China - North China Earthquake Reconstruction Project

Document of The World Bank FOR OFFICIAL USE ONLY Report No. 15805 PERFORMANCE AUDIT REPORT CHINA NORTH CHINA EARTHQUAKE RECONSTRUCTION PROJECT (CREDIT 2091-CHA) June 27, 1996 Operations Evaluation Department This document has a restricted distribution and may be used by recipients only in the performance of their official duties. Its contents may not otherwise be disclosed without World Bank authorization. Currency Equivalents Currency = Renminbi (RMB) Currency Unit = Yuan (Y) US$1.00 = Y 8.70 Y 1.00 = US$0.1149 (At appraisal 12/89 $1.00 = Y 3.71 and at completion 06/93 $1.00 = Y 5.7) Abbreviations and Acronyms MOF Ministry of Finance SSB State Seismological Bureau Weights and Measures I meter (m) 3.28 feet (ft) 1 cubic meter (CUM) 35.31 cubic feet (cu. ft.) I hectare (h) 2.47 acres I kilogram (kg) 2.20 pounds (lb.) Fiscal Year Government: January I - December 31 FOR OFFICIAL USE ONLY The World Bank Washington, D.C. 20433 U.S.A. Office of the Director-General Operations Evaluation June 27, 1996 MEMORANDUM TO THE EXECUTIVE DIRECTORS AND THE PRESIDENT SUBJECT: Performance Audit Report on China North China Earthquake Reconstruction Project (Cr. 2091-CHA) The Performance Audit Report (PAR) on the China: North China Earthquake Reconstruction Project (Credit 2091 -CHA, approved in FYI990) was prepared by the Operations Evaluation Department. The credit for SDR 23.4 million was approved in February 1990 and closed in June 1992. The objective of the project was to assist the Government of China to normalize economic activity in the Shanxi and Hebei Provinces after an earthquake that damaged housing, infrastructure, and productive activities. Specifically, the project aimed to: (i) assist and accelerate the normalization of economic activity through a village reconstruction program, and the replacement and re-equipping of essential social and economic infrastructure; (ii) encourage and facilitate the use of earthquake resilient construction standard; and (iii) support a program to improve earthquake prediction and monitoring, as well as a disaster preparedness plan for Yanbei Prefecture. The project was successfully implemented. All physical targets were achieved and some were exceeded. About 281 villages were rebuilt, 2,800 public facility rooms (classrooms, health centers, storehouses) were reconstructed, and village infrastructure (roads, power lines, wells irrigation channels) were repaired. Production by nine enterprises was restored, and two hospitals were rebuilt. Less successful aspects of the project include: (i) lack of agreement between IDA and the Borrower on housing replacement standards; (ii) the financing of earthquake prediction, an avenue of research whose validity is being questioned by the expert community worldwide; and (iii) a weak system of construction quality control during implementation. OED rates the project outcome as highly satisfactory, its institutional development-not a priority objective-as modest, and the project approach to reconstruction as sustainable. The Association's and Borrower's performances are rated as highly satisfactory. These ratings agree with those given in the PCR. The project made productive use of lessons learned in previous emergency recovery operations. Within the Bank the project helped to pioneer (i) simplified and innovative business processes; (ii) reduced processing times; (iii) streamlined project documentation; and (iv) a leadin project processing role for the resident mission. Attachment 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.  FOR OFFICIAL USE ONLY Contents Preface .............................................................. 3 Basic Data Sheet......................................................... 5 Evaluation Summary .................................................... 7 1. Background........................................................11 The Extent of the Damage...........................................11 The Bank's Response..............................................11 2. Project Description .................................................. 15 Objectives and Components .............................................. 15 Financing ...................................................... 15 3. Project Performance and Influencing Factors .................................. 17 Project Outcome ............................ ................ 17 The Implementation Modality was Based on a Local Management System ................ 18 The Financial Management for Reconstruction was Based on Recovery Needs ....................................................... 19 Special Account and Cash Flow: A Bottleneck............................. 19 Compliance with Covenants and Operational Directives .................. 19 Past Experience in Emergency Lending Helped in Project Design ...... ......... 20 The Reconstruction Project was a Pioneering Business Process............ ..... 20 4. Implementation Issues................................................. 23 Resettlement or In Situ Reconstruction? .................................. 23 Housing Replacement: Differences in IDA/Local Views Concerning Standards and Approach ................................................ 23 Technical Assistance: Did it Work? ........................................ 24 Quality Control was a Problem Area.............................. 25 5. Conclusions, Lessons Learned and Recommendations ............... .......... 27 Project Ratings ................................................... 27 Lessons Learned ........................................... ...... 27 Annexes A. Description of the Project...........................................31 B. Project Agreement ................................................ 33 C. Borrower Comments .............................................. 35 D. Translation of Borrower Comments .................................... 39 This report was prepared by Mrs. Alcira Kreimer (Task Manager) who audited the project in April 1995. Ms. Stacy Ward provided administrative support. The report was issued by the Infrastructure and Energy Division (Yves Albouy, Chief) of the Operations Evaluation Department (Francisco Aguirre-Sacasa, Director). 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 wiihout World Bank authorization.  3 Preface This is the Performance Audit Report (PAR) of the China: North China Earthquake Reconstruction Project. The Project was approved on February 8, 1990. The project was supported by Credit 2091-CHA in the amount of SDR 23.4 million. The Credit was fully disbursed on March 19, 1992, and was closed on December 31, 1992. The PAR was prepared by the Operations Evaluation Department (OED). It is based on the President's Report, sector and economic reports, special studies, Country Strategy and Policy Framework Papers, the credit documents, study of the project files, and discussions with Bank staff. A Project Completion Report (PCR) has been prepared by the Country Operations Division and the Resident Mission in China of the China and Mongolia Department, East Asia and Pacific Regional Office, and is attached. An OED mission visited China in April-May 1995 and discussed the effectiveness of the Bank's assistance with Government officials, management of private, state and local enterprises and local beneficiaries/stakeholders. Their kind cooperation and invaluable assistance in the preparation of this report is gratefully acknowledged. The PCR provides a satisfactory account of the Project experience with regard to design, implementation, the role of the Bank, achievements and sustainability. The PAR elaborates on the factors of performance and, in particular, on the use of a "fast-track" approach for project processing. Copies of the draft PAR were sent to the relevant Government officials and agencies concerned for their review and comments. Comments received have been attached to the report as Annex C.  5 Basic Data Sheet NORTH CHINA EARTHQUAKE RECONSTRUCTION PROJECT (CREDIT 2091-CHA) Credit Position (Amount in US$ million) Appraisal Estimate Actual/Reestimate Actual as % of Appraisal Estimate Total Project Cost 45.3 48.0 106.1 Credit Amount 30.0 31.8 106.1 Cumulative and Actual Disbursements 1990 1991 1992 Appraisal Estimate 5.0 25.0 30.0 Actual 7.7 24.3 31.8 Actual as% of 24.9% 81.1% 106.1% Estimate Date of Final Disbursement: March 19, 1992 Project Dates Original Actual Identification n/a 11/10/89 Preparation n/a n/a Appraisal n/a 12/03/89 Negotiations n/a 12/15/89 Board Approval n/a 02/08/90 Signing Date n/a 03/02/90 Effective Date 03/31/90 04/26/90 Closing Date 12/31/92 06/30/92 Completion Date 12/31/91 12/31/91 Staff Inputs Actual Identification 3.0 Appraisal/Negotiation 16.9 Supervision 13.9 Completion 2.0 6 Mission Data Month/ No. of No. of Performance Year Days Persons Ratings Identification 11/89 5 3 n/a Appraisal 12/89 18 6 n/a Supervision 1 04/90 5 2 1 Supervision 2 06/90 5 3 1 Supervision 3 10/90 5 3 1 Supervision 4 06/91 5 3 1 Supervision 5 10/91 5 3 1 Other Project Data Borrower: Ministry of Finance Executing Agencies: Governments of Shanxi and Hebei Provinces, State Seismological Bureau (SSB) 7 Evaluation Summary Objectives 1. The objective of this project was to assist the Government of China to normalize economic activity in Shanxi and Hebei Provinces in North China after an earthquake had damaged housing, infrastructure, and productive activities (paras. 1.1, 1.2 and 1.3). The project aimed to: (i) assist and accelerate the normalization of economic activity through a village reconstruction program, and the replacement and re-equipping of essential social and economic infrastructure; (ii) encourage and facilitate the use of earthquake resilient construction standards; and (iii) support a program to improve earthquake prediction and monitoring, and a disaster preparedness plan for the Yanbei Prefecture (Annex A). 2. The project was a first on two accounts: (i) it was the first IDA-funded activity to finance earthquake reconstruction in China, a country that is very vulnerable to seismic events; and (ii) it provided assistance to an impoverished area which never had access to IDA funds in the past. Nevertheless, the Bank and IDA have significant experience in Emergency Recovery Lending (ERL), and the project took advantage of lessons learned upstream (paras. 