Document of The World Bank

Transcription

1 Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized Public Disclosure Authorized Human Development ector Unit East Asia and Pacific Region Document of The World Bank IMPLEMENTATION COMPLETION REPORT (CL TF-37040) ON A LOAN IN THE AMOUNT OF U$300 MILLION TO THAILAND FOR A OCIAL INVETMENT PROJECT February 15, 2005 Report No: 30237

2 CURRENCY EQUIVALENT (Exchange Rate Effective October 2004) Currency Unit = Thai Bhat (THB) 1 Thai Bhat = U$ U$ 1.00 = Thai Bhat FICAL YEAR October 1 eptember 30 ABBREVIATION AND ACRONYM ADB Asian Development Bank OECF Overseas Economic Cooperation AID Acquired Immune Deficiency yndrome Fund of Japan BMA Bangkok Metropolitan Administration OED Operations Evaluation Department CA Country Assistance trategy PAD Project Appraisal Document CDP-P Country Development Partnership for ocial PCU Project Coordination Uni DLPW Department of Labor Protection and Welfare PIU Project Implementation Uni DOLA Departmentof Local Administration PRID Program for Rural Industrial DPW Department of Public Works Development FIRR Financial Internal Rate of Return PC Policy teering Committee FPO Fiscal Policy Office PTFs Provincial Task Forces GB Government avings Bank QAG Quality Assurance Group HIV Human Immunodeficiency Virus QEA Quality at Entry Assessment JBIC Japanese Bank for International Cooperation RCs Regional Committees MOF Ministry of Finance RTG Royal Thai Government MOInd Ministry of Industry RUDF Regional Urban Development Fund MOInt Ministry of Interior ECAL ector Adjustment Loan MOLW Ministry of Labor and ocial Welfare IF ocial Investment Fund MOPH Ministry of Public Health IP ocial Investment Project NCB National Competitive Bidding OFO ocial Fund Office NEDB The National Economic and ocial RP mall-cale Irrigation Improvement Development Board and Rehabilitation Project NGO Non-government Organization TAT Tourism Authority of Thailand Vice President: Country Director: ector Director: Task Team Leader: Jemal-ud-din Kassum, EAPVP Ian C. Porter, EACTF Emmanuel Y. Jimenez, EAHD Minna J. Hahn, EACTF

3 THAILAND TH-OCIAL INVETMENT PROJ I CONTENT Page No. 1. Project Data 1 2. Principal Performance Ratings 1 3. Assessment of Development Objective and Design, and of Quality at Entry 2 4. Achievement of Objective and Outputs 4 5. Major Factors Affecting Implementation and Outcome ustainability Bank and Borrower Performance Lessons Learned Partner Comments Additional Information 17 Annex 1. Key Performance Indicators/Log Frame Matrix 18 Annex 2. Project Costs and Financing 20 Annex 3. Economic Costs and Benefits 23 Annex 4. Bank Inputs 24 Annex 5. Ratings for Achievement of Objectives/Outputs of Components 27 Annex 6. Ratings of Bank and Borrower Performance 28 Annex 7. List of upporting Documents 29

4 Project ID: P Team Leader: Minna J. Hahn Project Name: TH-OCIAL INVETMENT PROJ I TL Unit: EACTF ICR Type: Core ICR Report Date: February 25, Project Data Name: TH-OCIAL INVETMENT PROJ I L/C/TF Number: CL-43730; TF Country/Department: THAILAND Region: East Asia and Pacific Region ector/subsector: Other social services (47%); General industry and trade sector (28%); Compulsory health finance (10%); Vocational training (10%); General water, sanitation and flood protection sector (5%) Theme: Improving labor markets (P); Access to urban services and housing (P); ocial safety nets (P); Participation and civic engagement (P); Municipal governance and institution building () KEY DATE Original Revised/Actual PCD: 03/09/1998 Effective: 08/01/ /09/1998 Appraisal: 03/26/1998 MTR: 12/31/ /07/2000 Approval: 07/09/1998 Closing: 04/30/ /30/2004 Borrower/Implementing Agency: Other Partners: Kingdom of Thailand/Ministry of Finance Ministry of Public Health, Ministry of Labor and ocial Welfare, Ministry of Interior, Ministry of Industry, Bangkok Metropolitan Administration, Tourism Authority of Thailand, Ministry of Agriculture, Government avings Bank, OECF (now JBIC), UNDP, AusAID TAFF Current At Appraisal Vice President: Jemal-ud-din Kassum Jean-Michel everino Country Director: Ian C. Porter Jayasankar hivakumar ector Director: Emmanuel Y. Jimenez Maureen Law Team Leader at ICR: Minna J. Hahn Christopher Chamberlin ICR Primary Author: Minna J. Hahn 2. Principal Performance Ratings (H=Highly atisfactory, =atisfactory, U=Unsatisfactory, HL=Highly Likely, L=Likely, UN=Unlikely, HUN=Highly Unlikely, HU=Highly Unsatisfactory, H=High, U=ubstantial, M=Modest, N=Negligible) Outcome: ustainability: Institutional Development Impact: Bank Performance: Borrower Performance: L U QAG (if available) Quality at Entry: Project at Risk at Any Time: No ICR

