershed development works; and (iii) the modern project management tools introduced under the project, such as computerized MIS, GIS, and other computer based planning and recording systems have significantly improved human and institutional capacity. 5. Major Factors Affecting Implementation and Outcome 5.1 Factors outside the control of government or implementing agency: 5.1.1 Severe droughts in much of the project implementation period (especially in 2000/2001) made the establishment of trees and shrubs extremely difficult, causing low survival rates which required replanting. The droughts also caused crops failures in some project areas. 5.2 Factors generally subject to government control: 5.2.1 Factors subject to government control have been generally favourable to project implementation. Macro-economic and sector policies have been very supportive of the development of integrated watershed development implemented under the project (see Section 3.5). A key policy measure was the government’s policy of providing long-term contracts for land. These tenure agreements, supported at all levels of government and by the Bank, provided strong impetus for implementation of project works. The high level of government commitment throughout implementation was also reflected in selection of top quality project staffing, the adoption of transparent project management system, and the rigorous project monitoring and evaluation process. The administrative arrangements made at each level from province to county have been proved efficient and effective. In addition, the government provided sufficient counterpart funds for the project. 5.3 Factors generally subject to implementing agency control: 5.3.1 Project Management : The project management was highly effective and contributed tremendously to the successful project implementation. The transparency and accountability of project management system adopted by the PMOs at different levels instilled and enforced discipline in the use of project funds, guaranteed the strict monitoring and evaluation of project activities and achievements, and ensured the beneficiary involvement in the project management. The technical assistance was effectively used by PMOs to acquire modern project management tools, such as computerized MIS, GIS, and other computer based planning and recording systems, which have significantly improved project management efficiency. 5.3.2 Procurement and Disbursement : The construction of over 90,000 ha of terraces involved the excavation and compaction of about 100 million cubic meters of earth. Private bulldozer operators according to designs prepared by the County PMOs carried out the work. Typical contracts were on the order of 20-30 ha. Each county advertised its annual work program for terraces and bids were invited for a number of contracts each working season. As the project progressed, bidders from outside the project area showed interest and this led to more competitive prices. The contractors used the small 90-hp bulldozers manufactured in Luoyang and found throughout China. The villages, on whose land the terraces were built, were responsible for supervision and quality control. The blocks of land to be terraced were so widely scattered that it was not feasible to package in them into larger contracts for National Competitive Bidding (NCB). Construction by county governments with equipment procured through International Competitive Bidding (ICB) was considered but rejected. The adopted approach had the advantages that it exploited local skills and resources, ensured competitive pricing, and placed quality control in the hands of the beneficiaries who would be responsible for loan repayment. Vehicles were procured through ICB, and computers and office equipment through Limited International Bidding (LIB). The process was slow because of the numerous levels of approval needed - 8 - Page 12 for such procurement. Disbursement was against unit prices per ha for the various components that were agreed with International Development Association (IDA) at appraisal. Some of the unit prices were adjusted during the course of the project. Before disbursement claims were forwarded to Central PMO, a detailed field check of the physical quantities was made by the provincial and prefecture PMOs. IDA missions did spot checks to ensure that the process was being followed. This procedure worked well and was far superior to a simple record of expenditures since there was a close link between disbursements and verified physical progress. 5.3.3 Quality Control and Physical Checking : The project introduced a complete system of quality control and checking at every level. Detailed maps, with all project activities clearly located and coded, allowed effective random checking of all project components at any time. This system has proven highly effective in supervising a very large and scattered project. Examples of those maps are shown in Additional Annex 9, pictures 23 to 28. Quality was assured by assigning clear responsibilities within the PMOs management structure. Standards were set for all works such as plant densities of trees and shrubs, survival rates in relation to planting time, size and depth of water harvesting pits, height and width of terrace lips, maximum slope tolerance. Frequent and well recorded checking system, which also carried the names of the responsible officers, were defined for each management level and each disbursement was subject to following those rules. 5.3.4 Compliance with Credit Agreement : The Borrower and the Project Executing Agencies and Management Offices essentially complied with the credit agreement and recommendations. 5.4 Costs and financing: 5.4.1 In the SAR, total project costs were estimated at US$248.7 million with IDA’s contribution at US$150.0 million. The actual project costs were US$252.1 million, only 1% over the SAR estimates. Of the total project costs, IDA disbursements amounted to US$148.7 million (99% of SAR estimates), while the government counterpart funds including farmers’ contribution totalled US$103.4 million (105% of SAR estimates). Detailed project costs and financing are described in Annex 2. 