1.4 and 3.13). One of the lessons from past experience is that the timing of the response is of the essence. The project was a pioneering attempt at implementing simplified and innovative business processes, some or which are similar to those the Bank is currently adopting as part of its Business Process Innovations (BPI) (paras. 3.14 - 3.17). Project Processing 3. The project followed a "fast-track" approach to increase the development effectiveness of emergency recovery (para. 1.9). It was client oriented, reduced processing time, collapsed project documentation, and included a significant role for the resident mission. Project preparation was done in one and a half months, negotiations were completed in the field, approval was sought also from the field, and presentation to the Board took place two months after the identification mission visited China. Savings in project preparation and supervision costs were substantial. The reduction in the elapsed time from approval to effectiveness, and from effectiveness to closing was significant; the project required a short preparation time, 19.9 staff weeks, less than the average for investment projects in the country, which is 162 staff weeks (para. 3.5). The elapsed time was short, 0.21 years from identification to approval (para. 1.9), and 2.68 years from approval to closing. Average projects take 0.51 years from identification to approval and 6.66 years from approval to closing (para. 3.17). Average emergency projects take 0.64 years from identification to approval and 2.89 years from approval to closing (para. 3.18). In the field, project preparation went smoothly (para. 3.19). At headquarters, the Bank/IDA culture prevalent at the time for project processing did not change enough to match the challenge (para. 3.20). Staff in the field worked as a team, their attitude was one of "let's do it". Management at headquarters, however, were reluctant to modify their standard approach to project clearance, an attitude that created a rift between headquarters and the field during project preparation. The disparity in attitudes led to the feeling on the part of staff in the field that their efforts were not appreciated. 8 Implementation Experience 4. The project achieved its physical targets (paras. 3.4 and 3.5). In fact, due to the devaluation of the US dollar against the SDR and local currency against the US dollar, the value of the credit increased by about 40 percent in local currency terms. Thus, the project went beyond the original targets. About 281 villages were rebuilt, 2,800 public facility rooms (classrooms, health centers, storehouses) were reconstructed, and village infrastructure (roads, power lines, wells irrigation channels ) were repaired. The production in nine enterprises was restored and two hospitals were rebuilt. The operational management skills in the involved agencies were improved. 5. Disagreements arose between local groups and the appraisal mission concerning housing replacement standards, mainly around the size and technology of replacement houses, the number of rooms in the rebuilt units, and the application of a construction technology that had been successful in another country, but which was perceived by the local engineers and the affected population as a "foreign" model. The Bank did not listen enough during appraisal (paras. 4.1 - 4.4). In the end, the Borrower's view on these issues prevailed. 6. Another aspect of the project that was not successful was the financing of earthquake prediction, an avenue of research whose validity is being questioned by the expert community worldwide (paras. 4.7 and 4.8). The audit concludes that the use of IDA resources, which only provided marginal support to a controversial research area, was not the best use of development resources. In a seismic-prone country like China, a more productive avenue would have been to support the reduction of vulnerability in areas like structural systems, land use planning, seismic designs, construction supervision, and insurance mechanisms. 7. Quality control for construction was another problem area, which received a great deal of attention from supervision missions, but in which further efforts are still needed (paras. 4.9 and 4.10). Project Ratings 8. The audit rates the project outcome as highly satisfactory, its institutional development as modest, and the project approach to reconstruction as sustainable (para. 5.1). It rates the Bank and Borrower performance as highly satisfactory. The success of the project is very much a result of the commitment of both IDA and the Borrower, a rapid response and reduced project processing time, work at the local level, excellent team work among IDA staff, realistic and limited objectives, and the extensive use of local teams to manage the project (para. 5.2). Lessons Learned 9. The Capacity in the Bank for Undertaking Reconstruction is Sound. The success of this project, as well as the satisfactory experiences in previous reconstruction projects, such as in Mexico and Colombia, should dispel any doubts concerning the Bank's capacity to undertake emergency projects, to react rapidly, and to cooperate fully with the affected communities. 10. The main lessons learned and recommendations for the future in similar projects are listed below. They highlight the difference in nature between the decisions for which local community participation is key (building technologies and project management) and those for 9 which the Bank's worldwide experience and technical knowledge should make it more assertive (in-situ reconstruction, impact reduction rather than earthquake prediction). (a) Promote In-Situ Reconstruction. Relocation needs to be strictly controlled. In- situ reconstruction makes better use of existing infrastructure and community facilities. Resettlement is more expensive both financially and socially; it needs to be strictly controlled and justified. Considerable self-help efforts in low-cost, largely in-situ reconstruction were stimulated by the appraisal mission and project design team, particularly because the IDA credit was directly targeted to the beneficiaries through special agreements signed between village and household owners on the basis of comprehensive reconstruction plans prepared by each participating village. Prior to this project, the advantages of in-situ reconstruction were shown in the earthquake recovery project in Mexico, which had a very successful housing component respectful of community networks. (b) Minimize Potential Disaster Impacts. Reducing vulnerability through architecture, urban planning and engineering should receive priority over earthquake prediction. Preparedness and prevention are important mechanisms for future disaster reduction, particularly in a country as vulnerable to seismic events as China. On very specific technical aspects, the Bank should seek specialized technical advice, since participatory mechanisms of decision-making are not appropriate for highly specialized technical issues. Earthquake prediction is an unproven approach and promoting it, even in China, where it benefited form a long tradition, proved not to be the best use of IDA's resources. (c) Participatory Design ofProject Components is an Important Factor of Success. Problems arose when there was an attempt by Bank staff and consultants to introduce building technologies from other regions of the world; these were found not to be appropriate to local conditions after expensive seismic studies and testing were conducted at the local University. (d) Secure Strong Local Leadership and Project Management Offices. A strong local leadership and an efficient system for reconstruction management were important factors in the efficient implementation of the project. The Project Management Offices at the pro'incial and local levels became strong implementation units. In "normal" development projects, Project Implementation Units are not always needed and often considered as artificial and unsustainable. The case is different for post-disaster programs where there are no functioning disaster response systems at the local level. Experiences with the Mexico and the Popayan (Colombia) Earthquake Reconstruction Projects confirm the importance of strong local project offices. (e) Minimize Bottlenecks to the Project's Cash Flow. Emergency projects require a strong positive cash flow and special attention to the design and implementation of disbursement arrangements. In this case, although a Special Account was established to help deal with cash flow needs, problems were encountered by the local Departments of Finance concerning timely disbursement from the Special Account. Some problems were found at the local level due to delays in the submission of documentation and reimbursement forms. Local disbursement 10 procedures were complicated, and proved to be a bottleneck. In addition, numerous project entities were borrowing for the first time and were not familiar with IDA disbursement procedures. Specific guidelines concerning disbursements from the MOF to local units may in the future help to avoid bottlenecks in project implementation. (f) Develop a "Fast-Track" Approach. The use of simplified procedures, fast cycle design and processing, a client orientation, and increased participation of the resident mission helped the success of the project and its development effectiveness. The close and productive cooperation between staff at local implementing agencies and IDA was an important ingredient in the success of the project. 11 1. Background The Extent of the Damage 1.1 China is an earthquake-prone country; sixty percent of its land area is in seismic zones vulnerable to earthquakes registering more than 6.0 on the Richter Scale. About 70 percent of metropolitan areas with populations of more than I million are in seismic zones. On the night of October 18, 1989, an earthquake measuring 6.1 on the Richter Scale struck Northern China at the boundary of Hebei and Shanxi Provinces and the Inner Mongolia Autonomous Region, near the city of Datong, approximately 220 km north-west of Beijing. A large number of aftershocks measuring around 5.7 on the Richter Scale occurred the next morning. The earthquake was one of the worst to hit Northern China since the very destructive Tangshan earthquake of 1976 where an estimated 400,000 persons were killed. 