ASIAN DEVELOPMENT BANK Operations Evaluation Department

ASIAN DEVELOPMENT BANK Operations Evaluation Department SPECIAL EVALUATION STUDY ON URBAN SECTOR STRATEGY AND OPERATIONS June 2006 In this electronic file, the report is followed by the Management response and Development Effectiveness Committee Chairperson s Summary of the Committee s Discussion on 26 July 2006.

Evaluation Study Reference Number: SST: REG 2006-03 Special Evaluation Study June 2006 Urban Sector Strategy and Operations Operations Evaluation Department

ABBREVIATIONS ADB Asian Development Bank ADBI Asian Development Bank Institute ADTA advisory technical assistance AUWSP Accelerated Urban Water Supply Program BOT build-operate-transfer CBO community-based organization CSP country strategy and program DFID Department for International Development DMC developing member country DPR detailed project report EA executing agency FYP Five-Year Plan GDP gross domestic product GFI government financial institution HDFC Housing Development Finance Corporation HUDCO Housing and Urban Development Corporation IA implementing agency ICICI Industrial Credit and Investment Corporation of India IUDP integrated urban development project JBIC Japan Bank for International Cooperation JFPR Japan Fund for Poverty Reduction KMC Kolkata Municipal Corporation KUIDP Karnataka Urban Infrastructure Development Project LGU local government unit MDG Millennium Development Goal MDP municipal development project MEPA Ministry of Employment and Poverty Alleviation MIC middle-income country MUD Ministry of Urban Development MR metropolitan region MWSP Municipal Water Supply Project MWSS Metropolitan Waterworks and Sewerage System NGO nongovernment organization NHB National Housing Bank NURM National Urban Renewal Mission OCR ordinary capital resources OED Operations Evaluation Department OEM Operations Evaluation Mission PCR project completion report PIU project implementation unit PM project manager PMU project management unit PPAR project performance audit report PPTA project preparatory technical assistance PRC People s Republic of China RM resident mission RP resettlement plan RRP report and recommendation of the President RSDD Regional and Sustainable Development Department

iii RUIDP Rajasthan Urban Infrastructure Development Project SES special evaluation study SPD Strategy and Policy Department TA technical assistance TCR technical assistance completion report UN United Nations UN-HABITAT United Nations Human Settlements Programme USAID United States Agency for International Development USS Urban Sector Strategy WSS water supply and sanitation WSSWM water supply, sanitation, and waste management NOTE In this report, $ refers to US dollars. Director General Director Team leader Team members B. Murray, Operations Evaluation Department (OED) R. Adhikari, Operations Evaluation Division 2, OED W. Kolkma, Senior Evaluation Specialist, Operations Evaluation Division 1, OED A. Anabo, Senior Evaluation Officer, Operations Evaluation Division 1, OED A. Alba, Operations Evaluation Assistant, Operations Evaluation Division 1, OED Operations Evaluation Department, SS-68

CONTENTS Page EXECUTIVE SUMMARY iii MAP ix I. INTRODUCTION 1 A. Background 1 B. The 1999 Urban Sector Strategy 1 C. Main Evaluation Questions 2 D. Organization of the Study 3 II. URBANIZATION IN ASIA 6 A. Trends 6 B. Needs 8 C. Relation of Sector with the Millennium Development Goals 10 III. ADB S FOCUS ON THE URBAN SECTOR 12 A. Genesis of the Urban Sector Strategy 12 B. Organizational and Other Changes in ADB 14 IV. ADB ASSISTANCE TO THE URBAN SECTOR 17 A. Lending Program in Numbers and Volume 18 B. Project Developments after the Urban Sector Strategy of 1999 22 C. Capacity Development in Urban Sector Projects 31 D. Technical Assistance Program 33 E. ADB s Knowledge Management in the Urban Sector 35 V. ADB ASSESSMENTS OF COMPLETED URBAN SECTOR PROJECTS 36 A. Assessment of Urban Sector Projects by Operations Departments 36 B. Assessment of Completed Urban Sector Projects by OED 39 C. Assessment of Completed TA Grants by Operations Departments and OED 42 D. Assessments of Benefits and Problems of Ongoing Projects 43 by Project Managers The guidelines formally adopted by the Operations Evaluation Department (OED) on avoiding conflict of interest in its independent evaluations were observed in the preparation of this report. The current Director General was involved in managing ADB s program in the People s Republic of China (PRC) from 1993 to 2004, and was the mission leader for two urban transport projects in Shanghai. He thus recused himself from the design and review of the PRC case study and from changing anything related to PRC operations in this report. James Arthur, Jose Antonio League, Narayanan Edadan, Edward Leman, and Zhang Rufei were the consultants. Imelda Baleta, Jennifer Simon, and Mary Grace Alindogan did background research and administered the questionnaire survey. Some of the consultants were earlier involved in ADB s urban sector technical assistance, but did not have responsibilities as loan project managers and did not review work in which they had been involved before. The team leader has not been involved in an operational capacity in ADB s urban sector before. To the knowledge of the management of OED, there were no conflicts of interest of the persons preparing, reviewing, or approving this report.

VI. ADB S ADDED VALUE: VIEWS FROM PEOPLE S REPUBLIC OF CHINA, 46 INDIA AND PHILIPPINES A. People s Republic of China 46 B. India 48 C. Philippines 50 D. Conclusions from the Country Case Studies 52 E. Potential Effect of ADB Policy Initiatives 53 VII. EVALUATION AND RECOMMENDATIONS 54 A. Key Questions 55 B. Rating of the Urban Sector Strategy and Urban Sector Operations 57 C. Suggestions for a New Urban Sector Strategy 59 D. Recommendations for ADB 60 APPENDIXES 1. Methodology of the Questionnaire Survey of Ongoing Urban Sector Projects 62 2. Urban Sector Projects Approved from July 1993 to June 2005 66 3. List of All Urban Sector Related Technical Assistance Grants 70 4. Urban Sector Needs 75 5. Asian Development Bank s Lending to the Urban Sector 80 6. Findings from All 51 Project Completion Reports for Urban Sector Projects, 83 Issued Between Mid-1998 and Mid-2005 7. Country Case Studies: People s Republic of China, India and Philippines 85 8. Comparison of Questionnaire Survey Findings in People s Republic of China, India 107 and Philippines 9. Innovation and Efficiency Initiative and Other Policy and Business Process Initiatives, with Comments on Relevance for Urban Sector 109 10. Lessons from Urban Sector Experience 113 11. Lessons Drawn from 51 Project Completion Reports Issued Between 1998 and 121 2005 in the Urban Sector SUPPLEMENTARY APPENDIXES (available on request) A. Urban Sector Trends B. Current Investment: Programs of Some Main External Agencies C. Names of Asian Country Urban Ministries/Departments D. Statistical Tables Derived from Study of 88 Reports and Recommendations of the President in the Urban Sector, by Period and by Subsector E. Asian Development Bank Publications in the Urban Sector, 1993 2005, Annotated F. Asian Development Bank Ratings of Completed Urban Sector Projects G. An Assessment of Urban Sector Strategies H. Executing Agency Questionnaire and Questionnaire Survey Response, by Subsector and by Country in the Urban Sector I. Performance Assessments of Two Centrally Sponsored Schemes in India J. Value Added and Lost as a Result of Asian Development Bank Involvement in Urban Sector Projects in India Attachments: 1. Management Response on the Special Evaluation Study on Urban Sector Strategy and Operations 2. Development Effectiveness Committee Chairperson s Summary of the Committee s Discussion on 26 July 2006

EXECUTIVE SUMMARY The urban sector strategy (USS) was one of several Asian Development Bank (ADB) sector strategies formulated around the end of the 1990s, and was intended to bring more coherence to ADB s work in the urban sector. At the time, ADB had two urban development, water supply, and housing divisions: one for East Asia and one for West Asia. The USS took a wide view of the scope of the urban sector and included urban water supply and sanitation, housing, urban drainage and flood control, urban transport, urban environment, slum improvement, sites and services, (urban) institutional strengthening and capacity building, urban land management, urban governance, and urban finance. Poverty reduction components and microcredit were also endorsed as part of the urban sector, when included in urban development projects. The USS summarized urban challenges and presented objectives, guidelines, and good practices. In response to rapid urbanization in Asia, the USS called for a systematic and incremental expansion of ADB s involvement in the urban sector in terms of the volume of lending, the range of subsectors to be addressed, and the interrelationships of project components. The USS also called for greater attention to poverty reduction; more community participation in projects; more institutional strengthening and capacity development; more private sector involvement; more policy lending; and more lending for urban transport, housing, and land management projects. Consistent with ADB s approach to sector strategy development at the time, the USS did not assess the resources needed to implement the strategy. The USS highlighted the following operational objectives: (i) maximizing the economic efficiency of urban areas, (ii) reducing urban poverty, (iii) improving quality of life, and (iv) achieving more sustainable forms of urban development. However, these objectives were neither quantified nor set out in a manner that could be monitored to assess progress. The USS was approved by Management in July 1999 and sent for information to ADB s Board of Directors. The Board did not discuss or approve it nor was there significant formal consultation with developing member countries (DMCs) or other stakeholders in preparing it. ADB s Regional and Sustainable Development Department (RSDD) planned to review and revise the USS by the end of 2005. This was viewed necessary as some important developments were not reflected, such as the adoption of poverty reduction as ADB s overarching objective, the emergence of the Millennium Development Goals, and ADB s joining of the Cities Alliance in 2002. The original USS was prepared before new guidelines on ADB policies and strategies had been formulated, so it was viewed as lacking the appropriate disciplined approach and procedural rigor. To provide the necessary analysis to revise the USS, it was agreed that the Operations Evaluation Department (OED) would look at the appropriateness and implementation of the existing strategy, while RSDD would identify and analyze emerging urban sector trends, directions, good practices, and related issues. Revision of the USS by RSDD is now expected to be completed in the fourth quarter of 2007. Given that the USS has been in effect for only 6 years and most projects approved after it was adopted are still in various phases of implementation, this evaluation pays limited attention to the on the ground impact of urban projects. Rather, the study analyzes ratings of completed projects and technical assistance (TA) grants, analyzes whether the USS s targets are being addressed, and assesses ongoing urban operations. Additional questions addressed in the evaluation are (i) why is ADB s urban sector program not much larger than it is today, given the rapid urbanization in the Asia Pacific region; (ii) what is the value added of ADB involvement in urban sector projects against the transaction costs, as perceived by ADB clients; and (iii) is a USS needed and under what conditions would it be most effective? The

iv assessment of the changes in urban sector operations was based on a comparison of projects approved in the 6 years before (mid-1993 to mid-1999) and after (mid-1999 to mid-2005) the USS was adopted. The special evaluation study (SES) analyzed ADB databases, documents of all 88 projects and 191 TA grants approved since mid-1993, and all available country urban sector strategies. The SES also conducted a questionnaire survey of all 52 ongoing urban sector projects (with an 85% response rate) and undertook case studies for People s Republic of China (PRC), India and Philippines. Urbanization is a mega-trend in Asia. ADB s Long-Term Strategic Framework (2001 2015) marked it as one of its main challenges. The region is in the midst of a transformation from a rural, agriculture dominated region to an urban, industrial/service sector dominated region. Asia has the fastest growing urban population in the world. Asia s share of global urban population has risen from 9% in 1920 to over 48% in 2005. It is expected to exceed 50% by 2020. Asia s urban population may grow from 1.35 billion (37% of the total population) in 2000 to 1.94 billion (48%) in 2015. By 2010, 12 of the world s 23 megacities are expected to be in Asian developing countries. Asia s urbanization has generally had positive effects on the countries economies but is also associated with serious negative externalities. There is strong evidence to suggest that cities increase productivity, generate economic growth, and jobs. However, productivity rates could be even higher if the efficiency of cities is enhanced. Unmanaged urbanization can lead to poor quality of life for urban dwellers even if income poverty declines, and to significant urban environment problems. With limited fiscal resources and poor planning, urbanization can lead to large urban subpopulations living in squalid living conditions and abject poverty. Analysis of current trends suggests that urban poverty and urban slum incidence in the poorer DMCs is increasing. Financing requirements for just water supply, sanitation, and waste management (WSSWM) and slum upgrading in urban areas may be $25 billion per annum for ADB s DMCs. This would double to $50 billion per year if urban roads are included. Using a conservative assumption that 100 kilometers of rail mass transit system are constructed in Asia s large cities only, a further $10 billion per annum would have to be added to urban funding requirements. The requirements in housing exceed all of these amounts by far. Current investment programs of governments, external agencies, and the private sector fund no more than half of the demand for water supply, sanitation, and urban transport. There is a pressing need for further investment and greater capacity development of urban bodies to enable them to effectively extend, operate, and manage these services. Problems related to urbanization are complex and cannot always be addressed effectively from the perspective of a single sector. The platform offered by urban ministries and urban governments for holistically addressing interrelated problems in different sectors offers opportunities for effective interventions. Given the existence of such large numbers of potential clients, there is ample scope for pursuing an urban strategy, provided that ADB can seize the opportunity. If ADB were to maintain its position as Asia s premier development institution, going forward would require a greater focus on urban development, which would also address many of the poverty reduction and environment related issues. A major finding of this SES is that, although urbanization is increasingly driving economic development in Asia, ADB s institutional focus on the urban sector has been lost. One and a half years after the USS was adopted, ADB underwent reorganization and introduced new business processes. ADB s professional expertise in the urban sector was dissipated. The two urban sector divisions were dissolved and staff were scattered over five new departments and four newly created regional social sector divisions, which included sector

v specialists in education, health, and other social themes. From an operational perspective, the link between urban development, health, and education is tenuous. Housing finance and all management and governance operations were shifted to four new governance and finance divisions in the regional departments. The urban transport operations were to be identified by the transport divisions, which historically focused on intercity and rural roads. The earlier division and sector strategies were overshadowed by an increasing focus within ADB on country strategies. ADB s water supply and sanitation operations, including those in urban areas, fell increasingly under the framework of ADB s Water Policy, approved January 2001 by the Board and managed by RSDD. As noted by a recent independent assessment of ADB s reorganization, ADB s sector expertise has declined. ADB s sector classification of projects and TA grants was streamlined in 2004. The new system did not include the urban development and housing sector as a distinct group, leading to fewer options for monitoring the urban sector portfolio. Urban development projects are now classified as multisector projects, a disparate group that includes rural development projects, environment projects, and a variety of other types of projects. Notwithstanding the loss of institutional focus, the urban sector portfolio grew modestly after the issuance of the USS. Before the USS was adopted, urban sector lending constituted 9.0% of the total loans approved between 1993 and 1999 (11.0% if some large emergency loans during the Asian Financial Crisis are excluded). After the USS was adopted, it accounted for 11.5% of total lending from 1999 to 2005. An analysis of the pipeline suggests that this share should increase in the future. However, in dollar terms, lending grew by only 5.9% (from an average of $551 million per year in the 6 years before the USS to $583 million per year after) much less than the 15% growth in urban population in ADB s DMCs over the same period. ADB s professional staff with an urban sector designation increased from 10 at the time the USS was approved to 16 at present. Urban sector projects often involve many components and many cities. These types of projects require intense supervision, which is facilitated by the increasing number and strength of resident missions (RMs). It could be argued that the urban portfolio has persisted in spite of the institutional dilution, and that it is due to continuing demand from DMCs in the face of urbanization. With more institutional focus on the urban sector, ADB may be better placed to respond to demands and opportunities and the urban portfolio could be larger than its current size. A stronger institutional focus on the urban sector might have expanded or strengthened ADB s knowledge products, TA, policy dialogue, project formulation, and portfolio management. By comparison, the World Bank, with a similar sized lending program in the Asia and Pacific region, has (i) 48 urban sector experts for this region, (ii) various urban theme groups in the infrastructure department, (iii) urban sections in many RMs, and (iv) retained urban development (like rural development) as one of its 11 main themes. Project completion reports (PCRs) show that urban multisector projects have generally performed well, with 81% rated successful or highly successful between 1995 and 2005. WSSWM projects have not performed quite as well (67% were rated successful). However, this level of success is still on a par with the average for other sectors. OED s project performance audit reports, issued on average 2 3 years after the PCRs, are somewhat less positive, and raise issues related to the sustainability of many urban sector projects. Urban sector projects were generally rated as relevant, effective, and efficient. Because of their complexity and the often lower capacity at local government level, urban sector projects require more TA support than some other sectors. However, increased TA was not forthcoming after the USS. In fact, it declined by about 40% after the USS, whereas TA funds in other sectors increased by 29%. OED reports in the period 2000 2005 confirm that the success rate of urban multisector projects is comparable to that of the other sectors on average, whereas the performance of WSSWM projects is somewhat weaker.

vi To broaden the context, the study also scanned ongoing operations in some countries. It found that the success of the ongoing project portfolio varies with the country. The three country case studies point to excellent progress with projects in the PRC, more uneven progress with the relatively younger portfolio in India (but still encouraging prospects) and disappointing progress with the current portfolio of complex projects in the Philippines. However, in all three countries, the project managers were satisfied with some key features of ADB operations. A questionnaire survey found that project managers in executing agencies were dissatisfied with national decision-making systems, except in the PRC. SES missions reviewed some evaluations of central government funded urban programs, and noted unsatisfactory performance, indicating generic difficulties in the urban sector and in DMCs. Project managers in DMCs also voiced some consistent criticisms of some elements of ADB s policies and practices. There is a widely shared view that many problems with urban sector projects are caused by ADB s project processing and implementation model. These concerns add to the occasionally lukewarm responses to the question regarding the clients willingness to engage in further urban borrowing from ADB, and a lack of lobbying for follow-on projects. Some of the concerns identified by this study are being addressed by ADB-wide reform processes, such as the Innovations and Efficiency Initiative. These should, in principle, increase the potential for engaging in more and better urban sector interventions. ADB needs to restore some of the institutional focus and urban sector leadership it has lost since 2002. As a result of the urban sector review process initiated, it is already taking steps in that direction. ADB s Medium-Term Strategy II (2006 2008), issued recently, identified urban infrastructure as one of the core operational sectors where ADB plans to build up a critical mass of expertise to act as a leading provider of assistance. The litmus test will be whether the pipeline of urban sector projects, relying on growth in India and the PRC in particular, and perhaps a resurgence of borrowing by Indonesia and Pakistan, will lead to the expansion in lending that the USS saw as appropriate given the growing demand for capital investment in Asia s rapidly growing cities. As to the issue of capacity building, partnerships and anticorruption measures, the SES findings imply that there is more desired. The use of consultants to support project implementation is still widespread, long term partnerships with city clients have been limited, and support to urban governance reform has been rare. Corruption is widely acknowledged to be a major threat to the efficiency of both projects and service provision in the urban sector in Asia. About half of the reports and recommendations of the President reviewed were found to have reported on and/or highlighted specific anticorruption measures or capacity building. OED assesses the USS of 1999 as partly successful given lower efficiency in influencing operations, in spite of good relevance to Asia and the Pacific. Its ambition to go beyond what ADB was already providing up to that time was not realized, except in a few countries, like India. OED assesses ADB s urban sector operations as broadly successful, mainly on account of the above average effectiveness of projects and TA, and in spite of their higher than average sustainability risks. Overall, ADB failed to seize the opportunity to significantly increase the size and quality of its urban lending portfolio and TA activity. The rationale for the rating of the USS and subsequent operations is summarized below, followed by a set of recommendations.