5 3. Assessment of Development Objective and Design, and of Quality at Entry 3.1 Original Objective: Following the onset of the financial crisis that hit Thailand in mid-1997, the Royal Thai Government (RTG) requested World Bank support in developing a ocial Investment Project (IP) on a rapid timetable. As stated in the Project Appraisal Document (PAD), the objectives of the IP were to: i) respond to the financial crisis through the rapid creation of employment opportunities and the provision of essential social services to the unemployed and poor and ii) use the crisis as an opportunity to support bottom-up service delivery through financing locally-identified and managed development initiatives and through promoting decentralization, local capacity building, and community development. The IP s objectives responded to critical social needs emerging from the crisis. By February 1998, the RTG estimated that 8.8% of the workforce was unemployed and that the long-term unemployment rate was twice the pre-crisis level. The IP s objectives also supported the RTG s longer-term development priorities of decentralization and community development as laid out in the Eighth National Economic and ocial Development Plan ( ). In addition, the IP was clearly linked to the Bank s Country Assistance trategy (CA), which was presented to the Board with the project. The IP primarily supported the CA theme of strengthening governance through its emphasis on decentralization and community-based development, and the theme of sharing the benefits of growth and improving the quality of life through a lessening of social impacts of the crisis and targeting of benefits. The IP s dual objectives did increase the complexity of the project and posed a significant challenge for a government in the midst of handling a national crisis. However, the project responded to RTG demand for support on both short-term and long-term priorities. 3.2 Revised Objective: The original objectives remained unchanged. 3.3 Original Components: The project was comprised of three components, summarized in Table 1. The first component, known as Channel 1, provided short-term support to existing government programs aimed at providing jobs and priority basic social services to the unemployed and poor. This channel involved seven executing agencies and was expected to close after 28 months. The second component, referred to as Channel 2, provided grant and loan financing for locally generated projects through two new institutions: a ocial Investment Fund (IF) and Regional Urban Development Fund (RUDF). The Government avings Bank (GB) was the executing agency for this channel, which was expected to be operational over a 40-month period. The third component focused on project management and supported the operation, staffing, and activities of the IP s Project Coordination Unit (PCU) and Policy teering Committee (PC)

6 Table 1. Thailand ocial Investment Project Project Description ummary Component Objectives ubcomponents/executing Agencies Est. Cost (U$M) Channel 1 To mitigate the immediate Ministry of Public Health effects of the crisis through 1. Public Assistance cheme short-term support to 2. Community-Based AID Program existing government Ministry of Labor and ocial Welfare programs that provide jobs 1. Local Job Training for Unemployed Workers and basic social services to 2. Local Job Training for Disadvantaged Women and the unemployed and poor 3. Persons with Disabilities Local Job Training for Women and Youth Ministry of Interior 1. Construction of Village Roads 2. Construction of mall-cale Weirs and Foreground Dredging Ministry of Industry 1. Program for Rural Industrial Development Bangkok Metropolitan Administration 1. chool Renovation Program 2. Occupational and Vocational Training Program 3. mall-cale Infrastructure Program Tourism Authority of Thailand (OECF/JBIC-financed) 1. Tourist Facility Construction 2. Tourism Training Royal Irrigation Department of the Ministry of Agriculture and Cooperatives (OECF/JBIC-financed) 1. mall-cale Irrigation Improvement and Rehabilitation Project Channel 2 To promote fiscal and ocial Investment Fund (5 subproject menus) administrative 1. Capacity and Learning Development in decentralization and build local capacity and social 2. Community Occupation and Community Economy Community Welfare and afety capital for longer-term 3. Restoration, Protection, Management and development; to finance local projects with employment and service 4. Promotion of Culture, Environment, and Natural Resources Community Capacity Building and Networking benefits during the crisis 5. Immediate Community Welfare for the Needy through Community Organization Networks Regional Urban Development Fund Project Management To support the Project Coordination Unit and Policy teering Committee PCU/PC 2.80 Channels 1 and 2 were clearly linked to the IP s short-term and long-term objectives, although the inclusion of both increased the project s complexity. The project design was underpinned by lessons learned from similar programs in other crisis-affected countries, particularly in the Latin America and Caribbean region. Notably, the IP came after a decade of very minimal Bank engagement in Thailand, so the rapid scaling up of support during the crisis meant working with agencies that were unfamiliar with Bank processes. Technical assistance for procurement, financial management, and evaluation was planned to help strengthen the capacity of implementing agencies. 3.4 Revised Components: As part of the RTG s response to the slow pace of disbursements in the early stages of the project, the - 3 -

