China - Shaanxi Provincial Highway Project

Document of The World Bank fOR OFFICIAL USE ONLY Report No. 15700 IMPLEMENTATION COMPLETION REPORT CHINA SHAANXI PROVINCIAL HIGHWAY PROJECT (LOAN 2952-CHA) May 15, 1996 Infrastructure Operations Division China and Mongolia Department East Asia and Pacific Regional Office This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.  CURRENCY EQUIVALENTS Currency Unit = Yuan (Y) 1987 (appraisal) $1 = Y 3.72 1988 $1 = Y 3.72 1989 $1 = Y 3.76 1990 $1 = Y 4.78 1991 $1 = Y 5.32 1992 $1 = Y 5.53 1993 $1 = Y 5.76 1994 $1 = Y 8.60 1995 $1 = Y 8.50 FISCAL YEAR January I - December 3 1 WEIGHTS AND MEASURES 1 meter (m) = 3.28 feet (ft I kilometer (km) = 0.62 mile (mi) 1 square meter (m2) = 10.76 square feet (ft2) 1 square kilometer (ki2) = 0.4 square miles (mi2) 1 hectare (ha)= 0.01 km2 = 2.47 acres (ac) = 15 mu I mu = 666.7 m2 = 0.0667 ha I kilogram (kg) = 2.2046 pounds (Ibs) I metric ton (mt) = 2,204 pounds (Ibs) ABBREVIATIONS AND ACRONYMS 8FYP - Eighth Five-Year Plan 9FYP - Ninth Five-Year Plan AADT - Annual Average Daily Traffic DF - Diversion Factor ERR - Economic Rate of Return GOVAI - Gross Output Value of Agriculture and Industry HPDI - Highway Planning and Design Institute ICB - International Competitive Bidding ICR - Implementation Completion Report MOC - Ministry of Communications NPV - Net Present Value PMS - Pavement Management System RDB - Road Data Bank SAR - Staff Appraisal Report SPHAB - Shaanxi Provincial Highway Administration Bureau SPHSDI - Shaanxi Provincial Highway Survey and Design Institute SPTD - Shaanxi Provincial Transport Department TOR - Terms of Reference VOC - Vehicle Operating Cost  FOR OFFICLAL USE ONLY CONTENTS PREFACE ........................................................ iii EVALUATION SUMMARY ........................................................ iv PART L IMPLEMENTATION ASSESSMENT .........................................................1 A. Project Objectives .........................................................1 B. Achievement of Objectives ......................................................... 2 C. Major Factors Affecting the Project ........................................................ 11 D. Project Sustainability ........................................................ 12 E. Bank Performance ........................................................ 13 F. Borrower Performance ........................................................ 14 G. Assessment of Outcome ........................................................ 15 H. Future Operation ........................................................ 15 I. Key Lessons Leamed ........................................................ 16 PART IL STATISTICAL TABLES ..................... ................................... 18 Table 1: Summary of Assessments .................................. ...................... 18 Table 2: Related Bank Group Loans/Credits ..................................................... 19 Table 3: Project Timetable ........................................................ 20 Table 4: Loan/Credit Disbursement: Cumulative Estimated and Actual ............. 21 Table 5: Key Indicators for Project Implementation .............................. ............ 21 Table 6: Key Indicators for Project Operations .................................................. 22 Table 7: Studies Included in the Project ........................................................ 22 Table 8A: Project Costs ........................................................ 23 Table 8B: Project Costs ........................................................ 24 Table 8C: Project Financing ........................................................ 24 Table 9: Economic Costs and Benefits ........................................................ 25 Table 10: Status of Legal Covenants ........................................................ 26 Table 11: Compliance with Operational Manual Statements .............................. 27 Table 12: Bank Resources: Staff Inputs ........................................................ 2 / Table 13: Bank Resources: Missions ........................................................ 28 ANNEX 1: ECONOMIC EVALUATION ........................................................ 29 ANNEX 2: BORROWER'S CONTRIBUTION TO ICR ......................................... 52 ICR MISSION'S AIDE MEMOIRE ........................................................ 61 MAPS: 1. Transport Network Shaanxi Province - IBRD 27778 2. Sanyuan-Tongchuan Highway - IBRD 27779 T ds document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization.   - iii - IMPLEMENTATION COMPLETION REPORT CHINA SHAANXI PROVINCIAL HIGHWAY PROJECT (LOAN 2952-CHA) PREFACE This is the Implementation Completion Report (ICR) for the Shaanxi Provincial Highway Project in China for which Loan 2952-CHA, for $50 million, was approved on June 9, 1988. The last disbursement was on October 28, 1994 and the loan was closed on December 31, 1994. The ICR was prepared by Messrs. Otto Raggambi (Consultant, Highway Engineer) and Han-Kang Yen (Research Analyst) in the Infrastructure Operations Division, China and Mongolia Department of the East Asia and Pacific Region of the Bank, and reviewed by Messrs. Richard Scurfield, Chief, EA2IN and Barry Trembath, Acting Project Adviser, EA2DR. The borrower's implementation evaluation summary for the project was annexed to the ICR. Preparation of this ICR began in February 1995, and an ICR mission visited the Province in July 1995. The ICR is based on materials available in the project file and additional data provided by the project executing agencies. The agencies also gave the Bank copies of their evaluation reports for the highway and rural roads components. Also, they prepared a manual for the operation of the highway and plans to expand and manage the road data bank and pavement management programs.   - lv - IMPLEMENTATION COMPLETION REPORT CHINA SHAANXI PROVINCIAL HIGHWAY PROJECT (LOAN 2952-CHA) EVALUATION SUMMARY Introduction and Project Objectives 1. The rapid development of the Chinese economy from 1980 onward greatly increased road transport demand. Therefore, after the mid-1980s, a major development objective has been to increase road transport capacity on the heavily trafficked segments of national and provincial roads. Another has been to gradually improve the road network and extend it into remote areas. The Shaanxi Provincial Highway Project reflects these priorities, together with the need to build institutions and transfer technology to provincial transport departments. 2. The project's primary objectives were to: (a) expand road transport capacity in the heavily developed corridor between Xi'an, the provincial capital, and the important industrial center of Tongchuan; (b) repair roads damaged by floods and improve the condition of some provincial roads in one of the less developed regions; (c) introduce a data bank and management system to guide road investment decisions; and (d) train staff and acquire equipment to transfer advanced technology. 3. The objectives reflected the physical and developmental needs of the Province's highway subsector, the central Government development strategy, and the Bank assistance strategy for the subsector. Implementation Results and Experience 4. The project's key objective, the construction of a new highway between Sanyuan and the northern end of Tongchuan, was achieved, although at a higher cost than estimated and, according to the reevaluated Economic Rate of Return (ERR), with a lower yield: 18.1 percent versus the SAR estimate of 27.8 percent. The design changes introduced during implementatioh, in aggregate, have contributed to the higher cost of construction. However some of the changes provide longer-term operational advantages, such as the provision of controlled access along the full length of the highway. Some other changes had the risk of introducing traffic capacity constraints by narrowing the designed width for a section of the highway from four traffic lanes to two traffic lanes. However, the adverse impact of this change has been greatly reduced by the slower than  _v_ expected build up of traffic on the highway, and by the measures taken to complete the construction of the omitted two traffic lanes by the end of year 2000. 5. The Rural Roads program is seen as a well-conceived component that was, after some delay, successfully implemented, achieved its anticipated impact on economic development, contributed to higher income levels, and increased mobility for the communities affected. 6. The equipment acquired, with some minor exception, was appropriate for the intended purposes. The procurement process, although delayed and perceived as cumbersome by the executing agencies, was accomplished according to Bank guidelines. 7. The staff training program differed somewhat from its original plan. Training and study tours abroad dominated, and the number trained was well below original estimates. Thus, the cost of training per person-month was substantially higher than envisaged. While it is difficult to measure the impact of training, most of the courses were regarded as beneficial. Nevertheless, planning, preparing and executing future training will need to be done more carefully to ensure it is more cost-effective and that more staff participate. S. The collection and storage of road data for highway planning progressed well within the pilot Road Data Bank (RDB) development and application exercise. So too did the Pavement Management System (PMS) for guiding road investment decisions. Still, both programs fell short of expectations, which, it seems, were over ambitious. Summary of Finding., Future Operations and Key Lessons Learned 9. The project was implemented satisfactorily and as planned, except for some design changes, which, overall, had adverse impacts on the cost of the highway but had either longer term operational advantages or the risk of trafflc capacity constraints, which, because of follow up actions, has now been substan.! ally reduced. Further, sufficient evidence already exists that the project's development objectives were substantially achieved and that the condition of the project roadsi and benefits (indicated in the ICR) can be sustained and are unlikely to experience substantive risks. WO. The condition of the investment-related components, such as the Sanyuan- Tongchuan Highway and five roads improved and built, is expected to be preserved through regular maintenance. Further, to help ensure the highway operates efficiently, the Shaanxi Provincial Transport Department (SPTD) prepared an Operations Manual to regularly monitor key performance indicators. The RDB and PMS programs, supported by the project, are also expected to be sustained and even broadened (to cover the whole Province) according to an Operation and Supervision Plan that SPTD recently produced. The impact of technology transfer through acquisition of modern equipment and staff training should also benefit institutional development for many years.  - vi - 11. The following lessons are relevant to future Bank-financed highway projects in China: * Design changes were needed under this project due to inadequate location studies. Therefore both land use and the development potential along major new highways need to be adequately assessed in location studies to avoid design changes that could be costly and disrupt the implementation of works; * The contractors' experience and available resources were not thoroughly evaluated during the prequalification period, causing project delays and adversely affecting its quality. Thorough contractor assessments are therefore needed in the prequalification process; * Chinese and Bank costing procedures for civil works are different, and agreeing to a common method would provide more reliable and consistent cost and ERR estimates for highway investments. The adoption of similar procedures is advocated; and * The staff training program, which incurred high person-month costs and had lower than expected participation, highlighted the need for a thorough review of training under Bank-financed highway projects.   - 1 - IMPLEMENTATION COMPLETION REPORT CHINA SHAANXI PROVINCIAL HIGHWAY PROJECT (LOAN 2952-CHA) PART I. IMPLEMENTATION ASSESSMENT A. PRojEcT OBJEcrIVEs 1. China joined the World Bank Group in 1980 and lending operations began soon after. However, B'nk involvement with highway infrastructure began only in April 1983, with the identification of the first Highway Project (Ln. 2539-CHA/Cr. 1954-CHA). That project provided funding to nine provinces to improve provincial roads and fill in the critical missing links, particularly those which constrained inter-provincial road transport. Although the involvement was limited, this initial step helped familiarize the Bank with the highway subsector issues in China and identify subsectoral priorities in subsequent projects. 2. The Bank's major objective for the highway subsector was to provide additional capacity on heavily congested sections of primary provincial and inter-provincial highways. After the first Highway Project, the lending program expanded rapidly, with five highway projects between May 1987 and May 1989 (see Table 2, Part II). 3. The objectives of the Shaanxi Provincial Highway Project, approved in June 1988, were to: (a) expand road transport capacity in the heavily developed corridor between Xi'an, the provincial capital, and the important industrial center of Tongchuan; (b) repair flood-damaged roads and improve selected roads in an economically less-developed region; (c) introduce road planning methods by creating a road data bank and a road management system to guide investment decisions; and (d) train staff and acquire equipment in order to transfer advanced technology to the subsector. 4. In retrospect, it appears that project objectives were generally realistic, well articulated and understood by the borrower and the project executing agencies. Further, they were consistent with both the Government's strategy and the Bank's assistance strategy for developing the road transport subsector. Although the project was not unduly complex some unforeseen design changes in the highway component contributed to cost increases, which combined with less than forecast level of traffic have adversely effected the economic rate of return.  -2 - B. ACHIEVEMENT OF OBJECTIVES 5. Project objectives were substantially achieved as indicated by the following review of various components. Construction of the Sanyuan-Tongchuan Highway 6. The 66.4 km highway was divided into two segments: a 51.3 km four-lane divided Class I standard highway between Sanyuan and Chuankou, a town just south of Tongchuan, and a 15.1 km two-lane Class II standard road between Chuankou and Houliu, a town just north of Tongchuan. The construction was considered a logical extension of the 34 km Xi'an-Sanyuan highway, which was completed under the first Bank-financed highway project. It was further justified because: (a) the existing road could not handle projected traffic growth; (b) traffic congestion and accident rates were exceeding tolerable levels; and (c) the economic evaluation indicated the proposed investment would yield an ERR of about 27.8 percent. 7. Engineering Design. The Shaanxi Provincial Highway Survey and Design Institute (SPHSDI), under the Shaanxi Provincial Transport Department (SPTD), designed the project. It was reviewed and cleared by the Ministry of Communications (MOC) as well as the Bank. Nevertheless, during implementation, a number of design changes were needed, including the following: (a) At project preparation, the Shaanxi Provincial Archeological Relics Department endorsed the construction of a four-lane highway near the archaeologically sensitive area situated between Yaoxian and Chuankou. However, further investigations shortly before construction revealed that the area containing the relics (Yaozhou Kilns) was more extensive than had been previously identified. In the interest of limiting the risk of interfering with the archeological relics, the Relics Department approved the construction of only two lanes of the originally designed four-lane, double carriageway highway across a 2 km long protection zone. (b) Although the original design was accepted by Yaoxian County in the project preparation phase, concerns emerged about the building of an embankment along the eastern outskirts of Yaoxian township because of its severe impact on land development: The original scheme would have required several industries to be relocated and 2,000 residents to be resettled. At the County's request, the original design of a four-meter high embankment segment of the four-lane highway was changed to a two-lane wide elevated structure for 2.7 km. (c) Originally, it was not envisaged that the highway would be built as a limited access toll highway. However, based on drivers' behavior and accident records for the Xi'an-Sanyuan section of the highway (opened in late 1989), the advantage of changing the new highway between Sanyuan and Chuankou (51.3 km) into one with limited access was clear. Thus, the  - 3- decision was made to invest in the construction of six interchanges, 10 structures to eliminate at-grade crossings, and fences for the highway along its length. This conformed with the Government's policy to build new high-class toll highways exclusively for motorized traffic in the interest of highway safety and cost recovery. 8. Due to the effects of the design changes described in para 7 (a) and (b), which restricted the width of the highway on two sections between Yaoxian and Chuankou, the substantial increase of cost associated with the design changes indicated in para 7 (b) and (c), and some constraints on SPTD's budget at the time, SPTD decided to build the 20.3 km section of the highway between Yaoxian and Chuankou as a two-lane Class II road, as an interim measure, instead of a four-lane Class I highway as originally intended. The Province was aware of the potential bottleneck created by the reduced highway width and moved ahead with the design of the second carriageway, the construction of which is now included in the Second Shaanxi Provincial Highway Project, which was approved by the Board in March 1996. Because less traffic has been diverted from the existing road to the new highway than expected, particularly on the section between Yaoxian and Chuankou [para. 16 (b)], the fact that only a two-lane width was provided for that highway section under the project has had no adverse impact on traffic operations so far. Moreover, with the scheduled completion of the second carriageway by the end of 2000, it is unlikely that capacity constraint would develop in the interim period. 9. The change to a limited access highway was expected to be highly beneficial in terms of smoothing the traffic flow, eliminating accidents with non-motorized traffic and pedestrians, and reducing the risk of accidents with motorized traffic. Although this section of the highway has only been open since 1993, it appears that these expectations are being realized. There were a number of other design changes with less impact on costs and traffic operations, for which the contractors received complete work variation orders. 10. Implementation and Supervision. Civil works were procured as agreed at negotiations. The only difference was minor and, in the Bank's view, positive: An 18 km section of the highway that passes through an archaeologically-sensitive area was awarded to local contractors after local bidding procedures under four contracts, instead of being executed through force account. The change reflects the Province's reorientation to competitive contracting practices. Highway construction was originally scheduled to start in February 1989 and be completed in 24 months, at the end of January 1991. However, the major contracts began in April 1990, after a 14-month delay; and, due to substantial design changes, some started only in March/April 1991. The first 31 km section of the highway (between Sanyuan and Yaoxian) was substantially completed and opened by September 1992, and the full highway was opened at the end of 1992-almost two years later than scheduled. The still-outstanding civil works were completed during 1993; delays were caused by slower-than-expected progress with pre-construction activities, particularly procurement, which was held up for about 10 months. Further, the contract period for larger contracts was increased from the original 24 months to 30 months.  - 4 - 11. Construction was generally satisfactory, although some problems occasionally occurred: (a) early in the implementation, the contractors could not get access to some sections of the highway which delayed the commencement of works and, later, some obstructions that remained in the highway alignment created work stoppages -- both problems were resolved with the local governments' help; (b) although contractors for the major contracts had been pre-qualified, some lacked the experience, resources and managerial skills to execute the works efficiently, and their shortcomings adversely affected the works' progress and quality; (c) some subcontractors had to be removed because of unsatisfactory performance; (d) the highway crosses terrain in predominantly "loess" soil, where landslides periodically interrupted the effort; and (e) the contractors' lack of experience, and the occasional use of poor quality machinery affected the quality of the paving on some highway segments. 12. Highway construction was supervised by a joint team of foreign and local consultants and provincial technical personnel. Intensive initial training in supervision and quality control measures helped overcome the local staffls lack of experience in supervising the construction of high-class highways. After initial shortcomings, the team became more effective in performing quality and environmental protection control, and it is to its credit that, by rejecting poor quality work and guiding the contractors, the highway was completed to a generally satisfactory technical standard, although with some deficiencies in the riding quality of the pavement on some sections of the highway. 13. Environmental Protection. During the highway design and construction, attention was paid to environmental issues, including roadside drainage and slope protection. However, because of the type of terrain and the soil, the drainage systems are still deficient on several highway sections and the risks of erosion and water pollution are substantial. Where possible, the embankment slopes and excavated areas were protected with grass and other vegetation. However, additional and, for some of the areas, repeated planting will be need before the slopes can be stabilized and protected against erosion. Unfortunately, the soil along the highway is such that the risk of slope failures and erosion cannot be fully eliminated. The High-Class Highway Administration Bureau of SPTD, which is responsible for operating and maintaining the highway, is aware of the deficiencies and environmental risks involved, and a program was prepared to continue the necessary improvements and maintenance. Regular reporting on protection and maintenance activities will be made according to the requirements in the highway Operations Manual (para. 46). 