1.2 The region affected by the earthquake of October 1989 is an impoverished rural area on the Loess Plateau, a windswept eroded plain about 1,300 m above sea level, with a 1989 per capita income of approximately US$66 (Yuan 375 in Yanbei Prefecture in Shanxi Province). Coal mining and agriculture (primarily growing corn and wheat in the dusty, infertile soil) constitute the main sources of local employment. 1.3 Casualties in Yanbei Prefecture in Shanxi and neighboring Yangyuan County in Hebei were low (approximately 20 dead and less than 200 injured). However, property damage was extensive. Approximately 25,000 privately owned houses and over 3,000 rooms in schools, clinics, and community facilities were destroyed. Infrastructure, including bridges and power and telecommunications lines, as well as 46 local industries were severely damaged and many more were partially damaged. The Shanxi Province Seismological Bureau estimated that direct losses in Yanbei exceeded US$150 million. Only 3 to 4 percent of assets were insured. Indirect production losses were primarily in industry and commerce; they amounted to approximately US$15 million in the first year. Approximately 1,500 workers were idled. Two months after the earthquake, approximately 15 percent of the population in Yanbei (125,000 people) and a smaller proportion in Yangyuan were in temporary shelter. The Bank's Response 1.4 The North China Earthquake Reconstruction Project (NCERP) was the first Emergency Recovery Loan (ERL) after an earthquake in China. Prior to this operation, the Bank had been involved in three other emergency disaster reconstruction activities in the country: after (i) rubber damage on Hainan island,(ii) typhoon damage to sea dikes in Zhejiang, and (iii) rehabilitation of the Da Xing An Ling Forest in the far northern portion of Heilongjiang Province and the Inner Mongolia Autonomous Region after a destructive fire in 1988. Those experiences occurred in different provinces. Although local agencies in the Northern Region were not familiar with Bank emergency procedures, the central agencies and the Ministry of Finance (MOF) were. 1.5 Major natural disasters attract world-wide attention, thus, the Bank's response is likely to have a much higher profile than normal operations. This was certainly the case with the earthquake in China, the first project approved in the country after a hiatus of several months in 12 the lending program following the events in Tianamen Square in 1989. The gap in IDA lending to China, the events that led to such gap, the humanitarian nature of the project, and the widespread nature of damages were indicators of the significant attention and visibility the project would command. Provincial authorities were very interested in working with IDA, since the project would provide an injection of funds to an area that had never had access to Bank/IDA funds in the past. Furthermore, the central government had an important commitment to this initiative. 1.6 At the time of the reconnaissance mission, questions were asked at the Bank concerning the justification for an emergency loan. Although the property damage was extensive for the affected prefectures, it was limited, in relative terms, for China. In the past, confronted with far larger disasters than this earthquake, China had expeditiously managed relief efforts without outside involvement. However, at the time this one occurred, the State disaster recovery attention was divided between this earthquake and three typhoons which had struck Hainan Province, an island off the country's southern coast, in early October. Thus, country resources to deal with extreme events were overstated. 1.7 The identification mission noted that damage in the affected area was considerable and used three major arguments to justify the project. First, IDA support would make an essential contribution to speed up recovery. Although the Chinese could undertake a 3.5 year reconstruction program, it was essential that at least the core of the program be implemented in one year, an important consideration given the heavy housing losses and the harsh cold climate in the affected area. Second, IDA involvement would enable resources to be transferred to the area on much more favorable terms than normal. This was essential if the cycle of poverty and hardship were to be broken. Third, IDA would contribute to more cost-effective design, and the village reconstruction experience could be replicated elsewhere, especially in rural resettlement programs or rural settlement improvement. 1.8 The main rationale for IDA involvement was to: (i) implement affordable and cost- effective reconstruction standards; (ii) expedite the mobilization of counterpart funds and beneficiary contributions to ensure prompt implementation of the reconstruction program; (iii) draw on substantial Chinese expertise in earthquake-related matters and seismic engineering; and (iv) strengthen the national agencies responsible for earthquake prediction. 1.9 IDA's response was rapid. Immediately after the earthquake, the Country Department Director convened a small advisory group to facilitate decisions on whether and how the Bank had to respond to the emergency. An identification mission visited Datong and the affected areas on November 9-11, 1989, approximately three weeks after the earthquake. Project appraisal and negotiations were held in the field just one month later, on December 15, 1989. The Project was submitted to the Board on January 18, 1990. The project agreement was signed 1. In response to a message from the Bank's senior management, on November 8, 1989, the Government of China expressed interest in receiving assistance for a possible emergency recovery credit. Internal Bank correspondence dating two weeks after the earthquake indicates that Bank officials were keen to link the earthquake rehabilitation operation to a wider earthquake prevention program or possibly a wider disaster prevention program, if the Chinese were amenable to the idea. The Bank memo noted that the Earthquake Engineering Institute in Harbin had predicted a period of high seismic activity in 1990 that would presumably affect all areas of China. Moreover, the 1990s had been designated by the U.N. as the International Decade for National Disaster Reduction, IDNDR, and this was seen an a timely opportunity to contribute to its goals. 13 on March 3, 1990. The project became effective in late April 1990, and disbursements began a few months later in the fourth quarter of 1990.  15 2. Project Description Objectives and Components 2.1 The objectives of the Earthquake Reconstruction Project were straightforward. They aimed to: (i) assist in and accelerate normalization of economic activity in the earthquake- affected area by undertaking reconstruction operations in Yanbei Prefecture of Shanxi Province and Yangyuan County of Hebei Province through a village reconstruction program, replacement and re-equipping of essential social and economic infrastructure, and rehabilitation of facilities or productive rural enterprises; (ii) encourage and facilitate the use of cost effective, earthquake resilient construction standards: and (iii) support through technical assistance the State Seismological Bureau (SSB) to improve earthquake prediction and monitoring, and Shanxi Province to develop a model disaster preparedness plan for Yanbei Prefecture. For a detailed description of the project see Annex A. 2.2 China has been involved in research efforts to forecast earthquakes for a very long time. Since 1966 Chinese scientists have made detailed observations of earthquake precursors leading to earthquake predictions. It has established agencies to deal with seismological research in the form of the SSB at the national level and the provincial seismological bureaus and their local units at the provincial and local levels. The Center for Analysis of Prediction (CAP) under the SSB deals with research and analysis procedures. The Beijing headquarters of the SSB collects data from every main city and province in the country. All Chinese institutions involved in earthquake prediction and research are served by SSB's national computer, database and satellite communications network. The Ministry of Construction (MOC) deals with earthquake damage investigation and mitigation. As part of China's efforts for the International Decade for Natural Disaster Reduction, the MOC drew up a program of studies and research into earthquake damage mitigation for its Earthquake Resistance Bureau. 2.3 The project financed a National Technical Assistance Component which included: (i) strengthening and expanding the national system of stationary and mobile earthquake monitoring stations; (ii) improving earthquake information collection, telenetry and processing, including data exchange systems in high seismic risk junction areas, and (iii) expanding the scope of seismic precursor observations at selected stations in high risk areas. Financing 2.4 The cost of the project, as estimated at negotiations, was US$45.3 million (1989 prices) of which IDA provided approximately 66.2 percent through a credit of US$30 million equivalent or SDR 23.4 million. Final completion cost was US$48 million (1992 prices), of which the IDA contribution was unchanged in SDR terms. 2.5 The Ministry of Finance was the Borrower and the arrangement was that it would on- lend the proceeds to Shanxi and Hebei Provinces in local currency, interest free, for 25 years, including 10 years grace, with a one percent annual service charge. The disbursement arrangements included a Special Account maintained in US dollars by the MOF. Local Special Accounts were maintained in local currency and opened at the provincial level in Shanxi and Hebei and at the county level in Yanbei Prefecture and Yangyuan County. Payments were made 16 by the counties from their domestic Special Accounts either as partial payments (65 percent) against invoices submitted by suppliers or as credits to village and household loan accounts for 65 percent of the value of work carried out by contractors or force account. 