China - Shanxi Poverty Alleviation Project

Document of The World Bank Report No: 29172 IMPLEMENTATION COMPLETION REPORT (IDA-28340 TF-29222) ON A CREDIT IN THE AMOUNT OF SDR 67.3 MILLION (US$100 MILLION EQUIVALENT) TO THE PEOPLE'S REPUBLIC OF CHINA FOR A SHANXI POVERTY ALLEVIATION PROJECT June 25, 2004 Rural Development and Natural Resources Sector Unit East Asia and Pacific Region  CURRENCY EQUIVALENTS (Exchange Rate Effective May 2004) Currency Unit = Renminbi(RMB) Yuan(Y) Y 1.0 = US$ 0.12 US$ 1.0 = Y 8.3 FISCAL YEAR January 1 - December 31 ABBREVIATIONS AND ACRONYMS ACWF All China Women's Federation CAS Country Assistance Strategy ERR Economic Rate of Return FAO/CP FAO/World Bank Cooperative Program FRR Financial Rate of Return ICB International Competitive Bidding IDA International Development Association MOF Ministry of Finance MWR Ministry of Water Resources N.B.F Non-Bank Financed NCB National Competitive Bidding NPV Net Present Value O&M Operation and Maintenance PCD Project Concept Document PLG Project Leading Group PMO Project Management Office PPMO Provincial Project Management Office PRC People's Republic of China PSR Project Status Report QAG Quality Assurance Group SAR Staff Appraisal Report SDR Special Drawing Right TOR Terms of Reference YIS Yuncheng Irrigation System YRCC Yellow River Conservancy Commission WTO World Trade Organization Vice President: Jemal-ud-din Kassum, EAPVP Country Director: Yukon Huang, EACCF Sector Director: Mark D. Wilson, EASRD Task Team Leader: Juergen M. Voegele, EASRD  CHINA SHANXI POVERTY ALLEVIATION 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 12 6. Sustainability 13 7. Bank and Borrower Performance 14 8. Lessons Learned 15 9. Partner Comments 17 10. Additional Information 17 Annex 1. Key Performance Indicators/Log Frame Matrix 18 Annex 2. Project Costs and Financing 19 Annex 3. Economic Costs and Benefits 21 Annex 4. Bank Inputs 23 Annex 5. Ratings for Achievement of Objectives/Outputs of Components 24 Annex 6. Ratings of Bank and Borrower Performance 25 Annex 7. List of Supporting Documents 26 Annex 8. Borrowers Comments and Assessment 27 Annex 9. Pictorial Overview of Selected Project Achievements 28  Project ID: P003649 Project Name: Shanxi Poverty Alleviation Project Team Leader: Juergen Voegele TL Unit: EASRD ICR Type: Core ICR Report Date: June 25, 2004 1. Project Data Name: Shanxi Poverty Alleviation Project L/C/TF Number: IDA-28340; TF-29222 Country/Department: CHINA Region: East Asia and Pacific Region Sector/subsector: General agriculture, fishing and forestry sector (68%); Flood protection (11%); Roads and highways (10%); Water supply (10%); Agro-industry (1%) Theme: Rural services and infrastructure (P); Gender (S); Land management (S); Water resource management (S) KEY DATES Original Revised/Actual PCD: 01/31/1994 Effective: 09/05/1996 09/05/1996 Appraisal: 06/18/1995 MTR: 09/22/1999 09/22/1999 Approval: 03/26/1996 Closing: 12/31/2002 12/31/2003 Borrower/Implementing Agency: CHINA/SHANXI PROVINCE Other Partners: STAFF Current At Appraisal Vice President: Jemal-ud-din Kassum Gautam S. Kaji Country Director: Yukon Huang Nikolas Hope Sector Director: Mark D. Wilson Joseph R. Goldberg Team Leader at ICR: Juergen M. Voegele Juergen M. Voegele ICR Primary Author: Josef Ernstberger 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: HL Institutional Development Impact: SU Bank Performance: S Borrower Performance: S QAG (if available) ICR Quality at Entry: S S Project at Risk at Any Time: No  3. Assessment of Development Objective and Design, and of Quality at Entry 3.1 Original Objective: The Project's main objective was to help alleviate poverty in 20 of the poorest counties of Yuncheng and Luliang Prefectures in Shanxi Province. The Project included a range of activities aimed at raising incomes for about 3 million poor through rehabilitation of an existing irrigation system, improvements of rural roads and construction of village water supplies, land development, expansion of livestock raising and horticulture, and expansion of primary agricultural processing capacity. 3.2 Revised Objective: There was no change in the Project's objective. 3.3 Original Components: The original components included: a) the improvement of the Yuncheng Irrigation System commanding an area of 123,000ha, b) the development of some 13,000ha river beach-land along the Yellow River, c) the construction of village water supply schemes, d) rural roads, e) soil and water conservation including terracing, afforestation and check dam construction, f) horticulture development, g) livestock raising, h) expansion of agro-processing, i) targeted support to women households, and j) project management support and training. This relatively broad mix of components was designed to address short-, medium- and long-term income generating options to achieve the Project objective, and the mix of components took into account the differences in natural resource endowment between the two project areas. Yuncheng prefecture has large areas of flat, fertile land, whereas Luliang prefecture is hilly with few tracts of flat irrigated land. Thus the irrigation and river beach-land development were appropriate for Yuncheng only, while the soil and water conservation component was suited to the upland areas of Luliang. 