6. Sustainability 6.1 Rationale for sustainability rating: 6.1.1 The sustainability of the project is rated highly likely for the reasons below: (i) Financial profitability (to farmers) and economic viability (to the society as a whole) of the project have created strong incentives to all the stakeholders involved. For farmers, the project has invested in a broad range of diversified income generating opportunities, including the production of a range of annual and perennial crops. The livelihood of the farmers, as a result, has been substantially improved through the diversified agricultural production and other off farm income sources (see Section 4.1); while at the national level the project has reduced the risks of floods downstream and has the potential to significantly reduce dust pollution in north China, including Beijing. (ii) The project promoted environmentally friendly production practices and reversed the soil deterioration that was taking place, and thus improved the potential for higher and diversified incomes. It has also reduced farmers’ vulnerability to droughts, and it will continue to have a profound effect on the sustainability of the farming systems in the future. (iii) The new institutional arrangements created by the project as well as the regulatory and - 9 - Page 13 policy support by government have contributed to the project sustainability, and extended the project impact well beyond project areas. The project activities and interventions are backed by strong policy reforms and are well integrated in the social, legal and administrative system of the country. Long-term land contracts provide a strong incentive for farmers under the project to adopt sustainable land management practices. The project’s proven concept and methodologies have already been introduced in several similar national projects. The highly successful grazing ban has also become a corner stone in the Chinese strategy to reverse soil degradation in the Loess Plateau areas and to re-establish natural vegetation. Furthermore, the accountability of public services and the human capacity built under the project will continue to extend its impact in many development activities in the area, and will be of growing value for the Government’s Western Development Program; and (iv) The on-going Bank follow-on project, Loess II, based on experiences gained from this project, is expanding the project activities and consolidating the project impacts. 6.2 Transition arrangement to regular operations: 6.2.1 Most project activities were contracted out and implemented by farmers and therefore do not require any specific transitional arrangement. However, for the arbor tree development, county PMOs have employed forestry guards to protect the treated areas. Assurances have been obtained from county PMOs that forestlands developed under the project will be well protected until they are eventually auctioned off to farmers. 6.2.2 The key dams are still under the management of the county PMOs although a pilot program has been initiated to test the hand-over to farmers. County PMOs, entrusted by local water conservancy bureaus, are likely to remain in charge of the key dams with financial support from the regular national soil and water conservation budget. The Yellow River Conservancy Commission (YRCC) has raised the point that the Central Government should bear a greater share of the costs for the key dams as the national environmental benefits (sediment reduction) outweigh the local benefits. Discussion is being initiated with the Ministry of Water Resources (MWR) and the Ministry of Finance (MOF). 7. Bank and Borrower Performance Bank 7.1 Lending: 7.1.1 The Bank’s performance in identification, assistance in preparation, and appraisal is rated satisfactory. 7.1.2 The project identified was consistent with the Government's and the Bank's strategies to develop environmentally sustainable agriculture, increase employment and income, and help alleviate poverty (see Section 3.5). The project was very well prepared in terms of technical specifications, particularly for terracing and dam construction components. The project preparation process was highly participatory and the borrower took the lead in the preparation process (Section 7.4). The Bank's inputs in terms of staffing and skill-mix were adequate for the preparation. Between identification and appraisal, 6 missions were fielded. The multi-disciplinary missions were composed of experts in the following aspects: agriculture, economic analysis, environment, project operations, water resources, and land use planning. 7.1.3 The Bank's performance at appraisal was also satisfactory. The Bank's team was an eight-member mission, which covered the key skills required by the project. The mission appraised, among other aspects of project design, the government commitment (including counterpart funding) and - 10 - Page 14 implementing agencies’ capacity in terms of financial management, project risks and mitigation measures, the incentives for stakeholders to sustain the project, and safeguard issues (environment, resettlement, dam safety). 7.1.4 However, one oversight of the project design was that the Bank should have, together with the central and local governments, worked out a more detailed cost sharing and repayment plan for construction costs for the key dams. 7.2 Supervision: 7.2.1 Bank supervision is rated highly satisfactory. The supervision missions were composed of a broad mix of international and local professionals, who were not only technically competent but also responsive to the client’s views and needs, and the local experiences and knowledge. The core supervision team was maintained throughout project implementation and an excellent working relationship was kept with their counterparts. Implementation progress was adequately reported and problems were identified early and addressed proactively. Performance ratings were realistic and adequate attention was paid to legal covenants and safeguards (dam safety). The supervision missions also helped to build the institutional capacity of PMOs at different levels and establish a transparent monitoring system using Geographical Information System (GIS). The team provided high quality technical and policy advice (such as initiating the grazing ban), which was highly valued by the PMOs. Size and frequency of supervision missions were on the low side given the complexity and the wide geographic coverage of the project--only 8 supervision missions, on average with 6 members, took place over 8 years of project implementation. However, the supervision has been proven efficient and adequate as the PMOs gradually developed their own capacity. The fact that the task manager was placed in the field allowed easy and frequent contact. These aspects of the supervision have had a significantly positive effect on the outcome of the project. 