14. Land Acquisition and Resettlement. Although land acquisition, resettlement and compensation procedures agreed upon at negotiations were followed and land area requirements changed only marginally, the cost of land acquisition and compensation increased from Y 34.55 million to Y 67.73 million. The substantial interval between the time when the original estimates were made and the time payments were actually made contributed to the almost 100 percent increase in compensation. Also, some farmers were dissatisfied about interference with irrigation facilities, access routes to farmland and other services. But, local governments helped SPTD resolve these issues satisfactorily. The number of persons resettled remained at about 1,400, as originally surveyed. The Bank's  - 5 - Resettlement Review Mission visited Shaanxi Province in November/December 1994, and, after assessing the resettlement process, confirmed that actions were carried out satisfactorily and in line with Bank policies and procedures. 15. Construction Costs. SAR cost estimates of the highway civil works were Y 228.30 million, including physical and price contingencies, while the final cost, based on certified payments to the contractors, was Y 372.75 million, an increase of about 63 percent. Of the increase, 58 percent was due to greater-than-expected physical quantities and extra work carried out through variation orders issued to the contractors, and 42 percent was due to price escalation being greater than assumed at appraisal. Although savings were gained by eliminating one of the carriageways, most were absorbed by the Y 27.6 million additional cost of the 2.7 km elevated highway (which was constructed instead of the earth embankment). The conversion of the highway to a limited-access toll facility cost about Y 30.0 million. 16. Economic Evaluation. The ICR economic evaluation is presented in Annex 1. The ICR estimate of the ERR for the Sanyuan-Tongchuan highway is 18.1 percent compared to the SAR estimate of 27.8 percent. The re-estimation of the ERR was based on final construction costs, revised indices for shadow pricing, actual corridor traffic data and the split of traffic between the existing arterial road and the new highway from 1993 to mid-1995, revised traffic forecasts from mid-1995 onwards, and updated vehicle operation costs (VOC). In general, the following explain the lower ERR: (a) The economic cost of the highway in the ICR, at July 1995 constant prices, is about 58 percent higher than the SAR estimate. (b) While traffic growth in the corridor between 1985 and mid-1995 was only slightly lower than the SAR forecast, the traffic diversion rate from the existing arterial road to the new highway was significantly lower than estimated, for the following reasons: (i) At appraisal, the highway was not expected to be operated as a toll facility, and no allowance was made for the impact of tolls on traffic diversion. (ii) The arterial road that parallels the new highway was upgraded during 1992-94. Its pavement was widened from about 7 m to 12 m, providing additional capacity and therefore reducing the attractiveness of the new highway. (iii) The VOC saving related to the shorter distance of travel on the new highway compared to the existing arterial road is, according to actual data, less than was assumed in the SAR. A closer analysis of the traffic pattern has revealed that, for other than long-distance through-traffic, the highway does not provide travel distance saving, which normally is regarded as an important incentive for diversion even for relatively shorter trips. The reason for this is that travel on access roads from population centers to the new highway adds to the overall travel distance, but was not considered in the SAR economic evaluation. Longer distance of travel even for shorter trips is also a disincentive for diverting to the new highway. (iv) It is generally recognized that utilization of new highways, particularly toll  - 6 - facilities, increases gradually (called the "ramp period" in recent technical literature). Therefore, the ICR forecast for traffic growth, which is based primarily on recent trends, could be regarded as more conservative than that indicated in the SAR. 17. In the SAR the highway investment cost used at arriving at the ERR was based exclusively on the estimated cost of civil works. To maintain comparatibility of the economic evaluations between the SAR and ICR, the ICR economic evaluation is also based on the final cost of civil works, amounting to Y 372.75 million (in current prices). A more common methodology is to include in the highway investment cost other well identifiable cost items, such as the cost of land acquisition and resettlement, construction supervision and possibly some others. When the final cost of land acquisition and resettlement and construction supervision is taken into consideration at Y 67.73 million and Y 6.0 million (Table 8A, Part II) respectively, they represent about a 20 percent increase in the economic cost, which, according to the sensitivity analysis, lowers the ERR for the highway investment to 16.1 percent. Arriving at more reliable estimates for highway investment cost, and thereby a better estimate for ERRs, is handicapped by the substantive difference between the Chinese and Bank approaches to the costing of highway investments. This is an area warranting close coordination between China's Ministry of Communications and the Bank in the interest of improving costing procedures and arriving at more reliable economic evaluation results for highway investments. Rural Roads 18. Originally, the Rural Roads program comprised the following: (a) Improved road sections: Shangnan-Yunxi (105 km), Shangxian-Zhashui Road (130 km), and Pingli-Zhenping (60 km); (b) Repaired flood damage: Shangxian-Shangou Road (101 km) and Sehe- Manchuan Road (73 km). 19. All five roads form part of the key arterial network and link seven counties/cities in Shangluo and Ankang prefectures, which, with a population of about 350,000, represent one of the poorest regions of Shaanxi Province. Lack of adequate transport infrastructure was found to be one of the major obstacles to economic growth in this mountainous region. The rationale for expanding and rehabilitating the road sections was that this could improve transport service and contribute to higher income levels and alleviate poverty. 20. Preparation of the program was based on the assessment of road network needs by Shaanxi Provincial Highway Administration Bureau (SPHAB) and SPHSDI. A feasibility study for a rural roads improvement program and preliminary engineering design for the selected roads were completed at the end of 1987. The engineering design was produced by prefectural and provincial design institutes, and was based on local experience and aimed for least-cost solutions obtained by applying appropriate technology and using locally available equipment, materials and man-power.  - 7 - 21. Some change in the program's scope occurred after SPTD's proposal in early 1990 to construct a new 50 km road link between Langao and Zhenping, instead of improving the Pingli-Zhenping Road, as originally planned. Based on the explanation for the change and the economic evaluation supporting the viability of a new road, the Bank accepted SPTD's proposal. 22. Implementation and Supervision. The repair of the flood damaged roads began in early 1988 under retroactive financing as originally scheduled, but was completed only in December 1991 with a delay of about 24 months. The delay was largely due to exceptionally heavy rainfalls during the fall of 1988 and the summer of 1989. The improvements of the Shangnan-Yunxi and Shangxian-Zhashui Roads commenced in mid- 1988 and were completed in March 1992 with 15 months delay. For both groups of roads, because of the damage caused by heavy rains, some of the already completed works had to be redone. The construction of the Langao-Zhenping road, under two contracts, started only in April 1990, was completed at the end of 1992 with only 3 months delay. 23. The mountainous terrain and adverse weather conditions contributed to implementation difficulties, as did shortages in equipment, which required major reliance on the labor force. However, particularly in the harvesting seasons, there was shortage even in manpower. Although with substantial delay, the Rural Roads program was satisfactorily completed, and as confirmed by Bank staff visiting ongoing works, construction quality was commensurate with the technology adopted, which satisfied the technical requirements for the designated road standards. 24. The construction was supervised by teams of technical personnel from the respective county/city transport units who operated under directives of the SPHAB and the prefectural transport bureaus. Since regular on-site supervision had not been practiced before on similar construction jobs in this part of the Province, it took some time for staff to become familiar with contract management and quality control procedures. However, during implementation, in addition to the experience gained on the job, a group of local technical personnel were able to obtain formal training in relevant fields to enhance the effectiveness of quality control. 25. Construction Cost. The SAR's cost estimate for the Rural Roads program was Y 159.1 million, including physical and price contingencies. The aggregate amount of the seven contracts was Y 105.57 million, while the final cost, including additional works and price increase, was Y 122.70 million, representing an increase of about 16 percent over the original contract but about 23 percent below SAR estimates. This lower cost could be explained by the economical approach to design and strict budget control during construction. 26. Economic Evaluation. As detailed in Annex 1, the ICR estimate of the aggregate ERR for the Rural Roads program is 41.5 percent, varying between 21.1 percent and 99.2 percent for individual roads. An overall comparison of ERRs for this program cannot be made between the SAR and ICR estimates since economic evaluation in the SAR was presented only for the improvement of three roads, one of which was deleted. For the  -8 - other two roads, ERR estimates are similar, for one road modestly higher and for the other modestly lower: (a) for the Shangnan-Yunxi Road (105 km), SAR and ICR estimates of ERR are 18 percent and 24.5 percent, respectively, and (b) for the Shanxian-Zhashui Road (130 km), estimates are 25.0 percent and 21.1 percent. 