3.4 Revised Components: No components were dropped or added, however, the relative importance of components was adjusted in response to emerging economic opportunities during the project implementation. When the Project started national food security and grain production was still high on the political agenda, while over time concerns over farm income gained overriding importance. Consequently higher value cash commodities gained importance over irrigated wheat and corn and the respective Yuncheng irrigation component was reduced in favor of the horticulture and livestock components. This process was accelerated by significant improvements in communication infrastructure (such as information systems and roads including a new expressway system), which greatly enhanced the market prospects for high value farm products. This also led to the inclusion of a large number of greenhouses in the horticulture component for production of fresh vegetables. 3.5 Quality at Entry: Quality at Entry was satisfactory. The Project objectives were consistent with the Bank's CAS in China, which called 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 experience and lessons learned from other Bank- and non-Bank financed land and water development projects in Shanxi Province. The Project also fully addressed the Bank's safeguard policies in project design. - 2 -  4. Achievement of Objective and Outputs 4.1 Outcome/achievement of objective: The overall outcome of the Project is rated satisfactory. Most components were completed ahead of the schedule envisaged in the SAR with the main physical targets of many components being exceeded. An estimated 4.1 million people1directly benefited from the Project as compared to an SAR expectation of 3 million beneficiaries. Farmers' incomes per capita in constant 1995 prices increased from Y 480 to Y 1,217, which also exceeds the SAR expectation of Y 700 by far. The development objective of the Project is clearly achieved. The success of the Project was supported by the favorable economic development in the country. However, the broad mix of components in the Project that addressed both short and medium-term income opportunities have proved to be the right choice to take advantage of this supportive overall economic environment. The priority areas such as livestock, high value fruit and vegetable crops, and value added processing together with improved rural infrastructure, proved to be forward looking at the time when this Project was designed ten years ago. It was only much later, when these became priority areas in the restructuring and reorganization of the agricultural sector in China. The Project has timely and flexibly addressed emerging opportunities and challenges from the opening of the Chinese markets and the changing food consumption pattern of a wealthier urban population. Adjustments made during the Mid-term Review have further focused the Project in this direction. The horticulture component was significantly increased, new elements such as greenhouses for fresh vegetable production and new livestock activities like dairy production and deer raising were included in the Project. This made this Project not only valuable by its direct poverty reduction impact in the 20 poor project counties, but the Project also provides lessons for agricultural restructuring and farm income generation in general. The Project demonstrates successful investment opportunities in agriculture using medium- to long-term loans, which can easily be scaled up in other areas. 4.2 Outputs by components: Yuncheng Irrigation System (US$ 17.1 million). This component is rated unsatisfactory, because it was significantly delayed and scaled down due to a lack of counterpart funds. However, the reduced investment effectively solved the irrigation problems of two of the three irrigation districts. The aim of this component was to improve irrigation of some 123,000ha and to reduce pressure on groundwater by providing a reliable water supply from the Yellow River to the existing Yuncheng Irrigation System (YIS). In the 1970s, three pumping stations Jiamakou, Xiaofan, and Zuncun were built to lift water from the Yellow River. These stations serve three independent irrigations areas of 18,400ha, 11,600 ha, and 93,000 ha respectively. The move of the river channel and the high sediment contents caused serious disruption in the water supply. Xiaofan was completely cut off from its water source; the other stations also lost frequently access to the water source and got choked up by sediment and entire sand-banks. This resulted in significant losses by farms incomes, which turned to the use of