17 3. Project Performance and Influencing Factors Project Outcome 3.1 Supervision missions rated the project to be free of major problems throughout implementation. The PCR rates the outcome as highly satisfactory, its institutional development as modest, and the sustainability of project benefits as likely. 3.2 The main findings of the PCR are: (i) despite relative inexperience with large projects, implementation capacity in the affected areas proved to be extraordinarily effective (PCR, para. 5.1); (ii) the appraisal targets in terms of physical scale were far exceeded (PCR, Part III, Tables 4 and 5), (iii) local ownership of the project was reinforced through participatory planning (PCR para 4.2): and (iv) projects of this type require a strong positive cash flow (PCR, para 5.7). 3.3 In the following, the audit confirms the good achievement of the project's main objectives and examines some of the factors that explain this good performance: local involvement, financial management, and application of lessons learned. A major factor was the quality of Bank performance which originated in part in the process innovations that were pioneered throughout this project. Section 4, on the other hand, deals with the less positive aspects of this experience. 3.4 While technical assistance did not achieve a substantial impact, the project surpassed its physical targets. Since December 1989, China has devalued its local currency several times. During the same period the US dollar also devalued significantly with respect to the SDR. Due to these realignments, the value of the credit for the project increased by over 40 percent in local currency terms. An extended village reconstruction program approved by the Bank's Executive Board on a no-objection basis and implemented in 1991 absorbed most of the credit savings. Implementation of the additional work (the project became 10 percent larger in real terms) was simplified by experience gained from the project in the first year. 3.5 Both Yanbei Prefecture and Yangyuan County undertook the reconstruction of 281 villages, including the reconstruction of 33,100 dwelling rooms and rehabilitation, repair or reinforcement of 31,700 others. Two thousand seven hundred and thirty-nine public facility rooms were rebuilt (village classrooms, health centers and village storehouses), and about 3,060 others repaired. Village infrastructure (roads, power lines, wells, irrigation channels and minor structures) and two hospitals were rebuilt. The production in nine enterprises was restored. Overall, the reconstruction program improved living standards in the damaged areas and provided assistance in restoring production and stimulating housing construction. Operational management skills in the involved agencies were improved. 18 3.6 The project made a positive impact on seismic monitoring, and disaster preparedness, but a negligible one on earthquake prediction. Also, through training activities, about one thousand persons at various levels in the project areas were trained in project design and planning, preparation, implementation, quality control, and loan administration, and have become important human resources in project implementation. In addition, thousands of farmers learned simple construction skills. Thirty seven engineers from SSB were trained abroad in the operation of equipment, and 230 persons participated in domestic training courses. Significant training was provided during project appraisal and implementation, as a result of an excellent team work among Bank and local staff. The Implementation Modality was Based on a Local Management System 3.7 The emphasis on local management as the basic implementation modality was an important ingredient in the project's successful implementation. Given the nature of the recovery needed, the large area damaged, and the different agencies involved, local management was a very appropriate approach. While the project was managed centrally (MOF) and the central govermment was the borrower, specific reconstruction activities were handed over to finance Bureaus at the Prefecture and county levels. Village recovery programs were executed by local governments. Following a Chinese practice of decentralization, project management offices were set up at each level of government (prefecture, county and township) to manage reconstruction activities. Each office had staff from line ministries, including the Planning Commission, Civil Affairs Bureau, Water Conservancy Bureau, Design and Planning Institute. After project completion, the local Project Management Offices, PMOs, were disbanded, and all staff returned to their original organizations. Thus, the newly created Offices only lived throughout project implementation. They became strong implementation units with effective leadership and coordination of the village reconstruction activities. An important ingredient in the success of the PMOs was that their heads were all persons with many years of experience and strong management and administrative ability. Project staff, selected for their professional skills and good working attitudes, performed excellently in promoting smooth project implementation, particularly under conditions which required a limited implementation time and which entailed a scattered distribution of project sites. The highly participatory modality used to implement the project, based on local management and decentralized decision making, proved to be very successful. Each village plan for the affected villages was prepared by the village and by the direct beneficiaries of the credit. Participation promoted an energized commitment at the local level. Although the local PMOs were disbanded, the lessons learned from this experience remained with the reconstruction management staff after they moved to other organizations.2 2. Comments received from the Borrower tsee Annex D) underline the importance of the local management system. "Strict supervision and efficient operation are the keys to project success. During the implementation of the project, we established a good institutional system. The Project Units were set at the levels of county, town and village, contracts were signed, duties and deadlines were specified. The county government established the Project Office with special units dealing with finance, purchasing, project management, and quality control. Following the standards of the World Bank and local conditions, these groups effectively managed the project and obtained superior outcomes. After the first earthquake, the reinforcement of earthquake-resistant measures for the new housing produced a significant result: none of these housing units fell or had any cracks on the walls when a earthquake with a magnitude of 5.8 Richter scale occurred in Yangyuan on March 26, 1991." 19 The Financial Management for Reconstruction was Based on Recovery Needs 3.8 The project set up a good system of financial management for reconstruction through an on-lending system among provinces, prefectures and counties. Loan allocations to prefectures were based on an assessment of their affordability and their needs for reconstruction works. Loan contracts were signed between townships and families, thus, loans were fully allocated to earthquake victims. The allocation system based on beneficiary needs guaranteed a proper use of loan funds. In China, past experience with funds for post-earthquake reconstruction has not always been positive. Those funds traditionally come from the central/local government through a financial allocation which does not always contemplate recovery needs, but which may include other investments as well. In many cases, it is very difficult to guarantee that funds are allocated to beneficiaries, particularly after the immediate crisis is over and priorities change. Typically, repayment of loans by the provinces to the central government is not required and funds may be diverted for other uses by local authorities. The project experience, including on-lending and contracts among the provinces and the prefectures, was positive on this score. It is an important experience that could be duplicated in future reconstruction activities in the country. Special Account and Cash Flow: A Bottleneck 3.9 Even though mechanisms such as retroactive financing and a Special Account were used, funding problems persisted. Retroactive financing for eligible expenditures incurred after December 15, 1989, were approved and a Special Account was established. Initially, implementation was held up due to a lack of working capital. The limited funds deposited in the Special Account proved to be a bottleneck and IDA agreed to increase the amount from the standard equivalent of three months average expenses, about US$4 million, to US$8 million. Bottlenecks in the disbursement process was a persistent issue, mentioned repeatedly in supervision reports, and discussed with the audit mission. Problems in disbursement were due to deficiencies in documentation, delayed fund application, and slow processing in the Provincial Finance Department and the Ministry of Finance (MOF). 3.10 The project included special disbursement features, not previously implemented in the China portfolio. Those special features included: (i) special procurement arrangements; (ii) special account management by the MOF's World Bank Department; (iii) on-lending by MOF in local currency for local expenditures, and (iv) multiple tiers of on lending agreements: MOF to Provincial Finance Bureau, (PFB); PFB to Prefecture Finance Bureau (PFFB); PFFB to County Finance Bureau (CFB); CFB to Township Finance Unit (TFU); TFU to Villages. The several layers of government involved in the SA withdrawal applications required a long path for disbursements. In practice, the cycle should have taken 30-45 days from the central to the local level. Many delays occurred that resulted in a cycle of about 60 days. Delays occurred due to deficiencies in documentation and tardy transmittal and processing. Disbursement problems required constant attention during implementation. In retrospect, a large amount in the Special Account upstream plus special training for implementing agencies and the MOF on the management of a Special Account would have been helpful. Compliance with Covenants and Operational Directives 3.11 Many Bank Operational Directives, Operational Manual Statements, Operational Policy Notes and Guidelines were applicable to the Earthquake Reconstruction Project. Of particular importance was the adherence to Operational Directive 8.50, "Emergency Recovery Assistance". 