27. The Rural Roads program greatly improved the condition of key arterial routes that now provide all-weather motorized vehicular access to communities in the region. Improved transport services contributed to the accelerated development of the regional economy and are expected to foster the exploitation of mineral resources and transport of farm products. The execution of the works created job opportunities in nearby communities. Also, the standard of living showed some improvement: Annual income per farmer increased from Y 164 in 1988 to Y 569 in 1994 and it is reasonable to assume that part of the increase was due to better access to road transport and increased mobility in the region. Equipment Procurement 28. The following items were purchased at an estimated $2.4 million, including physical and price contingencies: (a) Laboratory equipment, training tools and transport vehicles for vocational technical training schools ($860,000); (b) Instruments and equipment to evaluate road conditions ($290,000); (c) Equipment for RDB (Road Data Bank) and PMS (Pavement Management System) programs ($1.25 million). The final cost, $2.1 million, was less than appraised, with items (a), (b) and (c) being $1.32 million, $0.57 million and $0.21 million, respectively. 29. As stipulated in the loan agreement, about 60 percent of the equipment was to be procured through international competitive bidding (ICB) procedures. However, since the items to be purchased were relatively inexpensive and available off-the-shelf, the Bank waived the ICB requirement and agreed that all these items could be procured through international and local shopping. 30. The equipment acquired for SPTD's vocational-technical schools has been very well used. In recent years, about 2,600 technicians and 500 drivers were trained, and they greatly benefited from the new acquisitions. 31. The instruments and computer hardware and software chosen to help set up the RDB were also satisfactory, although the selection process was difficult because of uncertainty about whether the items available on the market would operate satisfactorily under Chinese conditions. As it turned out, they met all expectations. 32. However, the same cannot be claimed for the equipment chosen to measure road conditions. Some of the units still have operating problems that stem from the complex  - 9 - nature of the equipment and calibration problems. Also, the data output sometimes differs from Chinese standards for which no appropriate adjustment has yet been made and it appears the training for operators has not been very successful. Staff Training Program 33. The objectives of the staff training program were to: (a) improve the proficiency of engineers and others by sending some abroad for academic and practical training as well as international conferences; (b) strengthen the capacity of highway institutes and SPTD's technical schools by providing equipment, training aids and books, and (c) invite foreign experts to help upgrade the curricula. It was anticipated that the $400,000 budget be split for (a) English language training for 30-40 persons at $20,000; (b) professional training abroad, including study tours, of 10 SPTD officials, which amounted to 30 person-years at $280,000 and (c) seminars organized in China with the involvement of foreign exerts, at $100,000 (see Attachment 2 to the Minutes of Negotiations on April 27, 1989). 34. Soon after the project began, it was recognized that organizing training courses on various topics for a relatively small group of staff from just one provincial transport department was neither practical nor economic. At the Bank's initiative, and after consultation with MOC and other provincial transport departments, an understanding was reached in late 1989 that most of the staff training under ongoing Bank-financed highway projects would be coordinated by MOC. Accordingly, MOC prepared and administered training programs, sharing cost with the provinces receiving Bank assistance for highway projects being implemented more or less concurrently with the Shaanxi Provincial Highway Project. However, it was also agreed that some training programs could be arranged independently by any of the provincial transport departments for their own staff. 35. The training included 12 courses and study tours and involved 53 staff, for a total of 80.5 person-months at $505,000. Eleven of the 12 were coordinated by MOC, as outlined above. Training was offered on: (a) RDB fundamentals and operations as practiced abroad; (b) surveys of road conditions and evaluation methodologies for pavement; (c) the preparation of medium- and long-term road development plans for arterial and rural roads; (d) project preparation involving evaluation methods; (e) project management, including financial aspects; (f) advanced highway design, including computer applications; and (g) administering contracts and supervising construction. These areas of training were well chosen, and provided SPTD staff with both immediate and long-term benefits. Training for supervising construction involved some courses abroad (for a total of 928 person-months) and was carried out separately from the program under the foreign consultant services contract for construction supervisors. There were also extensive classroom and on-the-job training programs for local supervisors at the time the works were being constructed. 36. While the MOC-coordinated training was expected to be more economic than separate training for each province, the actual cost of $6,270 per person month is far higher than the SAR estimates of 30 person-years for $280,000, or an average of $780 per person-month. This seems to be attributable to the fact that contrary to expectations that some courses would be conducted in China, almost all were abroad, which greatly increased travel and subsistence  - 10- costs, and cost of consultants for organizing and facilitating study tours. In a few instances, courses were arranged through state road authorities and were less expensive. While these high costs point to the need to explore other ways to conduct training, the program for SPTD staff directly exposed them to advanced technologies abroad, which may have been more successful than classroom type presentations. Programs for Road Planning, Pavement Strengthening and Maintenance 37. When the project was prepared, it was thought the initial programs for data collection in the Province (to improve methods of road planning and practices of strengthening and maintaining pavements) would be part of a study whose objectives included: (a) establishing a road data base compatible with national guidelines that would include an inventory of permanent features and current conditions of existing paved roads; (b) beginning regular monitoring of current conditions of paved roads; (c) identifying pavement overlay and strengthening needs and resource requirements for about 7,500 km of paved roads; and (d) implementing a pilot study for planning and managing maintenance, compatible with national guidelines, so as to develop a systematic approach in this area. The study was expected to take 15 months and involve about 80 person- months of local and foreign personnel input (see the TOR in the SAR Annex 2). 38. However, the study was not carried out as planned because it was overtaken by central Government policies which instituted a nationwide approach to road planning, pavement strengthening and maintenance. A group was set up under MOC, headed by its Director of Highway Administration, to guide and coordinate the programs. MOC had already began to plan a nationwide RDB program in 1985 in each province and autonomous region of the country; the idea was to use the data to improve highway planning and project formulation, as well as for statistical purposes. The RDB framework was developed in MOC's Highway Planning and Design Institute (HPDI) in the late 1980s; the PMS began in MOC's Highway Scientific Research Institute, also in the late 1980s. 39. By the early 1990s, reasonably good progress had been made with both the RDB and the PMS programs. Therefore, MOC and the Bank agreed that it would be more efficient and consistent to develop local capability for both of these systems, rather than approach them separately in each of the provinces with external assistance. To introduce the PMS, 15 provinces/municipalities were selected, which included Shaanxi Province and the following nine provinces, that had Bank-financed highway projects scheduled for implementation during the Eight Five-Year-Plan (8FYP) and 9FYP periods: Hebei, Shandong, Jiangsu, Guangdong, Jiangxi, Zhejiang, Sichuan, Xinjiang and Henan. The Bank supported MOC's endeavor, and the highway projects helped finance the acquisition of equipment, including computer hardware and software, instruments for land-based land surveys and geographic information systems. Road Data Bank Development and Application 40. Once the RDB system was designed by HPDI, Baoji Prefecture in Shaanxi Province was selected to carry out a pilot study at prefecture level. The system was designed to operate at the municipal/prefectural, provincial, and state levels, designated as third, second and first  -11- levels, respectively. With the help of computer hardware and software and instrumentation provided under the project, SPTD was able to create a data base comprising 490 items and complete the Baoji pilot exercise satisfactorily. SPTD introduced the third level RDB system to four other prefectures/municipalities during 1995; the model can be replicated in other provinces. The pilot study was highly successful, largely through the use of local talent and resources. The Bank's contribution to it included periodic advice by consultations with Bank personnel on some technical aspects, acquisition of equipment, including computer software and hardware, and an extensive RDB study tour enabling a group of Chinese staff to be exposed to techniques and methods abroad. Pavement Management System Development and Application 41. Shaanxi is one of the pilot provinces included in MOC's program to introduce the PMS. Preparations began in February 1991, foUowed by training courses in December 1991 and August 1992 in Shandong Province, and coordinating sessions with MOC in Beijing in January 1993. The PMS was applied in Baoji prefecture in early 1993 with the collection of key road condition data for about 840 km of roads. This effort succeeded and the prefecture accepted the program as an effective planning tool. The PMS-generated outputs are being used to help authorities make investment decisions for road improvement and maintenance in the framework of the budgetary process. By the end of 1995, the system was extended to cover five of the Province's 10 prefectures/ municipalities, that had 4,754 km of paved roads, or 54.3 percent of Shaanxi's 8,715 km paved arterial roads. Extending the system to the remaining network will continue during the next three to four years. The PMS is administered by SPTD along with SPHAB's prefectural and municipal branches. 42. Setting up the RDB and PMS organizations, obtaining equipment and computer programs, and gaining experience using them in Shaanxi Province took longer than the 15 months originally assigned in the SAR. Also, because the concepts were so new, time was needed for those involved at the various levels of provincial and prefectural transport organizations to recognize the programs' usefulness. Thus, despite the progress in Shaanxi, more effort, budgetary support and overall-acceptance of using RDB and PMS output for planning and budgeting purposes will be needed if the programs are to expand and cover the entire provincial road network. In this respect, the Province's intention to expand the programs and take other measures in SPTD's Operation and Supervision Plans should help sustain the RDB and PMS (para. 49). C. MAJOR FACTORs AFFECTING THE PROJECT 43. Project implementation did not suffer from major adverse effects. No extraneous major factors on the national level affected the project, other than inflation, which reduced the budgetary resources of the project executing agencies. Budgetary constraint, in addition to the physical obstructions, had influenced the Province's decision in reducing the width of the highway from four lanes to two lanes between Yaoxian and Chuankou. Also, central Government sectoral policy affected the RDB and PMS development and implementation, but the change in approach from that outlined in the SAR was considered to be appropriate. Also, MOC's decision in early 1990 to establish a Technical  - 12 - Coordination Unit to liaise between the Bank and the provinces greatly helped implement this and other Bank-financed provincial highway projects. 