7 Channel 2 IF subcomponent was modified in This subcomponent, which originally financed community-based subprojects in four broad categories or menus, was expanded to include Immediate Community Welfare for the Needy (Menu 5) to support vulnerable groups who were not receiving assistance through other Channel 1 activities. The only required change to the IP legal agreement was an increase in the aggregate threshold for the community participation procurement category. The addition of Menu 5 demonstrated the flexibility with which the project responded to the country context and implementation experience. 3.5 Quality at Entry: Overall quality at entry is rated as satisfactory based on i) the responsiveness of the project to critical needs during the crisis, ii) direct linkages to the Government s immediate and longer-term development priorities, iii) consistency with CA objectives, and iv) the design features which addressed the risks inherent in a crisis-response operation. Although the design was complex, it likely befitted the evolving crisis. The inclusion of multiple subcomponents and the fact that most Channel 1 programs were already in existence mitigated some uncertainty as to how the crisis would evolve and which programs would be most effective. The project design also allowed for mid-term reallocations based on program performance. Adequate attention was given to safeguards issues, for example in terms of mitigating potential environmental impacts of subprojects and using social assessment findings to inform project design. Quality at entry was also rated satisfactory in the Quality at Entry Assessment (QEA) by the Quality Assurance Group (QAG). The QEA noted that while QAG/OED lessons on simplicity of design, rigorous economic analysis, and institutional assessment were overlooked to some extent, they were balanced by the innovative project approach, responsiveness to client needs, flexible design, and candid risk assessment. Quality at entry could have been strengthened by a greater emphasis on outcomes and their linkages to development objectives; explicit recognition of the tradeoffs involved in the project, such as rapid employment generation versus quality of infrastructure created; and a more realistic assessment of implementation capacity. hortcomings identified by the QEA included project complexity, unknown degree of Government ownership beyond the Ministry of Finance, and need for greater attention to financial management issues. The QEA also suggested that the project could have been separated into two operations to reduce complexity. It has been suggested that the IP should have been designed as an adjustment/investment hybrid operation, with an adjustment component delivering rapid short-term crisis relief (Channel 1) and an investment component supporting longer-term elements (Channel 2). However, the team had wanted very specific targeting of resources and community involvement which would have been difficult in an adjustment operation. The more fundamental issues to be considered are whether the types of programs being supported were best suited for a crisis situation, whether the procurement required for the specific activities could be handled quickly, and whether the implementing agencies had sufficient capacity to execute the activities on a rapid timetable. For instance, the selected instrument was appropriate for many of the programs but a sector investment operation or budget support could have helped with others. 4. Achievement of Objective and Outputs 4.1 Outcome/achievement of objective: The overall achievement of objectives and outputs is rated as satisfactory. Despite the challenging economic environment and the pressures of the crisis situation, the project enhanced employment opportunities and social services for the poor and unemployed as well as strengthened decentralized service delivery and community development

8 Regarding the first objective of responding to the crisis through the rapid creation of employment opportunities and the provision of essential social services, the IP appears to have delivered most of the intended results through Channel 1. While performance across subcomponents varied, the activities did create temporary employment and provided other social safety net benefits to the poor and vulnerable. However, implementation delays and capacity constraints affected the speed and extent to which assistance was delivered, which somewhat diminished the project s impact at the height of the crisis. Furthermore, the quality, sustainability, and longer-term benefits of training and infrastructure created under Channel 1 appear to have been limited, although these factors were not included explicitly as project objectives. The second objective of supporting bottom-up service delivery was also substantially achieved. Channel 2 helped build a foundation for locally driven development and generated employment and social service benefits through the financing of almost 7,900 subprojects and over 50 RUDF loans. Channel 2 activities appear to have had a significant institutional impact in promoting decentralization and strengthening the capacity of local governments, communities, and NGOs to participate in development. The IF, in particular, has been widely recognized for its contribution to strengthening social capital in Thailand. 4.2 Outputs by components: A. Channel 1 (estimated cost U$ million) 1. Ministry of Public Health Programs (estimated cost U$23.80 million) This subcomponent is rated satisfactory. Under this subcomponent, the Ministry of Public Health (MOPH) managed two activities: i) the Public Assistance cheme and ii) the community-based AID program. Both activities, which were relatively well-targeted to the poor and vulnerable, supported programs that came under pressure during the crisis due to high cost and/or budget cuts. The objective of the Public Assistance cheme was to sustain per capita subsidies to the low-income health card scheme. The IP provided financing for the scheme at 47 hospitals in six provinces, covering 1.15 million beneficiaries or 75% of the target. As the demand for low-income health cards was less than projected, the subsidy per card holder was not simply sustained but was increased from the 273 baht calculated originally to 400 baht. The PAD included only the target number of patient visits, but this indicator was not tracked. A secondary objective was to improve the management and efficiency of the Public Assistance cheme. The program introduced a new payment mechanism and a performance-based budgeting model which continues to be used under the new national 30-baht health care scheme. The community-based AID program aimed to continue MOPH support to NGO-led programs for prevention and care of HIV/AID. Although activities were delayed by almost one year due to unfamiliarity with the Bank s procurement procedures among PIU staff and NGO officials, the number of completed subprojects met or surpassed the number of planned projects in all regions. Actual beneficiaries totaled 447,886, greatly exceeding the target of 115,406. An independent evaluation found that benefits included: greater group participation of people living with HIV/AID and less discrimination against them; increased awareness of HIV/AID among local institutions; and strengthened capacity of local NGOs. NGO participation continues to be an important component of the RTG s overall HIV/AID strategy. 2. Ministry of Labor and ocial Welfare Programs (estimated cost U$27.43 million) This subcomponent is rated satisfactory. Under this subcomponent, the Ministry of Labor and ocial Welfare (MOLW) managed three programs: i) Local Job Training for Unemployed Workers ages 15 and - 5 -