20 The project was in compliance with the standard administrative, legal and procedural requirements relating to project preparation and implementation. 3.12 Guiding principles for project design, many derived from 0D8.50 "Emergency Recovery Assistance", included the following: * outward orientation towards the client, entailing an early involvement and ownership by the borrower. In this case, the client was already engaged in recovery activities as a matter of priority; * bringing the Bank close to the client through an expanded Resident Mission role in project preparation, negotiations, and supervision; * flexible upstream design. * clarity of purpose in terms of well defined and limited recovery goals within the context of development, i.e., restoration of assets and productivity and a realistic time schedule, * use of a simplified documentation process with only one master document-the President's Report-going to the Board. Past Experience on Emergency Recovery Lending Helped in Project Design 3.13 The China Project took advantage of the Bank's significant experience in Emergency Recovery Lending (ERL). Since its earliest days, the Bank has responded to emergencies by providing advice and financial support, mostly through (i) reconstruction loans and (ii) modifications of the existing portfolio and lending program. The Bank's experience in providing assistance for earthquake reconstruction is significant. It includes operations in Peru, Nicaragua, Guatemala, Romania, Yugoslavia, Colombia, Yemen, Mexico, Chile, Iran, the Philippines, Nepal, Ecuador, El Salvador, and India. Such experience indicates that important considerations in designing ERLs are: (i) early involvement of Bank staff, (ii) strong government commitment to recovery. (iii) limited objectives and realistic time schedule, (iv) rapid agreement with government on the strategy and scope of the overall recovery program, (v) conditionalities linked directly to the emergency, (vi) simple implementation arrangements, and (vii) use of disaster resilient reconstruction design standards. All these lessons were integrated in project preparation upstream. The Reconstruction Project was a Pioneering Business Process 3.14 Although many ERLs follow a "fast-track" approach, this project had innovations such as negotiations in the field immediately following project appraisal and project supervision from the Resident Mission. It was an early, unplanned "test-bed" for Business Process Innovation (BPI) processes, five years prior to their formal implementation. The principle underlying current BPIs is for the Bank to better serve its clients by taking full advantage of process innovations needed in a world that has changed. The processes selected as "test-bed" include the adoption of higher participation, fast cycle design, modifications in the procurement process, and use of communications/information technology as an enabling tool. 21 3.15 During project preparation, extensive use was made of information and communications technology by the Bank's team to prepare the combined Appraisal/President's Report and to conduct negotiations. Negotiations were completed in the field, approval was sought also from the field, and presentation to the Board took place two months after the identification mission visited China. A benefit of having negotiations immediately following appraisal was that they involved the same people on both the Bank and client's side. They operated as a team and ownership of the objectives was fully shared, although some differences of opinions (e.g., housing reconstruction) surfaced at appraisal. 3.16 IDA provided critically important technical assistance in the course of project supervision, during 1990 and 1991. Five supervision missions were carried out by the Resident Mission during that period. The missions provided strong guidance in detailed village design, in some instances less successfully than in others (e.g., housing standards). The good cooperation between IDA and the Borrower allowed for the sharing of views on how to proceed and sharing information about how the Bank/IDA works, and on procurement and disbursement mechanisms. The exchange of views was very important in solving problems along the way. 3.17 In the end, the proiect required only 19.9 staff weeks to prepare, appraise and negotiate and 40 staff weeks to supervise. In China, the average required by investment projects is 162 staff weeks to prepare, and 81.78 weeks to supervise. According to regional staff interviewed during the audit, anything under 120 staff weeks to prepare is considered cheap. The elapsed time for the project from approval to effectiveness was 0.21 years, while for the average investment project it is 0.51 years. The elapsed time from approval to closing was 2.68 years, for an investment project the average is 6.66 years. 3.18 The time required to prepare and implement the project was also lower than the average required for emergency projects. The average required by a sample of twenty projects prepared and implemented in FY85 through FY89 was 0.6 months from the emergency (identification) to approval and 2.89 years from approval to closing. 3.19 The audit noted however a range of varying reactions to the challenge of a fast-track approach. The discrepancy between perceptions and responses of staff in the field and at headquarters is particularly relevant at this time when efforts are being made to decentralize Bank operations. The experience with this project in China indicates that the transfer of responsibilities to the field may not be easy until some degree of decentralization and standards of quality congruent with efficiency requirements are formally established. 3.20 Staff in the Field. Preparation in the field worked very well. The experience with this project indicates that for small projects there is no need of doing the appraisal and then returning to headquarters to process the appraisal report and to negotiate the loan. It can all be done back-to-back while in the country. On this project IDA had a small and very experienced team with a "let's roll up our sleeves" attitude. A lawyer visited China after appraisal, in time for negotiations in the field. The team's goodwill helped dispel stereotypes prevalent locally, in Beijing, that it was very difficult to work with the Bank. The small identification/appraisal group consisting of six staff and consultants functioned more efficiently than have large groups on similar projects in other countries in the past. Good teamwork was established with the Borrower. Staff commitment and continuity in supervision was very important in the successful implementation of the project. 22 3.21 Staff at Headquarters. The "fast track" approach created a few bureaucratic problems. Standard management practices were not compatible with the modified project process. While managers expressed a strong commitment to the rapid processing and approval of the project, they hesitated to modify their standard approach to project clearance. After appraisal, management at headquarters expressed dissatisfaction to staff in the field. The President's Report was criticized as not being up to standards in presentational quality. Numerous corrections were requested. This perception left staff in the field feeling that their efforts were not being appreciated, and distressed considering the harsh conditions under which the project was prepared in the damaged areas, the reduced preparation time, which continuously required overtime work, and the fact that staff spent Thanksgiving and the end of the year holidays in the field. 23 4. Implementation Issues Resettlement or In situ Reconstruction? 4.1 Relocation of population was not justified on the basis of risk factors. However, there is a strong resettlement tradition in China to deal with populations affected by disasters. Thus, a resettlement approach was proposed by local authorities at the outset. The appraisal mission adopted a reasonable position in not following a generalized resettlement path, since resettlement not only implied a disruption in the lives of people who already had suffered losses, it meant also a loss of existing undamaged capital investment in infrastructure in the affected villages. In the end, the appraisal mission and the local authorities agreed that rebuilding in situ was to be the main approach for the project. In the case of only eight villages, relocation was deemed justifiable due to two factors: (i) ground motion had caused a significant loss of buildings and structures and over 90 percent of the total building floor area of certain villages was totally damaged, and (ii) there was lack of water supply and inconvenient communication in some of the affected villages. In all cases, inhabitants in the resettled villages were willing to move. Housing Replacement: Differences in IDA/Local Views Concerning Standards and Approach 4.2 The housing replacement standard proposed by the appraisal mission became a bone of contention during implementation. It created some resentment among local authorities, who stressed this issue to the audit mission. 4.3 The President's report recommended three types of replacement rooms for affected dwellings: (i) a concrete frame, adobe block walls, timber purlin, and rafter roof at cost of 159Y/m2; (ii) a wood frame structure, with, masonry walls, timber truss roof at a cost of 194 Y/m2; and (iii) an adobe vault with reinforced thin shell liner at a cost of 120Y/m2. Dwelling types (ii) and (iii) proposed by the appraisal mission against strong resistance by local engineers were finally not used in the reconstruction program, since they were not found to be justifiable in terms of earthquake resistance, relative costs and construction materials available. A local building technology of wooden structures rather than the steel mesh proposed by the appraisal mission, described in (iii) above, was finally adopted, but only after time consuming and costly testing (200,000 Yuan for test) at the local University. 