groundwater causing an over-exploitation of the valuable groundwater resources. The Project was to address these problems by: a) providing a reliable new water intake at Langdian, which is considered a stable site up-stream of the three existing stations and connecting all three stations though a lined and protected feeder canal, b) reducing the costs of sediment clearing by the construction of large-scale sediment ponds along the feeder canal, and c) rehabilitating and expanding parts of the up-stream distribution system. _________________________ 1 The number includes double-counting of people, which benefited from several components. Data about people benefiting from several project components are however not available. - 3 -  The implantation of this component has suffered from a severe shortage of counterpart funds throughout implementation, which has led to the elimination of some elements of the Langdian Complex and several years delay in its execution. The Project has nevertheless achieved to provide reliable water to Jiamakou and Xiaofan with the option to eventually connect Zuncun. A 28 cms pumping station (12, 1.25 cms pumps) and the 13.5 km feeder canal was put in operation in 2002, just in time to take the place of a temporary pumping plant that was no longer operable because of a shift in the river channel. The feeder channel was a major undertaking, since it was built along the active river channel and had to be protected from erosion by heavy revetment. The site of the Langdian station is at a point where satellite images show that the river channel has been stable since 1980. In the case of a major flood event there is a possibility that the main channel could move away from this site. The five amphibious dredges procured for use in sediment basin pond would then be used to maintain an intake channel to the pumps. Furthermore, the new pumps are mounted on floating platforms that can be moved to a new site along the feeder channel if necessary. The sediment basin to protect Jiamakou and Xiaofan was not built because of a shortage of counterpart funds, but there is now a plan to complete this feature by 2005. The 10 km section of the feeder channel from Xiaofan to Zuncun to supply the Zuncun area in part from the Langdian Pumping station has also been deferred because of a shortage of funds. Much of the area served by Zuncun continues to rely on tube-wells. Although the Langdian Complex was subject to protracted delays due to design changes, and a lack of funds, it is now a secure source of surface water for the Jiamakou and Xiaofan units of the YIS, and eventually the Zuncun unit. The main pumping stations for the Jiamakou and Xiaofan units were rehabilitated as planned and the 18,400 ha of Jiamakou and about 6,000 ha of the Xiaofan are adequately supplied with irrigation water. Both systems are also well managed and grow a large proportion of high value crops, such as cotton, or fruit trees. Grain production there takes currently only 8% of the land in Jiamakou and 22% in Xiaofan. A key problem with the implementation of this component was a decreasing political commitment during project implementation, which was clearly related to the overall developments in the agricultural sector in China. In the past, the Yuncheng plain was an important grain basket. This strategic role was still very much seen when the Project was appraised. However, during the mid- and late-1990s concerns over China's national level grain production lost importance. With a high self-sufficiency or even surplus production causing a depression in grain prices, political interest in investment in the irrigation scheme became low. Questions were raised about the appropriateness of using scarce and valuable water resources in a water deficient area for grain production in a country, which had become member of the WTO. These factors have contributed to the difficulties in the mobilization of counterpart funds. Understandably, investment was reduced and focused in those areas, which offered the highest farm incomes producing higher value crops, which was the case in Jiamakou and Xiaofan, but to a lesser extent in the Zuncun area, which is still much more grain oriented. A further concern for the future remains and this is the growing demand for water at all points along the Yellow River. This might in some years cause the Yellow River Conservancy Commission to require Yuncheng and other users to cut back diversions at certain times of the year. River beach development (US$ 16.9 million). This component is rated highly satisfactory. The Project developed some 16,820 ha (SAR target was 13,000 ha) of highly fertile land along the Yellow River (river beach land) in eight counties of Yuncheng. Much of this land did not exist until the 1950's when a change in the river's course left behind flood plains where the sediment gradually accumulated to form new land that was farmed by the nearby villages. Creation and erosion of land is a feature