9 older, under the Department of kills Development; ii) Local Job Training for Disadvantaged Women and Persons with Disabilities, under the Department of Public Welfare; and (iii) Local Job Training for Women and Youth, under the Department of Labor Protection and Welfare. This subcomponent was included because new labor market entrants plus the growing number of unemployed were creating a large pool of potential enrollees for existing training programs. This subcomponent achieved most of its outputs and objectives, despite the onerous administrative procedures involved, the instability/fluidity of the target population, and lack of MOLW experience in this type of training. Two departments had no prior experience delivering training, and the third had experience only in centralized training. The courses, which had a graduation rate of 91%, generated 467,300 person-months of training or 88% of PAD estimate. The courses also provided a temporary boost to trainee incomes, as 50-baht daily stipends were given as a small safety net provision to help finance subsistence of the unemployed. Beneficiary feedback suggests that the content and quality of some courses may not have been sufficient for generating post-training benefits, and the degree to which the acquired skills increased potential for employment is unknown. The PAD did include targets for person-months of employment generated, but these were not tracked. Only the training program for disadvantaged women and persons with disabilities included post-training employment assistance. 3. Ministry of Interior Programs (estimated cost U$52.51 million) This sub-component is rated satisfactory. Under this sub-component, the Ministry of Interior (MOInt) managed two activities: i) construction of village roads, under the Department of Public Works and ii) construction of small-scale weirs and foreground dredging, under the Department of Local Administration. This subcomponent aimed to create employment opportunities in rural areas, where migrants were expected to return as a result of job losses, and to improve basic infrastructure and services in rural areas. The programs were targeted to the unemployed and poor primarily through setting of wage rates. All 1,025 approved road subprojects well above the PAD target of 665 subprojects were implemented and resulted in the construction of 617 kilometers of new roads. A total of 2,365 weirs or 94% of the PAD estimate were constructed, and 87% of foreground dredging subprojects were completed. Employment generation benefits were less than expected due in part to the lower labor content of some activities and the fact that some contractors used their own labor rather than local labor as the project intended. All subprojects had difficulty employing local labor in certain areas where unemployment rates were low, wages were insufficient to attract laborers, and/or work was considered too menial. The PAD provided an indicator of person-months of employment generated, but it was not monitored. Beneficiary feedback suggests that the programs provided better road access and increased water supply and storage. However, some assessments found problems with construction standards and accessibility, and maintenance was considered difficult. Greater focus was given to short-term employment generation rather than the longer-term infrastructure benefits. 4. Ministry of Industry Program (estimated cost U$5.54 million) This subcomponent is rated unsatisfactory. Under this subcomponent, the Ministry of Industry (MOInd) managed one activity: the Program for Rural Industrial Development (PRID). The PRID aimed to stimulate industrial investment and employment in rural areas by financing NGO-provided services to facilitate industrial site location, match industrial sites with potential investors, train the local labor force, and facilitate arrangements between the investors and community leaders. Although the PRID was an existing program, it had been launched in 1997 just before the crisis. Its objectives were consistent with - 6 -

10 IP objectives only in the most general terms, and the program was not designed to provide immediate benefits. This subcomponent experienced implementation delays of about two years, and activities did not begin until late Causes of these delays included lower demand by investors to establish new facilities during crisis, difficulties in attracting private sector partners, and lack of NGO readiness to participate in the project. The delays led to the cancellation of 38 subprojects and reduction of the total commitment to U$3.9 million in February 1999 and later to U$1.32 million during the Mid-Term Review. In total, 57 subprojects were completed compared to the PAD target of 84. Beneficiaries expressed satisfaction with the increased income gained from the program, but additional impact evaluations are not available. 5. Bangkok Metropolitan Administration Programs (estimated cost U$11.73 million) This subcomponent is rated marginally satisfactory. The Bangkok Metropolitan Administration (BMA) managed three activities under this subcomponent: i) the chool Renovation program; ii) the Occupation and Vocational Training program; and iii) the mall-cale Infrastructure program. The objectives were to create employment and provide training for unemployed individuals in the Bangkok metropolitan area. The public works activities also had a secondary objective of improving infrastructure and facilities. This subcomponent proved to be important because the impacts of the crisis on the Bangkok metropolitan area were stronger than anticipated and less migration to rural areas occurred than was expected at the time of project design. The chool Renovation program was most successful in delivering the intended benefits and received an infusion of additional funds reallocated from other Channel 1 activities in July In total, 819 primary schools some of which were over 30 years old were rehabilitated. Approximately 5,300 laborers were hired for the program, although contractors sometimes used their own employees rather than workers from the unemployed labor pool. In terms of infrastructure quality, less than 1% of the work was judged unacceptable by BMA. The Occupational and Vocational Training program also received additional funds from intra-channel 1 transfers, allowing the number of training projects to be increased from 2,665 as originally planned to 5,007 and the number of trainees to be increased from 53,596 to 101,522, which represented about 2% of the area s labor force. The quality of training and its relevance to the labor market are unclear, particularly given the lack of follow-up data. The mall-cale Infrastructure program, which financed the renovation of district footpaths and sidewalks, was least successful in meeting its objectives. Only 15 out of 18 planned contracts were completed some only partially and the program created 7,165 jobs, considerably less than the 11,558 jobs expected. Many of the renovated sidewalks have shifted and/or buckled from erosion, making them unusable. 6. Tourism Authority of Thailand Programs (estimated cost U$75.02 million, OECF/JBIC-financed) This subcomponent is rated satisfactory based on inputs received during discussions with the Japan Bank for International Cooperation (JBIC), which was formed by the merger of the Export-Import Bank of Japan and the Overseas Economic Cooperation Fund of Japan (OECF), a IP co-financier. This sub-component, managed by the Tourism Authority of Thailand (TAT), supported local tourist facility construction and tourism training to improve the quality of services. The main objectives were to generate employment and improve basic infrastructure for tourism, as tourism was correctly projected to increase with the devaluation of the Thai baht during the crisis