vii Rating of the Urban Sector Strategy Relevance. The USS was relevant in terms of highlighting urbanization as one of Asia s mega-trends, and in identifying the need for ADB to focus on a response. All trends and issues identified by the USS are still relevant in 2005. The strategy appropriately anticipated ADB s increasing focus on poverty reduction. The 1999 USS was consistent with ADB s subsequent 2001 Long-Term Strategic Framework (2001 2015), which reflected urbanization as a major challenge for ADB, and one that ADB was particularly suited to address. However, the USS had little effect on the subsequent Medium-Term Strategy (2001 2005). The USS objective of increasing lending to urban areas was highly relevant given ADB s comparative strengths as an organization promoting infrastructure and capacity development. The strategy could have been more relevant as a guide to current operations if it had (i) made more distinction between regions and countries within the Asia and Pacific region in terms of urbanization, needs, and approaches; and (ii) given more guidance as to the role of the private sector in ADB s operations. Efficiency. The number of goals and targets make the USS hard to pursue in practical terms and hard to evaluate. After ADB s reorganization, the USS was even harder to pursue; urban sector staff were scattered and the new RSDD was not staffed with sufficient specialists to pursue special urban issues and coordinate effectively. The USS was not successful in providing a statement that focused ADB, as an institution, on the urban sector. Economic and sector work was not expanded. Country urban sector profiles and strategies have generally not resulted in a clear urban focus in country strategies and programs. ADB s TA for the urban sector diminished. ADB s 2002 reorganization and the new business processes led to much greater attention on country planning rather than sector planning. ADB s leadership and institutional focus on the urban sector, widely acknowledged in the 1990s, was due to the pre- 2002 division-centered organization. The efficiency of the USS within ADB is low, given its incompatibility with the present organizational structure. Growth in lending, TA, and knowledge products would normally need to be accompanied by appropriate levels of staffing. Given that ADB currently has a cap on increased levels of staffing, this would mean a redeployment of existing positions. The urban sector does not perform more efficiently in terms of preparing and implementing projects than before; there are as many delays and loan cancellations as (i) before the USS, and (ii) in many other sectors. Borrowers perceive ADB s transaction costs as high, although many executing agencies recognize significant ADB added value. Since ADB lost its specialized urban divisions and has few urban sections in RMs, there is a lack of focal point for urban clients. 1 This has led to the lack of an effective internal lobby to promote ADB involvement in the urban sector. Rating of Urban Sector Operations Effectiveness. Urban lending operations increased but not significantly. Intended expansions of operations into public-private sector partnerships, urban (mass) transit, and land management were not achieved, although there was some growth in the housing sector. The number of program loans and sector loans did not increase as planned. No more large urban forums were organized. However, other knowledge products increased. PCRs have generally rated the performance of urban sector projects as higher than the average of all ADB projects, especially integrated urban development (multisector) projects. OED ratings of completed urban sector projects have also been in the middle range, some sectors perform better and some 1 However, after the drafting of this study, one urban development division was created in a restructured South Asia Regional Department (May 2006).

viii perform worse. Given a reduced TA program in recent years and no growth in loan consulting services and loan components for institutional development or training, it is unlikely that the policy or capacity impact of current operations is higher than before the USS. Increased cofinancing sources may have partly offset this trend, but more TA resources could have made the impact larger. The USS did not specify outcomes across Asia and these are hard to quantify. To give an indication of the coverage of ADB s lending operations, ADB urban sector loans in the last 12 years covered 12% of the urban population in DMCs. Sustainability. The sustainability of urban operations varies according to subsector and country, in line with country contexts. OED evaluations have given lower ratings than PCRs with reference to the sustainability of urban sector projects and impact on institutions. There has been some progress with appropriate tariff setting and cost recovery in water supply and sewage systems, but practically all systems still need government subsidies, and are therefore less than sustainable. The housing finance subsector program, relying on financial intermediaries which add a margin when setting the cost of housing loans to final borrowers, has not been sustainable in some countries because of increasing availability of competing private sector sources and lower commercial interest rates. From ADB clients point of view, the need for urban sector projects continues to grow, but an exponential rise in demand is not foreseen, partly because of external debt considerations for many DMCs. ADB has yet to decide on a strategy for significantly increasing sub-sovereign lending. Doing so would increase the lending modalities that could be used to assist the financially stronger municipalities. Main Recommendations The SES recommends: (i) (ii) (iii) (iv) (v) ADB should formulate an Urban Sector Strategy II. The Urban Sector Strategy II should include a road map or business plan for implementation to translate the broad goals and objectives of the strategy into reality, given various constraints. The Urban Sector Strategy II should also include a results-based framework with clear, monitorable indicators. The new strategy needs to be grounded in a custodian division in RSDD, to coordinate across departments, monitor targets, and elaborate common themes. ADB needs to provide some guidance to staff on how to engage in anticorruption initiatives in the urban sector. Bruce Murray Director General Operations Evaluation Department

I. INTRODUCTION A. Background 1. The Asian Development Bank s (ADB) urban sector strategy (USS) was approved in July 1999 1 and was widely disseminated. Five years later, in 2004, ADB s Regional and Sustainable Development Department (RSDD) announced that it would review and then revise the USS by the end of 2005. 2 The adoption of poverty reduction as ADB s overarching objective, the emergence of the Millennium Development Goals (MDGs), and ADB s joining of the Cities Alliance in 2002, were seen as developments that were not adequately reflected in the USS. RSDD viewed the rate, scale, and pace of urbanization taking place throughout the Asia and Pacific region as having profound impacts on developing member countries (DMCs). ADB s recently issued staff instructions on the definition of directional documents and processing of policy and strategy papers were also mentioned as a reason for the revision. 3 The original USS lacked the disciplined approach and procedural rigor now envisaged by these instructions. For instance, it referred to "policy" at various places, but was not a policy document. 4 The USS also paid insufficient attention to the experience of other development banks, implementation arrangements, and human resource and cost implications required to implement the strategy. 2. An independent evaluation of the USS and related operations was included in the 2005 Work Program of the Operations Evaluation Department (OED) to contribute to the revision process. The OED evaluation was designed to examine the existing strategy and the effectiveness of urban sector operations. Work undertaken in parallel by RSDD would identify and review emerging urban sector trends, directions, good practices, and related issues. OED staff participated in an interdepartmental working group organized by RSDD to ensure that OED results were made available during the course of the study. This special evaluation study (SES) produced a number of intermediate outputs that have contributed to that review. The Strategy and Policy Department (SPD) also drew on a draft of this report when preparing ADB s Medium- Term Strategy II, 2006 2008 (MTS II). 5 OED s recommendation to postpone the revision of USS until after MTS II had provided guidance as to the role of the urban sector in ADB s operations, was reflected in a new target for the revision of the USS, by the fourth quarter of 2007. B. The 1999 Urban Sector Strategy 3. The USS codified the key subsectors in the urban sector as water supply, sanitation, solid waste, drainage, transport, environment, sites and services, housing, institutional strengthening and capacity building, land management, governance, and municipal finance. The USS also foresaw urban poverty reduction activities, including community development and microcredit, as possible components of urban sector projects. 4. The strategy highlighted the following operational objectives: (i) maximizing the economic efficiency of urban areas, (ii) reducing urban poverty, (iii) improving quality of life, and (iv) achieving more sustainable forms of urban development. The objectives were to be achieved through the promotion of urban sector policies to (i) encourage good governance; 1 ADB. 1999. Urban Sector Strategy. Manila. 2 ADB. 2004. Concept Paper for Policy or Strategy Paper: Urban Sector Strategy. Manila. 3 ADB. 2004. Staff Instructions on Definition of Directional Documents and Processing of Policy and Strategy Papers. Manila. 4 Policies (such as safeguard policies) carry mandatory elements; strategies do not. Existing sector and thematic policies are currently being revised and will be either retired or converted to strategies. 5 ADB. 2006. Medium-Term Strategy II, 2006 2008. Manila.

2 (ii) improve urban management; (iii) mobilize financial resources; (iv) reduce urban poverty; and (v) address urban development subsectors: water supply, sanitation, and solid waste management; land management, transport, and housing. Because of the complexity of the urban sector, the strategy was regarded as a flexible framework. 5. The USS proposed a systematic and incremental expansion of ADB s involvement in the urban sector in terms of the volume of lending, the range of subsectors to be covered, and the interrelationship of project components. The USS called for more attention to poverty reduction; community participation in projects; more institutional strengthening and capacity development; more private sector involvement; more program loans; and more projects in urban transport, urban housing, and urban land management. Integrated urban development projects were expected to continue to be the mainstay of urban sector lending. The USS also called for more staff resources for the sector and more research and knowledge products to support policy dialogue. C. Main Evaluation Questions 6. The goal of the study is to contribute to a discussion regarding urban sector interventions in ADB s program, and particularly to the revision of ADB s USS of 1999. The main questions that the SES addresses are as follows. (i) (ii) (iii) (iv) (v) (vi) (vii) (viii) Does urbanization in Asia need special attention by ADB? What is the nature of the USS, the context of its issuance in 1999, and what affected it afterward? What were the changes in ADB s program in the years after the issuance of the USS? Are the particular ambitions of the USS being achieved? What are clients perceptions of ADB s value added in the urban sector and transaction costs? How does the USS rate in terms of relevance and efficiency, and how do urban sector operations rate in terms of their effectiveness and sustainability? What are considerations in favor of a USS in ADB, including the (re-)designation of an urban sector, urban theme, and urban sector divisions? Under what conditions would a USS be most effective? What could be the suggestions for a revised USS, emerging from an evaluation of past and current operations? 7. The assessment of the changes in urban sector operations is based on a comparison of projects approved up to 6 years before the USS of mid-1999 with those approved afterward, up to mid-july 2005. A survey-based evaluation of the impact of the USS on beneficiaries was not undertaken because most of the projects approved post-uss are still under implementation and are not yet delivering the expected benefits. The evaluation assesses whether priorities set out in the USS have been followed in ADB s operations. Special attention is given to the question why, given increasing urbanization in the region, ADB s urban portfolio has not grown more rapidly. The evaluation also considers the possible effects of new initiatives in ADB, such as the Innovation and Efficiency Initiative, 6 the Enhancing Asian Development Bank Support to Middle 6 ADB. 2005. Cost Sharing and Eligibility of Expenditures for ADB Financing: A New Approach. Manila; ADB. 2005. Pilot Financing Instruments and Modalities. Manila; ADB. 2005. Local Currency Loan Product. Manila; ADB. 2006. Revising the Procurement Guidelines. Manila; ADB. 2006. Policies and Procedures on the Use of Consultants. Manila.

3 Income Countries and Ordinary Capital Resources Borrowers, 7 and the Review of the Policy on Supplementary Financing. 8 8. A question that the SES raises is whether ADB needs a USS, or whether it would be more effective if various sector or country strategies paid greater attention to spatial or urban aspects. Staff instructions on the preparation of strategies indicate that strategies should be prepared for countries and/or regions, sectors, and thematic areas. Urban development is no longer included in any of these three categories. Given that ADB has many policies and strategies, this also raises the question of a strategy s role in general. Some other regional development banks do not have urban sector strategies as such. The African Development Bank has none; the Inter-American Development Bank has a broader strategy on subnational development; 9 and the European Bank for Reconstruction and Development has an elaborate policy on municipal and environmental infrastructure. The World Bank has an urban and local government strategy. The European Bank for Reconstruction and Development and the World Bank have municipal divisions or urban theme groups; ADB does not have either of these. D. Organization of the Study 9. The study was headed by an OED evaluation specialist, and relied on short-term contributions of five urban planning consultants, three research assistants, and a national evaluation officer. A small-scale technical assistance (TA) funded studies to support the preparation of this SES. 10 The SES was also supported by funds from OED s TA for Selected Evaluation Studies for 2005, 11 particularly for a country case study in the People s Republic of China (PRC). 10. The evaluation relied on the following sources of information: (i) interviews with relevant ADB staff; (ii) study of ADB s policy and strategy documents; (iii) quantitative analysis of ADB databases on sector trends, OED databases on project success ratings, and documentation for all 88 urban sector projects approved between 1993 and 2004, for all 51 urban sector projects with project completion reports (PCRs) issued between 1999 and mid-2005, and for all 191 TA grants approved between mid-1993 and mid-2005; (iv) a questionnaire survey of all 52 ongoing urban sector projects (the response rate was 85%); (v) a selective review of urban sector evaluations undertaken by other agencies; (vi) an assessment of ADB urban sector profiles and studies and some other ADB publications; and (vii) country case studies in PRC, India, and Philippines. The country case studies included visits to all 15 ongoing projects in the sector during July and August 2005, and interviews with various institutional stakeholders, including project managers (PMs) and government officers in planning and finance agencies. A workshop was held in August 2005 with stakeholders in the Philippine urban sector. Appendix 1 provides methodological notes to the questionnaire survey. Appendix 2 lists all urban sector projects approved from July 1993 to June 2004, and Appendix 3 lists all urban sector related TAs. 7 ADB. 2006. Enhancing Asian Development Bank Support to Middle Income Countries and Ordinary Capital Resources Borrowers. Draft. Manila. 8 ADB. 2005. A Review of the Policy on Supplementary Financing: Addressing Challenges and Broader Needs. Manila. 9 The Inter-American Development Bank s Oversight and Evaluation Office concluded in 2003 that formal strategies provided relatively weak guidance in operations design (as a result, the office s future work would focus on evaluation of the projects themselves rather than sector and thematic strategies). 10 ADB. 2005. Small-Scale Technical Assistance for the Special Evaluation Study on Urban Sector Strategy and Operations. Manila (TA 6240-REG, for $150,000, approved 21 April 2005). 11 ADB. 2005. Technical Assistance for Selected Evaluation Studies for 2005. Manila (TA 6246-REG, for $1.25 million, approved 14 June 2005).

4 11. The urban sector represents a wide range of interventions, combining a number of currently independently classified functional sectors in ADB. Given time and resource limitations and current topical interests in ADB, this SES gives more attention to process and implementation oriented issues than to technical issues. This focus also reflected, in part, the fact that OED reviewed such issues in a 1997 impact evaluation, 12 and a 2002 impact evaluation of water supply and sanitation projects. 13 Many of the findings of those evaluations remain relevant. 12 ADB. 1997. Impact Evaluation on the Bank Assistance to the Urban Development and Housing Sector. Manila. 13 ADB. 2002. Impact Evaluation Study of Water Supply and Sanitation Projects in Selected Developing Member Countries. Manila.

5 o o 130 00'E 70 00'E Map 1 PERCENTAGE OF POPULATION IN URBAN AREAS, 1970 KAZAKHSTAN MONGOLIA ARMENIA AZERBAIJAN UZBEKISTAN KYRGYZ REP. TURKMENISTAN o 40 00'N Sea of Japan REPUBLIC OF KOREA Yellow Sea TAJIKISTAN PEOPLE'S REPUBLIC OF CHINA AFGHANISTAN PAKISTAN NEPAL o 40 00'N JAPAN East China Sea BHUTAN % Urbanization Rate BANGLADESH INDIA Arabian Sea MYANMAR Bay of Bengal TAIPEI,CHINA LAO PEOPLE'S DEMOCRATIC REPUBLIC THAILAND SRI LANKA 0 50 -- 75 VIET NAM CAMBODIA 75 -- 100 South China Sea PHILIPPINES International Capital Boundaries are not necessarily authoritative. P A C I F I C BRUNEI DARUSSALAM M A L A Y S I A MALDIVES 1 -- 25 25 -- 50 SINGAPORE o PAPUA NEW GUINEA N 0 500 I N D I A N 1000 Java O C E A N O C E A N Celebes Sea 0 o Sea I N D O N E S I A TIMUR-LESTE Kilometers o o 130 00'E 70 00'E o o 130 00'E 70 00'E Map 2 PERCENTAGE OF POPULATION IN URBAN AREAS, 2000 KAZAKHSTAN MONGOLIA ARMENIA AZERBAIJAN UZBEKISTAN KYRGYZ REP. TURKMENISTAN o 40 00'N Sea of Japan REPUBLIC OF KOREA Yellow Sea TAJIKISTAN PEOPLE'S REPUBLIC OF CHINA AFGHANISTAN PAKISTAN NEPAL o 40 00'N JAPAN East China Sea BHUTAN % Urbanization Rate BANGLADESH INDIA Arabian Sea Bay of Bengal CAMBODIA SRI LANKA 1 -- 25 25 -- 50 50 -- 75 75 -- 100 South China Sea PHILIPPINES International Capital Boundaries are not necessarily authoritative. P A C I F I C BRUNEI DARUSSALAM M A L A Y S I A MALDIVES 0 TAIPEI,CHINA LAO PEOPLE'S DEMOCRATIC REPUBLIC VIET NAM THAILAND MYANMAR SINGAPORE o PAPUA NEW GUINEA N 0 500 I N D I A N 1000 Java O C E A N O C E A N Celebes Sea 0 o Sea I N D O N E S I A TIMUR-LESTE Kilometers o o 130 00'E 70 00'E o o 130 00'E 70 00'E Map 3 ANTICIPATED PERCENTAGE OF POPULATION IN URBAN AREAS, 2020 KAZAKHSTAN MONGOLIA ARMENIA AZERBAIJAN UZBEKISTAN KYRGYZ REP. TURKMENISTAN o 40 00'N Sea of Japan REPUBLIC OF KOREA Yellow Sea TAJIKISTAN PEOPLE'S REPUBLIC OF CHINA AFGHANISTAN PAKISTAN NEPAL o 40 00'N JAPAN East China Sea BHUTAN % Urbanization Rate BANGLADESH INDIA Arabian Sea Bay of Bengal CAMBODIA SRI LANKA 1 -- 25 25 -- 50 50 -- 75 75 -- 100 South China Sea PHILIPPINES International Capital Boundaries are not necessarily authoritative. P A C I F I C BRUNEI DARUSSALAM M A L A Y S I A MALDIVES 0 TAIPEI,CHINA LAO PEOPLE'S DEMOCRATIC REPUBLIC VIET NAM THAILAND MYANMAR SINGAPORE o PAPUA NEW GUINEA N 0 500 1000 I N D I A N O C E A N O C E A N Celebes Sea Java 0 o Sea I N D O N E S I A TIMUR-LESTE Kilometers o 70 00'E o 130 00'E asia 06-0876 HR

6 II. URBANIZATION IN ASIA 12. This chapter summarizes urbanization trends in Asia, the financing needs of the urban sector, the current financing gap, and the relation of urban sector work to achieving the MDGs. This analysis provides a context in which to evaluate the relevance of the USS and ADB s urban sector operations. A. Trends 13. Viewed over a 100-year perspective, the region is in the midst of a transformation from a rural, agriculture dominated region to an urban, industrial/service sector dominated region. This transformation is being accompanied by major demographic shifts. Asia has the fastest growing urban population in the world (Maps 1, 2 and 3). With almost 3.9 billion people, Asia currently has 61% of the global population. Its share of the global urban population has risen from 9% in 1920 to over 48% in 2005. Based on current projections, 14 this proportion is expected to rise to 53% by 2030. Even if the United Nations (UN) 15 recent forecast of declining national and urban growth rates in Asia over the next 30 years proves to be correct, Asia s urban population will grow at a rate of 2.4% per year, increasing from 1.35 billion (37% of the total population) in 2000 to 1.94 billion (48%) in 2015, 16 a period when population is expected to grow at less than half this rate at 1.1% per year. The trend in some of ADB s major DMCs is shown in Figure 1. 17 14. There are currently 19 megacities in the world (i.e., with populations over 10 million), 10 of which are in Asian developing countries. By 2010, there will be 23 megacities, including 12 in Asian developing countries. These megacities will be home to 222 million people. By 2015, three of the world s five largest megacities will be in Asia: Delhi, Dhaka, and Mumbai. More significant in terms of urbanization will be the rise of metropolitan regions. By 2020, two-thirds of the urban population of the Association of Southeast Asian Nations (ASEAN) is expected to live in five metropolitan regions (MRs): the Java MR (100 million); the Bangkok-centered MR (30 million); the Manila MR (30 million); the Singapore Triangle (10 million); and the Kuala Lumpur Klang MR (6 million). Southeast Asian MRs will be relatively small, compared to those in East Asia, such as the Hong Kong Shenzen Guangdong MR (120 million); the Greater Shanghai MR (83 million); and the Tokyo Osaka Kyoto Kobe Nagoya MR (60 million). With the Beijing MR covering 16,870 kilometers and the Shanghai MR extending over 6,340 square kilometers, East Asian MRs are introducing urban issues on geographic scales never before experienced. The rapid development of small to medium sized cities places investment and management pressures on what are often relatively weak local administrations. In both India and PRC, small- and medium-sized towns and cities, which can have up to 1 million people, are growing at a faster rate than large or megacities. 14 Population projections 2004 based on United Nations (UN). 2002. World Urbanization Prospects: The 2001 Revision. Department of Economic and Social Affairs, Population Division. New York. 15 UN. 2002. World Urbanization Prospects: The 2003 Revision. Department of Economic and Social Affairs, Population Division. New York. 16 This compares to current estimated growth rates for ADB s developing member countries (DMCs) of 1.2% for total population and 3.4% for urban population. 17 More information is in Supplementary Appendix A.