4.4 The Audit team noted that the housing replacement standard proposed by the appraisal mission was not the same low-cost adobe technology used in the area, but a reinforced adobe structure designed by a Bank consultant during appraisal. The new technology was not "simple" and "low-cost" as the appraisal team argued, but complex and expensive. It would have required 3. On this issue, the members of the appraisal team note that the appraisal mission's aim was "to maximize use of repairable stock and utilize local low-cost technology". They note that "local official attitudes favored relatively expensive dwelling replacement over repair and village resettlement over reconstruction" (memo from Richard Scurfield to Yves Albouy, April 1, 1996). The government, on the other hand, in its comments to this audit (Annex D para. 2) note that "The Bank appraisal team proposed the use of steel cases for reinforcement (in the "cave" houses). However, this would require high costs in both materials and labor, reduce usable area of the rooms, and exceed the construction value of the original "cave" houses. This proposal did not match the local residents' reality and hence, was not accepted. An alternative method was used for the reinforcement of the units. 24 substantial training for construction workers before implementation, particularly because it proposed a structure based on the construction of vaults (not an easy task) with a reinforced shell liner. It was a construction technology imported from another country, Mexico, where the Bank financed a very successful housing reconstruction initiative. There was strong resistance by the local population and authorities to depart from established practices, and a strong resentment against what was perceived as the imposition of models from the "outside", foreign to China, which were found to be inappropriate to local conditions and traditions. 4.5 The experience with the cultural, technical and functional aspects of the housing component indicates that the Bank needs to listen more to local people. What is successful in one country may not be appropriate for another one in a different region. Furthermore, in a context in which local populations experience heavy losses, it is very difficult to depart from established models and traditions, and to attempt to impose new approaches to housing, an element of daily life which carries with it deep meanings that go beyond function and cost. Technical Assistance: Did it Work? 4.6 A main component of technical assistance consisted of financing earthquake monitoring and prediction at the national level through the State Seismological Bureau, SSB. This was found to be an inefficient use of IDA resources. The project failed to tackle the complexities of earthquake prediction, a research avenue whose usefulness has been questioned for over two decades by the expert community worldwide. Even though earthquake prediction appears as a tantalizing option, a recent piece in the New York Times( August 8, 1995) underlines the ongoing technical discussion: "Hopes for predicting earthquakes, once so bright are growing dim. Many of the leading seismologists now think that earthquakes are inherently unpredictable. While recent research suggests that some earthquakes may produce precursory signals involving shifts in the earth's crusts, those signals are so small, faint and hidden from view that detecting them in any practical sense may be impossible". The knowledge required for a sound evaluation of an earthquake prediction research system was absent from the project team, and the tradition of earthquake prediction in China is very strong. Thus, the Bank's project team went along with this route, without a proper assessment. 4.7 Research on earthquake prediction is nothing new in China; important resources have been devoted to this activity in the country for twenty years now. Overall, the support of IDA for this activity amounted to a small part of the overall SSB budget. It is debatable whether the best use of IDA resources was to provide marginal support to an area of research, such as prediction studies, which was controversial when the project was prepared. In fact, at appraisal there was discussion about whether this was the best use of IDA resources, or whether it would be better to support prevention and mitigation measures, like the ones that were being undertaken by the Ministry of Construction on planning and design. In an ill-advised move, the appraisal mission opted for earthquake prediction. Considering the vast experience and resources already devoted by SSB to earthquake prediction and monitoring, coupled with (i) the uncertainties surrounding prediction research and (ii) the seismic vulnerability in the country, the audit concludes that a more efficient use of IDA's development funds could have been provided by emphasizing disaster impact reduction through appropriate training in structural systems, land 4. For examples of substantial criticisms to earthquake prediction see publications by the National Academy of Sciences - National Research Council, Working Group on Earthquake Hazards Reduction, 1978 and Committee on International Disaster Assistance, 1980. 25 use planning, seismic appropriate architectural designs, construction supervision, quality control, and insurance mechanisms. Only a very limited activity was undertaken by the Project concerning the reduction of vulnerability. In the view of the audit, the latter would have been a more productive approach in the context of emergency recovery and development work. Quality Control was a Problem Area 4.8 The lack of local building codes and established construction inspection and quality control mechanisms posed problems at the outset. In fact, problems with quality control in the reconstruction work were noted during several supervision missions, and recorded in Aide- Memoires and Back-to-Office reports. The main problems identified included poor concrete quality in cast-in-place ring beams, low mortar strength in load-bearing brick walls, and loose brickwork underneath and between purlins on internal bearing walls. The affected prefectures attempted to resolve this problem with assistance from IDA and local universities. To guarantee project quality and to improve project benefits, a very strict system of quality control was set in place for reconstruction works during supervision, about one year after implementation started. The system included: (i) a quality control network, consisting of quality management and supervision organizations and authorized inspectors at the prefecture, county, township and village levels; (ii) selection of construction units for non-private buildings through bidding procedures; (iii) training of planners, construction managers, and inspectors through courses at different levels; (iv) provision of simple guidelines for constructing replacement houses; and (v) provision of standard construction drawings to organizations and persons requiring them. 4.9 Although there was improvement in construction quality during project implementation as a result of the assistance mentioned above, persisting problems similar to the ones noted by supervision missions (see para. 4.7) were noted by the audit mission. In spite of the shortcomings noted, however, overall, the jump in the quality control system from pre- earthquake conditions to the post-earthquake situation, played an important role in better seismic safety conditions. The magnitude of improvements achieved in construction became evident and were tested when an earthquake of magnitude 5.8. occurred in March 1991, and the rebuilt and repaired buildings and structures withstood the shock.  27 5. Conclusions, Lessons Learned and Recommendations Project Ratings 5.1 The audit corroborates the assessment provided in the self-evaluation of this project (Project Completion Report) concerning the success of this project. The audit rates the project outcome as highly satisfactory, its institutional development as modest, and the project approach to reconstruction as sustainable. The audit ratings concur with those in the PCR. 5.2 OED rates Bank and Borrower performance as highly satisfactory. The Project benefited from the integration of lessons learned in similar situations. It served also as a "test- bed" for BPIs introduced five years down the line, since IDA adopted a new way of doing business. The compressed process of project preparation and supervision entailed significant savings in terms of staff time. Compared with other projects in China, the elapsed time from approval to effectiveness to credit closing was also significantly reduced. This efficiency and the success of the project are very much the result of commitment from both IDA and the Borrower, excellent team work among IDA staff, realistic objectives, IDA's insistence on in-situ reconstruction, extensive use of local teams in project implementation, and the avoidance of new institutions, except for local project management units, which were dissolved after project implementation was completed. 5.3 The main problems identified were headquarters staffs reluctance to adjust to the modified requirements in a new way of doing business, IDA's acceptance of funding research on earthquake prediction, attempts at imposing component designs that were successful in other reconstruction contexts, quality control deficiencies, bottlenecks on the cash flow and in the operation of the Special Account. Lessons Learned 5.4 The Capacity in the Bank for Undertaking Reconstruction is Sound The success of this project, as well as the satisfactory experiences in previous reconstruction projects, such as in Mexico and Colombia, should dispel any doubts concerning the Bank's capacity to undertake emergency projects, to react rapidly, and to cooperate fully with the affected communities. 