11 A total of 165 subprojects were implemented, generating 97,131 person-months of construction employment or 85% of the target. The work was not as labor-intensive as expected, although facilities generated important ongoing employment (i.e. food vendors, ticket attendants) not captured by the data. Individual subprojects and overall subcomponent completion were delayed by months as the PIU s review of subproject designs took longer than expected, and often designs were not to standard and had to be redone. Most subprojects have continued to operate, generating revenues to sustain themselves. The subcomponent also delivered 434 training programs benefiting 15,782 people, but the quality of training and its impact are unclear. 7. Royal Irrigation Department of Ministry of Agriculture and Cooperatives (estimated cost U$48.32 million, OECF/JBIC-financed) This subcomponent is rated satisfactory based on discussions with JBIC as well as a consultants report prepared for JBIC. Under this subcomponent, the Royal Irrigation Department managed the mall-cale Irrigation Improvement and Rehabilitation Project (IRP). The objectives of IRP were to rehabilitate and improve existing small-scale irrigation projects and to extend distribution systems in existing projects, thereby creating jobs and raising incomes in rural areas. Of the 589 subprojects originally proposed to JBIC, 570 were included in the final workplan and completed. Employment creation totaled 2.27 million person-days or 91% of the revised target. The difference has been attributed to the need to use machines rather than manual labor for certain tasks. In terms of indirect benefits, the IRP helped raise crop productivity and net household incomes from the cultivation of land and livestock, and it increased public involvement in effective water management. B. Channel 2 (estimated cost U$ million) 1. ocial Investment Fund (estimated cost U$ million) This subcomponent is rated highly satisfactory. The objectives of the IF were to: i) improve the access of poor people to basic social and economic infrastructure, services, and employment opportunities and ii) to advance the decentralization process by strengthening decision-making and implementation capacities at the local level. In financing subprojects proposed and managed by community groups or other local organizations, the IF represented the first time the RTG channeled significant amounts of money to civil society. ubprojects included multipurpose community centers, cultural preservation activities, production of community master plans, and savings group capacity development. The IF supported 7,874 subprojects and reached about 10% of villages throughout the country. Although it did not achieve the PAD target of 11,590 subprojects, it appears that the target was overly ambitious, requiring the approval of roughly 66 subprojects per week and assuming a level of standardization inconsistent with IF objectives. IF subprojects appear to have created employment for 59,284 individuals (based on data from 7,052 subprojects) and benefited over 13 million individuals, including the poor, unemployed, children, senior citizens, and disabled. The IF met its regional targeting objectives, with the poorest region of the country, the Northeast, receiving the largest proportion of resources at 47%. The PAD indicators for person-months of employment created and for rural employment were not monitored. Beyond the more easily measured benefits of the subprojects, it is widely recognized that the IF s processes and investments helped strengthen community capacity and further governance objectives. The - 8 -

12 IF also instilled in the ocial Fund Office (OFO) and communities a level of discipline in operations by requiring detailed planning, transparent financial management and procurement systems, a management information system and unit cost database, and results-based management. The IF s social capital and economic impacts have been documented in a Participatory Impact Evaluation, and a more in-depth evaluation of the social capital impacts is underway and is expected to be completed by June Two key changes to the IF enhanced its impact. As described above, Menu 5 was added after nine months of IF operation to provide immediate assistance to the most vulnerable groups. The approval of Menu 5, which accounted for almost half of subproject disbursements, was an exemplary modification to project design in response to implementation experience and circumstances. The second change made around the same time was the decentralization of IF s operations. The ocial Fund Office (OFO) decentralized its activities to 76 Provincial Task Forces and 11 Regional Committees which were staffed by thousands of volunteers. In addition, the OFO established 11 Regional Offices throughout the country. This decentralization helped increase the IF s efficiency, enhance its responsiveness to community needs and ability to reach targeted beneficiaries, and improve subproject quality. The IF closed in late 2002 due in part to a change in government and the RTG s decision not to seek additional Bank financing for social protection programs. Local development is now being supported through a village fund program, introduced in Regional Urban Development Fund (estimated cost U$41.57 million) This subcomponent is rated satisfactory. The RUDF was established to on-lend money to local urban governments for subprojects of public benefit, such as markets, libraries, and child care centers. Its objectives were to: i) improve basic infrastructure, facilities, and services; ii) create employment in strategically important urban centers outside of Bangkok; iii) enhance capabilities at the national level to provide technical guidance to local governments; and iv) improve the skills of key local officials in administration, investment planning, project implementation, and financing. Like the IF, the RUDF aimed to strengthen the decentralization process. Despite being extended by almost two years, RUDF disbursed only U$13.7 million, half the originally planned amount of U$30 million and 91% of the revised target of U$15 million which was set in February low disbursement was caused by: a one-year delay in the handover of responsibilities to the GB due to slowness in establishing appropriate mechanisms to assess subprojects and disburse loans; a decision-making structure involving two separate committees, which slowed lending and disbursements; and an initial lack of local government capacity to prepare pre-feasibility studies and meet other requirements, which was then addressed by RUDF-provided technical assistance. Furthermore, demand for RUDF lending was lower than anticipated, due in part to the abundance of free grant money available to local governments from sources such as the Miyazawa fund. ome local governments used RUDF technical assistance to prepare proposals then switched to another source of financing. Despite these issues, the effectiveness of RUDF subprojects has been high. In total, 53 subprojects compared to the PAD output target of 20 have been completed or will be completed shortly under the RUDF, which has continued beyond the end of the IP. The PAD provided a target for person-months of employment created, but this indicator was not tracked for most subprojects. The average lag between subproject appraisal and approval was reduced from 16 months in 2000 to 9 months by 2002, although this is still higher than the PAD target of 60 days. The RUDF s most important contribution has been its institutional impact in laying the groundwork for sub-sovereign lending in Thailand