7 Figure 1: Urbanization Rates, 1950 2030 a 1,000,000 900,000 800,000 Urban Population (thousands) 700,000 600,000 500,000 400,000 300,000 200,000 100,000 281% 168% 60% 153% 388% a 0 1950 1955 1960 1965 1970 1975 1980 1985 1990 1995 2000 2005 2010 2015 2020 2025 2030 Year Bangladesh Indonesia People s Republic of China Brazil Japan Philippines India Pakistan Russian Federation United States of America From 1975 2010, People s Republic of China's urban population will have grown by 281%, although part of this is due to revision of administrative boundaries Source: United Nations. 2002. World Urbanization Prospects: The 2003 Revision. Department of Economic and Social Affairs, Population Division. New York. 15. Economic Importance of Cities. The rapid growth of urban Asia reflects the rapidly increasing importance of cities as (i) focal points of economic activity and engines of economic growth; (ii) centers of excellence for education, health care, innovation, entrepreneurship, business, commerce, industry, culture and social services; (iii) large markets for a wide range of products, goods and services; (iv) nodal points for transportation, telecommunications and information technology systems; and (v) the primary location for employment and livelihood opportunities. Given the positive correlation between levels of urbanization and gross domestic product (GDP), there is strong evidence to suggest that urbanization enhances economic productivity. There is also a strong positive correlation between city size and GDP per capita. The ratio of city GDP per capita to national GDP per capita is found to be 1.9 for Metro Manila, 2.5 for Kolkata, 3.5 for Bangkok, and 3.7 for Shanghai. Mumbai is generating one sixth of India s GDP, confirming the synergy of private entrepreneurs and investment in a single location, despite insufficient infrastructure. However, productivity rates would be even higher if city efficiency is enhanced.

8 16. Combinations of an urban core or cores with an urbanizing hinterland linked to the core by functional ties are becoming increasingly regarded as the engines of economic growth in a globalizing world, which demand development and implementation of new modes of economic governance. These city regions are of a scale which suggests they are the ideal point of entry for public policy interventions. This presents strategic and management challenges to national, regional, and city governments, and to ADB in seeking to support these regions. 17. Urban Poverty. Over the past three decades, there has been an unprecedented decline in poverty in the Asia and Pacific region. During the early 1970s, more than half the population of the Asia Pacific region lived below $1 per capita per day, average life expectancy was 48 years, and only 40% of the adult population was literate. By 2002, a quarter of the people live in income poverty, life expectancy has increased to 65 years, and about 70% of adults are literate. 18. Although there has been remarkable progress in reducing poverty in Asia, the continent still accounts for two-thirds of the world s poor (living on $1 or less a day per capita), affecting about 800 million people; 240 million 260 million of the poor live in urban areas. Both income and non-income urban poverty (quality of life poverty) are major problems in the region. Crime and corruption are often serious hazards. Street children begging and children used for sexual exploitation are perhaps the most heartrending examples of urban poverty. The Asia and Pacific region hosts about half of the world s urban slum dwellers. In 2001, 43% of the region s 1.2 billion urban residents were living in slums, with uncertain security of tenure, overcrowded conditions, poor housing, lack of adequate services, and poor environmental conditions. 18 The requirements of the urban poor in both low and middle-income countries (MICs) for decent living conditions and access to work need deeper consideration so that they too can enjoy the benefits of economic growth. The majority of the urban poor are concentrated in Bangladesh, PRC, India, Indonesia, and Pakistan. Nearly one third of the urban poor have no access to safe water and more than two thirds do not have access to adequate sanitation. 19. Environmental Problems. Because of lagging governance and lacking resources, cities have often have negative environmental externalities. Traffic congestion, crowded environments and housing conditions, overloaded infrastructure, flooding, water scarcity, and air and water pollution have constrained the growth of cities economic efficiency. 20. Natural disasters affected almost 2 billion people worldwide in the 1990s, and over 80% of the affected people lived in Asia. Populous parts of Asia are prone to earthquakes, storms and flooding. Climatic change is increasing the risks of flooding in coastal or riverine cities, such as Bangkok; Dhaka; Hong Kong, China; Manila; Mumbai; Shanghai; and Tokyo. Many poor people live in slums on land vulnerable to floods or pollution. 19 B. Needs 21. Urban sector development needs in the Asia and Pacific region span the full range of activities, infrastructure, services, and facilities required to (i) improve the efficiency of cities as places in which to invest and work, and (ii) improve the condition of cities as places in which to live and prosper. Improved governance and regulatory environments are required to address urban market distortions and attract private capital. Administrative reform and capacity building is required to improve urban planning and management, and to ensure that cities develop 18 UN-HABITAT (United Nations Human Settlements). 2003. Challenge of Slums-Global Report on Human Settlements 2003. Nairobi. 19 ADB. 2006. Review of the Urban Sector Strategy. Draft. Manila.

9 efficiently and maximize their development potential. Improved and extended infrastructure, utilities, and services are required to support economic development, improve peoples quality of life, and reduce poverty. In addition to the negative impact on quality of life, the economic impact of pollution in Asian cities, in terms of loss of productivity, has been estimated at between 1% and 5% of their GDP. Improved urban infrastructure and services are critical, requiring both improved management and an enabling environment for the private sector investment that is necessary to create economic growth and jobs. 22. Future Urban Infrastructure Demand. A World Bank model was used to predict future infrastructure demand in all sectors (net of requirements for rehabilitation). 20 The resulting estimation of the infrastructure investment needs for all developing countries over 2005 2010 amounted to $465 billion per year or 5.5% of their GDP (including both capital and maintenance expenditures). A similar methodology adopted for East Asia 21 and extended to include all ADB s DMCs for the same period indicated a requirement of $230 billion per year, for the following sectors: (i) electricity generation and supply, (ii) telephone landlines, (iii) mobile telephones, (iv) paved intercity roads, (v) rail lines, and (vi) access to improved water supply and sanitation (WSS). 22 23. A similar methodology was used to approximate urban development financing requirements (excluding power, telecommunications, railways and intercity roads, but including solid waste management and slum upgrading) for DMCs in the Asia and Pacific region. The same best practice unit costs were used for WSS, and similar figures were derived for solid waste management and slum upgrading based on recent field data. 23 The analysis suggested total financing requirements for just water supply, sanitation, solid waste management, and slum upgrading in urban areas will be $25 billion per annum (Appendix 4). This number would double to $50 billion per year if urban roads were included. The demand for mass transit systems is more difficult to assess. However, megacities and large cities across the region are going to find investment in such systems increasingly necessary as cities expand and traffic volumes exceed the carrying capacity of urban road networks. Even the conservative assumption that 100 kilometers (km) of rail mass transit system are constructed annually across the region, adds a further $10 billion per annum to the urban funding requirements. 24. This study estimates that ADB s DMCs may need to spend about $60 billion per annum on the development of urban infrastructure and services between 2006 and 2010 for water supply, sanitation, solid waste management, slum upgrading, urban roads, and mass transit systems. These figures agree quite closely with those generated for a number of countries, including India, where estimates for capital, operations, and maintenance were estimated at around $8 billion per annum for WSS and urban roads alone. 24 Asia Watch 2015 recently estimated the annual requirement for improved water supply for Asia (urban and rural) to meet 20 Fay, Marianne and Tito Yepes. 2003. Investing in Infrastructure: What is Needed from 2000 2010? World Bank Working Paper No. 2545. Washington DC: World Bank. 21 Cambodia, Federated States of Micronesia, Fiji Islands, Indonesia, Kiribati, Lao People s Democratic Republic, Malaysia, Marshall Islands, Mongolia, Myanmar, Palau, Papua New Guinea, Philippines, PRC, Samoa, Solomon Islands, Thailand, Timor-Leste, Tonga, Vanuatu, and Viet Nam. 22 The methodology was based on infrastructure stock trends using real demand data from eight Asian counties for which data was available, and adjusting for economic growth and geographic variables to extend this analysis to other countries within the same income group. Costs of increasing infrastructure stock and associated maintenance requirements were based on best practice unit costs derived from the actual cost of infrastructure and service provision, and were reduced to an average cost per capita per year for middle-income and low-income economies to derive the total funding requirement. The figures for PRC and Malaysia were adjusted slightly from the middle-income country figure to provide a better fit with observed data. 23 Figures taken from recent urban development projects with components in these sectors. 24 Rakesh Mohan Committee. 1996. India Infrastructure Report 1996. Delhi: Government of India.

10 the MDGs at $8 billion; access for all would need $16 billion $17 billion, and regulated in-house supply would require $85 billion. 25 25. Overall, the funding requirements for infrastructure translate to 6.9% of GDP for PRC, 3.6% for other MICs, and 6.3% for low-income countries. While this analysis provides some indication of the likely funding requirements, the cost of providing these services can be met by a combination of cost recovery from users and fiscal revenues. Urban infrastructure has a long operational life thus, the most appropriate type of financing is a loan with long maturity. 26. Current Investment. Data from a variety of sources show stagnation or a relative decline of the share of multilateral agencies in financing urban infrastructure contrasted with a rise in domestic financing. There is evidence of a recent recovery of confidence in infrastructure financing by the private sector, although this has yet to be reflected in capital flows. While decentralization has given more responsibilities to local governments in the region, there is no evidence that this has resulted in additional resources being made available to fund local infrastructure and services. Supplementary Appendix B summarizes the programs of some of the main external agencies involved in the urban sector in Asia. The main external agencies are the World Bank; ADB; the Japan Bank for International Cooperation (JBIC); Kreditanstalt für Wiederaufbau of Germany; a large number of bilateral agencies such as Australian Agency for International Development, European Union country agencies, EuropeAid, and the United States Agency for International Development (USAID); and agencies such as the United Nations Human Settlements Programme (UN-HABITAT). The three largest financiers (ADB, JBIC, and World Bank) account for 50% to 70% of the donor financed urban infrastructure. 27. The conclusion is that perhaps half of the investment need of $60 billion per year (exclusive of housing investments) is being covered by governments, external agencies, local financial institutions, and the private sector. This leaves a huge funding gap. C. Relation of Sector with the Millennium Development Goals 28. The urban sector is directly and indirectly related to ADB s overarching objective of poverty reduction and achievement of the MDGs. Urban interventions have a direct impact on some of the MDGs (Box 1) and an indirect impact on others. Urban economies particularly economies of large urban agglomerations generate proportionately significantly greater wealth per capita than rural areas. The recent history of development in Asia bears out the role of economic growth in achieving the MDGs. The MDG upon which urban development initiatives may have the greatest direct impact is the seventh: ensuring environmental sustainability. This MDG includes specific targets to halve the proportion of people without access to potable water and adequate sanitation between 1990 and 2015, and to significantly improve the lives of at least 100 million slum dwellers by 2015. Box 1: Urban Sector and Millennium Development Goals (MDGs) 1. MDG 1. The first goal to eradicate extreme poverty and hunger, halve the proportion of people living on less than one dollar a day, and halve the proportion of people who suffer from hunger between 1990 and 2015 is impacted by the ability of cities to work more effectively to create jobs and generate greater wealth. The positive impact which rapid growth in East and South Asia has had on poverty reduction is well documented. Urban areas are the generators of economic growth and job creation. 25 Asia Watch 2015. 2006. Are Countries in Asia on Track to Meet Target 10 of the Millennium Development Goals? New York: ADB, United Nations Development Programme, United Nations Economic and Social Commission for Asia and the Pacific and World Health Organization.

11 Continued support to urban areas, in terms of infrastructure provision and improved urban management and governance, will be critical in ensuring the increasing efficiency of cities thus, the sustaining of economic growth and the continuing reduction of extreme poverty and hunger in both urban and rural areas. 2. MDG 4. Urban development has a direct impact on the fourth MDG to reduce child mortality by two thirds by 2015. Improved water supply and environmental sanitation, especially when combined with hygiene awareness for mothers, has a positive impact on reducing infant and child mortality. While maternal health may only be marginally dependent on adequate service provision and environmental sanitation, urban investments in infrastructure and improved management can have a significant impact on the achievement of MDG5 reducing the maternal mortality ratio by three quarters through better environmental health resulting from improved access to better quality services. 3. MDG 6. Urban development may have some impact on the sixth MDG, which calls for combating HIV/AIDS, malaria, and other diseases; and specifically for reversing the spread of HIV/AIDS. Investments in drainage and sanitation remove stagnant water, reducing the opportunity for mosquito breeding, and improving public health which, in combination with improved hygiene practices, will significantly reduce the incidence of faecal-oral diseases. Prostitution and drug use, which increase the risk of HIV/AIDS infection, are often related to growing urban poverty. 4. MDG 7. The relevant sub-targets are to halve the proportion of people without access to potable water and adequate sanitation between 1990 and 2015, and to significantly improve the lives of at least 100 million slum dwellers by 2015. Connecting up poor people, with limited ability to pay, to piped water supply systems is not an attractive option to the private sector. Slum eradication is notoriously difficult. While the rapid economic growth of urban Asia has been responsible for lifting many out of poverty, the continuing attraction of these cities to economic migrants from rural areas, coupled with the absence of affordable serviced housing, means that there is a limitless supply of people for informal (slum) settlements. Source: Special Evaluation Study on Urban Sector Strategy and Operations. 29. There is limited analytical work available on urbanization, and this makes the MDGs hard to track. There are no solid figures as to whether reduced poverty levels in developing Asia over the years have also meant reduced poverty incidence, or even slum incidence, in urban areas. Certainly, urban poverty persists on a large scale and some of it is extreme poverty. Many urban slums are expanding. Data available indicate that the proportion of urban poor is increasing faster than the overall rate of urban population growth, especially in the poorer countries. 26 UN-HABITAT expects that (i) the estimated 30% level of global urban poverty will grow to 45% 50% by 2020; (ii) urban areas of the world s least developed regions will absorb nearly all the global population increase over the next three decades; and (iii) as a general rule, least developed countries, although least urbanized, are experiencing the highest growth in urban population and have the highest levels of urban slum incidence. 30. Urban water and sanitation service provision are not improving in all DMCs. While the proportion of urban population in DMCs with sustainable access to an improved water source is estimated to be 85%, the proportion of households with house connections declined from 55% in 1990 to 53% in 2002 in South Asia and from 69% to 67% in the Pacific, due to rapid urbanization and lagging expansion in systems. 27 While the proportions went up in Eastern (developing) Asia from 82% to 91% over the same period and from 37% to 45% in Southeastern Asia, in many cities water is not available 24 hours a day and quality is poor. 28 In many cities, 26 UN-HABITAT. 2004. The State of the World s Cities 2004/2005. Globalization and Urban Culture. Nairobi. 27 World Health Organization and United Nations Children Fund. 2004. Meeting the MDG Drinking Water and Sanitation Target. A Mid-Term Assessment of Progress. New York and Geneva. 28 ADB. 2005. Water Utilities Data Book. Manila.

12 householders only receive piped water for a few hours a day, or in some cases a few hours every two or three days. Fluctuating pressures contribute to contaminated water where network leakage is high (average 35%). It is often routine for householders to boil water before consumption, a costly and energy-wasting exercise, or to buy expensive bottled water. While statistics show that 76% of urban households have access to adequate or improved sanitation and the percentages are improving over time, 29 this simply means growing access to a sewer connection, septic tank, or improved pit latrine (footnote 27). All such systems have raw sewage or wastewater effluent which must be disposed of without further endangering public health or polluting the environment. Based on available data from 50 Asian cities, 30 about 40% of household sewage is treated prior to disposal. Frequently, only primary treatment is available, leaving the bulk of pollutants to be discharged untreated to urban water systems or the ocean. 31. ADB estimates (footnote 30) that while 75% of solid waste generated in urban areas is collected, less than 60% is taken to a solid waste disposal site. Most sites are open dumps, which sometimes provide a livelihood for poor urban scavengers. Only a fraction of solid waste (perhaps 10%) is disposed of in properly managed solid waste landfill sites. The scale of informal recycling is considerable: most developing cities achieve recycling of 5% to 25% of waste collected. However, while often sophisticated operations, recycling activities frequently involve manual picking by the poor at various stages in the waste disposal system, and present health risks to those involved. III. ADB S FOCUS ON THE URBAN SECTOR A. Genesis of the Urban Sector Strategy 32. The USS of 1999 had a long genesis. ADB was established in the mid-1960s when agriculture and the rural economy dominated the region. During its initial years of operation, the agriculture sector accounted for a large part of ADB s operations. In the late 1980s, ADB began paying more attention to urban issues. This was clearly reflected by increased levels of TA for urban subsectors beginning in 1988, and two theme chapters in ADB s annual reports (one on Asian urbanization in 1992 and one on Asian megacities in 1996). ADB sponsored five major regional seminars on the urban sector in the 1990s: urban poor (1991), water resources in megacities (1993), management of megacities (1995), urban infrastructure finance (1996), and Management of Cities: Establishing Networks and Cooperation (1998). In 1996, ADB issued one of its most widely read urban publications on urban infrastructure finance. 31 A number of groundbreaking urban poverty studies were published and ADB made major contributions to the second United Nations Conference on Human Settlements (Habitat II) in 1996. 33. During these years, the idea gained ground that ADB should have its own urban development and housing policy. A working group was produced in September 1996. In March 1997, it was decided that preparation of a strategy paper would be more expedient than a policy paper. A review of experience with some other policy papers had shown that such papers, as approved by ADB s Board of Directors, ought to follow thorough external consultations with governments, nongovernment organizations (NGOs), and private sector parties. The urban sector was undergoing rapid change; it was felt that such a process would be too complex and time-consuming. In addition, staff might be overly constrained by the resulting policy guidelines. Progress with drafting of the strategy document was slow. Three years were 29 Coverage went up from 64% in 1990 to 69% in 2002 in Eastern (developing) Asia, from 54% to 66% in South Asia, from 67% to 79% in Southeast Asia, and from 83% to 84% in the Pacific (footnote 27). 30 ADB. 2001. Cities Data Book: Urban Indicators for Managing Cities. Manila. 31 Brockman, Royston A. C., and Allen Williams, eds. 1996. Urban Infrastructure Finance. ADB: Manila.

13 required for the Water Supply, Urban Development and Housing Division in the Agriculture and Social Sectors Department (East) to finalize the document, which was sent for information to the Board on 19 August 1999. More than 30 years after its founding, ADB produced its first strategy statement on the urban sector. 34. The USS legitimized ADB s long-time involvement in a disparate set of investments for which no policy statement had been issued until that time. The document had an important public relations function at a time when many of the routine operations of multilateral development banks were being questioned. The USS gave a degree of coherence to many activities that ADB had been doing or supporting in urban areas in practice, which dealt directly or indirectly with urban public health, public space, urban governance, and urban poverty. The document gave ADB more legitimate access to explicitly urban forums such as that of the World Bank, UN-HABITAT, and the Cities Alliance. At a more practical level, the USS provided a common vocabulary for ADB s work in the urban sector; a set of subsectors commonly agreed to be part of the urban sector (although very wide); and a set of modalities, including integrated urban development projects (IUDP). The IUDP is indicated in this report as urban multisector project, in line with ADB s new classification system. 35. The USS was not designed to guide staff in identifying or developing specific projects. The USS codified and systematized practices already adopted. The report had excellent sections on urbanization challenges, and elaborated good practices regarding assistance to urban areas. Its recommendations were inclusive, rather than selective. As such, the USS was uncontroversial, did not suggest new directions to staff, and seemed not to require dramatic shifts in ADB s focus or ways of doing business in the urban sector. ADB staff interviewed suggested that the strategy was used primarily as a means to inform new staff, consultants, and external stakeholders unfamiliar with ADB s work in the sector. Staff viewed related documents, such as urban sector profiles and country urban strategies as being more relevant in giving direct guidance on identifying, designing, or implementing urban sector projects. At least 11 of these profiles and country strategy documents had been prepared before the 1999 USS. Those prepared afterward generally did not refer much to the USS. 36. The USS did not get much attention within ADB subsequent to its issuance. Apart from not benefiting from a Board discussion, there was no reference to the USS in the section on urban development in ADB s 1999 Annual Report, as attention had turned to poverty reduction as ADB s new overarching goal, the adoption of a health policy, and a review of the private sector strategy. Even subsequently, key ADB-wide publications did not give much attention to the urban sector strategy. Urbanization was seen as one of the eight main development challenges 32 in ADB s March 2001 Long-Term Strategic Framework (2001 2015) 33 and experience with urban governance was regarded as one of ADB s particular strengths. 34 However, there was no special attention to the urban sector in ADB s Medium-Term Strategic Framework (2001 2005) of October 2001. 35 The only reference is the statement that 32 The others were (i) poverty reduction, growth, and sustainability; (ii) social development; (iii) environment; (iv) demographic shifts; (v) information and communications; (vi) regional ties; and (vii) private sector resources. 33 ADB. 2001. Moving the Poverty Reduction Agenda Forward in Asia and the Pacific: The Long-Term Strategic Framework of the Asian Development Bank (2001 2005). Manila. 34 See para. 43 of the Long-Term Strategic Framework: One important area where ADB has developed considerable expertise is the governance of the region s rapidly growing cities and municipalities. With continuing population pressures on urban areas, the development and administration of the necessary urban physical and social infrastructure are essential to maintain an acceptable quality of life, maintain social harmony, and address the needs of the rapidly growing numbers of urban poor. The expertise ADB has developed in dealing with governance issues for urban development will be broadened and applied to addressing problems in all sectors. 35 ADB. 2005. Medium-Term Strategic Framework (2001 2005). Manila.