44. The fact that the project was an expansion of one undertaken earlier (in part in the same Province) helped SPTD administer it in a relatively smooth fashion. Key personnel were reasonably familiar with the Bank's policies and procedures, and no measurable conflict or delay emerged on this account. However, the major design changes needed for the Sanyuan-Tongchuan Highway point to the lack of adequate land use planning and coordination among provincial and local government entities during the highway alignment location studies. 45. The SAR expected the project to be completed by the end of 1993 and this target has been substantially achieved, although some components were delayed when compared to the detailed schedule in Annex 3 of the SAR. The start of construction of the Sanyuan- Tongchuan highway was delayed by about two years, and instead of two years it took about three years to complete the works. The delay occurred mainly in finalizing the bidding documents and during the procurement process. The schedule for procuring equipment could be seen as overly optimistic in light of the executing agency's limited experience with Bank's procurement requirements and the country's internal clearance constraints. Also, the construction period was longer than planned, partly because of design changes. The Rural Roads program was completed almost two years later than expected, mostly because of adverse weather conditions. However, the SAR made allowance for such delays by setting the completion date of the project as December 31, 1993, and the loan closing for December 31, 1994; in addition, after delays in the earlier years, disbursements caught up with the SAR schedule in FY92, and remained on target thereafter (Table 4, Part II). D. PROJECT SUSTAINABILITY 46. Provided the roads are properly maintained the benefits generated by the project's civil works components are expected to be sustainable. Although the executing agency for the Sanyuan-Tongchuan highway component was not obligated to provide an operational plan as part of the ICR, it is an indication of SPTD's commitment to the sustained and satisfactory operation of the highway, that it has prepared an Operation Manual for monitoring the operational features of the highway, as suggested by the Bank. The Manual contains 22 monitoring formats with which to gauge the highway's operational performance. 47. The sustainability of the rural roads will depend largely on regular and appropriate routine and periodic maintenance for these roads, using resources earmarked from the road maintenance funds. Nevertheless, because the roads are located in a mountainous region and exposed to frequent and severe rains that can cause extensive damage and interrupt traffic, emergency maintenance funds might be needed periodically to restore the roads expeditiously. 48. The equipment procured for the maintenance and operation of the highway had been carefully selected to meet quality and up-to-date technology requirements. Local  - 13 - personnel had been trained to operate the units, and spare parts and maintenance facilities are available to ensure that the units will provide many years of service. Because of rapid technological advances, some of the computer hardware and software may succumb to early obsolescence, and will need to be upgraded or replaced. The operational problems encountered with the road condition survey equipment (para. 32) are expected to be overcome as technical staff and operators become more experienced with the handling of the units. 49. The Province is aware of the need to consolidate the accomplishments of the RDB and PMS programs; at the request of the Bank, SPTD prepared Operation and Supervision Plans to help ensure the upkeep and further expansion of both, in the Province. 50. The key areas of training (para. 35) were selected according to priorities for transferring technical know-how to the trainees and their work units. Traditionally, Chinese engineers have been very receptive to advanced technologies and have the qualifications and perseverance to apply most of the new methods. Also, the Chinese style of working in teams is particularly conducive to spreading new ideas, even though only a relatively small number were trained. E. BANK PERFORMANCE 51. The project was consistent with the Government's development objectives, the Province's priority to invest in the subsector, and the Bank's country assistance strategy. In general, the identification of the components and the preparation, appraisal and supervision of the project were satisfactory, although through hindsight, the following weaknesses have been noted: (a) The economic evaluation omitted the inclusion of justifiable investment components, such as the cost of land acquisition and resettlement, and construction supervision. Furthermore, the assumptions made about the diversion of traffic from the arterial road to the new highway turned out to be overoptimistic, although the relatively low diversion factor partly emanates from the tolling of the highway which was not considered at the time of appraisal. In addition to the higher than estimated cost of construction, these two factors contributed to the lower than estimated ERR for the highway. (b) The staff training program could have been prepared in greater detail, but more importantly, during supervision the Bank could have been more alert to the fact that more training was supposed to be organized in China for more trainees. This could have reduced the relatively high cost of training and study tours abroad for a relatively small number of participants. (c) The expectation that road data information and pavement management systems could be established in 15 months, without SPTD's management  - 14 - and staff having had previous experience in these areas, and considering the complexity of the study's objectives, was overoptimistic. 52. During implementation, the Bank's involvement was adequate and commensurate with the project's size and complexity: On average, there were two Bank missions a year over the five-year implementation period. The missions' visits of construction sites helped highlight noticeable deficiencies and alert the executing agencies and supervisors to the need to exercise more effective quality control. These suggestions were generally well received by the executing agencies and helped sharpen their focus on some of the deficiencies noted by the missions. 53. The Bank showed flexibility in dealing with the design changes, so as to avoid interrupting and delaying the execution of works. These were pragmatic responses to problems, which were unforeseen during project preparation. Some of the changes added to the cost of the project. However, some of these will have beneficial impact on the operation of the highway, such as controlling access along the full length of the project highway. The acceptance of constructing only one carriageway of the four-lane highway between Yaoxian and Chuankou carried the risk of encountering eventual traffic capacity constraints on this section of the highway. However, such a risk has been greatly reduced because of the slower than forecast build up of traffic on the highway and the construction of a second two-lane carriageway under the follow-up Second Shaanxi Provincial Highway Project by the end of year 2000 (para. 8). 54. Overall, the project was implemented within the terms of the loan agreement and in substantial compliance with the Bank's procedural requirements. F. BORROWER PERFORMANCE 55. SPTD was exposed to Bank-financed project management in the first Highway Project. The benefit of that experience could be seen in the way the project was prepared and administered. Moreover, SPTD, through the implementation of the project, gained further experience in the organization and management of the construction of high-class highways, which imposed on the Department an even wider range of coordinating and managerial tasks. SPTD was able to cope well with the added responsibilities and demonstrated that it is a well-organized institution with high-caliber technical staff. Its administrative system reflects good record keeping and comprehensive reporting. The progress report format developed with the assistance of the Bank was distributed as a sample to be followed by other provinces involved with similar highway projects. SPTD also demonstrated its keen interest to improve quality control in the execution of highway works, and provided strong support to the construction supervision team in the enforcement of technical specifications and other quality control measures. The Bank's policies and procedures were well understood by the Department, which helped reduce excessive delays, and greatly contributed to the timely completion of the project. SPTD's cooperation and well-prepared documents were also very helpful in preparing this ICR. 56. Project implementation also benefited from MOC's regular involvement in Bank- financed highway projects. Following consultation with the Bank, in early 1990, a  - 15 - Technical Coordination Unit was set up in MOC to liaise with the Bank and the provincial entities. Such liaison was particularly useful in resolving the major design change issues on a consensus basis. G. ASSESSMENT OF OUTCOME 57. The implementation of the project can be regarded as satisfactory. Some of the components suffered some implementation delays but within the parameters envisaged at project appraisal. Consequently, the project highway and the rural roads were opened for traffic at the end of 1992, and the project as a whole was substantially completed by the end of 1993, as well reflected at the level of disbursements at that time. The loan was fully disbursed in late October 1994, ahead of the loan closing date of December 31, 1994. 58. The contractors' performance on the Sanyuan-Tongchuan Highway had been adequate, despite their noticeably limited experience on similar types of high-class highway civil works. The contractors' unfamiliarity with intemational contract conditionalities was also apparent. These shortfalls put added responsibility and burden on the construction supervision team, which stood up well to the challenge. Overall, the highway was constructed to a satisfactory standard but with less than desirable surface quality of the pavement on some sections of the highway. Although the implementation of the Rural Roads program encountered difficulties due to exceptionally heavy rain falls, lack of equipment and manpower, the program as executed met the expectations with regard to administration control and technical requirements and the economic viability of the program. 