13 C. Project Management (estimated cost U$2.8 million) This component is rated satisfactory. As described above, this component supported the operation, staffing, and activities of the PCU and PC. Their responsibilities included: i) overseeing implementation of the IP; ii) setting its overall policies and evaluating and reporting on its performance; iii) providing technical assistance and training to all PIUs in accounting, procurement, and monitoring; iv) commissioning and financing a series of technical field reviews, beneficiary assessments, and social assessments; v) studying and analyzing the social impacts of the crisis, including developing a targeting map; and vi) developing a strategic framework for strengthening the RTG s social safety net programs. As the PAD did not provide indicators for this component, the assessment of this component is based on the performance of the PCU and PC in carrying out the responsibilities outlined above. Almost all activities were implemented, despite the newness of the PCU and PC, the number of activities and agencies involved, and the challenges encountered in Bank procedures. The PCU coordinated closely with the PIUs, although it gave relatively more support to Channel 1 agencies despite the fact that the Channel 2 institutions were newly created and arguably in greater need of assistance. The Bank helped fill this gap to some extent through its continuous field-based supervision and prompt assistance to the Channel 2 institutions. In terms of governance-related achievements, the PCU helped install competitive bidding procedures that increased transparency and helped keep unit prices low. Governance objectives were also advanced by the IP s innovative web-based management information system. Although data entry by PIUs was delayed and information on Channel 2 activities was not included, this database has been viewed as revolutionary for promoting transparency. The PCU and PC were less successful in the remaining activities. Project reporting focused heavily on disbursements, inputs, and outputs, and the evaluation of project performance was not as comprehensive and rigorous as envisioned during design. The social impact assessment was terminated and the targeting map prepared after more than half of the project resources had already been spent. In addition, only modest progress was made in developing the social protection framework, a task that has been incorporated into the Country Development Partnership for ocial Protection (CDP-P) which is currently underway. The CDP-P is described in greater detail in the Thailand Country Assistance trategy for Net Present Value/Economic rate of return: The economic rate of return was not calculated at the time of appraisal due to the emergency nature of the project, nor was it calculated at project close due to data constraints. No rate of return analysis was required for subcomponent activities related to infrastructure. For all Channel 1 activities for which unit cost targets were calculated, actual costs were lower than projected. For the IF, a unit cost database was developed, and cost-effectiveness ratios guided subproject approval. For the RUDF, no systematic analysis of benefits has been conducted, although it appears that most subprojects have met project feasibility estimates of economic rates of return. 4.4 Financial rate of return: Financial internal rate of return (FIRR) analysis was carried out for RUDF subprojects only. FIRRs varied significantly among the subprojects. Among the 36 subprojects for which FIRRs have been calculated, three had negative expected FIRRs, and the others ranged from 5.17% to 33.86%. No assessment has been conducted of how forecasted FIRRs are being met. Regarding RUDF subproject repayment, the PAD target of 100% repayment has been achieved to date. Because of the three-year gap before repayment of subproject loans starts, the full repayment stream will not be realized by the RUDF until fiscal year