7.3 Overall Bank performance: 7.3.1 The overall Bank performance is rated highly satisfactory given the above-mentioned lending and supervision performance. Borrower 7.4 Preparation: 7.4.1 Project preparation by the borrower was rated highly satisfactory. The borrower committed adequate resources to the preparation efforts. A national preparation team, composed of top professionals in areas of agriculture, water resources management, forestry and institutional and farm management was assembled by MWR/YRCC. The project experienced two years of participatory project planning, directed by individual watershed communities. Project staff gathered knowledge about villagers’ successful experiences with local erosion control measures. Based on these, each county’s project management office prepared a watershed plan with modules of integrated packages. The small watershed communities discussed the plans, and decided upon sequencing, payment and implementation procedures. Maps were prepared on a 1:10,000 scale showing existing and planned land use. This occurred with more than 1,000 project management offices for over 2,000 villages, involving hundreds of thousands of people. The project preparation work, adopting a fully participatory approach, covered major technical, institutional, economic and environmental aspects of the project design, and the views of all stakeholders at central, provincial and local level were all reflected in the process. 7.5 Government implementation performance: - 11 - Page 15 7.5.1 The Government implementation performance is rated satisfactory. As stated in Sections 5.2 and 3.5, the government macro-economic and sector policies have been very conducive to the development objectives. The government’s policy of providing long term contracts for land provided strong impetus for implementation of major project activities. The high level of government commitment throughout implementation was also reflected in selection and retention of top quality project staffing, the adoption of transparent project management system at central, provincial and local levels. And the government also fulfilled its commitment to providing sufficient counterpart funds for the project. 7.6 Implementing Agency: 7.6.1 The performance of the implementing agency is rated highly satisfactory. As discussed in Section 5.3, the PMOs at all levels are well staffed and highly effective. The transparency and accountability of project management ensured rigorous monitoring and evaluation, and strict compliance with disbursement and procurement rules and regulations. The management and institutional capacity of PMOs have been progressively strengthened through the technical assistance and interaction with the Bank’s supervision missions. This enabled the implementation of such a technically complex and geographically wide project slightly ahead of SAR projection, and allowed the Bank to reduce its inputs for supervision. 7.7 Overall Borrower performance: 7.7.1 The overall borrower’s performance is rated highly satisfactory, considering the strong government commitment to the project in terms of policy support and counterpart fund provision, and the excellent performance of the implementing agencies described in Section 7.6. 8. Lessons Learned 8.1.1 On the basis of the above review and assessments of the project, the following lessons can be drawn for similar projects: Watershed management project requires well designed technical packages which generate l incomes for the local communities. The comprehensive technical packages supported by the project included afforestation, halting cropping on steep slopes, establishment of large-scale terracing and sediment control structures, and reduction of overgrazing. Those technically sound project interventions combined sustainable soil and water conservation practices with gains in agricultural production and farm incomes and were the foundation and prerequisite for project success. Strong ownership at different levels makes a project truly participatory. Project interventions l were adapted to the requirements of each watershed. Local communities were in charge of developing their own plans. Equally important was the support from different levels of government who saw clear benefits for the down stream flood control, reduced soil and water erosion and sandstorms. As a result, high quality technical and management staff were assigned to the project and issues such as counterpart funding were speedily resolved. Effective project management with rigorous monitoring and evaluation contributes to the l successful project implementation. Despite its large scale and scattered nature, this project remained well managed. The key elements for successful project management include: (i) making use and developing local capacity and institutions for implementation, (ii) applying simple but strict and transparent procurement and disbursement mechanisms with a high degree of beneficiary control, and (iii) imposing a strong and transparent monitoring system, which allows efficient and - 12 - Page 16 cost-effective internal and external control like the system of ‘maps and tables’ introduced in this project. Physical progress was checked against maps for each sub-watershed showing the project plan and completed areas, with a table for each map listing all completed areas with their sizes and dates of development. For terraces, sketch maps or detailed design drawings show the layout for each plot. Successful pilot activities can be scaled up with good project planning and management. The l project components were based on activities successfully tested by similar projects funded by other donors and government. The major factors underlying their successful up-scaling were attributable to adopting integrated watershed based production system and overall strong project management at field level, with clear responsibility assigned to county PMOs. Consistency in the composition of a competent core Task Team greatly enhances Bank-Client l cooperation. The task team, which cohesively worked with the project, was composed of a small but highly committed, and productive team. The task team’s working style and commitment have stimulated the same spirit on the Borrower’s side and have resulted in a remarkably open, cooperative and mutually respective working relationship between the Bank and Borrower with a high level of mutual trust. Only through such a working relationship a dynamic process of project implementation became possible. The grazing ban offers an example for such a dynamic process. On this issue the Bank’s team worked with the client over several years towards a policy change and reacted to the emerging success of the grazing restrictions with immediate adjustments in project activities. The "public goods" nature of the project should be fully recognized with the financial l obligations accepted by the different stakeholders. As a lot of the environmental impact from increased vegetation cover and reduced sediment and sand storm created benefits outside of the project area. The costs of major project inventions, which generate such benefits, should be jointly borne by project farmers, and local and central governments. If this government responsibility had have been fully accepted, the project’s achievement in sediment dam construction could have been higher. 