5.5 The main lessons learned and recommendations for the future in similar projects are listed below. They highlight the difference in nature between the decisions for which local comunity participation is key (building technologies and project management) and those for which the Bank's worldwide experience and technical knowledge should make it more assertive (in-situ reconstruction, impact reduction rather than earthquake prediction). (a) Promote In-Situ Reconstruction. Relocation needs to be strictly controlled. In- situ reconstruction makes better use of existing infrastructure and community facilities. Resettlement is more expensive both financially and socially; it needs to be strictly controlled and justified. Considerable self-help efforts in low-cost, largely in-situ reconstruction were stimulated by the appraisal mission and project design team, particularly because the IDA credit was directly targeted to the beneficiaries through special agreements signed between village and 28 household owners on the basis of comprehensive reconstruction plans prepared by each participating village. Prior to this project, the advantages of in-situ reconstruction were shown in the earthquake recovery project in Mexico, which had a very successful housing component respectful of community networks. (b)- Minimize Potential Disaster Impacts. Reducing vulnerability through architecture, urban planning and engineering should receive priority over earthquake prediction. Preparedness and prevention are important mechanisms for future disaster reduction, particularly in a country as vulnerable to seismic events as China. On very specific technical aspects, the Bank should seek specialized technical advice, since participatory mechanisms of decision-making are not appropriate for highly specialized technical issues. Earthquake prediction is an unproven approach and promoting it, even in China, where it benefited form a long tradition, proved not to be the best use of IDA's resources. (c) Participatory Design ofProject Components is a Factor of Success. Problems arose when there was an attempt by Bank staff and consultants to introduce building technologies from other regions of the world; these were found not to be appropriate to local conditions after expensive seismic studies and testing were conducted at the local University. (d) Secure Strong Local Leadership and Project Management Offices. A strong local leadership and an efficient system for reconstruction management were important factors in the efficient implementation of the project. The Project Management Offices at the provincial and local levels became strong implementation units. In "normal" development projects, Project Implementation Units are not always needed and often considered as artificial and unsustainable. The case is different for post-disaster programs where there are no functioning disaster response systems at the local level. Experiences with the Mexico and the Popayan (Colombia) Earthquake Reconstruction Projects confirm the importance of strong local project offices. (e) Minimize Bottlenecks to the Project's Cash Flow. Emergency projects require a strong positive cash flow and special attention to the design and implementation of disbursement arrangements. In this case, although a Special Account was established to help deal with cash flow needs, problems were encountered by the local Departments of Finance concerning timely disbursement from the Special Account. Some problems were found at the local level due to delays in the submission of documentation and reimbursement forms. Local disbursement procedures were complicated, and proved to be a bottleneck. In addition, numerous project entities were borrowing for the first time and were not familiar with IDA disbursement procedures. Specific guidelines concerning disbursements from the MOF to local units may in the future help to avoid bottlenecks in project implementation. (f) Develop a "Fast-Track"Approach. The use of simplified procedures, fast cycle design and processing, a client orientation, and increased participation of the resident mission helped the success of the project and its development effectiveness. The close and productive cooperation between staff at local 29 implementing agencies and IDA was an important ingredient in the success of the project.  31 Annex A Description of the Project According to Schedule 2 of the Development Credit Agreement, the objectives of the Project were to: (i) assist in earthquake reconstruction operations in the Affected Areas; and (ii) Strengthen SSB capabilities in earthquake prediction and mitigation. The Project consisted of the following parts: Part A: Shanxi: 1. Provision of civil works, construction materials, and goods for reconstruction works in about 80 villages in Yanbei Prefecture (including about 4 villages to be relocated), including: (i) reconstruction of about 19,400 collapsed dwelling rooms, and rehabilitation of about 21,000 other rooms; (ii) reconstruction of about 900 public facility rooms (village classrooms, health centers, village storehouses and other public buildings), and repair and reinforcement of about 1,600 other rooms; (iii) re-establishment and repair of critical village infrastructure of power lines and water supply; and (iv) repair of damaged buildings of about 13 rural enterprises. 2. Drawing up an emergency preparedness plan for Yanbei Prefecture. Part B: Hebei: Provision of civil works, construction materials and goods for reconstruction works in about 70 villages in Yangyuan County, including: (i) reconstruction of about 2,400 dwelling rooms, and rehabilitation, repair and reinforcement of about 600 other rooms; (ii) repair of about 35 village's water supply and irrigation works, and (iii) repair of two seriously damaged industrial and commercial enterprises. Part C: Strengthening earthquake prediction capability of SSB through provision of equipment, consultant services and training.  33 Annex B Project Agreement Agreement dated March 2, 1990, between International Development Association (the Association) and Shanxi Province and Hebei Province (the Provinces). Whereas the Development Credit Agreement of even date herewith between Peoples' Republic of China (the Borrower) and the Association, the Association has agreed to make available to the Borrower an amount in various currencies equivalent to twenty-three million four hundred thousand Special Drawing Rights (SDR 23,400,000), on the terms and conditions set forth in the Development Credit Agreement, but only on condition that the Provinces agree to undertake such obligations toward the Association as are set forth in this Agreement; and Whereas the Provinces, in consideration of the Association's entering into the Development Credit Agreement with the Borrower, has agreed to undertake the obligation set forth in this Agreement; Now therefore the parties hereby agree as follows: Article I. Definition Section 1.01 Unless the context otherwise requires, the several terms defined in the Development Credit Agreement, the Preamble to this Agreement and in the General Conditions (as so defined) have the respective meanings therein set forth. Article II. Execution of the Project Section 2.01 (a) Each of the Provinces declares its commitment to the objectives of the Project as set forth in Schedule 2 to the Development Credit Agreement, and, to this end (i) Shanxi shall carry out Part A of the Project, and (ii) Hebei shall carry out Part B of the Project, both with due diligence and efficiency and in conformity with appropriate administrative, financial and engineering practices, and shall provide, or cause to be provided, promptly as needed, the funds, facilities, services and other resources required for the Project. (b) Each of the Provinces shall make available to the Beneficiaries, in accordance with the arrangements satisfactory to the Association, the proceeds of the Credit made available to the Provinces by the Borrower under sections 3.01 (c) and (d) of the Development Credit Agreement. Section 2.02 Except as the Association shall otherwise agree, procurement of the goods, works and consultant's services required for the Project and to be financed out of the proceeds of the Credit shall be governed by the provisions of Schedule 3 to the Development Credit Agreement. 34 Section 2.03 Each of the Provinces shall carry out the obligations set forth in Sections 9.03, 9.04, 9.05, 9.06, 9.07, and 9.08 of the General Conditions (relating to insurance, use of goods and services, plans and schedules, records and reports, maintenance and land acquisition, respectively) in respect of the Project Agreement and its part of the Project. Section 2.04 (a) Each of the Provinces shall, at the request of the Association, exchange views with the Association with regard to the progress of its part of the Project, the performance of its obligations under this Agreement and other matters relating to the Purposes of the Credit. (b) Each of the Provinces shall promptly inform the Association of any condition which interferes or threatens to interfere with the progress of its part of the Project, the accomplishment of the purposes of the Credit, or the performance by such Province of its obligations under this Agreement. Section 2.05 Each of the Provinces shall promptly carry out the reconstruction and repairs under the Project in accordance with the planning, design and standards, and in accordance with the time table agreed with the Association. Article M. Financial Covenants Section 3.01 (a) Each of the Provinces shall maintain records and accounts adequate to reflect in accordance with sound accounting practices its operations and financial condition in respect of the activities related to its parts of the Project. (b) Each of the Provinces shall: (i) have the accounts referred to in paragraph (a) of this Section, including the Special Account, for each fiscal year audited, in accordance with appropriate auditing principles consistently applied, by independent auditors acceptable to the Association. (ii) furnish to the Association as soon as available but in any case not later than six months after the end of each such year: (a) certified copies of its financial statements for such year as so audited, and (b) the report such audit by said auditors, of such scope and in such detail as the Association shall have reasonably requested; and (iii) furnish to the Association such other information concerning said records, accounts and financial statements as well as the audit thereof, as the Association shall from time to time reasonably request. 