14 investments in several key areas include priority physical infrastructure necessary to promote growth and productivity in both rural and urban areas, and essential social infrastructure and services for human development to support pro-poor growth. 37. In retrospect, the USS shared the same limitations as most ADB sector strategies. It was oriented more towards an external than an internal audience. In the absence of clear guidance from ADB-wide strategy documents or other pertinent Board decisions, the USS could not provide strategic choices or set priorities. It took such a wide view of the urban sector that it included all that was needed for more efficient and effective development. It thereby exceeded the scope of activities generally engaged in by ADB s urban sector divisions at the time. The USS included a comprehensive set of urban development goals and objectives, and strategic policy areas, but did not provide strategic options and a rationale for priority setting and focus; nor did it set time-bound quantitative targets for cities or countries, staff deployment, or loan or TA amounts. As such, it implied but did not specify a significant increase in loan funds for the urban sector. Since a significantly different allocation of staff resources and loan funds would have had consequences for other sectors, negotiation across ADB and with DMCs would have been necessary to achieve this objective. This broader consultation did not take place. The USS did not exclude any urban subsector. It defined functional sectors, but it could not decree that specific urban sector divisions or departments would need to be expanded to meet the expectations set out in the USS. It did not state that more staff time would be devoted to identify and administer more urban sector projects. It did not estimate the demand for urban sector projects from ADB s main borrowers. There was criticism from some quarters that IUDPs, like integrated rural development projects, were less efficient. However, the USS stressed that such projects would be the mainstay of the strategy. Given its rationale, the USS could not do otherwise the theoretical interrelatedness of investments in public health and public space appeared to make such types of projects unavoidable. If uncollected solid waste clogged storm water drains, then the project needed to address not only drainage but also waste management. If public health and public space oriented subsectors were not by their nature interrelated, then urban chapters in different sector strategies would have sufficed. Thus, the USS stressed the interrelations between sectors as an argument in favor of the need for holistic assessments of the urban sector and multisector or multicomponent projects. 38. The USS did not anticipate the changes that were coming its way, such as ADB s reorganization and changes in business processes beginning 2002. B. Organizational and Other Changes in ADB 39. The Operations Evaluation Mission (OEM) believes that the USS was a long overdue product of some major contributions that ADB had made to the urbanization debate in the 1990s. When the USS was issued, it turned out to be the closing session of this debate within ADB, rather than a new dawn of activity. No further major urban conferences were held. Within ADB, attention shifted to the related water policy, approved by the Board on 16 January 2001, 36 which led to an expanded program of research, TA, and international forums in that area. The international development discourse had moved on to different topics, such as poverty reduction, governance, and later the MDGs, Management for Development Results agenda, and anticorruption. Some measure of the urbanization discussion returned with some presentations at ADB s 2005 Annual Meeting. The ongoing discussion of the role of infrastructure in Asia s development, and ADB s role in it, may have contributed to this. 36 ADB. 2005. Water for All: The Water Policy of the Asian Development Bank. Manila.

15 40. A significant unforeseen organizational change was the abolition of the two Water Supply, Urban Development and Housing Divisions in ADB (East and West) during ADB s 1 January 2002 reorganization and the establishment of the four social sector divisions, one in each of the four new regional departments. The result of this reorganization was a scattering of urban sector expertise and a reduction of the beneficial effects from their consolidated management and daily interaction. WSS projects and integrated urban development projects were isolated from urban environmental projects and urban housing finance projects. These henceforth came under the responsibility of four agriculture, environment and natural resources divisions, and a Pacific operations division; and four governance, finance and trade divisions. In the social sector divisions, urban water supply and urban development were grouped with other sectors such as education and health. As a result of this reorganization, the specific urban focus that existed in the two previous divisions was diluted. The revised business processes taking effect in 2002 gave more weight to country strategy and program (CSP) preparation processes. The agriculture, natural resources and social sectors division of RSDD, which had one specialist with an urban designation, was disbanded; and a new transport, energy and water division was created in August 2005. Consequently, the USS no longer has strong technical divisions as its natural home nor a central custodian in ADB. 41. In July 2004, ADB revised its sector and theme classification systems for projects, eliminating the macro-sector of social infrastructure and, with it, the urban development and housing sector. Integrated urban development projects were henceforth to be classified as multisector, and mixed with other categories of unrelated multisector projects, including rural projects. 37 Within the new list of seven ADB thematic clusters, there is no urban development cluster, such as the World Bank has (among the 11 themes that it pursues). Instead, there is an urban development theme under the sustainable economic growth theme cluster, and an urban environment theme under the environmental sustainability theme cluster. Under the theme of governance, municipal governance or urban governance was not identified as a subtheme. Rules to distinguish between water supply, sanitation, and waste management (WSSWM) and multisector projects are absent. The new classification, with 27 sometimes competing themes, and allowing 3 per project, makes tracking ADB s urban sector work more difficult, and has led to inconsistent results. 38 42. In spite of the above developments, the number of professional staff officially designated as urban specialists increased after 1999 10 in October 2000, 13 in May 2003, and 16 in April 2005. 39 All urban specialists were posted in the Social Sector Divisions of the regional departments. 40 While there has been a growth in the number of urban specialists, there has been considerable staff turnover. Of the 10 urban specialists in 2000, 4 had left ADB by August 2005; of the remaining 6, none continued to hold an urban designation. The principal urban 37 Housing finance projects were classified under the finance sector, housing construction projects under the industry and trade sector. Water supply and sanitation was joined with waste management (WSSWM) as a new sector, with three subsectors: (i) water supply and sanitation, (ii) waste management, and (iii) integrated. 38 For instance, the Mindanao Urban Services Sector Project (ADB. 2001. Report and Recommendation of the President to the Board of Directors on a Proposed Loan to the Republic of the Philippines for the Mindanao Urban Services Sector Project. Manila), is currently classified under the human development theme. 39 In April 2005, there was one principal urban specialist (in the India Resident Mission), three senior urban development specialists, one senior housing and urban development specialist, one senior water supply and sanitation specialist, six urban development specialists, two urban economists, one housing and urban development specialist, and one urban development and planning specialist. 40 Not all projects in the urban sector are administered by specialists with an urban designation. Many projects are implemented by staff with a more general designation, such as project engineer. ADB does not have, in the operational divisions, a designation water supply specialist or sanitation specialist. In 2005, ADB had six water resources specialists, but these are in the agriculture, environment, and natural resources divisions.

16 specialist assigned to RSDD in 2002 retired in 2005 and has not been replaced, since RSDD s focus is on the water sector rather than the urban sector. The number of urban specialist staff in ADB is not large. By comparison, the World Bank has 48 urban specialists for Asia alone and has various urban theme groups. 43. The number of resident missions (RMs) in DMCs increased from 11 in 1999 41 to 26 by August 2005. 42 This, in turn, led to a substantial increase in the number of national officers involved in the administration of urban sector projects (although without a specific urban designation). Some RMs use TA funds to fund semipermanent consultants in the RMs to supplement their resources, such as the India RM, where the urban portfolio has expanded substantially over the period of the USS. 44. In November 2004, an independent panel to assess the effectiveness of ADB s reorganization concluded that ADB had lost critical mass of technical specialists. 43 It recommended increasing the number of technical specialists; and establishing communities of interest (guilds, committees, and/or panels) to help manage sector knowledge, promote innovation, and maintain professional standards. The advice to create new committees has been adopted, although with little benefit to the urban sector. In September 2005, at the time of drafting this SES, ADB Management approved the establishment of 10 Communities of Practice, to replace the 10 sector and 9 thematic committees instituted along with ADB s reorganization in 2002. An urban sector Community of Practice was not among these; the sector committee on urban development, municipal services, and housing was disbanded. However, a voluntary urban Community of Practice has been created. ADB has started a study of the adequacy of its technical expertise in 2005. 45. Assessment. The view of most urban sector specialists interviewed for this evaluation is that ADB has been unable to maintain or improve an overall focus on the urban sector. The designation of one urban specialist in RSDD after the reorganization was not sufficient to ensure the needed overall perspective. Combining urban sectors with education, health, and some other subsectors in the social sectors divisions has contributed to reducing ADB s urban focus. Directors of such divisions would not necessarily have a background in the urban sector. The creation of an urban sector committee and a network in 2002 was not effective in sharing best practices and stimulating professional debate and feedback among urban specialists. Generally, urban sector profiles and strategies were not updated, although a few reviews continued to be conducted and urban road maps are now routinely being produced as part of the CSP process. For the urban sector, ADB s new organization did not develop natural synergies with government structures in DMCs, many of which have urban or housing ministries, departments, or agencies (Supplementary Appendix C). 46. This experience is not unique to ADB. World Bank experience is particularly instructive in helping ADB to identify a way forward for its urban sector work. The World Bank s Urban and Local Government Sector Strategy 44 states the following: 41 Bangladesh, Cambodia, India, Indonesia, Kazakhstan, Nepal, Pakistan, South Pacific Regional, Sri Lanka, Uzbekistan, and Viet Nam. 42 Afghanistan, Azerbaijan, Bangladesh, Cambodia, India, Indonesia, Kazakhstan, Kyrgyz Republic, Lao People s Republic, Mongolia, Nepal, PRC, Pacific Liaison and Coordination, Pacific Subregional, Pakistan, Papua New Guinea, Philippines, South Pacific Regional, Sri Lanka, Timor-Leste, Tajikistan, Thailand, Uzbekistan, Viet Nam, and some extended missions and other offices (Gujarat and Sumatra). 43 ADB. 2004. Independent Assessment of the Effectiveness of the Reorganization of the Asian Development Bank. Manila. 44 Infrastructure Group Urban Development. 2000. Cities in Transition. The World Bank Urban and Local Government Strategy. Washington DC: World Bank.

17 In the first two decades of urban development lending (1972 92) the (urban) portfolio was rated among the best in the Bank. But the Bank s reorganization in 1987 dispersed the central urban projects staff and left many of their projects in the charge of managers who lacked knowledge of urban issues. The strong body of urban sector work, policy studies, and research received inadequate operational follow-up and was poorly reflected in country assistance strategies. Little new analytical work was initiated after the early 1990s. The result of all these factors was a sharp dip in the performance ratings of completed projects around fiscal 1995. 47. The MTS II, issued March 2006, groups sectors into three categories: (i) sectors where ADB will focus its operations and for which it will build up a critical mass of expertise; (ii) sectors which are not core but which are nevertheless important in enabling ADB to meet the diversity of needs across different countries, and to ensure country focus in ADB operations; and (iii) sectors in which operations should be gradually wound up. Partly as a result of OED feedback from the present study, the urban sector was classified in the first group. 45 If ADB expects to increase its activities in the urban sector, the results of this evaluation suggest that organizational changes will be required to be in harmony with this strategic objective. In deciding on the organizational context for MTS II, the study identified the following options, a combination of which would likely be required: (i) (ii) (iii) (iv) (v) The reformation of the operational urban sector divisions, to lead ADB s work in the urban sector, including in WSSWM, multisector, urban transport, governance, and housing finance sectors, which are currently diluted or dispersed. A matrix organization with macro sector divisions in operations departments and theme divisions in RSDD, one of which is devoted to the urban sector or alternatively local government, and is made custodian of USS. The institution of a critical mass of specialists in RSDD pursuing an urban development or local government theme (this implies the re-designation of some of the current more dispersed specialists or drawing in new specialists or long term consultants). For countries with a large urban portfolio, establishing urban units in RMs such as presently the case in the India RM, mainly consisting of national staff, and to be concerned with both preparation and implementation of urban sector projects. The creation of an ADB thematic cluster on urban development (or urban and local government), and the reinstitution of an interdepartmental theme group, to improve sharing of information and informal coordination within ADB. IV. ADB ASSISTANCE TO THE URBAN SECTOR 48. ADB s assistance takes the form of (i) loan products and associated administration, (ii) grant funded TA products and associated administration, and (iii) knowledge products produced by ADB staff. This chapter considers each component individually, sketches relevant trends and links these to the USS prognosis and targets. 45 Next to road transport, energy, rural infrastructure, education and the financial sector.

18 A. Lending Program in Numbers and Volume 49. ADB has been active in the urban sector since the late 1960s. The first loan in the sector was the fourth loan that ADB approved: the Penang Water Supply Project 46 for Malaysia in 1968. The first loan for a sewage treatment project was in Hong Kong, China, approved 7 years later. The first complex urban development loan was approved soon after, in August 1976, for the Bandung Urban Development and Sanitation Project 47 in Indonesia. Lending for housing started in the following year, for the Sha Tin Urban Development (Housing) Project 48 in Hong Kong, China. After its approval in 1980, the sector lending modality quickly became an important feature of ADB s urban operations, and the first sector loan was for the Small Towns Water Supply Sector Project 49 in Indonesia, approved in December 1980. Housing finance loans were processed mostly in the second half of the 1990s, mainly for India. ADB s involvement in urban transport has been limited and sporadic. ADB has approved five such projects (Shanghai Nanpu Bridge Project 50 in 1991, Shanghai Yangpu Bridge Project 51 in 1992, Bangkok Urban Transport Project 52 in 1992, Sukkur Bridge 53 in 1994, and the Xi An Urban Transport Project 54 in 2003). However, road improvement constituted a sizeable portion of many integrated urban development projects. In general, ADB did not succeed in developing large numbers of publicprivate partnerships or in catalyzing large private sector investments in urban infrastructure. ADB s work with the private sector to provide urban infrastructure remained sporadic confined to some housing finance and a few infrastructure investments in city expressways (Bangkok Second Stage Expressway in 1990 and 1994 and the Manila North Tollway and water supply (Chengdu Water Supply Project and Lyonnaise Viet Nam Water Company). 50. Lending Trends. The overall trend in annual urban sector lending is shown in Figures 2 and 3. Similar to the World Bank, around 10% of ADB s lending was for urban sector projects, both in amount and number of loans. As of the end of June 2005, total lending was $10,215 million, under 201 loans for 173 projects. Lending was mainly for investment projects and sector projects, with three quarters of the aggregate loan amount originating from ordinary capital resources (OCR). 55 Although there have been wide annual variations, there has been an 46 ADB. 1968. Report and Recommendation of the President to the Board of Directors on a Proposed Loan to Malaysia for the Penang Water Supply Project. Manila. 47 ADB. 1976. Report and Recommendation of the President to the Board of Directors on a Proposed Technical Assistance Loan and Grant for the Bandung Urban Development and Sanitation Project and a Proposed Technical Assistance Loan for the South East Sulawesi Transmigration Area Development Project in Indonesia. Manila. 48 ADB. 1977. Report and Recommendation of the President to the Board of Directors on a Proposed Loan to the Government of Hong Kong for the Sha Tin Urban Development (Housing) Project. Manila. 49 ADB. 1980. Report and Recommendation of the President to the Board of Directors on a Proposed Loan to the Republic of Indonesia for the Small Towns Water Supply Sector Project. Manila. 50 ADB. 1991. Report and Recommendation of the President to the Board of Directors on a Proposed Loan and Technical Assistance Grants to the People s Republic of China for the Shanghai Nanpu Bridge Project. Manila. 51 ADB. 1992. Report and Recommendation of the President to the Board of Directors on a Proposed Loan and Technical Assistance Grants to the People s Republic of China for the Shanghai Yangpu Bridge Project. Manila. 52 ADB. 1992. Report and Recommendation of the President to the Board of Directors on a Proposed Loan and Technical Assistance to the Kingdom of Thailand for the Bangkok Urban Transport Project. Manila. 53 ADB. 1994. Report and Recommendation of the President to the Board of Directors on a Proposed Loan and Technical Assistance Grant to the Islamic Republic of Pakistan for the Sukkur Bridge Project. Manila. 54 ADB. 2003. Report and Recommendation of the President to the Board of Directors on a Proposed Loan to the People s Republic of China for the Xi An Urban Transport Project. Manila. 55 It is a challenge to estimate total ADB lending for the urban sector because of the lack of an urban category in ADB s loan classification system. Lending for the urban sector may be larger than this amount because there are other multisector projects which cover water supply and sanitation (WSS) or urban environment components. Some rural WSS projects also cover small towns. Irrigation projects may also cover elements of drinking water supply.

19 increasing trend in urban lending over the past 35 years. However, urban lending has remained constant as a share of total annual lending at about 10%. Figure 2: Lending for the Urban Sector in Constant 2005 Prices 1,200 1,000 $ million 800 600 400 200 0 1968 1971 1973 1975 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 2003 Year of Approval Source: Asian Development Bank s internal databases. 25 Figure 3: Public Sector Lending for the Urban Sector as Proportion of Total Lending, 1968 2004 20 Percentage 15 10 5 0 1968 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 Year of Approval 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 Source: Asian Development Bank s internal databases. 51. Multisector projects accounted for 44% of the total loans for urban sector projects, followed by WSSWM projects at 43% of total lending. The trends in lending by urban subsectors are shown in Figure 4.

20 100 Figure 4: Shares of Different Categories in Urban Sector Lending (1968 2004) Percentage 80 60 40 20 0 1968 1970 1971 1972 1973 1974 1975 1976 1977 1978 1979 1980 1981 1982 1983 1984 1985 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 Year of Approval WSSWM Multisector Housing Transport MS+Housing MS+Housing = multisector combined with housing; WSSWM = water supply, sanitation, and waste management. Source: Asian Development Bank internal databases. 52. Trends by DMC. Thirty-one of 43 DMCs have borrowed for urban sector projects, but most of the lending has gone to only a few countries (e.g., Bangladesh, PRC, India, Indonesia, Pakistan, Philippines, Sri Lanka, Thailand, Viet Nam) (Appendix 5). The PRC and India together accounted for 47.6% of lending to the urban sector between 1993 and 2004. The PRC started borrowing for the urban sector in 1991 and India began in 1995. Bangladesh, Indonesia, and Philippines borrowed for the urban sector from the 1970s onwards, but the volume declined after the 1990s as did total ADB lending to those countries. The forward pipeline for Indonesia includes three urban loans for $210 million; for the Philippines, two loans for $100 million; and for Bangladesh, two loans for $120 million. Pakistan s pipeline is growing; by mid-2005, 6 loans for $400 million were identified. Future urban sector lending for other countries is much smaller, although the significance of modest levels of lending could still be high, given the smaller population of many of these countries. 53. Before and After the USS. When comparing the periods before and after the USS, this study concluded that the USS objective to significantly increase urban sector lending has not been achieved. Urban lending increased only marginally from $3,306 million in the 6 years before the USS to $3,500 million in the 6 years after the USS a rise of 5.9% in nominal terms. Average annual urban sector lending was $550 million before the USS and $582 million afterward. From mid-1993 to mid-1999, urban sector lending accounted for 9.0% of total lending; afterward, until mid-2005, this proportion increased marginally to 11.5%. The increase is, however, somewhat misleading, as it is influenced by a surge in lending in 1997 1998 related to the Asian financial crisis. If the large policy loans provided to Indonesia, Korea, and Thailand, totaling $6.2 billion, are excluded, urban sector lending was around 11% in the period 1993 1999, about the same level as post-uss. The greatest expansion in urban sector lending volume was achieved in multisector projects (22.6%), principally in India. 56 The other sectors, such as housing and environment, increased by 19%. WSSWM lending, however, fell by 56 This excludes the Gujarat Earthquake Rehabilitation Project (ADB. 2001. Report and Recommendation of the President to Board of Directors on a Proposed Loan to India for the Gujarat Earthquake Rehabilitation Project. Manila), which includes a sizeable urban sector rehabilitation component.