9. Partner Comments (a) Borrower/implementing agency: Comments from CPMO We think the assessment of the project in the ICR generally reflects the actual situation of the project and we appreciate very much the efforts made by all of the ICR mission members. We would like to emphasize some aspects: The project has achieved its targeted objectives with highly satisfactory ecological, economic and social 1. benefits at all levels - national, regional and local communities. The project design is very focused and practical, which is a major contributor to the great success of 2. the project and has developed an Integrated Watershed Management Model being replicated nationwide. The project management throughout the implementation of the Project has been highly satisfactory. 3. The state-of-the-art expertise for project management introduced by the IDA team has been a - 13 - Page 17 tremendous gain for the PMOs at various levels, which is more important than the funding itself to some extent. Many successfully implemented project interventions and policies, like grazing management and 4. enforcement of long-term land contract have strategic impact and hence will make contribution to the sustainability of the Project. The excellent cooperation between the World Bank and the Chinese counterparts is an outstanding 5. example for partnership building, which not only has made this particular project a great success, but also contributes substantially to the long term collaboration. (b) Cofinanciers: N.A. (c) Other partners (NGOs/private sector): N.A. 10. Additional Information List of Research and Extension Topics Shaanxi Province Large-scale soil and water conservation agriculture and integrated collocation technology 1 Caragana-licorice root intercropping 1&2 Forest species collocation and high-yield forestry technology in Yanhe River watershed 1&2 Forest establishment in Jialu River project area 2 High-efficiency and ecological fruit industry development technology in gullied and rolling Loess Plateau 2 Grass species establishment and animal development for soil and water conservation in rolling Loess Plateau 1 High yield alfalfa planting technology and goat improvement 1&2 Engineering works technology of warping dam flood prevention 1 Dam system optimum distribution and operating model in small watershed 1 Runoff and sediment changes and flood water and sediment reduction by soil conservation measures in the Yanhe and Jialu Rivers 1 High yield and high quality apple orchard establishment 1 Comprehensive dry cropping agriculture and high yielding technology 1 Forest system establishment for soil and water conservation 1 Engineering works distribution for soil and water conservation 1 Grass land improvement and establishment 1 Economic forest development of soil and water conservation 1 Soil and water conservation using cultivating measures 1 Optimum collocation of soil and water conservation measures 1 - 14 - Page 18 Gansu Province Plant seedling selection for soil and water conservation 1 Water-force sluicing and filling technology for gully retaining works 2 Advanced afforestation technology for timber forests 2 Mixed arbor tree and seabuckthorn planting 1&2 Elimination of viruses of fruit trees and high-yield planting technology 1&2 New terrace fertility 1&2 Integrated techniques of artificial grass and shed-bred small tail sheep 1&2 Application of computer technology for soil and water conservation planning and management 1 Efficient development and utilization of soil and water resources in small watersheds 1 Scattered land and terrace ridges utilization in small watersheds 1 Optimum design of machine-built terraces 1&2 CPMO Fodder production and processing 1 Collective apple management in Longdong Loess Plateau 2 Benefits of monitoring and evaluation in the project area 1 High-quality and high-yield of Chinese date 2 Economic plants 2 Animal breeding and green forage storage 2 Central nursery establishment 1 Shanxi Province Drought protection afforestation technology 1 Dam land warping, flood prevention and water drought protection technology 1 Mountain land micro-irrigation technology 2 Zhunger Saline-alkali land improvement and salinization prevention measures 1 Yih Ju League Eco-economic benefits of small watershed integrated treatments 1 Selection of superior seabuckthorn types and high-efficiency planting 1 Plant edge 1 Gully head protection 1 Land management information systems 1 - 15 - Page 19 1 Research topic 2 Extension/demonstration topic - 16 - Page 20 Annex 1. Key Performance Indicators/Log Frame Matrix Outcome / Impact Indicators: Indicator/Matrix Projected in last PSR 1 Actual/Latest Estimate Annual Cropped area (28%) 21% 21% Perennial plants (37%) 54% 48% Forest cover (14%) 38% 41% W asteland (28%) 17% 15% Annual sediment discharge reduction 60 million tons 57 million tons Annual Grain output (427,000 tons) 660,000 tons 698,600 tons Annual fruit production (80,000 tons) 320,000 tons 345,000 tons Annual per capital grain output (357kg) 507kg 532 kg Annual per capital net incomes (¥360) ¥900 ¥1,263 1 Numbers in parenthesis indicate actual estimate at project appraisal Output Indicators: Indicator/Matrix Projected in last PSR 1 Actual/Latest Estimate a) Terracing/Irrigation/land formation (ha) Not applicable Terrace 90,446 Irrigated land 7,098 W arping land 4,010 ha Land formed by dams/gully ctrl works2 2,214 ha b) Afforestation (ha) Arbor tree 9,0951 Shrub tree 136,041 Economic tree 57,687 c) Grassland and livestock support (ha) d) silt retention structures (ha) 100,411 Key dam 149 W arping dam 1,140 Check dam 1,956 1 End of project 2 Estimates for warping land and land formed by dams and gully control works will be completed in the future when the process of