59. The project's development objectives (through its highway/road investment components) were achieved and are considered sustainable. The stream of benefits is expected to be maintained, although at a somewhat lower level than originally forecast for the Sanyuan-Tongchuan Highway. For the Rural Roads program, the reestimated ERR confirms the expected level of viability of the investments made and highlights the high priority of investing in areas with economic potential but with poor road infrastructure. 60. Sectoral development under the project, in the context of the RDB and PMS programs, was highly successful. It was achieved primarily because of the dedication and perseverance of the personnel in the provincial highway administration and technical institutions. Furthermore, the success of the Shaanxi pilot exercise provides incentives and practical experience in the introduction of these programs to other provinces. 61. The level of technology transfer was not easy to measure because of a lack of well- defined baseline parameters. The training program, although at a price, provided a satisfactory level of exposure to new concepts and incentives for adopting new approaches. H. FUTURE OPERATION 62. Because the Sanyuan-Tongchuan highway narrows to two lanes between Yaoxian and Chuankou, this section of the highway will be widened to four lanes under the Second  - 16 - Shaanxi Provincial Highway Project to eliminate a potential bottleneck. Also, the second project will convert the Xi'an-Sanyuan section of the highway into an access-controlled highway, which, combined with the extension of the highway from Chuankou to the city of Huangling (73.6 km), is expected to contribute to a more rapid buildup of through traffic and a higher rate of traffic diversion to the new highway. 63. To ensure that the highway is operated efficiently, SPTD prepared an Operations Manual that can be used as a reference for regularly monitoring key performance parameters (para. 46). It also provides information about warning signals that will alert staff of the need for interventions. 64. SPTD outlined a comprehensive program to expand the RDB and PMS to the entire Province by 2000 (para. 49). It also created a rational plan for the upkeep of the RDB; data would be updated as often as needed, based on the way it is used. The progress made with applying the RDB and PMS in Shaanxi Province seems to justify the Bank's continued support under future provincial highway projects for such programs. 65. Similarly, including a poverty alleviation component (RIPA) in the Second Shaanxi Provincial Highway Project was justified because the small-scale Rural Roads program under the first highway project was very successful. L KEY LESSONS LEARNED 66. These lessons are: (a) The major design changes for the Sanyuan-Tongchuan Highway were needed to avoid adverse effects to community and industrial development. This points either to the lack of adequate land use planning at the county and municipal levels or to poor coordination with local government entities during the highway alignment location studies, or both. Similar problems, which were noted during the implementation of other Bank-financed highway projects, led to substantial cost increases and disrupted the execution of works. Thus, the issue of appropriate land use planning and better coordination between local governments and provincial highway administrations warrants nationwide attention. (b) The executing agency could have been more stringent in prequalifying some highway construction contractors, ensuring they had enough experience and resources to carry out the works efficiently and satisfactorily. The lack of firm enforcement of prequalification criteria has been frequently noted in Chinese highway projects and an issue requiring national regulations, possibly by creating a classification system for the highway construction firms based on well-established performance criteria for constructing high-class highways. (c) The Chinese and Bank approaches to costing Bank-financed projects, particularly civil works components, are substantially different, and the  - 17 - Bank and MOC should make an effort to agree to common methods. A standard methodology would produce more reliable and consistent estimates of ERRs and investment priorities. (d) As indicated, the cost of staff training per person-month was high, which has been also the experience in other Bank-financed highway projects in China. Thus, it would seem highly desirable for the Bank, Ministry of Finance and MOC to address the issue as soon as possible, perhaps by appointing a technical training institution in China to coordinate highway subsector training programs nationwide.  - 18 - PART II. STATISTICAL TABLES TABLE 1: SUMMARY OF ASSESSMENTS A. Achievement of objectives Substantial Partial Negligible Not applicable Macro policies X Sector policies X Financial objectives X Institutional development X Physical objectives X Poverty reduction X Gender issues X Other social objectives X Environmental objectives X Public sector management X Private sector development X Other (specify) X B. Project sustainability Likely Unlikely Uncertain x Highly C. Bank performance satisfactorv Satisfactory Deficient Identification X Preparation assistance X Appraisal X Supervision X D. Borrower performance Preparation X Implementation X Covenant compliance X Operation (if applicable) X Highly Highly E. Assessment of outcome satisfactory Satisfactorv Unsatisfactory unsatisfactory x  - 19 - TABLE 2: RELATED BANK GROUP LOANS/CREDITS Year of Loan/credit title Purpose approval Status Preceding operation 1. Highway Project provincial road network FY85 completed Ln.2539-CHA/Cr. 1954-CHA expansion/improvement Following operations 1. Beijing-Tianjin-Tanggu Expressway Project see footnote /a FY87 completed Ln.281 1-CHA/Cr. 1792-CHA 2. Sichuan Provincial Highway Project see footnote /a FY88 ongoing Ln.2951-CHA/Cr. 1917-CHA 3. Jiangxi Provincial Highway Project see footnote /a FY89 completed Cr. 1984-CHA 4. Shandong Provincial Highway Project see footnote /a FY89 ongoing Ln.3073-CHA/Cr.2025-CHA 5. Jiangsu Provincial Transport Project see footnote /a FY91 ongoing Ln.3316-CHA/Cr.2226-CHA 6. Zhejiang Provincial Highway Project see footnote /a FY92 ongoing Ln.3471-CHA 7. Henan Provincial Highway Project see footnote /a FY93 ongoing Ln.3531-CHA 8. Guangdong Provincial Highway Project see footnote /a FY93 ongoing Ln.3530-CHA 9. Fujian Provincial Highway Project see footnote /a FY94 ongoing Ln.3681-CHA 10. Hebei/Henan National Highway Project see footnote /a FY94 ongoing Ln.3748-CHA 11. Xinjiang Highway Project I see footnote/a FY95 ongoing Ln.3787-CHA 12. Shanghai-Zhejiang Highway Project see footnote /a FY96 ongoing Ln.3929-CHA 13. Second Shaanxi Provincial Highway Project see footnote /a FY96 ongoing Ln.3986-CHA /a These highway projects, typically, have had the following components: (a) construction of a major highway; (b) improvement of provincial roads; (c) procurement of road maintenance equipment; and (d) institutional development components, such as RDB and PMS programs, staff training and selected subsector oriented studies.  - 20 - TABLE 3: PROJECT TIMETABLE Steps in project cycle Date planned Date actual Identification (Executive Project Summary) N/A Jun. 9, 1987 Preparation N/A Jul. 1987 Appraisal Feb. 1988 Jan. 1988 Negotiations Apr. 1988 Apr. 1988 Board presentation Jun. 9, 1988 Jun. 9, 1988 Signing Feb. 27, 1989 Feb. 27, 1989 Effectiveness May 31, 1989 May 31, 1989 Project completion Dec. 31, 1993 Dec. 31, 1993 (substantially) Loan closing Dec. 31, 1994 Dec. 31, 1994  - 21 - TABLE 4: LOAN/CREDIT DISBURSEMENT: CUMULATIVE ESTIMATED AND ACTUAL ($ mnillion) FY89 FY90 FY91 FY92 FY93 FY94 FY95 Appraisal estimate 5.00 15.00 35.00 40.00 45.00 49.00 50.00 Actual 2.99 11.19 21.87 38.69 45.98 49.55 50.00 Actualas%ofestimate 59.8% 74.6% 62.5% 96.7% 102.2% 101.1% 100.0% Date of final disbursement Oct. 28, 1994 TABLE 5: KEY INDICATORS FOR PROJECT IMPLEMENTATION Estimated Actual Start Complete Start Complete 1. Highway Construction Feb-88 Jan-91 Apr-90 Jun-93 2. Rural Roads Jun-88 Jun-90 Feb-88 Dec-92 3. Equipment for Pavement Evaluation Mar-89 Dec-89 Feb-91 Sep-94 4. Laboratory Equipment for Training School Mar-89 Apr-90 Nov-91 Nov-94 5. Road Data Bank Jun-88 Jul-90 Jan-88 Nov-92 6. Consulting Services Feb-89 Dec-90 Jun-88 Dec-92 7. Staff Training Nov-88 Dec-90 Jun-90 Apr-94  - 22 - TABLE 6: KEY INDICATORS FOR PROJECT OPERATIONS Length (kcm) Traffic (1994 AADT) Estimated Actual Estimated Actual 1. Sanvuan-Tongchuan Highway Sanyuan-Yaoxian Section 32.00 31.00 5,698 2,925 Yaoxiao-Chuankou Section 19.00 20.30 7,032 2,718 Chuankou-Houliu Section 14.00 15.10 640 1,522 Total 65.00 66.40 2. Rural Roads Program Shangxian-Shagou Road 101.00 100.78 - 1,951 Sehe-Manchuan Road 73.00 73.00 - 2,082 Shangnan-Yunxi Road 105.50 105.48 341 929 Shangxian-Zhashui Road 129.90 129.85 801 1,007 Pingli-Zhenping Road /a 59.50 - 297 - Langao-Zhenping Road - 50.00 - 930 Total 468.90 459.11 Grand Total 533.90 52551 /a During the project implementation, Pingli-Zhenping road was replaced by the Langao-Zhenping road. TABLE 7: STUDIES INCLUDED IN THE PROJECT A Shaanxi Highway Data Bank Preparation, Pavement Evaluation and Strengthening Program and Maintenance Management Study was included in the Project (Terms of Reference, Annex 2 of the SAR). However, as elaborated in paras. 37-42 of Part I of the ICR, the Study has not been executed as envisaged, although the objectives of the study have been partially achieved through the development and application of the RDB and PMS programs by SPTD with the assistance of MOC. The programs were supported by the Bank through the acquisition of equipment for road condition survey, land survey and geographic location, and computer hardware and software for data processing and storage.  - 23 - TABLE 8A: PROJECT COSTS (Y million) Appraisal estimate ActualAatest estimate Item Local Foreign Total Local Foreign Total Sanyuan-Tongchuan Highway 99.56 98.44 198.00 186.37 186.38 372.75 Consultant services for supervision 1.76 1.76 3.52 3.00 3.00 6.00 Rural roads 82.20 54.80 137.00 73.60 49.07 122.67 Equipment and expertise for RDB 0.00 4.44 4.44 0.00 1.05 1.05 Equipment and expertise for PMS 0.00 1.00 1.00 0.00 2.85 2.85 Equipment for training schools 0.00 3.00 3.00 0.00 6.60 6.60 Staff Training 0.00 1.48 1.48 0.00 2.50 2.50 Base Cost 183.52 164.92 348.44 262,97 251.45 514.42 Physical contingencies 18.35 16.49 34.84 0.00 0.00 0.00 Price contingencies 22.55 20.29 42.84 0.00 0.00 0.00 Subtotal 224.42 20170 426.12 0.00 0.00 0.00 Land acquisition cost 34.55 0.00 34.55 67.73 0.00 67.73 Total Project Cost 258 97 201.70 460 67 330.70 251.45 582.15  - 24 - TABLE SB: PROJECT COSTS ($ million) Appraisal estimate Actual/latest estimate Item Local Foreign Total Local Foreign Total Sanyuan-Tongchuan Highway 26.90 26.60 53.50 34.38 34.38 68.76 Consultant services for supervision 0.48 0.48 0.96 0.60 0.60 1.20 Rural roads 22.23 14.81 37.04 17.08 11.38 28.46 Equipment and expertise for RDB 0.00 1.20 1.20 0.00 0.21 0.21 Equipment and expertise for PMS 0.00 0.28 0.28 0.00 0.57 0.57 Equipment for training schools 0.00 0.82 0.82 0.00 1.32 1.32 Staff Training 0.00 0.40 0.40 0.00 0.50 0.50 Base Cost 49.61 44 59 94.20 52 06 48.96 101.02 Physical contingencies 4.96 4.46 9.42 0.00 0.00 0.00 Price contingencies 2.58 2.32 4.90 0.00 0.00 0.00 Subtotal 7.54 6.78 14.32 0Q00 0.00 Land acquisition cost 9.34 0.00 9.34 14.66 0.00 14.66 Total Prgect Cos 642 SI2 117.86U 6672 4&896 115.68 TABLE SC: PROJECT FINANCING ($ million) Appraisal estimate Actual/latest estimate Local Foreign Local Foreign Source costs costs Total costs costs Total IBRD/IDA 0.0 50.0 50.0 0.0 50.0 50.0 Central Government 13.5 0.0 13.5 14.4 0.0 14.4 Provincial Government 53.0 1.3 54.3 51.3 0.0 51.3 TQW 66 S S1 3 117.8 6S 7 Ii  - 25 - TABLE 9: EcoNOMIC COSTS AND BENEFITS (Y million) Costs /a Benefits /b NPV /c ERR (%) The ICR Santuan-HouliuHighway 459.99 32.91 329.00 18.1 Rural roads (5) 159.42 120.07 735.20 41.5 Total 619.41 152.98 1.064.20 27.9 The SAR Santuan-Houliu Highway 291.30 59.73 n.a. 27.8 Rural roads (3) 246.37 30.06 n.a. 16-25 /d Total 537.67 89.79 na. na. /a July 1995 prices. /b First year benefits (July 1995 prices). /c Discount rate = 12%. /d Between 16% to 25% for the rural roads. J  - 26 - TABLE 10: STATUS OF LEGAL COVENANTS AU..