21 12.6%. 57 The growth in housing finance loans may not be sustainable. ADB rates in the 1990s and early 2000s were seen as advantageous by financial intermediaries. Now, however, the margins are insufficient for such loans to be profitable for financial intermediaries in OCR countries. 54. The volume of urban sector lending may be about to change (Figure 5). The forward pipeline includes 40 new projects and indicates a growth potential. It is expected that aggregate urban lending will be in the order of $3.4 billion in the next 3 years or over $1 billion per year, almost double previous urban lending. If this pipeline materializes, it indicates that, consistent with the USS, there is a growing appetite for urban lending in DMCs. However, some caution is needed in assessing the forward pipeline. Several urban sector specialists recalled that, after the USS, the pipeline portfolio looked good but that it did not materialize as anticipated. Figure 5: Urban Lending Before and After the Urban Sector Strategy, and Pipeline, by Main Category $ million 3,500 3,000 2,500 2,000 1,500 1,000 500 0 681 810 470 743 1,496 1,308 2,174 1,129 1,382 July 1993 June 1998 July 1999 June 2005 Pipeline 2005 2007 Multisector WSSWM Other urban WSSWM = water supply, sanitation, and waste management; Other urban = mainly housing and urban environmental projects. a Note: Standby pipeline loans have been excluded. a This study reclassified three projects: Loan 2026-SAM, from multisector to WSSWM, Loan 1663/65-PHI from multisector to other urban, and Loan 1560-MON from WSSWM to multisector. Source: Asian Development Bank internal databases. 55. The pattern of past and expected future urban lending for all countries that borrowed more than $100 million from mid-1993 to mid-2005 is shown in Figure 6. Loans to two of ADB s largest clients, India and the PRC, dominate urban lending. Urban lending to both countries increased substantially after 1999 and there are strong pipelines for continued urban lending. ADB is engaged in addressing major urban challenges, especially in the larger cities in the two most populous countries in the world. The future urban pipeline in Pakistan is growing. Pakistan s urban lending pipeline may be larger than that indicated in Figure 6, as a result of recent project identifications in Karachi, a megacity. Although ADB s urban operations were significant in Indonesia and the Philippines between 1993 and 1999, lending fell thereafter and the future urban pipeline remains modest. Both countries substantially reduced their total 57 Loans for the other sectors would have been higher had it not been for two housing loans to India which were approved by the Board but cancelled before signing (ADB. 2000. Report and Recommendation of the President to the Board of Directors on Four Proposed Loans to the Housing and Urban Development Corporation, National Housing Bank, Housing Development Finance Corporation and ICICI for the Housing Finance II Project in India. Manila.

22 borrowing from ADB after 1999, experienced some political uncertainty, and pursued a decentralization strategy. Figure 6: Patterns of Urban Sector Lending in Selected Developing Member Countries $ million 1,400 1,200 1,000 800 600 400 200 0 IND PRC PHI INO VIE THA PAK SRI 2005 2007 July 1999 July 2005 July 1993 June 1999 BAN NEP LAO Other Others s July 1993 June 1999 July 1999 July 2005 2005 2007 BAN = Bangladesh, IND = India, INO = Indonesia, LAO = Leo People s Democratic Republic, NEP = Nepal, PAK = Pakistan, PHI = Philippines, PRC = People s Republic of China, SRI = Sri Lanka, THA = Thailand. Note: Projection 2005 2007 as of March 2005. Sources: Operations Evaluation Department data and Asian Development Bank Project Preparation Information System. 56. Two conclusions stand out. (i) Lending for the urban sector has increased over the past 35 years in real terms, but enormous funding gaps remain. Lending increased by a marginal 6% during 1999 2005, compared to the preceding period. This increase is modest compared to 15% growth in Asia s urban population over the same period. (ii) USS qualitative lending target to significantly increase urban lending was not achieved. The larger future pipeline could indicate that there is a growing appetite for urban sector lending, particularly in the PRC, India, and Pakistan, if ADB can respond effectively and efficiently. B. Project Developments After the Urban Sector Strategy of 1999 57. The USS proposed not only to increase the volume of lending, but also to expand (i) the range of subsectors to be addressed, (ii) interrelationships of project components, (iii) number of policy-based loans, (iv) cofinancing opportunities, (v) support for private sector investments, (vi) ADB s role in institutional development and capacity development, (vii) cooperation with civil society organizations, (viii) community participation, and (ix) gender development activities. To assess actual developments against these expectations, pre- and post-uss patterns were analyzed, without a rigid focus on attribution. In the 6 years before the USS, 45 urban projects were approved, and 43 were approved in the 6 years after the USS. All figures below are drawn from statistical analysis of the reports and recommendations of the President (RRPs) approved from July 1993 to the end of June 2005 (Supplementary Appendix D).

23 58. Shift to More Multisector Projects and Other Urban Projects? There is little evidence of any significant change in the urban portfolio attributable to the USS. Figures 4 and 5 suggest a de facto move in urban lending away from WSSWM projects and towards larger lending volume for integrated urban development/multisector projects. The project designs often incorporated WSSWM combined with other related investments (e.g., storm drainage or urban roads). In terms of quantities of ADB-supported projects, however, these trends are less marked. The number of multisector projects fell from 17 before the USS to 16 afterward; WSSWM projects declined from 24 to 21. The number of other urban projects increased from 4 to 6. A number of housing loans and urban environmental loans were included in this category. The USS target to expand urban transport operations was not achieved: there was only one urban transport project post-uss. The average number of urban transport components in projects declined. In the area of land management, there was only one small cadastral survey and land registration project in Mongolia. 58 59. When analyzing components within urban sector projects, there was a decline in the number of projects with water supply components after the USS and an increase in the number of projects with sewerage components (Figure 7). The number of components related to urban shelter (slum upgrading, housing, and land development), land management and regulation, and urban finance increased after the USS. Urban renewal Slaughterhouses Street lights Bus terminals Air pollution control Market development Land admin/tenure/regulation Public housing Housing finance Slum development Other WSS/environment Municipal taxation Urban finance Toilet/latrines Policy development Training, workshops Urban transport Environmental management Tariff and user charges work Solid waste management Public awareness campaigns Drainage/flood control Sewerage Water supply Institutional/capacity development Figure 7: Percentage of Projects with Urban Sector Components, Before and After the Urban Sector Strategy 1993 1999 1999 2005 0 10 20 30 40 50 60 70 80 90 % of projects Source: Review of 88 reports and recommendations of the President approved between July 1993 and June 2005. 58 ADB. 2000. Report and Recommendation of the President to the Board of Directors on a Proposed Loan to Mongolia for the Cadastral Survey and Land Registration Project. Manila.

24 60. More Projects with Interrelated Subcomponents? The USS implied that the number of different subcomponents included in the projects would increase. This has not happened and urban projects have generally become less complex when measured by the number of components. The number of components went down by 12%, from an average of 12.4 before the USS to 10.9 after. The pattern held for multisector projects (from 17.4 to 14.1) and other urban projects (from 5.0 to 4.2). WSSWM remained similar in terms of complexity, with 10.1 subcomponents before the USS and 10.4 afterward. WSSWM was often found to include storm drainage components but generally do not include roadworks, market development, bus terminals, and municipal tax advice. 61. The proportion of urban sector projects with water supply components (covering new, extended, and rehabilitated networks and water treatment plants) fell from 71% before the USS to 54% after. The share of urban sector projects with one or more sewerage and sewage treatment components rose from 42% to 49% (most water supply projects now include sewerage components). The share of projects with components supporting construction of public toilets and private toilets and/or latrines fell from 31% before the USS to 19% after. 62. The proportion of projects with urban transport subcomponents declined from 40% before the USS to 26% after while the proportion with drainage subcomponents remained constant at 49% of projects. Urban road subcomponents are usually successful. The procurement process and the extra supervision that ADB support induces often make their quality better than other urban roadworks. The wisdom of reducing road subcomponents could be questioned. However, this study has not fully evaluated this issue and the reasons for the decline in the number of urban road components. 63. The prevalence of housing finance components doubled from 7% to 14% of projects. Two loans to India helped set up an urban finance facility. However, overall, the housing finance outcome was lower than the USS intended. Urban sector projects supported by ADB seldom included components to support bond market development, microcredit, land tenure, land development, crime prevention, and safety. Overall, there is no evidence that the interrelationships of project components have grown post-uss. However, there seems to be an appropriate trend of more focus within projects on either public health or public space as a nucleus, and a modest shift away from the Christmas tree approach of multiple subcomponents in different sectors outside the context of public health. 64. Overall, the evaluation concludes that the trend toward fewer components was appropriate. Lessons have been learned that trying to do too much and including too many disparate components in the design often results in projects that are not always rated satisfactory after project completion. 65. Loan consulting services and attached TA components for institutional development and capacity development were included in most projects before the USS (82%) and continued to be so afterward (84%). Some of this assistance was designed to strengthen project implementation units (PIUs) 59 and part was focused on broader policy/institutional issues. From cost estimates available, the size of the components remained similar. This suggests that the objective of increasing ADB s focus on strengthening institutional capacity was not achieved. 66. More Work on Policy Reforms such as Tariffs, User Chargers, and Municipal Taxes? The analysis of the RRPs approved over the period shows that ADB planned more 59 ADB. 2005. Special Evaluation Study on the Role of Project Implementation Units. Manila.

25 work on water or sewerage tariffs and collection efficiency after the USS. The percentage of WSSWM projects with clearly identified work on tariffs/user charges almost doubled from 29% before the USS to 57% after. There was not a significant corresponding increase in such a component in integrated urban projects. The proportion of projects, including work on municipal taxes such as municipal property tax and road tax, fell from 20% of projects before the USS to 14% afterward. In many DMCs, preparatory work on legislation on property tax and municipal taxes has now been completed and new systems are being implemented. 67. Greater Variety in Lending Modalities? The USS called for the use of a greater variety of lending modalities for the urban sector, including program and sector lending modalities. Policy-based lending has remained limited in the urban sector. There was only one such loan before the USS (Metro Manila Air Quality Improvement Program) 60 and two after (Kathmandu Valley Water Services Sector Development Program 61 and Pasig River Environmental Management and Rehabilitation Sector Development Program 62 ). The number of sector projects doubled, from 5 before the USS to 10 after, but the associated loan volumes declined from $510 million to $352 million. The number of project loans fell from 36 to 27, but loan volumes increased from $2,197 million to $2,604 million. The number of urban development finance institution projects remained limited. While there was a modest increase in the variety of lending modalities, the objective of increased urban policy-based lending, and by extension greater ADB impact on urban policy reform, was not achieved. The reasons may be various. The two largest urban borrowers, the PRC and India, have shown little interest in program lending. Also, experiences with the three programs in the urban sector have been mixed. 68. More Projects Cofinanced with Other External Agencies? Progress was made in increasing cofinancing in the urban sector. The number of projects cofinanced by external agencies increased slightly from 8 (19%) to 11 (25%) between the two periods. The amount of cofinancing rose more significantly, however, from $50.6 million to $330.7 million. 69. Greater Involvement of Other Financing Sources? The USS aimed to mobilize more funds other than ADB and government sources. The evaluation assessed progress in achieving this objective by determining whether the share of ADB loans in total project financing fell. The financing of projects before and after the USS is shown in Figure 8. The share of project costs financed by ADB remained constant 51% before USS, 52% after, despite some more cofinancing with external agencies which implies that the objective was not achieved. However, there was a change in the mix of financing sources. The proportion of central government financing declined from 18% to 13%. Local government financing increased from 9% to 15%, state-owned banks from 1% to 7%, commercial banks from 1% to 3%, and other external agencies from 2% to 5%. However, the proportion of financing provided by water utilities, financing corporations, and direct beneficiaries of projects declined in importance from 23% of the total project finance before 1999 to 7% after. These changes in financing patterns may reflect increasing decentralization in many countries, including decisions to make local governments responsible for financing urban development. While this increases accountability and transparency, the fiscal position of many municipalities is weak. Financing urban sector investments is a major challenge in most DMCs. 60 ADB. 1998. Report and Recommendation of the President to the Board of Directors on the Proposed Loans and a Technical Assistance Grant to the Republic of the Philippines and the Land Bank of the Philippines for the Metro Manila Air Quality Improvement Sector Development Program. Manila. 61 ADB. 2003. Report and Recommendation of the President to the Board of Directors on the Proposed Loans to the Kingdom of Nepal for the Kathmandu Valley Water Services Sector Development Program. Manila. 62 ADB. 2000. Report and Recommendation of the President to the Board of Directors on the Proposed Loans and a Technical Assistance Grant to the Republic of the Philippines for the Pasig River Environmental Management and Rehabilitation Sector Development Program. Manila.

26 Figure 8: Financing of Projects Before and After the Urban Sector Strategy $ million 4,000 3,000 2,000 1,000 0 3,500 849 1,048 447 3,306 213 80 226 1,187 552 94 23 ADB loan 0 106 Local government Commercial banks IFIs Bilateral agencies Central government State owned banks Mid-1993 to Mid-1999 Mid-1999 to Mid-2005 ADB = Asian Development Bank, IFIs = international financial corporations. Source: Review of 88 reports and recommendations of the President approved between July 1993 and June 2005. 70. More Private Sector Operations? The USS called for more support for private sector urban development projects using loans without government guarantee and/or equity investments. Inadequate progress has been made in this area. Only 5 loans were for private sector urban projects (2 for housing or home mortgage finance, 2 for water companies, and 1 for an urban expressway company). In addition, some other operations, such as assistance for the local government unit (LGU) guarantee corporation in the Philippines and an equity investment in the China Environment Fund, should have an overall beneficial impact on private participation in the urban sector. An example of a public-private sector partnership is the Chengdu Water Supply Project in the PRC (Box 2). Box 2: Example of a Public Private Partnership The Chengdu Water Supply Project is a build-operate-transfer (BOT) scheme to address the need for treated water by Chengdu, the capital of Sichuan Province. The project entails the construction of a treatment plant and the laying of a 27-kilometer transmission pipeline to deliver water to the city s homes and industries. With a population of more than 10 million, Chengdu is one of the important industrial and scientific research bases in the southwest People s Republic of China (PRC). The Asian Development Bank (ADB) extended a loan of $26.5 million and helped in structuring finance for the $107.6 million project. Through complementary financing, ADB also made possible commercial debt funding of $21.5 million. Traditionally funded by the government on a grant basis, water supply projects were opened up to private investment because of the large investments needed. As the first BOT water supply project in the PRC, the project demonstrated that the BOT scheme can be successfully implemented at the municipal level and externally funded without any central government guarantee. ADB s involvement ensured that transparent competitive bidding was observed, the lowest possible tariff was obtained, and project fundamentals were in place. The project won three international awards in recognition of its innovative financing structure, which enabled international lenders to take long-term municipal risk in the PRC for the first time. Source: Adapted from ADB s website. Available at http://www.adb.org/privatesector/finance/chengdu.pdf 71. Growth in urban private sector operations has been modest, during a time when ADB s private sector operations overall have expanded considerably. There has been no lending to municipalities without government guarantees. ADB is currently considering how sub-sovereign and non-sovereign lending to the many semi-government agencies is to be handled. Overall,

27 the public and private sector parts of ADB have not worked together to promote private sector investment in urban infrastructure. Given the objectives of the USS, the OEM expected to find more evidence of public private partnerships, and the public side of ADB undertaking more policy dialogue to create an enabling environment for the private sector. Given the difficult state of municipal finances in many countries, increased private sector participation holds the potential to strengthen urban infrastructure and provide urban services when the public sector cannot. There have also been a number of cases where private sector investments in urban infrastructure have experienced difficulty and some contracts have not been honored. This suggests that there are opportunities to use ADB s credit enhancement and risk mitigation tools, which have not been fully exploited. 72. More EAs at the Subnational Level? As foreseen by the USS, a shift occurred post- USS to reduce the predominance of national government ministries and departments in urban projects. Prior to the USS, two-thirds of urban projects involved national governments or national authorities. The proportion fell to half after 1999 (Table 1). The shift to greater involvement of local agencies versus national governments implies a greater opportunity for creating more sustainable results, as national government projects usually involve project management units (PMUs) and PIUs that are dissolved upon project completion. This is less likely to be the case when local governments and their authorities and corporations run the project. Even if there are PIUs in such cases, these will generally be composed of staff that later operate and maintain the assets created. The expansion of the urban sector programs in India (predominantly with state governments and delegated agencies) and the PRC (local agencies created for the project) contributed to this shift. Table 1: Executing Agencies of Projects Approved Before and After Urban Sector Strategy (% of projects) 1993 1999 1999 2005 National Government 62.2 41.9 National Agency 4.4 9.3 Provincial Government 6.7 16.3 Provincial Agency 6.7 0.0 Local Government 2.2 14.0 Local Agency 15.6 7.0 Development Finance Institution 2.2 11.6 Source: Review of 88 reports and recommendations of the President approved between July 1993 and June 2005. 73. Shift in Type of Urban Area Addressed? The USS did not address the issue of the size of municipalities to be covered by ADB s urban operations. There are four types of urban projects: (i) multiple towns or cities; (ii) single cities with populations above 1 million; (iii) single cities with population below 1 million; and (iv) beneficiaries of whom the location cannot be easily specified (e.g., housing finance). Over the period reviewed, 49 of the 88 projects covered multiple towns, 10 projects were for single cities with a population below 1 million, 24 projects were for cities with populations above 1 million, and 5 were finance projects. Figure 9 demonstrates that little changes occurred in the type of urban areas addressed. About 180 million people live in municipalities involved in ADB-funded urban projects (44 multi-town projects covered 57 million people, 9 small cities covered 1.7 million, 19 projects for large cities covered 120 million, and 2 finance projects covered 1.4 million).

28 Figure 9: Type of Urban Area Addressed Before and After Urban Sector Strategy ($ million) 2,000 1,500 1,464 1,538 1,439 1,501 $ million 1,000 500 0 300 365 103 95 Multi-town Single city < 1 million Single city > 1 million Finance July 1993 June 1998 July 1999 June 2005 Source: Review of 88 reports and recommendations of the President approved between July 1993 and June 2005. 74. Larger Populations being Targeted, and More Poor Households? The USS had no objective related to the size of municipality to be targeted or the incidence of poverty. Post-USS projects did not indicate that larger population centers were being supported. Before the USS, 45 projects covered cities with an aggregate urban population of 102 million. The 43 post-uss projects covered an urban population of 63 million. These figures indicate the reach of ADB support. The total Asian urban population was about 1.35 billion in 2000. This means that about 12% of Asia s urban population lives in cities that have benefited from an ADB financed urban project. For 27 projects approved before the USS, on average, a 33% poverty incidence was estimated; the corresponding figure for 35 post-uss projects was 25%. It is difficult to draw firm conclusions about the apparent decline of coverage of poor households. It may be due to increased urban sector projects in the PRC where the incidence of poverty is relatively low, and growing incomes over a 6-year period. 75. Longer Relationships with Cities? There is little evidence that the USS objective of building long-term relationships with selected cities was achieved. ADB has ended its urban sector relationships with cities such as Shanghai and Beijing because the PRC country strategy placed priority on supporting projects in poorer interior provinces rather than wealthier cities along the east coast. In India, when an urban project is finished, the next project may be in the same state, but in a different city or town. However, this is partly a consequence of government decisions. 76. More Community and Civil Society Participation? In line with the USS, there were more attempts after 1999 to involve communities, NGOs, and community-based organizations (CBOs) in projects in a participatory manner. Half of the projects foresaw the participation of NGOs, CBOs, and other elements of civil society in project activities; before the USS, the corresponding figure was 30%. Multisector projects tended to have more NGO and CBO involvement.