of the - 4 -  Yellow River. This land was eventually protected by flood embankments and about 80 percent of the land is now under crops and plantations, but only 20 percent was originally irrigated mainly by tubewells (the area has ample groundwater resources). The land is leased to the villages on the high ground bordering the river. The Project has provided 1,682 tube-wells and associated electrification and constructed 318 km of roads. This investment yields significant incremental income for some 40,117 farm families. The formerly drought-prone land producing single crops is now double cropped and the cropping intensity is as high as 180% with a large proportion of land also used for high value crops (such as asparagus, or fruit trees). The component also included fishpond improvement and aquaculture supporting activities. The significant wealth generated by this component can be seen in the villages on the high ground bordering the beach-lands, where many new farm houses have been constructed and the local economy is particularly vibrant. Village Water Supply (US$ 11.4 million). This component is rated highly satisfactory. The component addressed the severe shortage of water in many villages in both prefectures. In Yuncheng, wells to produce a reliable supply of good quality water (20-30 m3 /hour) have to be between 150 and 300 meters deep. In Luliang, groundwater can be found in shallow aquifers (10-15 m deep) in the beds of gullies, but these can only be pumped for a few hours at 5-10 m3 /hour. Most villages got their water from pits in the ground (cisterns) filled with storm runoff, and from shallow wells in the beds of gullies. The latter are very low yielding and in the dry season people had to wait for hours for the wells to recharge after the first villagers had filled their buckets. With the villages located at higher elevations close to the plateau the time spent to fetch daily water was often several hours per family. Particularly in Yuncheng water was often sold in the villages by people who had access to a well in their own village. The average family in water-short areas had to spend around Yuan 100 in a year on drinking water - this is about 10 percent of the average family income. The Project successfully addressed the problem of water supply in many villages. In Yuncheng, 79 systems were built supplying water for 350,000 people, 45,000 animals, and 202 enterprises. In Luliang, 307 project built systems supply 120,000 people, 15,000 animals and 150 ha of land is irrigated. The better water supply had far reaching positive implications such as: (a) expanding possibilities for villagers to own draft animals for farming and transportation; (b) reducing the incidence of water-borne diseases; and (c) allowing for watering of seedlings in orchard establishment. Adjustments were made to the original design of this component in terms of: (a) replacing three large systems, financed from other sources, with three smaller systems in more poor and difficult areas in Yuncheng, and (b) including a larger number of smaller systems because of other programs financing the larger systems in Luliang. This has resulted in a cost increase of 24%, while the number of beneficiaries was reduced from an expected 684,000 to 470,000. However, this adjustment was well justified as it took advantage of other financial resources, which only became available during project implementation and focused the project funds in more difficult and very needy areas inhabited by particularly poor people. This adjustment had no negative implications on the economic rate of return. Both, people's willingness to pay and their time spent to fetch water was much higher in the difficult supply areas. Concerns raised by various IDA missions about appropriate O&M arrangements have been addressed by the local PMOs. Most schemes have management boards and collect water fees sufficient to operate and maintain the systems. The total annual fee collection is Y 3.0 million in Luliang, and Y 4.26 million in Yuncheng, which is enough to maintain the systems. In 2003, a full survey was made of all water supply schemes, which showed that over 90% of the schemes are very well managed and maintained. - 5 -  Rural Roads (US$ 27.2 million). This component is rated satisfactory. The component planned to upgrade existing rural roads and tracks to Class 4 Standard for Rural Roads (7.5 meters wide, a 6 m pavement of 20 cm of compacted gravel, and 0.75 m shoulders). The project plan was to upgrade six roads with a total length of 208 km in Yuncheng and five roads totaling 118 km in Luliang. All of the roads planned under this component were built already in the first three years of the Project. The design and construction of the roads was entrusted to the Shanxi Highway Department. The original plan was to build the roads to Class 4 standards. But in view of the expected heavy traffic they were provided with an asphalt surface, and in some cases they were