- Covenant Original Revised ment Section clss(es) Status Date Date Dewcription of Covenant Comments Loan 2.02(b) 1 C Open Special Account and operate it according to Schedule 5 of the Development Credit Agreement 3.01(a) 5 C Commitment to project objectives and meeting obligations by Beneficiary under the Project Agreement 3.01(b) 3 C Satisfactory onlending arrangement to Beneficiary 3.02 5 C Cornpliance with Schedule 4 of the Loan Agreement 3.03 5 C Meeting requirements of the General Conditions, Sections 9.04-9.09 4.01 1 C Maintaining adequate fuiancial recording and auditing reporting requirement Project 2.01 5 C Commitment to project objectives and meeting obligations by Beneficiary under the Project Agreement 2.02 5 CD Compliance with Schedule 4 of the Loan Delays had been Agreement encountered in the procurement proceas 2.03 5 C Meeting requiremnents of the General Conditions, Sections 9.04 -9.09 2.04(a) 9 C Information on interference with project implementation 2.05 7 C Carry out resettlement actions as agreed Status C - Complied with CD - Compliance after delay NC - Not complied with SOON - Compliance expected in reasonably short time CP - Complied with partially NYD - Not yet due Coveant Clas: I Accounts/audit 2 - Financal performance/generate revenue from beneficiaries 3 Flow and utilization of Project funds 4 - Counterpart funding 5 Maagement aspects of the Project or of its executing agency 6 - Environmental covenants 7 - Involuntary resettlement S - Indigenous people 9 Monitoring., review and reporting 10 - Implementation 11 - Sectoral or cross-sectoral budgetary or other resource allocation 12 - Sectoral or cross-sectoral regulatory/institutional action 13 Other  - 27 - TABLE 1 1: COMPLIANCE WfTH OPERATIONAL MANUAL STATEMENTS Indicate any significant lack of compliance with an applicable Bank Operational Manual statement (OD or OP/BP): Statement number and title Describe and comment on lack of compliance None None TABLE 12: BANK RESOURCES: STAFF INPUTS (Staff Weeks) FY 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 Total Preappraisal 3.5 16.3 19.8 Appraisal 12.8 12.8 Negotiations 7.8 7.8 Supervision 4.0 10.2 8.1 6.3 4.9 1.6 35.1 Completion 4.2 16.2 20 4 Tota /a 3.5 36.9 4.0 10.2 8.1 6.3 4.9 1.6 4.2 16.2 95.9 /a As of April 30, 1996.  - 28 - TABLE 13: BANK RESOuRCES: MISSIONS Performance rating Number Days Specialized Implemen- Develop- Stage of Project Month/ of in staff skills /a tation ment Types of Cycle year persons field represented status /b objectives problems Identification 06/87 -- -- EN,EC Preparation 07/87 3 5 PO,EC,FA Preappraisal 10/87 3 5 EN,EC Appraisal 01/88 4 5 EN,EC Negotiation 04/88 6 -- EN,EC,PO LC,DO,RA Board approval 06/88 -- - - Signing 02/89 -- -- -- Effective 05/89 - -- -- Supervision 1 03/89 3 5 EN 1 1 Supervision 2 11/89 2 5 EN,TS 2 1 Supervision 3 04/90 1 5 EN 2 1 Supervision 4 10/90 2 5 EN,RA 2 1 Supervision 5 05/91 1 5 EN 2 1 /c Supervision 6 10/91 1 5 EN 2 1 Supervision 7 05/92 1 5 EN 2 1 Supervision 8 04/93 1 5 EN 2 1 Supervision 9 04/94 1 5 EN 1 1 Supervision 10 09/94 1 5 EN 1 1 Supervision 11 03/94 1 5 RA 1 1 Completion 07/95 2 5 EN,CO - - /a PO: Project Officer; EN: Engineer; FA: Financial Analyst; EC: Economist; OS: Operations Specialist; LC: Legal Counsel; CO: Country Officer; DO: Disbursement Officer; RA: Research Analyst; TS: Training Specialist. ib 1: Highly satisfactory; 2: Satisfactory. L Typical problems included: design changes, quality deficiency and delays.  - 29 - ANNEX 1 ANNEX 1: ECONOMIC EVALUATION Preface 1. In the Staff Appraisal Report for the Shaanxi Provincial Highway Project (Report No. 7167-CHA) the economic evaluation was presented for the (a) construction of the Sanyuan-Tongchuan highway, and (b) improvement of three roads included in the Rural Roads component of the project. 2. The economic analysis presented in this ICR is based on updated data on traffic, vehicle operating cost (VOC), economic cost, and road user benefits, taking into account revised assumptions regarding future traffic growth rates. The methodology used in the economic analysis for this ICR is similar to that used in the SAR and is summarized in the following: (a) capital investments and maintenance costs have been revised to reflect July 1995 prices and are included in the cost streams; (b) the benefit streams presented, also reflect July 1995 prices, and consists of savings in VOC, reduced traffic congestion on the existing road, and cost savings on account of accident reductions; (c) a project life of 20 years has been assumed and the capital investment period for all roads has been taken from 1989 to 1993; and (d) full benefits started to accrue the Sanyuan-Tongchuan Highway in 1994 and to the rural roads in 1993. 3. During the implementation of the project, there were some design changes in the Sanyuan-Tongchuan highway and the Rural Roads program. The impact of these changes had been taken into consideration in the economic analysis for this ICR. PART I: SANYUAN-TONGCHUAN HIGHWAY 4. As in the SAR, the highway was divided into three sections for economic evaluation: Section 1 Sanyuan to Yaoxian (31.0 kmi); Section 2 Yaoxian - Chuankou (20.3 kmn); and Section 3 Chuankou (Tongchuan South)-Houliu (Tongchuan North) (15.1 km).  - 30 - ANNEX 1 Cost and benefit analyses were carfied out separately for each section as well as the highway as a whole. Corridor Traffic 5. The 1994 corridor traffic and its distribution between the old road and the new highway is presented in the following table, indicating SAR forecast and actual traffic census data: NUMBER OF MOTORIZED VEHICLES PER DAY (AADT) FOR 1994 Diversion The Old Road The New Road Total Corridor Factor (DF) (1) (2) (3)=(1)+(2) (4)=(2)/(3) SAR: Section 1 2,442 5,698 8,140 70.0% Section 2 1,759 7,032 8,791 80.0% Section 3 2,562 640 3,202 20.0% ACTUAL Section 1 4,188 2,925 7,113 41.1% Section 2 6,059 2,718 8,777 31.0% Section 3 3,361 1,522 4,883 31.2% (ACTUAL(SARM RATIOS Section 1 171.5% 51.3% 87.4% Section 2 344.5% 38.7% 99.8% Section 3 131.2% 237.8% 152.5% Sources: Shaanxi Provincial High Class Highway Administration and Bank staff. 6. As shown above, the 1994 corridor traffic (i.e., the sum of traffic on the parallel sections of the existing arterial road and the new highway), for other than Section 3 does not differ significantly from the SAR estimates. However, the expected distribution of corridor traffic between the existing arterial road and the new highway is substantially different from that expected at project appraisal, as reflected in the Diversion Factor (DF), which indicates that a high percentage of the traffic, as of 1994, prefers to travel on the existing arterial road. The most likely reasons for the less than expected rate of traffic diversion to the new highway are: (a) at the time of project appraisal the new highway was not expected to be operated as a toll highway; (b) the existing road paralleling the highway was widened and improved during 1990-92 resulting in an increase in traffic capacity; (c) the perceived benefits of using the new highway from the road user's perspective may not have been adequate to offset the cost of tolling.  -31- ANNEX 1 7. As indicated in the above table, instead of the expected 70-80 percent of corridor traffic Diversion Factor (DF) to Sections 1 and 2 of the new highway, after more than two years of the highway's operation, the DF has been only about half (31-41 percent ) of the forecast DF. Consequently, traffic reached only 51.3 percent and 38.7 percent, respectively, of the expected level of traffic on these two sections. Regarding Section 3, the traffic conditions are somewhat different from that of Sections 1 and 2. First of all, this section was built as a two-lane road, and it is not a toll road. Furthermore, this road between Chuankou and Houliu functions primarily as a bypass route for Tongchuan City. After the opening of the new road, most of the truck traffic through the city has been restricted, and thereby diverted to the new road. Thus, the 1994 traffic on this section of the highway is significantly higher than SAR's projection. Traffic Projection 8. It is recognized that, normally, the buildup of traffic on a new highway, particularly when it operates as a toll facility, is gradual in the early years and then the diversion accelerates. Notwithstanding this, the ICR's traffic forecast for the highway is more conservative than SAR estimates because, even after three years of operation, the DF remained well below expectations. The main reasons for this are indicated in para. 6, and there seems to be no obvious reason for expecting a sudden change in the present traffic pattern. Furthermore, a closer analysis of the traffic pattern has revealed that for other than long-distance through traffic the highway does not provide travel distance saving, which, normally, is regarded as an important incentive for diversion even for relatively short-distance trips. For shorter trips, the new highway does not provide distance savings because the travel on the access roads to the highway increases the overall travel distance. 9. The long-term traffic growth rates (1993-2015) for Sections 1 and 2 of the highway has been estimated at 7.8 and 7.7 percent, respectively, and for Section 3 at 5.8 percent, compared to the SAR's estimate of a uniform growth rate of 9.5 percent for all three sections. In estimating the growth rate for Section 3 it was taken into consideration that the two-lane road through hilly terrain will be faced with capacity constraints, and furthermore, the extension of the four-lane section of the highway from Chuankou to Huangling, as part of the Bank-financed Second Shaanxi Provincial Highway Project, will be attracting some of the traffic presently using Section 3. The traffic projections by sections are summarized as follows:  - 32 - ANNEX 1 TRAFFIC FORECAST COMPARISON FOR THE NEW HIGHWAY (AADT), BY SECTIONS Section 1 Section 2 Section 3 Sanyuan-Yaoxian Yaoxian-Chuankou Chuankou-Houliu (Tongchuan) SAR 1993 5,133 6,334 576 1994 5,698 7,032 640 1995 6,326 7,808 710 2000 10,707 13,222 1,204 2010 24,817 30,614 2,787 2015 37,939 46,761 4,257 Average growth p.a. 9.5% 9.5% 9.5% 1993-2015 ICR 1993 2,805 2,544 3,621 1994 2,925 2,718 1,522 1995 4,180 3,681 2,451 2000 6,142 5,409 3,974 2010 11,527 10,151 8,921 2015 14,711 12,956 12,512 Average growth p.a. 7.8% 7.7% 5.8% 1993-2015 10. The projected traffic for the three sections of the Sanyuan-Tongchuan Highway for years 1993-2015 is detailed in Table 1 of this Annex. For 1993 and 1994, actual traffic data have been used. Traffic projections for 1995 are based on the actual traffic data for the first six months of the year. For the subsequent years the following traffic growth rates have been assumed: (a) For Section 1 of the existing arterial road and Section 1 and 2 of the new highway normal traffic growth forecast is estimated to increase by 8 percent per year between 1995-2000, with the growth rate reduced by 1 percent for each of the five years thereafter to 5 percent in 2015. (b) For Section 2 and Section 3 of the existing arterial road, considering the future planning of other alternative routes in this area, the growth rate is estimated to be 2 percent lower than that for the new highway; 6 percent per year between 1995-2000, and reducing to 2 percent by 2015. (c) The generated traffic in the ICR was assumed to be 10 percent of normal traffic. In the SAR, by comparison, for normal traffic a single growth rate of 9.5 percent per year was applied to all three sections of the highway, and the generated traffic growth rate was assumed to be between 10 and 20 percent of the normal traffic.  - 33 - ANNEX 1 Economic Costs 11. Following the pattern of the SAR, the economic cost reflects the financial cost of civil works according to the final cost of the civil works contracts (Table 2). Financial costs have been converted to economic costs by shadow pricing for each of the three sections of the highway as shown in Table 3. The overall effect of the difference in the shadow pricing of the financial costs in the SAR and ICR, and the higher than SAR estimate of the financial cost is that the total economic cost of the highway, according to ICR estimates, is about 58 percent higher than the SAR estimates, at constant July 1995 prices. ECONOMIC COST COMPARISON (Yuan million) SAR ICR ICR/SAR Jan 1988 Jul 1995 Jul 1995 in% Sanyuan-Yaoxian 121.0 163.2 207.6 +27.2% Yaoxian-Chuankou 65.0 87.7 181.4 +107.0% Chuankou-Houliu 30.8 40.4 71.0 +75.4% Total 216.8 291.3 460.0 +57.9% Economic Benefits 12. The economic analysis includes the benefits derived from (a) VOC savings on the new highway for normal and generated traffic, (b) relieved congestion on the existing road, and (c) savings in accident cost. The updated VOC per km for the various types of vehicles is presenterl in Table 4. 13. It was conservatively assumed that about 30 to 40 percent of traffic from the existing arterial road would divert to the new highway. It was assumed that less traffic would reduce congestion on the existing road, resulting in 15 percent savings of the VOC applied for the existing arterial road between 1993 and 1995. Thereafter, because of the gradual buildup of traffic on the existing arterial road, the savings would be reduced by 1 percent for each of the remaining five year periods to 11 percent by 2015. 14. In 1985, there were, on the average, seven accidents per km per year on the existing road. After the opening of the new highway for traffic, despite the higher traffic volume, the number of accidents in 1994 was lower; 5.2 per km per year. On the new highway, the number of accidents in 1994 was 2.9 per km per year. For estimating road accident cost savings, it was assumed that the reduction in the number of accidents,  -34 - ANNEX I resulting from the construction of the highway, would be two accidents per km per year at a cost of Y 10,000 for each accident. Economic Evaluation and Sensitivity Analysis 15. Total costs and benefits streams, Economic Rate of Return (ERR) and Net Present Value (NPV), for the highway are presented in Table 5A, and for the three individual sections of the highway in Tables 5B-5D. The overall ERR on the highway investment is estimated at 18.1 percent versus the SAR estimate of 27.8 percent. Sensitivity Analysis and comparative SAR figures are presented in the following table. Section I Section 2 Section 3 Sanyuan- Yaoxian- Chuankou- Total Yaoxian Chuankou Houliu Highway ICR: Best estimate of rate of return 19.7% 16.7% 16.8% 18.1% NPV (12%, million Yuan) 191.7 95.0 42.3 329.0 20% increase in costs and best estimate of benefits 17.5% 14.8% 15.0% 16.1% NPV (120/o, million Yuan) 154.8 62.8 29.6 247.1 20% decrease in benefits and best estimate of costs 17.1% 14.4% 14.6% 15.7% NPV(12%, millionYuan) 116.4 43.8 21.1 181.3 20% increase in costs and 20% decrease in benefits 15.1% 12.6% 12.9% 13.8% NPV (12%, million Yuan) 79.5 11.5 8.4 99.4 SAR: Best estimate of rate of return 28.1% 30.0% 20.6% 27.8% NPV (12%, million Yuan) n.a. n.a. n.a. n.a. 20% increase in costs and best estimate of benefits 25.5% 27.3% 18.5% 25.2% NPV (12%, million Yuan) n.a. n.a. n.a. n.a. 20% decrease in benefits and best estimate of costs 24.9% 26.7% 18.0% 24.7% NPV (12%, million Yuan) n.a. n.a. n.a. n.a. 20% increase in costs and 20% decrease in benefits (%) 22.6% 24.3% 16.0% 22.4% NPV (12%, million Yuan) n.a. n.a. n.a. n.a. Evaluation of Improvement to the Existing Arterial Road 16. Coinciding with the implementation of the Bank-financed highway project, the Shaanxi Provincial Transport Department carried out the widening and pavement improvement of the existing arterial road, paralleling the new highway, by using domestic funds. During years 1992-94 a total of Y 12.7 million was invested for such improvements. It is assumed that the improvements made to the condition and capacity of  -35- ANNEX1 the existing arterial road was one of the reasons for the less than expected traffic diversion to the new highway. Based on the data provided by SPTD, the improvement of the existing arterial road was estimated to yield an ERR of 76.8 percent. The cost of improvement has been Y 6.5 million per km, which can be regarded as a marginal investment for a relatively high ERR. The combined ERR for this highway corridor, i.e., the Bank-financed new highway, and the locally financed improvement of the existing arterial road is estimated at 19.5 percent. PART II: RURAL ROADS COMPONENT Background and Traffic 17. The original Rural Roads program included the upgrading of three roads and to repair two flood-damaged roads in the mountainous prefectures of Shanglou and Ankang, both are located adjacent in the southern part of the Province adjacent to each other. These two prefectures are classified by the State as some of the poorest areas in China. During project implementation, one of the proposed road (Pingli-Zhenping--60 km) in Ankang prefecture was replaced by a new road for inclusion in the Rural Roads program (Langao-Zhenping-50 km), a missing link in the prefecture's road network. 18. Since all the five roads included in the Rural Roads program are located in the same mountainous area, the local economic development and traffic patterns are quite similar. Based on the actual traffic data, the following assumptions have been made in the economic analyses regarding future traffic growth: a 7 percent per year traffic increase in 1995, reduced by 1.5 percent for each of the next five years, and thereafter by 1.0 percent until year 2012. In the SAR, the average growth rate was estimated as 8 percent per year between 1989 and 2006). Economic Costs and Benefits 19. Costs and benefits have been converted to July 1995 economic costs. Financial costs have been converted to economic costs by shadow pricing, using a factor of 0.97, which was the overall conversion factor used for the highway component. The overall effect of the shadow-pricing on the financial cost of the five rural roads at July 1995 prices has been that the economic cost became about 30.0 percent higher than the financial costs (Table 2). 20. The benefits have been derived from (a) VOC savings (i) for normal traffic due to shorter distance and/or road condition improvement, and (ii) for generated traffic; (b) time savings on freight and passenger transport; and (c) savings in accident cost. 21. Without the project, all five roads would have remained either class IV or substandard roads. With the project, all five roads had been upgraded to Class III roads. The better pavement of the road surface would reduce VOC, which yielded the major portion of the project benefits. Generated traffic was assumed to be 5 percent of the normal traffic. In the SAR, it was 15 percent.  - 36 - ANNEX 1 22. The better pavement of the road would increase vehicle speed and improve the turnaround time for freight and passenger transport. The benefits to freight transport (assumed at Y 2.0/hour/vehicle) and passenger transport (assumed at Y 1.0/hour/ passenger) were calculated on that basis. 23. It was estimated that the accident reduction on the improved roads would be one accident per four km per year at a cost saving of Y 3,500 per accident. Economic Evaluation and Sensitivity Analysis 24. The economic evaluation of the five roads is presented in Tables 6a-6e indicating the stream of costs and benefits for years between 1990 and 2012, pertinent VOC data, and the characteristics of the roads. The Economic Rates of Return (ERRs) for the five rural roads are ranging from 21.1 to 99.2 percent. The overall ERR for the five roads is 41.5 percent. The detail calculation of the ERRs are summarized as follows: ECONOMIC RATE OF RETURN ON THE FIvE RURAL ROADS Name of roads ERR NPV /a Main Purpose 1. Shangxian-Shagao (Shangsha) Road 41.9% 105.8 (for flood repairs) 2. Sehe-Manchuan (Seman) Road 27.0% 65.7 (for flood repairs) 3. Shangnan-Yunxi (Shangyun) Road 24.5% 46.2 (for improvement) 4. Shangxian-Zhashui (Shangzha) Road 21.1% 40.2 (for improvement) 5. Langao-Zhenping (Lanzhen) Road 99.2% 477.3 (for missing link) Total Rural Roads 41.5% 735.2 /a Y million, July 1995 prices, 12 percent discount rate. 25. Table 7 of this Annex summarizes key data for the roads included in the Rural Roads program, estimated and final cost of construction, traffic volumes and the estimates for ERRs. The table also compares the SAR and ICR sets of figures. Total length of construction in the ICR was 459 km or about 10 km shorter then the SAR's plan (469 kmn) due to the new Langao-Zhenping road being about 10 km shorter than the road it has replaced in the program. The final construction costs of the Rural Roads program, has been about 35.5 percent lower than the SAR's estimate. The latest available data (1994) show that, in aggregate, the actual traffic volumes (AADT) were higher than SAR's forecast. 26. The majority of the Rural Roads component show robust economic returns indicating the high priority of the investments made, particularly for the construction of the missing link and the repair of the flood damaged roads. This seems to indicate that marginal capital investments in road improvements in economically depressed regions could lead to high, quantifiable economic benefits.  - 37 - ANNEX 1 PART m: THE OVERALL ECONOMIC RATE OF RETuRN AND THE PROJECT RISKS Overall Economic Rate of Return 27. The overall economic rate of return on the Sanyuan Tongchuan Highway component and the five Rural Roads component) is 27.9 percent and the NPV (at 12 percent) is Y 1,064.2. In the SAR, an overall ERR for these components, was not presented. SAR ICR Best estimate of rate of return (%) 27.8/a 27.9 NPV (12%, million Yuan, July 1995 prices) n.a. 1,064.2 /a 27.8% for the Sanyuan-Tongchuan Highway and 16% to 25% for the rural roads.  Table 1: Sanyuan-Tongchuan Highway Traffic Summary (AADT) Width(m)/_a No ofof lanes Gradent,(%)/_b Road class Road condition Avg. speed (kn./how) _ , _ _ . _ _ _ . _ .........~~~~~~~~~~~~.......... . ... ,____._... - --- -. --- - --- . .... _.... __... ..... .. I II Ii I II II I Ii II I II ll I II III I II illl Theoldroad: a 6 5 l1S 110 2 2 2 F M M 11/111 It It Fair Fair Fair 40 43 25 b 75 12.0 120 to 809/. to 3.0%% t 3