29 77. More Attention for Resettlement and Land Issues? Given the intended increasing orientation towards environmental issues, poverty reduction, and comprehensive urban sector projects outlined in the USS, the number of RRPs identifying resettlement issues and formulating resettlement plans (RPs) was expected to rise after mid-1999. As a matter of ADB policy, the preparation of RPs became mandatory after 1995 in all cases where resettlement was anticipated. After the USS, in 31 RRPs, one or more RPs were prepared (72% of projects), although only 24 RRPs raised a resettlement issue in their main text; before the USS, the number was 13 (29%). Land and right-of-way issues became more prevalent, perhaps indicating the increasingly critical nature of such issues as urbanization grew. Before the USS, almost no private land had to be acquired for ADB-supported projects; after the USS, in 19% of urban sector projects, more than 100 hectares had to be acquired. The increased focus on resettlement after the USS reflects both the increasing attention given to ADB safeguard policies and the changing design of urban projects. 78. More Attention for Gender Issues? Before the USS, 51% of RRPs included a discussion of gender considerations in project design; afterward, the proportion rose to 77% surely related to the approval of the gender and development policy in June 1998. In many cases, the nature of the considerations remained at the level of identification of project effects on gender issues. However, a targeted approach was increasingly adopted in project design. Gender action plans identified special measures to ensure women s participation in the project. Before the USS, such plans were elaborated in 24% of RRPs; this rose to 52% post-uss. 79. Attention for Corruption in RRPs? The USS stated that specific anticorruption measures should be put in place and enforced to promote transparency and accountability. After the USS, 30% of urban sector RRPs reflected that ADB had discussed its anticorruption policy with local officials; 24% highlighted specific anticorruption measures or capacity building. In response to more attention to the negative effects of corruption measures, a mandatory paragraph on anticorruption policy was added to the RRP in late 2005. It remains to be seen whether this becomes largely a format issue or actually results in the inclusion of measures to support anticorruption initiatives or measures to mitigate the risk that ADB-funded urban development projects will be infected by corruption. Corruption is widely acknowledged to be a major threat to the efficiency of both projects and service provision the urban sector in Asia, as shown by the material in Boxes 3 and 4. The challenge going forward is for ADB to more explicitly identify potential corruption risks in urban projects and then to include mitigation measures in the project design and monitor the effectiveness of these measures during implementation. Box 3: Corruption and Irregularities in Projects in the Urban Water Supply Sector Many examples of corruption and irregularities are elaborated in a book published by ADB: Asian Water Supplies: Reaching the Urban Poor: a Collusion among bidders. Contractors work together or with politicians to win projects with their local service providers on favorable terms. In some cases, contractors take turns to win bids by agreeing on an amount which has already been padded to incorporate the cost of bribes to other bidders. Favoritism in appointing contractors. This can include favoritism in pre-qualifying contractors and consultants using inappropriate techniques when appointing technical evaluation committees and releasing of documents and estimates to contractors prior to bidding.

30 Kickbacks to executing agencies. Asian Water Supplies (footnote a) estimates the amount in kickbacks in the sector at 10% 15% of the contract price in the procurement of goods, 25% for civil works, 20% for engineering design, and 12% 15% for technical assistance. It was, however, not clear which countries were being referred to. b Other forms of kickbacks can include: sponsoring foreign tours; covering the costs of educating children in developed countries; purchasing luxury homes and vehicles; entertainment costs for government officials during study tours and inspection visits. Payments are usually made out of the mobilization or advance payment received from the development agency. Collusion with contractors. This includes extending contracts for unnecessary work, protecting unfit contractors, complicity of agency staff in the use of substandard materials; approvals of falsified materials invoices, use of construction shortcuts in the field, drawing up specifications to favor certain suppliers. Extorting funds from contractors by executing agencies. Harassing contractors to receive bribes by delaying approvals; direct negotiation with contractors over claims; delaying payment on contractor invoices until bribes are paid and terminating and suspending contracts to exact revenge. a b McIntosh, Arthur C. 2003. Asian Water Supplies: Reaching the Urban Poor. Asian Development Bank and International Water Association. Manila. See also: Davis, Jennifer. 2003. Corruption in Public Service Delivery: Experience from South Asia s Water and Sanitation Sector. World Development. Vol. 32, No. 1. Box 4: Corruption and Theft in Water Supply Service Provision Bribery needed to obtain a water supply connection. A recent survey in India a found that 12% of the customers of a water utility had made payments to expedite new water or sanitation connections in the previous six months. Studies in the Philippines corroborated in 1996 that illegal connections can be obtained from the water supply concessionaire at about ten times the price of a legal connection. b The Philippine Center for Investigative Journalism (footnote b) reported several cases of bribery in the sector. Employees of the former Metropolitan Waterworks and Sewerage System have been reported to accept bribes to install or accept illegal connections or to settle arrears. Regarded as common knowledge is the alaga system where water thieves give corrupt officials and employees regular bribe money to ignore their offenses. Executing agency personnel laying pipes that pass through depressed communities would offer to install illegal connections, at high prices. c An article in World Development reports that the same scenarios apply in India. d Water sellers defrauding customers. According to the Philippine Center for Investigative Journalism (PCIJ) (footnote b), some water sellers in the Philippines turn off water valves resulting in no-water situations in some areas, in order to force ordinary consumers to buy water for as high as thirteen times the amount charged by the two main concessionaires (i.e., Maynilad and Manila Water) in Metro Manila. Theft/illegal connections by water users. IIlegal connections are widespread in many Asian cities. For instance, PCIJ reported that in 1997, Manila Water discovered 11,000 illegal connections in its service area and that in 1999, Maynilad disconnected 700 large and small illegal connections. Several big companies were found to have illegal water supply connections. Collusion executing agency staff and customers. The survey in India (footnote a) found that 41% of the customers had paid small bribes in the previous six months to falsify meter readings and thereby lower bills. Thirty percent had paid bribes to expedite repair work. Unaccounted for Sale of Water by Executing Agencies to Water Vendors. ADB s Asian Water Supply publication e ventured that there may be a link between low piped water coverage in some areas and high levels of non-revenue water (NRW), especially since the latter is often attributed to illegal connections and leakage/technical problems. The book deems it possible that those with vested interests allow the illegal sale of utility water to small scale water providers at high prices (low volumes). The illegal sale is part of the NRW. The total revenue obtained from this may amount to the same earnings from official sale

31 of water by utilities. It is possible that the high revenue generated from the illegal sales to SSWPs is the reason for the long-standing status quo (the urban poor not receiving service). a Corruption Hinders Potable Water Supply. 2006. ABC News Online. Available: http://www.abc.net.au/news/ newsitems/200603/s1589048.htm b Rimban, Luz. 2000. Big Companies Steal Water. Philippine Center for Investigative Journalism (PCIJ). Manila. Available: http://www.pcij.org/stories/2000/water.html c PCIJ. 2000. Water Thieves Operate with Impunity. Manila. Available: http://www.pcij. org/stories/2000/water2.html d See also: Davis, Jennifer. 2003. Corruption in Public Service Delivery: Experience from South Asia s Water and Sanitation Sector. World Development. Vol. 32, No. 1. e McIntosh, Arthur C. 2003. Asian Water Supplies: Reaching the Urban Poor. Asian Development Bank and International Water Association. Manila. C. Capacity Development in Urban Sector Projects 80. ADB approves many infrastructure loan proposals on the strength of their added value in terms of policy reform initiatives and capacity development impacts. While this study did not concentrate on measuring actual impacts on the ground, the SES used a number of input indicators as proxies for institutional and capacity development. A questionnaire survey of projects provided perceptions of clients on capacity development. This section poses the question: were more institutional and capacity development activities included in project design since the USS? This section also investigates the use of (i) foreign and domestic project management consultants, (ii) design and supervision engineering consultants, and (iii) use of externally driven and temporary PIUs. 63 81. Were more Advisory TAs Approved for Project-Related Capacity Development since the USS? The 45 projects approved in the 6 years before the USS had 17 attached TAs. 64 In the 6 years after the USS was adopted, the 43 approved projects had 11 attached TAs, although some stand-alone TAs also addressed project-related capacity. The average size of the attached TA approved was $0.8 million both before and after the USS. The use of domestic consultants in TAs increased after the USS from an average of 18 to 44 personmonths. The number of person-months of international consultants declined slightly from 21 to 18 person-months. Contributing factors may include an increasing recognition of the capacity of domestic consultants, the desire of EAs to work more closely with local consultants, limited TA budgets, and a perception that expensive international consultants do not always deliver value for money. 82. Were Greater Amounts of Consulting Services included under Loans? There is no evidence to indicate that more of the loan proceeds are being allocated to consulting. This is in spite of the fact that international consulting fees have gone up significantly over the years. Judging from the cost estimates in RRPs, the proportion allocated to consulting services declined slightly from 3.2% to 2.4%. Similarly, loan proportions for institutional strengthening or capacity building fell from 3% to 1%. However, the varying definitions of line item costs in RRPs introduce some uncertainty into these estimates. In addition, cofinancing by bilateral and multilateral agencies increased, partly for capacity development. 83. Were More Consultants Used in Urban Sector Projects? The evidence indicates that almost all projects involved consultants in varying degrees. The average number of personmonths per million dollars of ADB lending remained constant before and after the USS across project types, fourteen person-months for integrated urban development projects and 8 to 63 The base tables for this analysis are in Supplementary Appendix D. 64 Approved concurrent with the related loan.

32 9 person-months for WSSWM. Multisector projects were most consultant-intensive per million dollars of ADB loan. 84. Were the Consultants Providing More Operational or Advisory Services? Operational support accounted for two-thirds of consultant time, and advisory and training services made up one-third. In multisector projects, person-months for advisory services rose significantly from 2 to 5 person-months per million dollars of loan, whereas the ratio for operational consultants decreased from 12 to 9. This is in line with USS and with more recent ADB targets in capacity development, which call for more effort to reduce the reliance on operational consultants in favor of more involvement of the agencies own staff, on the assumption that learning-by-doing will engender more effective capacity development, and will reduce capacity erosion through substitution of agency staff. 85. Were More Domestic Consultants Used under Loan Financing? The number of domestic consultant person-months financed under the loans rose by a quarter after the USS, from 785 to 982 person-months per project, while the input of international consultants remained stable at around 95 person-months per project. 86. Were More Consultants Involved in Capacity Development? Figure 10 reflects the types and average number of consulting person-months used per project before and after USS. The use of consultants for training and community programs remained more or less the same; consultant use for other services, mainly advisory, increased. Prior to the USS, half of the urban sector projects had capacity development and community development components; this proportion rose slightly, to 55%, after the USS. Figure 10: Use of Consulting Services in Urban Sector Projects Before and After the Urban Sector Strategy Person-months per project 406 450 400 350 288 250 300 250 200 106 94 92 103 150 44 100 50 0 Project Management Training/Community Program Design and Supervision Engineer Other consultancies Types of Services Mid-1993 to Mid-1999 Mid-1999 to Mid-2005 Source: Review of 88 reports and recommendations of the President approved between July 1993 and June 2005. 87. Conclusion. The findings do not all point in the same direction, indicating mixed results. The predominance of design and supervision engineers fell. Project management consultants have become more prevalent, indicating both greater availability of domestic expertise and increasing complexity of ADB-supported projects. The use of consultants to support projects is

33 still widespread, partly because ADB has not been successful in developing long-term partnerships with city clients. Thus, capacity needs to be built for every project. D. Technical Assistance Program 88. Comparing TA amounts provided before and after the USS, the downward trend in ADB s grant-funded TA program in the urban sector is marked: the TA amount approved after the USS ($37.2 million) is 37% lower than the amount approved before ($59.2 million). This compares with 29% growth in the amount of TA for other sectors over the same period. The number of urban sector TAs almost halved from 112 to 65, whereas the number of TAs in all other sectors increased by 12% from 1,235 to 1,386. The amount of urban TA is shown in Figure 11. The amount of urban TA for both the Philippines and the PRC fell, as it did for many other countries. In contrast to the general pattern, the amount of urban TA increased for India. This is consistent with the large expansion of the urban lending program in that country. Since much of ADB s policy dialogue and capacity building is financed by TA, the declining trend in declining TAs implies that, despite the intentions outlined in the USS, ADB has been less active in these strategic areas than before the USS. Figure 11: Amount of Technical Assistance Approved for Urban Sector, 6 Years Before and After the Urban Sector Strategy 10.5 5.1 4.2 17.4 Jul 1999 Jul 2005 Period Jul 1993 Jun 1999 6.0 11.1 5.9 36.2 10.0 20.0 30.0 40.0 50.0 60.0 Amount ($ million) India TAs PRC TAs PHI TAs Other TAs PHI = Philippines, PRC = People s Republic of China, TA = technical assistance. Source: Asian Development Bank internal databases. 89. The downward trend in TA funds assigned to the urban sector affected both project preparatory TA (PPTA) and advisory TA (ADTA), but particularly PPTA, which fell by 45% (Figure 12). ADTAs also declined significantly, by 27%. From this perspective, it is surprising that project and program lending expanded modestly after 1999, despite the sharp fall in PPTA. However, there may be a delayed impact: some of the post-uss lending was in fact prepared by PPTA that was approved before the USS. 65 If there is no adverse impact on project quality, 65 Before the USS, 58 PPTAs at a value of $31.65 million were used to prepare 42 projects financed by 46 loans totaling $2.8 billion (1.3 PPTAs per loan and $11,000 in PPTA per million dollars of loan). In the last 6 years, 30 PPTAs at a value of $19.04 million were used to prepare 46 projects financed by 54 loans totaling $4.5 billion (0.65 PPTAs per loan and $4,200 in PPTA per million dollars of loan). The projects approved after 1999 are still under implementation, so it is too early to draw conclusions on whether the cutback in PPTA will have an impact on project quality.

34 these figures would suggest that ADB s urban sector processing became more efficient and less resource-intensive after the USS. Given the concerns raised by many MICs about ADB s transaction costs and cumbersome procedures, such a finding would be unexpected. More PPTA funds might have enabled preparation of more projects over the period. Figure 12: Amounts for Technical Assistance, Before and After the Urban Sector Strategy Period July 1999 June 2005 July 1993 June 1999 18.8 18.4 25.1 34.2 10. 0 20.0 30.0 40.0 Urban ADTA Amount ($ million) Urban PPTA ADTA = advisory technical assistance, PPTA = project preparatory technical assistance. Source: Asian Development Bank internal databases. 90. Urban PPTAs accounted for 15% of all PPTAs in the 6 years prior to the USS compared with only 6% in the last 6 years. Urban-related ADTAs fell from 6% to 4% of total ADTAs, in terms of numbers of ADTAs. Similar patterns emerge when financial allocations are considered. 66 One more positive development has been ADB s continuing membership of the Cities Alliance, which allows ADB to access TA funds from alternative sources (Box 5). Box 5: ADB s Membership of the Cities Alliance The Cities Alliance is a coalition of cities and their development partners committed to scaling up successful approaches to poverty reduction, and brings cities together in direct dialogue with donor agencies. A main function of the partnership is to promote coherence of effort among donors, bringing them around shared strategy that defines a vision for the city and its economic growth, while highlighting clear priorities for actions and investment. ADB was the second international financing institution to join this growing partnership, which to date has provided $60 million in grants for City Development Strategies and Cities without Slums activities. The Cities Alliance has leveraged some $6 billion in urban sector investment globally. ADB has so far developed two regional technical assistance projects to support Cities Alliance initiatives in India, Philippines, and Viet Nam. Several stand-alone Cities Alliance proposals are currently being pursued. ADB = Asian Development Bank. Source: Special Evaluation Study on Urban Sector Strategy and Operations. 91. Urban Sector Strategy Studies. The amount of ADTA resources devoted to updating urban sector profiles, urban country strategies, and urban country reviews declined in the last 6 years. This is contrary to the intentions of the USS. Of the 84 urban-related ADTAs approved since 1993, 15 funded urban sector or subsector studies, profiles, or strategies. Since ADB s 2002 reorganization, only a few such studies have been prepared. While 10 such documents were produced between 1999 and 2001, only 3 have been prepared in the 4 years since. 66 Urban PPTAs fell from 17.3% of the total amount of PPTAs to 7.3%. Urban ADTAs fell from 5.3% of the total amount of ADTAs to 3.3%.

35 92. Japan Fund for Poverty Reduction. Since 2000, the Japan Fund for Poverty Reduction (JFPR) has supported ADB-administered projects targeting poverty reduction. JFPR funded 65 projects up to the end of September 2005, an average of $2.5 million in grant funds per project. Eleven JFPR projects were in the urban sector, involving $17.5 million or 10% of total JFPR commitments (Appendix 3, Table A3.4). There are 3 urban sector JFPR projects in the 2005 and 2006 pipeline. E. ADB s Knowledge Management in the Urban Sector 93. In addition to project lending and grant funding of TA, ADB staff produce and disseminate research documents and publications (e.g., books, handbooks, reports, working papers, articles for periodicals) to explore and contribute to a better understanding of critical development issues, and help guide ADB and its DMC clients. Over the past 12 years, there have been 83 ADB and ADB Institute (ADBI) publications, which fall into a broad definition of urban sector-related (Supplementary Appendix E). The publications cover a wide range of topics from fiscal transfers and local government finance, through urban and subsector country-level strategies, to subsector information sources. The majority of these publications are on the subject of urban development and housing (56%), followed by WSS (10%), and urban poverty (7%). 94. Analysis of the list shows that the range and number of urban sector documents published by ADB (and ADBI) increased from an average of about 5 per year to about 9 per year around the time of the USS publication in 1999. During the 6 years prior to the launching of the USS, 94 public documents were issued (web-based directory), of which 31 were ADB publications (paper directory). In the 6 years after the USS, 134 public documents were issued, of which 52 were in print. However, increasing numbers of these publications are ADB review articles and CD-ROM reviews. Thus, part of the increase in urban-related knowledge products may owe more to the growing success of ADB s dissemination efforts than a resurgence of interest within ADB in the urban sector. 95. The USS called for research on the following themes: (i) the urbanization process, (ii) urban poverty reduction, (iii) decentralization and urban governance, (iv) community participation in urban development, (v) private sector involvement in urban infrastructure, (vi) informal settlements, (vii) urban environmental management, and (viii) local government capacity building and finance. An analysis of the research-oriented publications published since the USS was issued shows that all of these themes were, to variable degrees, addressed: (i) 6 papers were written on urbanization issues of which 4 were country-specific urban sector strategies; (ii) 5 on poverty reduction of which 1 was a specific country poverty assessment; (iii) 3 on decentralization and urban governance 1 country-specific, (iv) 2 on community participation (participatory decision making); (v) 2 on the potential for private sector involvement (in roads and water supply); (vi) 1 on extending services to the poor, 1 on housing for the poor, and 1 on water for the poor; (vii) 2 on environmental strategies for cities, and 1 on solid waste management; and (viii) 11 on municipal management and finance with an emphasis on financial management and revenues. Other documents addressed specific infrastructure studies, reports and proceeding of conference seminars and workshops, the urban indicators data book, and work on gender and malnutrition. 96. Although each of the key themes identified in the USS was addressed, the list of publications is highly diverse and lacks focus, depth, and continuous ADB involvement with the topic. More focus should be placed on areas to provide deeper insight. An example would be the cluster of issues related to the increased autonomy of urban areas under the

36 decentralization process, and mechanisms for mainstreaming the benefits of development to the urban poor. The former is an issue that has had implications, sometimes adverse implications, for the urban portfolio in some countries that used to borrow significant amounts for urban development. The latter is directly related to ADB s overarching objective. 97. More fundamental is the issue of the use of documents and knowledge within ADB. Staff turnover in ADB has often led to ADB staff moving out of the urban sector relatively quickly. Because of weaknesses in ADB s knowledge management system, it is not clear how much of the information in these publications is put to optimal use for a better urban sector program. This problem of knowledge management extends well beyond the urban sector. However, the lack of an effective knowledge network among urban specialists and limited knowledge sharing across the relevant divisions in the regional departments compounds the problem in the urban sector. ADB has recognized that steps need to be taken at the institutional level to improve knowledge management. A vice-president of knowledge management was created in 2002. A knowledge management framework was approved in 2004 and a knowledge management center was created at the same time. V. ADB ASSESSMENTS OF COMPLETED URBAN SECTOR PROJECTS 98. This chapter begins by summarizing assessments of processes, outputs, and outcomes of urban sector projects made by PCRs, project performance audit reports (PPARs), TA completion reports (TCRs), and TA performance audit reports. This is followed by a section which presents assessment made of ongoing urban sector projects by their project directors, as based on a questionnaire survey. Many of the answers are instructive to gauge demand for ADB loans and the benefits and the transaction costs perceived by ADB s main clients. A. Assessment of Urban Sector Projects by Operations Departments 99. Ratings. For the past 8 years, ADB s methodology for rating projects in PCRs has been similar to that of PPARs, and is based on assessments of relevance, efficacy, efficiency, sustainability, and other impacts. 67 Figure 13 is derived from a database maintained by OED since 1995, when PCRs started rating projects more systematically. During the past 10 years, according to PCR ratings, urban sector projects have a higher probability of being rated successful than other projects in ADB s portfolio. Integrated urban development projects tend to perform better, on average, than WSSWM projects. Although all other urban projects were rated successful, the number of projects in this category was small. 67 This is done on a scale of 4 (e.g. highly relevant, with a value of 3, relevant [2], less than relevant [1] and irrelevant [0]), and weighing these assessments at 20%, 25%, 20%, 20% and 15% of the total respectively, to calculate an overall project success rating between 0 and 3. As a last step, ADB rates the projects with values >0 and <0.6 as unsuccessful, those with values >0.6 and <1.6 as partly successful, those with values >1.6 and <2.5 as successful, and those with values >2.5 as highly successful. These values can be expressed as a proportion of a maximum score of 100%.