14 4.5 Institutional development impact: The IP s institutional development impact was substantial. The IP introduced several new project and financial management innovations, such as a new procurement framework for national competitive bidding. However, institutional development impact appears to have varied across Channel 1 agencies. In the case of BMA, IP procedures regarding procurement and internet monitoring of inputs/outputs were dropped after IP support ended. In contrast, the MOPH adopted the performance-based budgeting model introduced by the IP for the current 30-baht health scheme. The MOLW noted that the IP provided valuable experience in the delivery of training, selection of target groups, and budget utilization. Channel 2 activities also had far-reaching institutional development impacts. The IF enhanced the capacities of over 300 community network organizations, helped communities develop skills in project planning and management, and mobilized thousands of volunteers to help those affected by the crisis. ocial capital, inspired by the new Constitution promulgated in 1997 and operationalized by the IF, is now one of the main pillars of the national development agenda. The IF s lessons have been incorporated into policy initiatives, particularly in the Ninth Economic and ocial Development Plan, and the IF experience has been informing the philosophies and operations of non-government and government organizations. The RUDF s institutional impact has also been substantial, particularly in pioneering the establishment of sub-sovereign finance mechanisms. Domestic banks are establishing sub-national lending programs, and the Fiscal Policy Office is designing a Local Development Financial Corporation. The RUDF has also helped local governments become more adept at preparing effective subprojects. Furthermore, it has helped encourage learning among urban centers, funding demonstration subprojects for which local officials from implementing municipalities are being sought for advice by other interested municipalities. Finally, the RUDF has furthered governance objectives, for example in its requirement that civic forums be held to inform the design of subprojects which has now become normal practice. 5. Major Factors Affecting Implementation and Outcome 5.1 Factors outside the control of government or implementing agency: The evolution of the crisis had both positive and negative effects on demand for IP programs. While demand for Channel 1 programs such as low-income health cards, MOLW training, and public works-type activities was lower than expected, the IF faced an overwhelmingly large demand for its resources when it launched. In its first six months, the IF received over 5,000 applications totaling U$460 million or four times the total available financing, which clogged the pipeline and temporarily limited the OFO s efficiency. Macroeconomic conditions after the crisis affected the RUDF, which faced an initial lack of interest because commercial rates in Thailand fell more rapidly than RUDF rates could be adjusted. Local conditions also affected implementation. ome infrastructure subprojects were delayed by heavy rains and flooding, and other activities faced difficulties mobilizing subcontractors, consultants, and NGOs in certain areas. For the RUDF, the availability of new grant funding and the local elections in 2000 which resulted in new mayors and councils in many local governments also affected implementation. In the initial stages, problems with insufficient pecial Account authorization and delays in reimbursement from the Bank contributed to slow implementation. The Bank requirement that contracts and other documentation be prepared in English placed additional burdens on implementing agencies. Other aspects of Bank performance are discussed below

15 5.2 Factors generally subject to government control: The IP experienced a critical four-month delay between approval and effectiveness, during which time preparation activities by most PIUs were effectively suspended. The lag was caused by a delay in the RTG s legal opinion for the signed loan agreement. low Government approval or agreement on other issues, such as the formal structure and processes of RUDF or labor content issues in the public works-type activities, also resulted in some implementation delays. The 5% cap on OFO administrative expenses negotiated by the Government restricted the numbers of staff employed as well as technical assistance, but this constraint was partly mitigated by the mobilization of thousands of volunteers. 5.3 Factors generally subject to implementing agency control: Implementing agency performance was commendable given the magnitude of additional responsibilities they were charged with handling and the project s tight timeframe. The decentralization of IF s operations, described earlier, provides a positive example of how PIUs adjusted to constraints during the course of the project. A critical factor causing initial implementation delays was lack of familiarity with Bank procedures among the PIUs, but technical assistance from the Bank, cofinanciers, and PCU helped mitigate these problems to some extent. Frequent staff turnover and insufficient allocations for staffing, administrative expenses, and supervision of works affected PIU management of some subcomponents. In a couple cases, inability to provide counterpart funding also affected implementation, but not significantly. 5.4 Costs and financing: The total amount disbursed from the Bank loan was U$253.5 million, about 85% of the appraisal estimate. The lower cost was due mainly to the slower launch and implementation of subcomponents than anticipated, which resulted in a reduction of project activities and cancellation of U$46.5 million as the economy recovered. 6. ustainability 6.1 Rationale for sustainability rating: Overall project sustainability is rated as likely. ustainability can be considered in two ways: sustainability of activities and sustainability of benefits created. Most Channel 1 programs were already ongoing, and those programs (or similar ones) have continued after the close of the IP. The longer-term benefits of Channel 1 training and infrastructure activities are more difficult to gauge, since their impacts have not been not closely monitored. ustainability of the Channel 2 subcomponents also appears likely. While the IF ceased operations in 2002, its impact remains evident at many levels. The community assets created are reportedly being used, maintained, and extended to new activities. Many communities benefiting from Menu 1 allocated a portion of their profits to the establishment of community welfare funds, and community network organizations continue to operate revolving funds with monies disbursed originally through Menu 5. The community planning process was continued by the NEDB, and as noted above, social capital has become a pillar of the national development agenda. The RUDF as an institution will continue to operate, and its impact on sub-sovereign lending in Thailand is expected to be sustained. 6.2 Transition arrangement to regular operations: For Channel 1, programs were already existing and integrated into the regular operation of the executing agencies, and the programs have continued with RTG funding. For Channel 2, the IF was closed prior to the end of the IP. The GB is continuing the RUDF s operation with a capital infusion of 1 billion baht. The RUDF can use current and future repayment flows to meet operational costs and re-lend any remaining