built to higher standards than Class 4. The related cost increase of some 59% was fully covered by additional counterpart resources. The improved roads have: (a) lowered transport costs of inputs and outputs for farmers and the primary and secondary industries; (b) attracted more buyers for farm produce; (c) allowed local products (such as medicinal herbs) access to markets, (d) attracted new bus services, or more frequent services, that expanded access to education, health centers, and opportunities for off-farm employment and education. Road access has led to better market information and cash inflows in the rural areas served. The benefits of the road component have been substantial as revealed by interviews carried out by the monitoring unit as well as the ICR mission. Agriculture has become increasingly mechanized and sideline businesses have increased substantially. Many households own three wheel tractors, and motorcycles are common. A survey shows that the number of cars and small trucks owned by residents along the roads has increased by 3,671. Shanxi Province is rich in mineral resources and coal and ore mines are common in rural areas. Therefore, some roads have attracted frequent trips of heavily loaded trucks. For example the road in Pinglu county carries bauxite trucks that have damaged the surface. In March 2004, the road was closed to trucks to allow the county to repair and resurface the road by June 2004. The burden of road repair apparently falls on the county budget, but taxes on the bauxite mine are a significant contributor to the budget. This road connects 50,000 people to the county town by a short bus trip compared to a 10 hour trip on a dangerous road served by an infrequent bus service. There is a need for the counties to enlist the support in periodic repair and maintenance of those companies whose vehicles make frequent and heavy use of the roads. The original plan was to levy tolls, but that has not so far been implemented. A possibility would be a tax surcharge on the enterprises that already pay county taxes on their output. Soil and Water Conservation (US$ 23.3 million). This component is rated satisfactory. The component was limited to Luliang Prefecture and aimed to implement integrated watershed management on about 20,000 ha. It included terracing on slope land, afforestation on slope land that is too steep for terracing or crop cultivation, and construction of sediment control dams. The terraces were built on 16,107 ha compared to the target of 13,400 ha. The conversion of slopelands to level fields virtually eliminated erosion and helped maintain soil moisture, allowing higher crop yields and more diversified cropping patterns. Most of the terraces were constructed by local bulldozer operators, with some manual labor for shaping the riser and terrace lips. Roads built by the bulldozer operators to reach the areas that they terraced have proved of value to farmers. With better access mechanized soil preparation became possible and transport of inputs and outputs became easier. The standard of terrace construction was very high. The benefits of the terraces were highlighted in a dramatic way during the drought which affected the area in 1999, 2000 and the first half of 2001. On the terraces farmers obtained - 6 -  good yields while non-treated lands suffered complete crop failures. Over 100,000 poor farm households living in some of the most difficult natural environments in Luliang have benefited from project terraces. Afforestation (Chinese pine, larch and poplar) was planned for 4,200 ha of steep slopes and wasteland. This sub-component was particularly adversely affected by prolonged droughts during the course of the Project that caused low survival rates for seedlings. Although replanting took place, the area of afforestation was 1,677 ha compared to the target of 4,200 ha. This component also included plans for the construction of 56 sediment control dams. These are earth dams with heights between 10 and 25 meters built in the gullies. Gross initial storage capacities range between 100,000 m and 1,000,000 m The dams reduce flooding and sediment loads in the downstream rivers, and 3 3 the sediment deposits behind the dams create high quality agricultural land. The reservoir sites are in deeply eroded gullies and are devoid of agriculture and human habitation. During the first few years after construction, the water stored behind the dams can be used for irrigation and village water supplies. The number of check dams was reduced to 18 from the target of 56, because the villagers took the view that they should not be asked to pay for the significant downstream benefits in the Yellow River. Instead, it was decided, to repair 123 existing dams that had good local benefits. Horticulture (US$ 53.7 million) This component is rated highly satisfactory. The component has been the most successful component of this Project with the by far largest contribution to income generation. The