37 Figure 13: Percentage of Projects Rated Successful or Higher in PCRs, for Urban and Nonurban Projects (1995 2005) Percent of Projects Successful 100 75 50 25 0 81.3 Multisector (48) 66.7 WSSWM (36) 100.0 Other Urban (4) 76.1 67.3 Total Urban (88) Category of Urban Sector Projects Other Projects (459) PCR = project completion report; WSSWM = water supply, sanitation, and waste management. Note: Number of PCRs in parenthesis. Source: Operations Evaluation Department s Post-Evaluation Information System for all PCRs issued from January 1995 to July 2005. 100. The SES assessed the scores on the individual rating criteria for all 51 PCRs completed between mid-1998 and mid-2005 (Table 2). Fifty were approved before the USS, and implemented during the time that the USS was in effect. On average, the sample is somewhat younger than that discussed in para. 99. Table 2: Average PCR Scores (% of Maximum) on Five Evaluation Criteria, by Subsector of Urban Sector Projects (1998 2005) Urban Subsector (Projects) Relevance Efficacy Efficiency Sustainability Other Impacts Average Composite Rating Percent of Projects Successful Multisector (23) 78.3 60.9 59.4 47.8 49.3 59.2 82.6 WSSWM (22) 83.3 60.6 53.0 50.0 53.0 59.8 77.3 Other (6) 83.3 72.2 66.7 55.6 61.1 67.7 83.3 Total (51) 81.0 62.1 57.5 49.7 52.3 60.5 80.4 PCR = project completion report; WSSWM = water supply, sanitation, and waste management. Source: Fifty-one PCRs issued between mid-1998 and mid-2005. 101. A comparison of Tables 1 and 2 confirms the perception that the quality of ADB s projects has improved over time. The more recent sample of PCRs reflected in Table 2 also confirms that multisector projects are slightly more likely to be successful than WSSWM projects. WSSWM projects are assessed as somewhat less efficient, although sustainability of project outcome is more of an issue with multisector projects. 102. Capacity and Institution Building. The SES assessed statements made in PCRs regarding the project s success in capacity and institution building. All projects reviewed had such components. Thirty-five TA grants were attached to the projects and most projects included loan financed consulting services in their design. The PCRs rated 65% of the capacity building efforts as successful and 50% of the institutional development initiatives as successful. In 11 of the 51 PCRs, critical statements were made about capacity and institutional

38 development. The best performance occurred in the PRC and worst in the Philippines. The subsequent departure of trained staff from the agency, sometimes to the private sector, continued to result in disappointing capacity and institutional development results in the Philippines. 103. Much of the capacity building was designed to improve project management and implementation. Institutional development was generated by implementing TA recommendations (e.g., incorporation of planning standards into urban development plans and building codes, water tariff revisions, revisions of master plans, improvement in billing and collection from computerization, organizational changes, increased collection of taxes, private sector participation in projects, and management information systems). Loan financed consulting services helped with the transformation from public utilities to more autonomous enterprises, corporatization, adoption of commercial accounting systems, improved garbage collection systems following solid waste management training, and improved revenue collection. Many PCRs indicated that there were further needs in capacity development, such as in financial management, operations and maintenance, project performance and management systems, and urban economic development planning. 104. Irregularities. Irregularities in urban sector projects were reported in a few PCRs for Indonesia. These were eventually dealt with by the Government of Indonesia and ADB in close collaboration. The Sumatra Urban Development (Sector) Project is an example (Box 6). 68 The Samut Prakarn Wastewater Management Pollution Control Project in Thailand 69 is another example of irregularities, where lack of community consultation and alleged irregularities in land acquisition led to the Government of Thailand s suspension of project implementation by February 2003, and an ADB inspection. 70 Given the apparent pervasive nature of corruption in the urban sector in Asia (para. 80 and Boxes 3 and 4) as in other sectors, it is important that ADB and governments pay more attention to preventing corruption in the urban sector, through training of staff, incentives and better systems. Box 6: Irregularities in Sumatra Urban Development (Sector) Project In April 2000, the Government of Indonesia transferred the supervision of many projects to local governments. To accommodate the decentralization process, the Government produced its own set of procurement regulations. Unfortunately, procurement problems irregularities and more basic administrative issues increased significantly. The Asian Development Bank (ADB) began requesting more bid and contract documentation for review. Some local governments were not fully aware of ADB procurement guidelines, while others sought to equivocate on compliance. Still more found it incomprehensible that government regulations could be superseded by ADB guidelines and were afraid of the consequences from state auditors if they contravened the presidential decree. In June 2002, ADB suspended withdrawals from the loan account, citing many procurement irregularities, declaring that the executing agency (EA) lacked adequate internal controls, and concluding that the EA was in breach of the Loan Agreement. The move followed several meetings to discuss procurement irregularities. The EA accepted the existence of irregularities, but noted difficulties of managing a project with so many participating local governments and associated contracts, as well as its lack of authority 68 ADB. 2005. Project Completion Report on the Sumatra Urban Development (Sector) Project in Indonesia. Manila. 69 The project was supported by Loan 1410-THA (ADB. 1995. Report and Recommendation of the President to the Board of Directors on a Proposed Loan to the Kingdom of Thailand for the Samut Prakarn Wastewater Management Project. Manila). 70 Inspection Request (28 Feb 2002); Sixth Semiannual Report on Implementation of the Recommendations of the Board Inspection Committee (24 June 2005).

39 to control procurement in the newly decentralized environment. The EA offered to work closely with ADB to resolve these problems and enable project implementation to resume. In October 2002, the EA agreed to an external assessment of its internal management controls and oversight systems, the establishment of local government procurement committees with independent members, a joint procurement review team in the Ministry of Public Works, and consolidation of all remaining equipment to be procured under the Project into two international competitive bidding packages. Shortly afterward, ADB removed the suspension of withdrawals. Source: Adapted from ADB. 2005. Project Completion Report on the Sumatra Urban Development (Sector) Project in Indonesia. Manila. 105. Other findings from the study of 51 PCRs are summarized in Appendix 6. B. Assessment of Completed Urban Sector Projects by the OED 106. PPARs prepared by OED 71 provide ratings based on the situation encountered, usually 2 years after the PCR is prepared. These may take a more independent look at the project; the further elapsed time may also play a role in arriving at ratings that are different from those of the PCRs. Table 3 shows that, contrary to ratings given by operational departments through PCRs, OED s ratings for the urban sector are on average somewhat below the ratings for projects in many other sectors, especially urban WSSWM projects. One reason may be that urban sector projects are usually operated by weaker local governments and utilities and are more likely to have long-term sustainability issues, which may not have been as evident at the stage of the PCR mission. Table 3: Comparison of Project Performance Audit Report Ratings for Urban Sector Projects and Other Projects, 2000 2005 Item Relevance Efficacy Efficiency Sustainability Other Impacts Composite Rating % of maximum rating urban 75.6 62.2 55.6 46.7 44.4 57.8 sector projects (N=15) of which % of maximum rating 73.3 63.3 60.0 46.7 36.7 56.6 urban WSS projects (N=6) % of maximum rating other urban 80.0 60.0 46.7 46.7 60.0 60.0 projects (N=9) % of maximum rating of other 75.1 60.9 56.3 58.6 55.2 62.1 sectors (N=88) % difference between urban sector and other sectors 0.6 2.1 (1.4) (20.4) (19.4) (7.0) WSS = water supply and sanitation, N = sample size. Note: Sixty percent of urban sector projects were rated successful (of which 50% were WSS projects, and 66.7% were multisector projects); 71.6% of nonurban sector projects were rated successful over the same period. Source: Operations Evaluation Department project performance audit report databases. 107. Lower PPAR ratings stem mainly from the lower degree of (i) project sustainability, and (ii) other impacts that urban sector projects have on institutions and the environment. OED s 1997 SES on urban development and housing 72 corroborates the lower long-term sustainability of urban sector projects. That study of 12 urban multisector projects led to the conclusion that such projects generally succeeded in improving living conditions, but that local capacities to operate and maintain project facilities were not always in place, and that more projects covered in that study had not given much support to developing such capacities. From this perspective, it 71 By end 2005, the format of the PPAR was changed, and the new OED project evaluation report was called the project (or program) performance evaluation report (PPER). 72 ADB. 1997. Impact Evaluation Study of Bank Assistance to the Urban Development and Housing Sector. Manila.

40 is a matter of concern that the level of capacity development support has not increased since the USS. 108. Ratings of the multisector projects are sometimes influenced positively by minor components in financial terms, such as urban road resurfacing, slum improvement, self-help groups, and community development. Although requiring relatively less funds, such components can deliver substantial benefits to the target beneficiaries and create goodwill and appreciation by addressing some of the more intractable issues related to poor quality of life. 109. Water supply projects have sometimes suffered from the lack of data on groundwater resources; resistance to increase water user fees; and insufficient, well qualified human resources needed to ensure the technical sustainability of the project outputs. There now seems to be an increasing recognition in many DMCs of the need to charge for adequate water consumption and wastewater treatment to ensure that the required financing is available for operations, maintenance, and capital expansion. It is likely that the sustainability of these projects will improve in the future. 110. Single City versus Multicity Projects. Both PCRs and PPARs concluded that projects located in one city, usually a large city and/or a capital, are more likely to be successful than multicity projects (Figure 14 and Supplementary Appendix F). However, multicity projects can also be successful. Investment projects prepared in detail before project approval have been slightly more successful than sector projects, which usually have only a few city subprojects prepared at the time of project approval. PPARs, prepared a few years after the PCRs, rate many more single city projects as successful than multicity projects. All of this implies that multicity projects, whether sector projects or not, need better preparation and more support. These findings may also suggest that projects that have relatively simple institutional arrangements are more likely to succeed than projects with complex institutional arrangements. These findings suggest that ADB should explicitly identify institutional arrangements as a potential risk and put more effort into designing risk mitigation measures. If suitable risk mitigation measures cannot be developed, particularly in countries with weaker institutions, the complexity of the project design should be reduced to increase the probability of project success.

41 Figure 14: Rating of Single City versus Multicity Projects in PCRs (1995 2005) and PPARs (2000 2005) 71 67 Rating (scale 0 to 100) 75 70 65 60 55 50 45 40 60 Single City Projects (23) Multicity Projects (26) 48 PCR rating PPAR rating PCR = project completion report, PPAR = project performance audit report. Source: Forty-nine PCRs completed between 1998 and mid-2005, and 15 PPARs completed between 2000 and 2005 (8 single city and 7 multicity). 111. Box 7 summarizes urban sector findings in Indonesia, Lao People s Democratic Republic and Pakistan as reported by recent OED country and sector assistance program evaluations. Box 7: Urban Sector Findings from Recent OED Country Evaluations Indonesia. The Operations Evaluation Department s (OED s) Country Assistance Program Evaluation (CAPE) a rated Asian Development Bank s (ADB) large urban sector program, completed within the period 1990 2004, as broadly successful. The overall performance in all sectors was noted to have gone down after 1999, when Indonesia s decentralization process began. Local level project management capacities proved a bottleneck. In the uncertain operating environment of the transition years (2001 2004), ADB was slow to channel adequate resources, delegate authority, and develop new modalities to suit the changed nature of the client. This is now improving. The evaluation paid special attention to urban water quality and sanitation. The water quality was deteriorating in Indonesia due to lack of sewerage treatment, dumping of untreated liquid waste, and sedimentation caused by a lack of upper watershed management. ADB s activities in water pollution control were integrated as subcomponents under selected urban development initiatives but these had not all been successful. Although Indonesia was in need of more wastewater treatment facilities, the enabling environment remained weak. Enforcement was lacking, and the tariff structure remained ineffective. The evaluation provided the following lessons for consideration in designing and implementing future projects: (i) emphasize a package of services, not just infrastructure, (ii) avoid funding water supply projects which are not based on strong household demand, (iii) water supply enterprises need to be privatized, but only after the adoption of a comprehensive water supply policy; and (iv) ADB needs to give greater attention to waste management. Lao People s Republic. OED s CAPE b assessed the urban development sector program which consisted of four projects, two of which were completed, and two ongoing at the time of evaluation. The overall performance was rated as successful. A major outcome achieved was the Urban Development Administration Authorities ability to raise their own funds for Operation and Maintenance of urban infrastructure. Another outcome was the institutionalization of the Authorities, through the passage of the Law on Local Administration, which was formulated under an ADB TA. Other outcomes registered included (i) access to improved urban environment increasing to 33% of total urban population, (ii) reduced damage and destruction due to flooding, and (iii) increased number of businesses in the downtown urban areas. ADB was judged to have made a significant contribution to access of the

42 urban population to improved sanitation, which increased from 11% to 61% (1990 2002). In the water supply and sanitation sector, in which there were also four projects, the effectiveness was lower, ADB s contribution to long term impacts was seen as modest, and the overall performance was rated as partly successful. Pakistan. OED s Sector Assistance Program Evaluation for the Social Sectors in Pakistan c included a review of the water supply and urban development projects. Overall, the performance of the social sectors over the period 1985 2002 was assessed as poor: achieving the Millennium Development Goals would require a turnaround from negative progress in the 1990s or much faster progress than what was achieved in the past. ADB s focus on investing directly in public provision of infrastructure and facilities including in urban areas had generally led to disappointing results. The evaluation judged that no water supply, sanitation or urban development projects had been successful; five projects were rated as partly successful and 4 as unsuccessful. a b c ADB. 2005. Country Assistance Program Evaluation: Indonesia. Manila. ADB. 2006. Country Assistance Program Evaluation: Lao People s Democratic Republic. Manila. ADB. 2005. Sector Assistance Program Evaluation for the Social Sectors in Pakistan. Manila. C. Assessment of Completed TA Grants by Operations Departments and OED 112. Aggregating all TAs classified as urban development or WSSWM, ADB s TCRs rate these sectors as having performed similar to other sectors overall. Whereas 84% of all urban sector TAs completed since 1995 (until the end of 2004) have generously obtained a successful or highly successful rating through self-assessment by operations departments, 82% of all other TAs have obtained such rating. The remaining portions in all categories were rated partly successful or unsuccessful. 113. Between 1995 and 2004, OED evaluated 18 TAs in urban development or WSS in 5 TA performance audit reports. Only 13 had been rated by the operational departments earlier. The TCRs rated 92% of these urban development and WSS TAs successful. In comparison, OED rated 78% of these TAs successful. Although OED s ratings are lower than those recorded in TCRs, OED s success rating for urban TAs (78%) exceeded the 57% for the TA portfolio as a whole. 114. Assessment of ADB s Strategy Studies. The SES made a special assessment of ADB s strategic studies funded by TA grants over the past 10 years: country urban sector profiles, country urban sector strategies, and country urban sector reviews (Supplementary Appendix G). The evaluation confirms that they provide sector data, information, and analysis for the formulation and preparation of sector TA and lending programs. The sector profiles frequently set out elements of a sector strategy for the Government, ADB, or both. However, neither the profile nor strategy documents (i) set out clear strategic choices; (ii) identify sector targets or monitoring and evaluation mechanisms; or (iii) assess the constraints to, and risks of, the strategy or the potential for ADB intervention. 115. Most of the strategy and profile publications are based on sector work carried out by consultants under ADTAs or staff consultancies; there is less direct involvement of ADB staff. In many cases, the usefulness of the document is compromised by the time elapsed between the implementation and completion of the TA, and the publication of the profile or strategy document. For those studies for which information was available, this elapsed time was between 9 and 18 months. 116. Fifteen TAs over the past 12 years have financed the preparation of urban sector profiles or strategies. Operations department issued TCRs for 7 of these, of which 6 were rated

43 satisfactory and 1 partly satisfactory. Based on the findings in the TCRs, the main issues affecting the quality and utility of these documents relate to (i) the difficulty of finding the right institutional home for the study in the DMC; and (ii) developing a suitable methodology which can accommodate, within the short time span usually available to ADTAs, adequate stakeholder involvement in the preparation process. D. Assessments of Benefits and Problems of Ongoing Projects by Project Managers 117. This section focuses on the ongoing urban sector portfolio, and the value added and transaction costs of ADB involvement. PMs of all urban sector projects in the DMCs were requested to provide their views through a multiple choice questionnaire. Table 4 summarizes answers of (i) PMs in the main offices, and (ii) respondents from both main and subsidiary offices. The questionnaire is in Supplementary Appendix H, including a tabulation of all the responses. Table 4: Main Advantages of Asian Development Bank Involvement in Project (Rating by DMC respondents on scale of 0 100) Advantage Suggested Main Office Total More certain and/or steady supply of funds 85 82 Good and transparent procurement processes 75 79 Better design of the project 66 70 Access to advice or operational support by loan-funded consultants 66 69 External quality control and/or monitoring by ADB project officers 63 66 Intellectual or technical leadership, good practice, and/or policy dialogue 58 66 Large investment project has catalytic effect 56 61 ADB can persuade government decision makers 51 57 ADB can leverage additional resources (access to cofinancing) 50 55 Project administration or project salaries funded through loan 31 39 Other advantages 7 6 ADB = Asian Development Bank. Note: 44 responses from project directors (85% of all projects in the survey) and 44 from others (generally from project implementation unit managers). The special evaluation study (SES) rated the three response categories provided as follows: major 100, minor = 50, none = 0. Source: SES questionnaire survey of 52 ongoing urban sector projects. 118. Benefits Connected with ADB Involvement. The relatively high positive responses on the advantages of ADB involvement (Table 4), as compared to those on a question on drawbacks (Table 5), demonstrate that many PMs believe that there are advantages of ADB involvement in projects. This was in spite of reservations, which were also regularly voiced, and which point to room for future improvement. The most highly rated benefit of ADB involvement was the more certain and steady supply of funds for the Project. This response is significant as it revealed how difficult access is to secure project financing in many DMCs. 119. A second most important benefit was the use of good and transparent procurement processes resulting from ADB involvement (notably the two envelope system). This suggests that ADB s rigorous procurement process helps to reduce fraud and corruption. This positive feedback was given despite concurrent complaints about the ADB system leading to procurement delays. ADB is working to streamline and simplify its procurement procedures. 73 73 In India, several states have adopted ADB's tendering procedures for larger locally financed projects. Similarly, the quality control testing, monitoring, and reporting is considered a major improvement over locally financed projects where suppliers and contractors regularly "cheat" on specifications and installation quality. Under ADB financing,

44 A variety of other benefits from ADB involvement included improved project design, the TAs attached to loans and the consultant support funded by it, the quality control exerted by ADB project officers, and ADB s good practices in general. 120. Perceived Drawbacks to ADB s Involvement. Table 5 identifies relative problems with the involvement of ADB as perceived by the survey respondents. DMC project staff rated the enforcement of the ADB agenda as the biggest drawback of ADB s involvement, followed by delays as a result of ADB involvement, and the extra paperwork involved. However, the severity of these problems was lower; they were usually viewed as minor. A number of respondents mentioned problems associated with consultants. Delays in ADB disbursements, problems of continuity of ADB staff, and a lack of suitable ADB staff were mentioned by 10% to 15% of respondents. Table 5: Categorization of Main Problems in Project (Rating by DMC respondents on scale of 0 100) Problems Main Office Total Associated with ADB involvement: Difficult ADB forms and procedures and excessive paperwork 42 39 Enforcement of ADB agenda (e.g., resettlement, environment, participation, etc.) 39 35 Delays in ADB responses, decisions, and/or approvals 35 34 Improper staffing by consultants recruited for the project 42 33 Delays in ADB disbursements 14 15 Lack of staff continuity in ADB 12 12 Lack of suitability of ADB project officers for task at hand 11 12 Associated with Government involvement: Delays in government responses, decisions, and/or approvals 60 49 Lack of effective coordination, or opposition from other agencies 44 35 Government policies or decisions obstruct and/or delay project activities 47 33 Insufficient government budget made available to project (or late release) 31 33 Difficult and/or unwieldy government systems and procedures 40 32 Lack of (qualified and capable) staff to implement the project 38 28 Involvement of politicians in the administrative domain 30 23 Problematic division of responsibilities between project officers and other agencies 31 22 Lack of effective zoning/building regulations 21 17 Associated with deficient design/preparation: High and rising cost of land or other issues with acquisition of land and/or rightof-way 35 31 Project has too many components and implementing agencies 28 25 Project has lack of effective monitoring system and/or lack of data 18 18 Project has more donors than ADB and this demands time 12 14 Opposition to project activities from civil society or private sector parties 15 13 Lack of demand for project (components) from intended beneficiaries 12 13 ADB = Asian Development Bank, DMC = developing member country. Source: Special evaluation study questionnaire survey 2005. 121. Problems with Government Involvement. Problems traceable to the nature of government involvement in the projects were seen as more prevalent than the problems traceable to ADB involvement. The most common problem was delayed decision making by the the EAs are able to use ADB's monitoring as a reason to implement third-party inspection of contractors works, leading to much better finished product.