sediment deposition is completed . - 17 - Page 21 Annex 2. Project Costs and Financing Project Cost by Component (in US$ million equivalent) Appraisal Estimate Actual/Latest Estimate Percentage of Appraisal Component US$ million US$ million Land Development Terracing/Irrigation 76.40 96.75 126.6 Afforestation 50.80 72.21 142.1 Grassland 11.90 16.84 141.5 Dams 12.00 26.70 222.5 Institutional Development Support Services 17.80 28.75 161.5 Training/management 11.30 10.79 95.5 Total Baseline Cost 180.20 252.04 Physical Contingencies 8.80 Price Contingencies 59.70 Total Project Costs 248.70 252.04 Total Financing Required 248.70 252.04 Appraisal estimates include a lumpsum contingency, while the Actual/Latest estimates include contingencies by component. Project Costs by Procurement Arrangements (Appraisal Estimate) (US$ million equivalent) Expenditure Category ICB Procurement NCB Method 1 Other 2 N.B.F. Total Cost 1. Works 0.00 5.60 200.90 0.00 206.50 ( 0.00 ) ( 3.60 ) ( 122.90 ) ( 0.00 ) ( 126.50 ) 2. Goods 16.70 0.00 2.00 0.00 18.70 ( 16.70 ) ( 0.00 ) ( 2.00 ) ( 0.00 ) ( 18.70 ) 3. Services 0.00 0.00 7.60 13.80 21.40 ( 0.00 ) ( 0.00 ) ( 4.80 ) ( 0.00 ) ( 4.80 ) 4. Other 0.00 0.00 0.00 2.10 2.10 ( 0.00 ) ( 0.00 ) ( 0.00 ) ( 0.00 ) ( 0.00 ) Total 16.70 5.60 210.50 15.90 248.70 ( 16.70 ) ( 3.60 ) ( 129.70 ) ( 0.00 ) ( 150.00 ) Appraisal estimates include a lumpsum contingency, while the Actual/Latest estimates include contingencies by component. - 18 - Page 22 Project Costs by Procurement Arrangements (Actual/Latest Estimate) (US$ million equivalent) Expenditure Category ICB Procurement NCB Method 1 Other 2 N.B.F. Total Cost 1. Works 0.00 17.96 193.32 0.00 211.28 ( 0.00 ) ( 10.79 ) ( 112.55 ) ( 0.00 ) ( 123.34 ) 2. Goods 20.70 0.05 0.40 0.23 21.38 ( 20.70 ) ( 0.05 ) ( 0.31 ) ( 0.00 ) ( 21.06 ) 3. Services 0.00 0.00 6.73 9.08 15.81 ( 0.00 ) ( 0.00 ) ( 3.52 ) ( 0.00 ) ( 3.52 ) 4. Other 0.00 0.00 1.17 2.43 3.60 ( 0.00 ) ( 0.00 ) ( 0.75 ) ( 0.00 ) ( 0.75 ) Total 20.70 18.01 201.62 11.74 252.07 ( 20.70 ) ( 10.84 ) ( 117.13 ) ( 0.00 ) ( 148.67 ) 1/ Figures in parenthesis are the amounts to be financed by the Bank Loan . All costs include contingencies. 2/ Includes civil works and goods to be procured through national shopping, consulting services, services of contracted staff of the project management office, training, technical assistance services, and incremental operating costs related to (i) managing the project, and (ii) re-lending project funds to local government units. Project Financing by Component (in US$ million equivalent) Component Appraisal Estimate Actual/Latest Estimate Percentage of Appraisal Bank Govt. CoF. Bank Govt. CoF. Bank Govt. CoF. Land development Terracing/irrigation 63.90 44.50 55.31 41.44 86.6 93.1 Afforestation 40.88 30.42 40.75 31.47 99.7 103.5 Grassland 12.25 7.66 9.81 7.03 80.1 91.8 Dams 10.77 5.87 16.61 10.10 154.2 172.1 Institutional development Support Services 14.45 6.71 20.71 8.04 143.3 119.8 Training/management 7.75 3.53 5.47 5.32 70.6 150.7 Total 150.00 98.69 148.66 103.40 99.1 104.8 Appraisal estimates include a lumpsum contingency, while the Actual/Latest estimates include contingencies by component. - 19 - Page 23 Annex 3. Economic Costs and Benefits Project Benefits: The project’s major benefits occurred in form of: (A) substantive raise in farm income and living standard on-site within the project area, (B) improvements in the river regime downstream in the Yellow River, (C) global or regional ecological improvements beyond the project areas and the Yellow River Basin, and (D) significant efficiency gains in public administration and public spending under similar national and international investment projects: A. On-site Benefits within the project area The majority of project benefits likely occurred within the project area, where about one million farmers directly benefit from this project. The project areas belong to the poorest in China and the project raised average per capita net-incomes from annually Yuan 306 to Yuan 1,263 with over one million farm households directly benefiting from the project. The various project components contributed to the significant reduction of poverty and tripling of net-incomes from different angles addressing a range of short-, medium-, and long-term income generating and income stabilizing measures: Direct Income Increase. Terracing of loess soils usually allows to double crop yields with only l marginal higher input costs, and will therefore lead to an almost immediate raise in incomes. In some of the drier areas and where sufficient surface or groundwater can be secured, terracing will be combined with irrigation, which secures optimal yields, since soils, temperature and sunshine hours make the Loess Plateau to one of the most fertile areas in the world. The new terraces, and access roads to the terraces, gave farmers the opportunity to grow a wider range of crops with higher yields than is possible on unimproved slopes or old inaccessible terraces. Particularly temperate fruit and nut trees grow very well and have a more medium and longer-term, however, very significant impact on agricultural incomes, too. Employment and Labor Productivity. Labor saving in crop production and the diversification of l agriculture and livestock production allowed engagement in new employment and income opportunities. This includes both on-farm and off-farm employment. While part of the farmers used the opportunity to have more time available for outside employment, others entered into new farming activities particularly in livestock or fruit production; this led to an increase in the labor rate from 70% before to 87% after project and a more balanced use of labor over the year. In the past labor demand was high in the months of crop planting and harvesting with significant underemployment in winter and early spring. Tending of trees, orchards or livestock provides work in the off-season. Employment of women and their labor productivity has particularly benefited from new on-farm employment. Food Security and Risk Reduction. Farmers in the Loess Plateau face the threat of severe l droughts. In dry years, crops on slopes often failed completely and the Government had to supply the farmers with food. Terracing lowered the variability of yields significantly as proven during years of severe drought during project implementation, but in years with average rainfall, grain yields on terraces reached two to three times those on slope-lands. Since farmers typically store grain over years, the high yields in the