original plan was to plant 69,200 ha of fruit and nut trees mainly Chinese date, kernel apricot, prickly ash, chestnut and walnut trees. In Luliang, trees were generally planted on slope land on narrow terraces 1 m wide. In Yuncheng, large quantities of date trees were inter-planted with field crops on flat land. Depending on the tree crop planted, farmers were able to intercrop for 3-7 years before the trees reach maturity allowing for a smooth transition period from low-value grain to high-value fruit production. The choice of varieties and the establishment and management of the plantations drew on successful experience in the past. During project implementation there have been significant changes on the original design. Plantings of fruit and nut crop varieties has been reduced and replaced by large areas of grafted dates on wild date rootstock. The total area reached 73,914 ha compared to the target of 69,200 ha. Much of the intercropped date planting took place in the first two years of the Project. These seedling trees have been planted out in wheat fields where they do not compete much with the grain crop. The introduction of grafting improved scion wood on to wild date rootstock has led to the significant expansion of the date sub-component. Although not introduced into the Project until 1996-97, farmers have been quick to learn the grafting technique and this has allowed for the rapid growth of the date area. Grafted dates did not displace other crops (they usually grown on waste land) and produce fruit from the first year. Their success has led to a sharp reduction in the planned areas for prickly ash, walnut and chestnut as farmers chose a more easily established, managed and faster producing crop. At the mid-term review it was decided to support investment for the construction of greenhouses. By the end of the Project 19,344 greenhouses have been built. The greenhouses are typically constructed with mud walls surrounding three sides and covering an area of about 5 m width and 100 to 120 m length with one of the long sides open to the south direction. Concrete or steel frames support a roof made with plastic cover. Access to the house is through a sidewall that usually has a small living/sleeping/storage area built on to it. Construction costs including labor for a 1 mu house are about Y13,000. The average yearly net profit from a 1 mu greenhouse growing vegetables ranges between Y5,000-10,000. There are risks involved: for example some of the greenhouses were damaged by the torrential rains in 2003 and needed - 7 -  significant investment in repair. Nevertheless, intensive use of land and labor provides farmers with highly attractive financial returns. Greenhouses are usually sited in blocks with as many as 300 in one place. This allows the owners to share the cost of a reliable water supply, usually a deep tubewell, and to benefit from a collective marketing point and extension services. Irrigation is essential for maximizing crop yields and the greenhouses provide an environment for highly efficient water use. Marketing during the main growing season is done on a daily basis with numerous traders directly purchasing crops from individual farmers benefiting from better roads and telecommunications with the markets in surrounding cities in Shanxi and other provinces. Livestock (US$ 23.6 million) This component is rated highly satisfactory. The component includes a beef cattle component (mainly two animals per household) and components for deer and dairy cattle added during the mid-term review. The Project basically helped financing the initial breeding stock, while the farmers themselves constructed the livestock shelter, water trough, silage pits, and other needed facilities. Beef cattle raising proved popular with farm families in both prefectures. The model of one or two breeding cows was adopted by 17,171 families in Yuncheng compared to the appraisal target of 9,400. In Luliang, 4,187 families participated compared to the target of 2,100. The Project also provided for improvements to the facilities and the equipment of about 148 existing veterinary service stations, support for the Shanxi Frozen Semen Center in Taiyuan, and a frozen semen and liquid nitrogen production center in Lishi. The opening of a new dairy processing plant in Yuncheng created an opportunity for the Project to finance 1,000 farm families to purchase dairy cows. This sub-component has been an outstanding success. Household units have been established with in-calf Holstein cows that were within 3-4 months of calving. High quality cows have been procured from sources within and outside Shanxi at Y6,000-9,000 per head. The cows are managed at the household with feed provided on a cut and carry basis. Milking is done three times a day and processed concentrate feed is provided to maximize production. Individual cow production is reported to average between 3,000 - 5,000 kg per lactation. Farmers report a net profit of Y4,000