河北省阳原县地震通建项目的经济发展起到了积极而有效的促进作用，1,9即 同毓刀有亚沁5年我县国民生产总位达到35214万元日9二 一‘’一”--9。年不，价），比1 99。年增长了，4 .4寒 河’匕省阳原，地震，建项”在世界银行的““时政收入完成，&6’万元，比，,90年呈 ”一，、在各级各部门的，”下，经过，9,0年增镼，&“·‘&,‘农民人均纯收’、达’110,，罗 ,&&”年两年的努力工作·已经全面完成·元1’匕，99”年增长了’&,’·”％。里 1石曰剐，．系．.&：遭·占二里不口甲习‘灾点宜男二。人，电二之汤人石刁 该项目共为阳原县84个村庄的1 481户断建一”&”了叫‘何为践”,&&‘俘’！一’卜，,,.‘什成””臼 “房3,2‘比为，1 19，户修络和加因住肯定，项目的执行经脸和教训有待于时问的检验爵 房26&,7问；为，8个村庄新打机井70眼、和今后工作的不断总结，问顾项目的执行，成果 新’城防渗水扫，万未修复灌扣·、万米和效互我刃感觉有以下几点值得评沈 ：为61听学校新途校舍640间，修复加固校卜地装灾害紫急援助的必冬l生。阳原地哀 舍，93问、购籮课奈和1 75，套及部分设灾害造成了巨划嗡失、时于一个贫困的地区，吕 备·新建八处、，4问敬老院；为两所医院新建、无疑是雪上加霜。如果得不，lj外来的援动、用两 8 40平方术的住院、门诊楼：为两个企业修年时l&0l完成地戏重建工作是不可能的。所以在阳 复商店两处；祈建一处面粉加工丰间。超顺1．原县发生地宋灾害之后，得到了世界银行的关汪． 4％完成项目总投资。及时地进行了实地考一蔡、认为这一贫挤的地区应 项目的完成产生了巨大的效益：为近5万人该得到世行的贵助，并且立为华北地衷重建紧急 ．。‘二＿,,，。，二，，。二、～,，八．*‘忍．，。项目．把建设周期定为一年左右．给予了及时的 解决和改善一了居住问题；为1 3594名学生住入”&,“、“&，明入“’一甲’&&&~&.&&& ．。，“小，一，、，，、，.,,,*，。。，,n,*J、＿，帮助和支持，取得了预期的效果、我们感列做为 进了其有杭震设施的校舍．伙2 3 5 06名学生’&~&，人．&J’一尸‘切“入，．碑’、，一”’、”“扮们‘&‘子 ．。二，,，。，,&．。二二。．,．。。。‘。。，．、二，世弃性金融姐织的支持是必要的。 用土了崭断的课桌觉；新增加和改善水浇地6．卜”&&&~~&~”碑人．&J~~‘一‘& --－·，-－一·－一－－·＊·－---，枯2、项目的实施是实事求是的．项目的确立 5万亩；为30万人改善了医疗防疫灭病条件；“、一界”&J不～～乃下”&&J&&&&&”一 .~,‘、。＊、．&．、．。。，&，。．。二．,＿人和实施既要符合执襄设施的要求，又要切合当地卜 为250多名孤寡老人解决了住房；改德一了三个’‘、～~～一’&~~&”一’&’一’一‘&&&&’一百 &-．·，，一升·．&,JL .n-L一、，一‘,,，、、。的实际，这样才能保证项目的顺利有效地实施．里 企业的经营生产条件。同时，项目资金的注入和”J7、’.&.&～一”&,&&”一八”&&&&&”门～～外～”舀 ,.0.,,～洁．人，&=，…二I&～二。．，二…我们在项目的实施中感觉到，由于世界恨行的成6 佼用及共回收资金的再佼用，对阳原县各行各业’、”’一户””一‘7、～’玛为”’川“&”一’&&&~&J 12 员．来自世界各地、成镼过程、生活经历、学识建立责任制，签仃一责任状，明确具体任务和完成 水干、风土人情的不同，所以，接触到阳原的实时间。县成立了项目办公室：没立了财务组、物 际情况，显然会叶阳原重建项目会提出不同的见贵朱购组、项目节理组、质黄监替检查组，严格 解，这也是正常的，合手情理的，但必须与中国按照世行的要求：结合阳原实际，进行了有效地 的阳原县的实际情况相结合、才能有效地实施项管理，取得了很好的效果。如务反的浙建和加固、 目。比如时土窑洞的加固，世行评估组提出用钢接受地震灾害的经验教训．加强了杭炭设防措绝， 筋籑加固：这样投资高用工多，簩小了原有住房取得了明显效果，在190）年3月26日，阳 面积，超过了原窑洞的建设价值，完全不符合当原再次发生里氏5.8级中强地震时．无一裂缝、 地居民的生活实际需要，所以没有彼接受。而为倒梢。 别的维修方式所代替。5、有效快捷的贷款支付保汪了项目的顺利 飞、确立项目要有镼远观点，要考虑项目的实施。世行．是实行报帐支付制，这在阳原县是没 效曲性。在这方面·我们布一个教训，阳原县根有实行过的，’士于一个贫困地区，特别是遭受地岁 油加工牟同的新建。在重建项目中，投资50万衷灾害的农民、自己的支付能力有限，所以及时 元断建了～个。l,8乎方米的依油加工牟问，地得刘世行的资款是如士fL，叔渴的。在这方而．我 预计年加工能加05万p屯。但由于当时处于依们按照世行的要术，次时报项目开工准备的追溯 食统一调拨，茧不是根食生产区，但有土级调运，报帐，及时地收到货欲，为农民早日完成重建注 有根源可加工，而现在是市场经济，效益是第一入资金。农民高兴地说，地采灾害．因祸得福 位的，现在已经满足不了其加工能力，处于仔产起房盖崖全家忙，经济不是靠世行。 半停产状态，影响了投资效益的发挥。所以说，6、项目的完成，离不开各级各部门的帮助 时投资的效益性必须做为一个重要的问题、首要和支持。阳原利用世行货款进行地震重建项目． 的问题加以重视。是一项新工作，时于世界鉴卜行的要求和工作程序 4、项时勺严格管理和有效运作，是搞好项一切都是末知教，在各级各部门的支持下，我们雪 目的关键．在买施项目中，我们建立了完整的从J页。的准备、谈判、立项、实、和验收进行了昊 组织体系。县、乡、村项月单位层层建立组织，认城一地学习和研究，从不懂到懂，列熟练掌握。。 34 比如支付报帐，时政部召集了专门的培训班，省 时政厅、市时政局手把手地教，力．顺利支付货款 起到了有力的保证作用。 总之，利用世行货款进行地震重建，在我县 是第一次利用外资，取得了较好的效果，我们仍 需不断．慈结，不断．提高，时其成果仍需不断评j丸 一九九六年五月十六日 七J ．习 》 口 口 抢 冰 6 5 38 39 Annex D Translation of B6rrower Comments Yangyuan County, Hebei Province Earthquake Reconstruction Project Retrospect and Evaluation The North China Earthquake Reconstruction Project, financed by the World Bank and supported by all levels of the government, has been fully completed after the efforts of 1990 and 1991. The project provided 1,481 households in 84 villages in Yangyuan County with 3,724 new residential rooms, fixed and reinforced 26,037 rooms for 11,195 households, drilled 70 new wells in 58 villages, repaired flood-control dikes of 150 kilometers and irrigation dikes of 35 kilometers, built 640 class rooms for 61 schools, repaired and reinforced 592 class rooms, added 11,753 class room table and chair sets and other educational equipment, built 154 rooms in eight welfare housing units for the elderly, built 5,840 square meter hospital rooms and clinic offices for-two hospitals, repaired two stores for two enterprises, and built a flour-processing factory. The project achieved 1.4% more than planned. The project produced enormous benefits: new and improved housing for nearly 50,000 people, new earthquake-resistant buildings built for 13,594 students, new class room furniture that is now being used by 23,506 students, added or improved 4,300 ha. of irrigated land, up-graded health facilities serving 30,000 people, housing for 250 elders, and infrastructure for three enterprises. Furthermore, the use of the project fund and the use of the proceeds of the fund played a very active and effective role in the economic development of all sectors in Yangyuan County: the 1995 GDP of our county reached 35.2 million yuan (1990 constant yuan), up 54.4% from the year 1990; fiscal income was 3.2 million yuan, up 158.4% from the year 1990; farmers' net annual income per capita was 974 yuan, an increase of 305.8% from the year 1990. The successful completion of the project won praise from the World Bank and government officials. Project implementation experience and lessons are yet to be tested in the future, and summarized in our continued work. With regard to the implementation, achievements and effectiveness of the project, we feel the following is worthy of being evaluated. 1. The necessity of urgent relief after the earthquake. The Yangyuan Earthquake incurred tremendous damages, adversely impacting an already poor area. It would not have been possible for us to complete the reconstruction in two years without outside assistance. After the earthquake. the World Bank made a timely field investigation in the earthquake affected area, and decided to prepare a relief project for the poor area. The Bank regarded the project as urgent, and the project should be completed in about one year. The timely help resulted a positive outcome. We feel the relief effort from an international development agency was necessary. 2. The implementation of the project was practical to the local people. To ensure smooth and efficient implementation. project objectives and design should meet the requirements for earthquake-resistant structures, as well as appropriately take the local reality into 40 Annex D consideration. During the implementation process. we felt that, due to the very diverse background of the World Bank staff, it was normal for the Bank staff to have different opinions on the Yangyuan Reconstruction Project. However, it must take the local conditions of Yangyuan County in China into account in order for the project to be effectively implemented. The reinforcement of the "cave" houses is such a example. The Bank appraisal team proposed the use of steel cables for reinforcement. However, this would require high costs in both materials and labor, reduce the usable area of the rooms, and exceed the construction value of the original "cave" houses. This proposal did not match the local residents' reality and hence was not accepted. An alternative method was used for the reinforcement of the units. 3. The design of a project should have a long-term view, and consider the effectiveness of the project. The new construction of the Yangyuan food/oil processing factory is a lesson. The project invested 500,000 yuan to build a new 1,198 square meter food/oil processing factory with an estimated annual processing capacity of 1.05 million tons. Although a non-food producing area, Yangyuan was allocated with food to be processed by the government under the old system when the project was designed. Now we are under the market-oriented system and effectiveness is the first priority; the factory cannot be operated in full capacity and has partially shut down, reducing the effectiveness and benefits of the project investment. Therefore, the effectiveness of the investment should be considered high priority. 4.'Strict supervision and efficient operation are the keys to project success. During the implementation of the project, we established a good institutional system. The Project Units were set at the levels of county, town and village, contracts were signed, duties and deadlines were specified. The county government established the Project Office with special units dealing with finance, purchasing, project management and quality control. Following the standards of the World Bank and local conditions, these groups effectively managed the project and obtained superior outcomes. After the first earthquake, the reinforcement of earthquake-resistant measures for the new housing produced a significant result: none of these housing units fell or had any cracks on the walls when a earthquake with a magnitude of 5.8 Richter scale occurred again in Yangyuan on March 26, 1991. 5. Prompt disbursement ensured smooth project implementation. The Bank uses the reimbursement system for disbursement, which was never used before in Yangyuan. For a poor area, especially for farmers affected by the earthquake, payment ability was limited, and thus the Bank's prompt disbursements were urgently needed. Following the Bank's requirements, we quickly sent the retroactive bill of project preparation and received loan disbursement promptly, providing funds for the reconstruction of our region. The general feeling is that the earthquake resulted a boon because of the financial help from the World Bank and ensuring reconstruction. 6. The completion of the project relied on support from government officials at all levels. It was a new task for Yangyuan to utilize a Bank loan for reconstruction after an 41 Annex D earthquake: the requirements and procedures of the Bank were all unknown to us in the beginning. With the support of government, we learnt and studied how to prepare, negotiate, design and implement the project, we advanced from knowing nothing to operating efficiently. For example, the Ministry of Finance organized training classes on project cost claims, and Finance Bureaus of province and city taught us with their own experience. The education was beneficial in ensuring a smooth fund flow the Bank. In summary, the experience of applying the loan from the Bank for earthquake reconstruction is unprecedented in our county in terms of utilizing foreign funds, and the overall result is rather good. We need to improve and internalize the experience, and evaluate the accomplishments further. May 16, 1996   IMAGING Report No: 15805 Type: PPAR