good years provide insurance for the drought years. The diversification of production as supported by the project reduced the variability of income. Livestock and tree provide a buffer in difficult times. Irrigation, although a small part of the - 20 - Page 24 project, also provides drought protection. Natural Resource Protection and Securing of Long-term Income Opportunities. Large areas in l the project area were severely degraded and past agricultural practices were clearly unsustainable due to uncontrolled grazing, gathering for fuel, and cropping on slopes that were too steep for sustainable farming. The project succeeded taking a large proportion of these areas entirely out of production for natural regeneration and turning unsustainable use of slopes into sustainable use by planting trees and shrubs, which protect soils and provide returns. This practice secured long-term productivity of those areas and raised incomes for the local people. B. Downstream Benefits In addition to increasing production and income within the project areas, the project provides significant benefits to the downstream areas: Reduced Sediment Loads in Rivers. The project reduced sediment inflow to the Yellow River and l its tributaries with several benefits. First, irrigation systems suffer less from large inflows of sediment that choke the canals and often make it necessary to stop diversions when sediment loads are high. Second, river channels are more stable and maintenance costs for river training works are reduced. Third, sediment build up in downstream reservoirs is slowed. Fourth, the rise of the riverbed in the Yellow River will be slowed and cost of raising the flood embankments can be deferred. Lower variability in water flows and reduced flood risk. Retaining rainfall water on terraces, l behind key dams and check dams and in millions of small water harvesting pits around each tree planted under this project lower the peak run-off. This reduces severity of down-stream floods, reduces flood damage, and lowers the cost of flood protection works. Due to the difficulty in quantification the effect is not included in the analysis. C. Regional and Global Benefits Global and regional environmental benefits of the project are also likely to occur in the long run, although it is difficult to fully specify the magnitude and time-horizon. Reduction of Greenhouse Gases and Carbon Sequestration. Vegetation and soils are widely l recognized as carbon storage sinks. Sequestration of carbon in terrestrial ecosystems offers a low-cost means of reducing carbon emissions. The ‘Kyoto Protocol’ makes provision for direct human-induced land use change and vegetation recovering activities to be considered in relation to each country’s Greenhouse Gases reduction target. The project resulted in a significant incremental sequestration of carbon in sub-surface and surface biomass in an estimated range between 20 and 100 tons per hectare. The analysis has taken an extremely conservative approach in the quantification of these benefits limiting the impact to the project watersheds only and using carbon sequestration of 20 tons per ha for regenerated closed areas and 40 tons per ha for afforested land valued with Yuan 120 per ton, which represents about the lowest value used in similar calculations. Nevertheless, the economic benefits are substantive and raise the ERR by 7%. Using a more realistic approach the economic benefits from carbon sequestration would outweigh all other project benefits by far. Biodiversity. Almost as a collateral benefit resulting from the re-vegetation of large formerly l barren areas the project contributed to a restoration of biodiversity, and restored the habitat for plants and wildlife. The impact is already well noticeable, but its full magnitude is yet difficult to - 21 - Page 25 fully assess. Long-term Impact on Climate and Rainfall. A dense vegetative cover in the Loess Plateau is also l expected to have a positive impact on the climate in the region particularly regarding increased and more evenly distributed rainfall and more moderate temperatures. While the trend is very likely to be positive, this effect is very difficult to quantify and is not included in the analysis. D. Efficiency Gains in Public Administration and Public Spending The project’s impact on institutional development and accountability of public services has significant economic benefits. Monitoring and supervision tools developed under the project have been introduced in large-scale national programs. Even if the project had resulted in only marginal efficiency gains under these programs the overall benefits would be substantive due to the large-scale national investments. Quantification is, however, difficult and beyond the scope of the ICR work. Project Costs: Total project investment costs in financial terms have been RMB 2,092 million. The current stage of trade liberalization in China does not suggest significant price and cost distortion for most traded commodities and services. However, a significant under-employment in rural areas suggests economic labor costs being well below current market wages. The economic analysis uses economic opportunity cost for farm and unskilled labor of Yuan 3 per day for the winter months (November to April, when underemployment is particular severe) and Yuan 4.5 per day for the summer months (May to October) during the eight years of project implementation. In the future opportunity costs for labor will be higher assuming growing overall labor productivity in the country. With these cost adjustments the investment cost in economic terms are estimated at RMB 1,785 million. - 22 - Page 26 Overview of Major Benefits by Components, Interdependence of Components and Quantification Assumptions Component, Sub-Component Major direct Economic Benefits by Category 1 (by decreasing importance) Cat. Major linkages with other Components and indirect benefits (grouped in ·dependent on· and ·mutually supported/supportive·) Quantification Method/Module in Area Analysis Terraces A A B B Increased crop yields, Reduction of risk, Reduction (elimination) of water run-off (flood protection benefit) Full sediment retention Dependent on : none, terracing is one of the few components, which could be implemented independently of other components. 