16 funds, and it expects to be able to solicit new loan applications by the end of For at least the next three years, its loans will be syndicated loans funded partly from IP funds (perhaps 10%) and partly from the new infusion of GB money (perhaps 90%). 7. Bank and Borrower Performance Bank 7.1 Lending: Bank performance during project preparation and appraisal was satisfactory. The Bank was very responsive to the RTG s concern with mitigating the effects of the crisis quickly, preparing the IP in only four months. Although the short preparation time did not allow for thorough analysis or economic and sector work, the Bank did use examples from other crisis-affected countries to inform project design and prepared a strong risk analysis with specific risk minimization measures. In addition, the Bank took into account the policy conditionalities under an existing sectoral adjustment loan (ECAL) supported by the Asian Development Bank, which the IP complemented with additional financing for selected programs. In terms of support to the borrower, Bank pre-appraisal and appraisal missions included sector specialists to work closely with implementing agencies. Bank performance could have been improved by further appraisal of line agency capacity, potential difficulties with implementing Bank procedures, and the Channel 1 programs themselves to determine how quickly and effectively assistance could be delivered. This assessment may have required more time, but the initial lag in implementation was a critical gap at the height of the crisis. In addition, stronger indicators and mechanisms for evaluating impact would have been useful. 7.2 upervision: Bank supervision was continuous and intensive, particularly in the early stages of the project, and most implementation problems were identified early and addressed proactively. Beyond supervision missions that included staff with social fund experience, the Bank s field-based staff provided day-to-day procurement and financial management support to the PCU and PIUs. The Bank demonstrated prudent flexibility in allowing implementing agencies to address disbursement problems or other constraints with pragmatic alternatives and in working collaboratively to introduce Menu 5 under the IF. upervision reporting could have placed greater emphasis on outcomes and impact, although this issue should have been addressed from project preparation as noted above. Differences between Project tatus Report (PR) subcomponent ratings and ICR ratings may be due partly to the focus on outputs, processes, and disbursements during supervision. Only some PAD indicators were monitored during supervision, and not always on a regular basis. Furthermore, context issues such as the evolution of the social impacts of the crisis appear to have been largely overlooked as Bank staff members, like the PCU, were focused on implementation details. Washington-based managers seemed to be less involved than Bangkok-based staff, as illustrated by the fact that no manager comments were provided in PRs until August Overall Bank performance: The Bank s overall performance was satisfactory. Bank staff worked closely with local counterparts and responded rapidly and flexibly to issues in both project design and supervision. The project s design and implementation schedule could have been better informed by a more explicit assessment of the implications of using an investment instrument to deliver support during the crisis and by a more rigorous evaluation of implementation capacity, although this may not have been feasible given the tight timeframe. In both design and supervision, increased attention to evaluating the outcomes and development impact of the project was needed

17 Borrower 7.4 Preparation: The MOF was clearly engaged in the preparation process and demonstrated ownership through important actions undertaken prior to loan approval, such as the appointment of the PC and PCU. Channel 1 agencies participated in preparation and appraisal missions as well as prepared their own subcomponents, with guidance from the Bank on standards and targets. Channel 2 was owned to a high degree by a civil society working group which shaped the subcomponent activities, especially in implementation. 7.5 Government implementation performance: As noted above, some delays in the RTG s decision-making process slowed implementation, particularly the four-month lag between approval and effectiveness during which most PIU activities were suspended. To a lesser extent, reorganization and turnover of staff in the MOF and PCU affected implementation in the early stages of the project. Timely audit reports were a challenge for the IP and for the Thailand portfolio overall. In general, provision of adequate counterpart funds was timely, and in some cases, the RTG even provided up-front funding for activities when Bank funds were delayed. 7.6 Implementing Agency: In an environment of intense public pressure and scrutiny, implementing agencies effectively managed their subcomponents and made use of technical assistance. While staffing constraints and personnel changes did slow implementation to some degree, it does not seem to have affected project outcomes. Overall, Channel 2 agencies were stronger in monitoring and evaluation. The OFO continually revised its procedures to improve implementation and impact and prepared comprehensive performance reports which were disseminated across the country. In addition, the OFO thoroughly investigated, took appropriate action on, and publicly reported the small number of cases of misuse of funds by project contractors. Likewise, the RUDF has maintained a database on disbursements, subproject status, and other indicators, and it intends to undertake an impact evaluation based on an extensive field survey of all subprojects. 7.7 Overall Borrower performance: The Borrower s overall performance was satisfactory. The RTG demonstrated a high degree of ownership for the IP, and the implementing agencies were effective in managing their activities despite the pressures of the crisis and the agencies limited prior knowledge of Bank procedures. 8. Lessons Learned Key lessons learned from the project include: In crisis situations, project design should take into account the speed with which assistance can be delivered. Complex structures and processes that slow activities and disbursements should be avoided, and agencies should have a clear understanding of the requirements involved with Bank operations. During project preparation, an up-front assessment of implementing agency capacity is needed. The assessment should also evaluate the capacity and likely participation of partners such as local governments and NGOs. uch an assessment, although time-consuming, would help inform project design and the development of a realistic timeline, as well as identify necessary technical assistance to ensure smooth project implementation. For a crisis or emergency operation, in particular, the Bank must provide substantial on-the-job training and assistance at the beginning of the operation, which would likely require budget allocations higher than the norm. Risk management and flexibility are important, particularly for crisis operations in which support needs to be scaled up quickly. The structure of a crisis-response project should allow sufficient flexibility to accommodate changing circumstances and provide incentives which help mitigate risk. If the crisis abates faster than expected, the project should have reallocation options to direct funds to