2 Mutually supported/supportive: Livestock raising provides draft power l and manure for cropping; Increased crop production on terraces l provides feed and crop residues facilitate changes in grazing pattern; Terraces extend the lifetime of gully l structures. Standard crop models valuing inputs and outputs at economic prices and comparing ·with· and ·without· project situation; by-products such as crop residues and non-traded inputs such as animal draft and manure are not valued in a comprehensive area analysis, since they are only intermediate products; (Sediment retention see next line) Silt Retention and Warping Land Structures (check dams, key dams, warping dams, and protection works in gully bottom) B B A Sediment retention Balancing of water run-off Income through agricultural production by creation of new crop land Dependent on: Gully structures are usually built as a l system of structures, where key dams depend on check dams above, warping land development is inter-linked with dam structures, or protection works in the gully bed, etc. Mutually supportive: Irrigation, many irrigation systems under l the project take surface water directly from gully structures; most groundwater systems depend in the l long-run on water infiltration from gully structures, terracing and other contour measures implemented under various components For sediment retention the SAR assumption is followed, which used the embankment raising cost and the reservoir live time as the major reference value. A cost figure of Yuan 1.5 per ton of sediment was calculated and a conservative figure of 1 Yuan was used in the SAR. The ICR used a figure of 2 Yuan per ton brining the original 1.5 Yuan up to 2002 prices. Incremental agricultural production on new land is assessed by standard crop models Irrigation A A Increased agricultural production and output through increased yields and production of higher value crops, Reduction of risk Dependent on: many irrigation systems are combined l with gully structures Mutually supportive: terracing, particularly in very dry areas terracing without irrigation does not secure sufficient yield increase to make it economically viable Crop models Arbor trees A B B C Income Sediment retention Balancing of water runoff Climatic benefits Dependent on: Livestock component as changing l grazing pattern are essential for protecting new plantations Mutually supportive: Trees provide fodder for livestock, l Incremental fuel wood releases crop l residues for use as fodder Crop models for the income from timber; Fodder from trees enter livestock models as sub-model; Global benefits are valued at Yuan 120 per ton of fixed carbon, assuming that a carbon sink of up to 10 tons per ha would be built up over 25 years. - 23 - Page 27 Shrubs B A C Sediment retention Income mostly indirect through forage for animals or fuel wood Limited climatic benefits Dependent on: Livestock component as changing l grazing pattern are essential for protecting new plantations Mutually supportive : Shrubs provide fodder for livestock, l Incremental fuel wood releases crop l residues for use as fodder The fodder from shrubs is not valued, crop models for shrubs are used as sub-models in the animal production models. Artificial Grassland (plantation of alfalfa, astragalus and other grass) No direct economic benefit in the A category, however, a chain of benefits through changing pattern in livestock raising with again indirect benefits on trees, shrubs and natural grassland development. Dependent on and mutually supportive: This component is closely linked to the objective to change of livestock raising from natural grazing to stall-feeding Crop models for grass production are used as sub-models in the animal production models. Natural Grassland protection (closing of areas) B B C Sediment retention Limited water run-off reduction Climatic benefits Dependent on: Livestock component and acceptance of l farmers to change grazing pattern and close natural grassland areas from grazing. Sediment retention at Yuan 2 per ton Carbon sequestration of 20 ton per ha over 10 years Livestock Development (animal pens, breeding cows, fodder processing equipment) A Income Dependent on: Artificial grassland development, shrub l and tree production, and crop residues as alternative fodder source, Crop residues from incremental crop l production on terraces Mutually supportive : Terraces and horticulture by providing l manure and draft power Livestock cash flow models with several sub-models for grass and shrub production and fodder from trees. Horticulture (fruit orchards, date and nut trees, top-working of existing orchards) A B B C Income Sediment retention Limited water run-off reduction Climatic benefits Dependent on: Terracing to free up currently cropped l slope land for orchards on slope land, Crop residues from incremental crop l production on terraces Mutually supported : Livestock provides manure l Crop models Nursery Development No direct economic benefits, nurseries produce an intermediate product, which is costed at economic prices in the production models Mutually supportive : Arbor tree, l Shrub, and l Horticulture component l Economic prices for seedlings are used in the production models 1 A - on-site benefits, B · downstream benefits, and C · regional and global benefits. 2 Only in very dry areas, such as most places in Inner Mongolia, terracing alone will not result in sufficient yield increases and needs to be combined with irrigation. - 24 - Page 28 Annex 4. Bank Inputs (a) Missions: Stage of Project Cycle Performance Rating No. of Persons and Specialty (e.g. 2 Economists, 1 FMS, etc.) Month/Year Count Specialty Implementation Progress Development Objective Identification/Preparation 5/90 2 A, AG 11/91 7 A, AG, 2E, ESD, F, PS 7/92 6 AG, ES, ESD, 2F, PS 6/93 1 WCS 8/93 8 A, AG, 2E, OP, 2PA, ES 11/93 8 A, AG, 2E, OP, 2PA, ES Appraisal/Negotiation 5/94 8 A, AG, 2E, 2OP, PA, ES Supervision 7/95 6 H, 3E, 2ES S HS 9/96 4 AG, ES, E, F HS HS 9/97 7 A, 2AG, 2ES, WRS, F S HS 10/98 6 AG, E, M, WRS, H, F S S 11/99 6 ES, OP, M, WRS, FN, AG S S 9/00 9 AG, 2E, 2M, ESD, FN, 2F S S 9/01 4 AG, E, WRS, F S S 9/02 4 AG, E, WRS, F S HS ICR 09/02 3 A, AG, E HS HS A=Agronomist AG=Agriculturist E=Economist ES=Environment Specialist ESD=Engineer for the Safety of Dams F=Forestry Specialist FN=Financial Official H=Horticultural Specialist M=Monitoring Specialist OP=Operation Officer PA=Project Analyst PS=Procurement Specialist WCS=Water Conservation Specialist WRS=Water Resources Specialist Note: Several mission members were internationally acknowledged specialists in more than one discipline. Livestock, Land Use Planning and Agro-processing as well as marketing have been covered repeatedly by the mission. (b) Staff: Stage of Project Cycle Actual/Latest Estimate No. Staff weeks US$ ('000) Identification/Preparation 119.6 300.8 Appraisal/Negotiation 37.8 96.4 Supervision 93.32 240.6 ICR 10 50 Total 260.72 687.8 Note: According to the Bank's accounting system, consultant staff weeks (excluding FAO/CP) are recorded - 25 - Page 29 only until Fiscal Year 1999. Input value includes all expenses, including consultant's input after Fiscal Year 1999. - 26 - Page 30 Annex 5. Ratings for Achievement of Objectives/Outputs of Components (H=High, SU=Substantial, M=Modest, N=Negligible, NA=Not Applicable) Rating Macro policies H SU M N NA Sector Policies H SU M N NA Physical H SU M N NA Financial H SU M N NA Institutional Development H SU M N NA Environmental H SU M N NA Social Poverty Reduction H SU M N NA Gender H SU M N NA Other (Please specify) H SU M N NA Private sector development H SU M N NA Public sector management H SU M N NA Other (Please specify) H SU M N NA - 27 - Page 31 Annex 6. Ratings of Bank and Borrower Performance ( HS=Highly Satisfactory, S=Satisfactory, U=Unsatisfactory, HU=Highly Unsatisfactory) 6.1 Bank performance Rating Lending HS S U HU Supervision HS S U HU Overall HS S U HU 6.2 Borrower performance Rating Preparation HS S U HU Government implementation performance HS S U HU Implementation agency performance HS S U HU Overall HS S U HU - 28 - Page 32 Annex 7. List of Supporting Documents Aide Memorie of the ICR Mission Economic Analysis Working Paper Detailed ERR Calculations (Farmod and Spreadsheet Files) Borrower's Comments Borrower's ICR - 29 - Page 33 Additional Annex 8 . Pictorial Overview of Loess Plateau and Project Achievements Picture 1: The Loess Plateau covers an area of some 640,000 sq. kilometers in the upper and middle reaches of Yellow River. This aerial photograph shows a section in Central Shanxi Province, which is already treated to a large extent (foreground) and lesser extent (background), (aerial photograph taken June 2002) Picture 2: Former plateau-type landscape is clearly visible on this picture with gully erosion starting to cut into the plateau (upper right) and heavy erosion (lower left), treatment is typical for project type works such as water harvesting structures, contour bunds with shrub and tree planting, gully protection works, (aerial photograph, June 2002) - 30 - Page 34 Picture 3: Severe soil and water erosion and low water retention results in very low output farming system. Picture shows a typical landscape as it was seen in many areas before the project started Picture 4: Picture shows a project site in Northern Shanxi Province near the Yellow River with comprehensive treatment done under the project - 31 - Page 35 Picture 5: Cropping on highly erosive slopes with low water retention capacity was one of the key problems addressed by the project Picture 6: Terracing of crop land effectively stops erosion, retains almost all rainfall and yields reliable crops, picture shows a sub-watershed comprehensively treated under the project in Gansu Province - 32 - Page 36 Picture 7: Early stage of treatment in a project sub-watershed in Central Shanxi, terracing partly finished, contour bunds constructed, the picture still clearly shows the severe deterioration before project Picture 8: Picture taken during ICR mission in 2002 shows a different project sub-watershed in Northern Shanxi Province, where comprehensive watershed treatment was finished in 1998. Watershed is largely stabilized, crop yields are high and erosion is largely controlled, farmer interviews confirmed high satisfaction with the project and significant income increase - 33 - Page 37 Picture 9: Although not the same location, picture 9 and picture 10 provide a good comparison of slope-land cropping and terracing Picture 10: Project site in Gansu Province - 34 - Page 38 Picture 11: Low fertility and high erosion, crops highly affected by annual droughts Picture 12: Picture taken by the ICR mission shows project terraces in Pianguan, complete water retention guarantees good yields - 35 - Page 39 Picture 13: Extremely low vegetative cover in pre-project areas Picture 14: Picture taken in the drought year 2001, on the treated areas, where grazing bans were implemented to protect the project works, vigorous natural re-vegetation sustained the drought. Such areas with low survival rates of trees and shrubs after the drought were often replanted using contour bunds and water harvesting structures constructed before - 36 - Page 40 Picture 15 : Erosion in gullies is particularly severe Picture 16 : In addition, damage by uncontrolled grazing prevents any natural re-vegetation - 37 - Page 41 Picture 17 : Picture shows a gully treated under project and demonstrates that erosion in even steep and severely erosive gullies can be tackled with tree and grass plantation to be harvested by local farmers Picture 18 : A strict grazing ban as implemented in this area also brings back a dense vegetation cover usually within 2-4 years, soil erosion can be tackled this way - 38 - Page 42 Picture 19: Project served as model and demonstration area for other national soil and water conservation projects Picture 20: Project provided employment at large scale with a high proportion of women - 39 - Page 43 Picture 21: This picture taken during a supervision mission in 1995 exemplifies the challenge of unsustainable farming and resources use addressed by the project Picture 22: The vision of a sustainable watershed providing income for the local people and protection of natural resources has become reality in many project areas. Picture shows a project sub-watershed in Gansu - 40 - Page 44 The following pictures illustrated method project progress was recorded and supervised using "Maps and Tables" with clearly verifiable project activities all the way down to individual farmer's fields: Picture 23 : Project progress shown on overall watershed Picture 24 : Typical sub-watershed progress map, map for orientation (here 6 sub-watersheds shown) showing treated areas with acreage and year of development Picture 25 : Detail of sub-watershed map with areas under Picture 26 : Plot maps show each field developed different treatment (e.g. terraces, orchards, etc.) under project (e.g. a terraced hill-top, each field numbered) Picture 27 : Corresponding Tables show size of each Picture 28 : Topo-sheets show precise location of field, tables add-up to sub-watershed progress tables, project activities for verification by monitoring and and overall project progress supervision teams, recently GPS is also used - 41 - Page 45 - 42 -