45 Government, which adversely affected project progress. This was followed by a lack of coordination between agencies involved, and government policies obstructing project activities. Cumbersome government systems were also mentioned. Almost half of project directors mentioned a lack of qualified and capable staff to implement the project. About 30% stated that involvement of politicians in administrative decisions already taken was a problem despite the fact that many directors also stated that one of the advantages of ADB involvement was that this curbed political decisions overtaking technical and administrative decisions. 122. In total, only 17% of the respondents did not mark any of the problems listed in the questionnaire. At the other end of the spectrum, 15% registered all or almost all of the problems mentioned. The average number of problem categories marked was 2.3. 123. Perceived ADB Value Additions. Table 6 summarizes the results of the questionnaire survey regarding perceptions about issues related to project design and preparation. Except in the PRC, these areas are often dominated by ADB staff. Table 6: Project Preparation Related Statements as Confirmed by Survey Respondents (% of DMC respondents) Item Main Office Total There is no resistance from intended beneficiaries 61 63 There were no complaints from other stakeholders about project set-up 52 58 There was enough input from our agency into project design 56 56 Loan covenants are helping and/or have helped significantly to achieve reforms 55 57 There is no resistance from non-beneficiaries 51 49 Project is completely uncontroversial in the city and is universally seen as beneficial 58 51 This project is better able to guard against corruption than nationally funded projects 39 41 All stakeholders perceive this project as driven by EA agenda, not by ADB agenda 26 29 Project is confronted with more attempts at corruption than nationally funded projects 5 6 ADB = Asian Development Bank, EA = executing agency. Source: Questionnaire survey of 44 project directors and 44 other respondents (project implementation unit managers). 124. The responses are generally positive but it should be borne in mind that project staff may be favorably inclined towards their project. A response rated below 60 on a scale of 0 to 100 may well be indicative of problems. If the positive inclination is discounted, then it becomes clear that respondents were ambivalent about the effect of the input from their agency in project design and the beneficial nature of the loan covenants, and more skeptical about the ability to guard against corruption and especially the prominence in the project of the ADB agenda. 125. How Project Clients Viewed Capacity Development. The questionnaire survey elicited views from PMs and subsidiary PIU managers on capacity development. PMs saw lack of qualified and capable government staff to implement the project as an important problem. They also frequently noted incidences of improper staffing by consultants. Underpayment of staff and resulting underperformance was seen as a main staff capacity problem by 46% of PMs; lack of staff in the agency (either positions or filled positions) by 25%; demoralization due to political decision making, corruption, and lack of incentive to move to field by 18%; and short tenure of senior staff by 14%. On the positive side, over half (56%) of PMs saw improved capacity of agencies involved in the ADB project as a major benefit of the project.

46 VI. ADB S ADDED VALUE: VIEWS FROM PEOPLE S REPUBLIC OF CHINA, INDIA AND PHILIPPINES 126. This chapter starts with a brief assessment of operations and views expressed by stakeholders in three of ADB s four largest borrowers in the urban sector: PRC, India, and Philippines. 74 The section is a summary of a more extensive digest of the findings of the three country case studies (Appendix 7). The case studies were intended to bring out the influence of different country contexts and the need for country specific urban sector strategies. They focused on (i) urbanization, (ii) the capital investment requirement, (iii) other programs in the sector, (iv) results from past ADB involvement, (v) an assessment of current ADB operations, (vi) feedback on perceived value added and transaction costs of ADB involvement, and (vii) perceptions by the OED team. The chapter concludes with an assessment of the relevance of some recent ADB reform initiatives to the urban sector. A. People s Republic of China 127. A senior official at the Ministry of Finance stated to the study team that the Government of the People s Republic of China did not regard itself as dependent on ADB financing for the urban sector; sufficient project finance was currently available from state-owned banks and commercial banks in the country. The ministry and municipal borrowers were looking to ADB to provide significant added value in support of knowledge generation and transfer to address the PRC s formidable urbanization challenges. The official stated that ADB s added value currently left much to be desired in terms of bringing best international practice and knowledge transfer. ADB needs to perform better in this area to justify the transaction costs that the Government incurs when dealing with ADB loans. Although the study registered a more positive view of ADB s added value by other stakeholders, unless satisfactorily addressed, this perception may limit the potential for expansion of ADB s urban sector lending. The urban sector has nevertheless been one of the four sectors for which the PRC borrows evolving from early focuses on mitigating severe environmental pollution in cities and addressing pressing problems (such as restoring city or district heating systems), into more water supply and sewerage systems. Currently, ADB s urban portfolio in the PRC includes 9 loans at $994 million; 11 loans are under preparation with indicative loan amounts totaling $1,221 million. 128. Success of ADB s Lending. PCRs of operating departments as well as OED reports have invariably rated projects as successful or highly successful (an overall success rating of 100%). The PRC s overall good loan portfolio performance extends to the urban sector. In the cities where the projects took place, ADB has contributed to significant results, and the scores given for the sustainability of the investments have been higher than in many other DMCs. However, not everything has gone perfectly. For instance, some sewage treatment plants are underutilized in the first years due to a lagging development of sewage networks. Considerable transaction costs are incurred both by ADB and the PRC. Loan cancellations for the five completed projects have been extraordinarily high, amounting to $263 million, or 25% of the loan amounts. This is at least 5% higher than the ADB average for loan cancellations. 75 Contingent factors may have played a role, such as the fall in equipment prices and heavy competition for civil works contracts in the late 1990s. But PIUs also reported that they did not like seeking the reallocation of savings because the process required approval by ADB headquarters in Manila, and was tied to pre-set reimbursement ceilings that seemed difficult to 74 The other large borrower for the urban sector is Indonesia. It was not included as a country study for this evaluation as Indonesia, including the urban sector, was being covered in another OED product: the Country Assistance Program Evaluation (para. 112). 75 ADB. 2004. Special Evaluation Study on Project Cost Estimates. Manila.

47 change. The more flexible financing from domestic sources, although not financially competitive with ADB loans, required far less administration costs and could finance expenditures directly, rather than reimburse, such as was the case with some of ADB s financing. It was easier to get replacement financing from local banks. The dropping of components from the project design, while occasionally unavoidable, suggests that the preparation of such components may have been rushed. 129. The ongoing ADB urban sector project portfolio in the PRC is widely acknowledged as performing well. It is almost exclusively in the WSSWM sector; the straightforward nature of the project design may be one reason for success, although more important factors include the EAs strong project ownership, technical strength, and good planning and preparation. There is growing demand from cities for multisector projects, in order to deal with complex and pressing issues. Demand from smaller cities has led to the preparation of the first sector loan. The new demands find growing support for more sector loans and multisector projects within the Ministry of Finance and the National Development and Reform Commission. The trend is towards greater complexity but also greater responsiveness to city needs, and larger potential impact on poverty reduction and capacity building. 130. ADB s Added Value. The OEM registered the following client perceptions: (i) (ii) (iii) (iv) (v) (vi) (vii) The good reputation of ADB helps the EAs to raise additional loan financing from domestic commercial banks at lower costs. ADB projects often receive more support from local government for land acquisition, resettlement, staffing, and tariff adjustment; and counterpart funds are more easily secured. ADB helps EAs develop capacity; projects are mainly implemented through internally staffed PIUs and specially created companies that later operate and maintain the project outputs. There are fewer influences from local government during implementation regarding issues that have been decided (requests to change the schedules or the project scope, etc.). Contractors in the PRC are more interested in bidding for ADB project contracts due to ADB s good reputation with respect to transparent procedures and secure payment. Contractors put in lower bids as the risks are perceived as lower. ADB s interest rates are competitive, and long-term loans on similar conditions cannot be obtained from PRC banks. ADB s environmental safeguards have generally been appreciated and are acknowledged to have improved the quality of the projects. 131. Recommendations. The OEM noted the following made by ADB s project clients in the PRC: (i) (ii) (iii) ADB mission leaders should be given more responsibility, accountability, and authority to improve efficiency and effectiveness, particularly at the preparation stage, and to minimize approval procedures in ADB headquarters. Responsibilities for procurement and payment control should be transferred to the ADB s RM in Beijing, and ideally to Chinese speaking officers and/or staff. Problems due to language barriers and the need for translation should be minimized. ADB s support in the form of capacity building TA, and policy and strategy studies for local governments, is highly appreciated and should be strengthened

48 (iv) (v) (vi) (vii) (viii) (ix) further. This could then be followed by loans targeting the priorities and challenges identified. The strict reimbursement mechanism should be replaced by a more flexible mechanism that would allow the reallocation of loan funds among major expenditure categories (civil works, equipment, training, etc.) so that any loan savings are easily utilized. The size and performance of imprest accounts should be increased to minimize the need for seeking bridging loans from local banks. Project preparation procedures should be simplified and shortened. It should be possible to make more use of retroactive financing in ADB projects. ADB should accept the PRC s own resettlement policy. This provides for negotiation and payment to village and town governments. ADB s resettlement policy was criticized because negotiating compensation directly with households was seen as unworkable in the PRC where land is collectively owned. ADB needs better branding and promotion of its urban capacity. 132. ADB s TA Program in the PRC. ADB s support of institutional and capacity development was substantial (section IVD) but has declined after the USS. The same has been the case for ADB s PPTA resources. Loan consulting services for similar objectives has been at around 1.8% of the loans for the PRC before the USS; this was below the ADB average, and declined to 0.8% after the USS. These ratios for consulting assistance cannot be compared to other countries because most of the local consulting services are financed by the EA rather than by the ADB loan. Much of the loan financed consultant services was directed towards management improvement of the PIUs. The World Bank, while not having devoted similar resources to grant-based TA, has often succeeded in securing bilateral grant funds for loan preparation, while its loan consulting services have been between 3% and 5%, and directed more at local policy reforms. ADB s ADTA program shows that TAs have performed important reform-oriented functions at the national level, and often kick-started the capacity building of and institutions at the local level. TAs on water tariff policy, urban heating tariff reforms, and townbased urban development are examples of this. ADB s role in the creation of a body of knowledge on urban poverty and social security issues is well respected and considered useful by the Government. B. India 133. ADB lending for the urban sector in India began in 1995 with the approval of two loans of $85 million and $20 million respectively, for the Karnataka Urban Infrastructure Development Project (KUIDP). 76 ADB regarded this as a breakthrough, as the urban sector was part of the social infrastructure sector, for which India had chosen not to borrow OCR funds before. Fifteen loan approvals for the urban sector followed for another eight projects, for a total of $1,850 million. Two approved housing finance loans were not signed and eventually were cancelled. 77 Two projects have been completed so far. ADB has a considerable pipeline of new urban sector projects that is to lead to loans totaling $1,250 million in 2005 and 2006 (section IVB). Between 1993 and 2004, ADB approved 31 urban TAs, for $17 million: 52% were for urban sector profile studies and project preparation activities, 22% for institutional strengthening 76 ADB. 1995. Report and Recommendation of the President to the Board of Directors on Two Proposed Loans and Technical Assistance to the Republic of India and the Housing Development Finance Corporation Limited for the Karnataka Urban Infrastructure Development Project. Manila. 77 ADB. 2000. Report and Recommendation of the President to the Board of Directors on Four Proposed Loans to the Housing and Urban Development Corporation, National Housing Bank, Housing Development Finance Corporation, and ICICI for the Housing Finance II Project in India. Manila.

49 of urban local bodies and housing finance institutions, 16% for capacity building of project implementation agencies, and 10% for poverty reduction and community development objectives. 134. Multisector Projects. ADB s urban sector project interventions in India, starting with the KUIDP in 1995, have been generally in line with the multisector integrated urban development framework used in urban development planning in India. Except for the housing and urban finance credit lines, all ADB s loans for the urban sector have been for multisector projects. A consequence of the government s policy to assign focal states for external assistance was that state governments viewed the externally aided project as a one-time opportunity, demanding a wide range of investments in many sectors all at once. This increased project complexity, especially for the initial urban projects. PPTA consultants were often required to incorporate many components in the project design (e.g., urban water supply, drainage and sanitation, roads, traffic management, low-income housing, slum improvements, residential and industrial sites and services, truck and bus terminals, microcredit mechanisms, and community development). The approach delivered sometimes impressive results in high profile activities such as water supply, slum improvement, and self-help groups. However, there were also incidences when large sewage treatment plants remained insufficiently utilized due to lack of connections with households. It is too early to assess the extent to which the multisector project modality works as compared to the single sector modality: KUIDP is the only completed project so far. It was rated successful by ADB s PCR, and OED s draft PPAR has confirmed this rating, mainly because it has generated significant positive externalities within Karnataka State, and contributing to the national debate on urban development. It was the first major integrated multicity urban investment project in India in many years. As such, it generated substantial experience with the modality, and led to further statewide and India-wide policy initiatives. It also helped establish a professional urban investment center. After the start of KUIDP, the World Bank developed similar multisector projects, and other agencies followed with similar types of projects for the urban sector. ADB s urban sector project in Rajasthan is likely successful as well, and expectations are even higher with respect to an added component which will construct a 2.3 meter diameter pipeline over 72 km to a reservoir which will save Jaipur s water supply. 78 Projects in Madhya Pradesh, coastal Karnataka, and Kolkata appear to be well designed to address major problem areas, but are at relatively early stages and are facing implementation delays. 135. Credit Lines for Urban Sector Projects. Housing finance loans, targeted at low-income households, were an important component of ADB s urban sector programs during the late 1990s. The results varied significantly, from highly successful to unsuccessful projects and loans that were cancelled. Determination of the eligibility of low-income households for loans was one of the main difficulties experienced. With the exception of the Industrial Credit and Investment Corporation of India (ICICI) component of the second housing finance loan, the performance of the ongoing housing and urban infrastructure finance loans is less than satisfactory. Difficulties in determining low-income households, monitoring the end use of funds, the relatively high cost of ADB funds, and the Reserve Bank of India s guidelines on priority lending are some of the reasons for the current lack of demand for ADB housing loans. 136. Added Value and Transaction Costs. A senior official of the Ministry of Finance noted the Government of India s unhappiness with ADB s commitment charges, which are not 78 The Bisalpur component. Rajasthan State had unsuccessfully tried for 10 years to finance this water supply system through various modalities: local financing, build-operate-transfer, and public-private partnerships. Through ADB's initiative, the state was finally able to proceed with these essential works.

50 structured to reflect planned implementation schedules. Rather, a standard structure is used for all projects. Secondly, the high transaction costs associated with ADB safeguards has reduced the ministry s enthusiasm to borrow from ADB. 137. Interviews held with senior project staff identified the following main reasons for the generally slow absorption of ADB urban sector loans: (i) poor local capacity to design and implement projects; (ii) institutional complexity involved in designing and implementing integrated urban projects; (iii) inadequate local government ownership in many cases projects were driven by central Government and state governments; (iv) weak eligibility criteria for the inclusion of urban local bodies in the project; (v) lack of market discipline by local governments in formulating revenue-generating subprojects; (vi) lack of advance actions to sustain investments such as institutional reforms, particularly those related to cost recovery, acquisition of land, and procurement of technical expertise; (vii) inadequate community participation; and (viii) weak communication and media relation strategy of projects. 138. In spite of the drawbacks discussed, the performance of ADB-supported projects may not be worse than that of projects exclusively supported by the Government of India. Two such centrally supported schemes are discussed in Supplementary Appendix I. 139. As in the PRC, the argument often brought forward was the need for clear added value provided by ADB to complement the provision of loans. Project financing by commercial banks and semi-government financing organizations has recently become much more competitive; the difference between ADB and commercial and semi-government financing organizations is sometimes summarized as 7% real interest rate (including swap) for OCR loans, and 7.5% to 9% for other project financiers. State governments, in particular, are sensitive to this comparison. Although PPTA and bridging TA is appreciated, they are often seen as too short. The insistence on the use of foreign consultants is seen as inappropriate, as is the rigid application of social and environmental safeguards within sector projects, regarded as prone to delays and extra costs. Supplementary Appendix J lists further views from various stakeholders on value created and value lost as a result of ADB involvement. 140. ADB s strategy for urban multisector projects in India is evolving. ADB and the Government of India are experimenting with a variety of project modalities and implementation structures: the first project in Karnataka was decentralized, the second centralized. The project in Rajasthan is centralized as well. Projects in Kolkata and Madhya Pradesh are driven by PMUs and PIUs, which are based in local bodies. The Northern Karnataka Urban Infrastructure Development Project, which is under preparation, will have much more involvement from the private sector. Some of the lessons learned include reliance of projects on (i) involvement of finance corporations to identify and manage urban sector investments; (ii) e-procurement and open tenders for contracts rather than on prequalification, which can be manipulated by contractor coteries; (iii) bridging TA consultants to initiate advance actions; and (iv) preparation of management action plans for urban local bodies during and after the projects. C. Philippines 141. ADB in the Sector. The Philippines is ADB s fourth largest borrower in the urban sector and ADB s lending program has historically been concentrated on water supply. The first urban sector project in the Philippines was approved in 1976. Consequently, the urban portfolio in the Philippines is much older than the urban portfolios in India and the PRC. ADB s involvement consisted of 2 water supply projects in the second half of the 1970s, 4 in the 1980s (and a first

51 sewerage project), and 6 more water supply projects in the period 1990 1995. OED evaluations of projects approved before the 1990s found that 4 were successful and 2 unsuccessful. 142. In the last 10 years, the program has diversified. ADB s first multisector project was approved in 1995: the Regional Municipal Project. 79 Since then, 4 more water supply projects were started, 3 multisector projects, 2 environmental management projects, 1 project for shelter and poverty reduction, and 1 project for private sector finance. Since 1992, ADB s program has had to adjust to an uneven and uncertain process of decentralization of tasks from central Government to local governments. 143. Overall, ADB has approved $1,159 million urban sector loans in the Philippines from 1975 to 2003. From 1993 to 2002, ADB provided the Philippines with 18 TAs, 7 of which were ADTAs and 11 PPTAs. 144. Past Experience. Over the past 6 years, ADB has issued PCRs for five projects started in the 1990s: (i) Metropolitan Cebu Water Supply Project; 80 (ii) Municipal Water Supply Project; 81 (iii) Manila South Water Distribution Project; 82 (iv) Angat Water Supply Optimization Project; 83 and (v) one multisector project, the Subic Bay Area Municipal Development Project. 84 The projects were completed between 1999 and 2003. The PCRs rated the Municipal Water Supply Project successful while the projects in Angat and Cebu were rated partly successful. The projects in Manila and Subic Bay were rated unsuccessful. Taking the performance of the projects completed earlier into account, the proportion of unsuccessful urban projects is much larger in the Philippines than elsewhere. As the description of ADB s portfolio in Appendix 7 indicates, the progress of ongoing projects is generally disappointing except for two projects which have started only recently. 145. Client Views on Added Value and Costs of ADB Involvement. Discussions with officials and staff of the EAs and implementing agencies (IAs) of these six projects provided mixed feedback regarding ADB s involvement in their projects. This coincided with responses to the EA questionnaire survey PMs in the Philippines indicated generally double the number of problems compared to managers in India and the PRC. 146. PMs indicated advantages of involvement with ADB. Five of the six EAs saw ADB s primary contribution as support for financing urban infrastructure. ADB had a competitive advantage in this area because it could package TA with the loans. Three EAs saw ADB projects as generally better designed than locally prepared projects, although they were sometimes too optimistic about project inputs and outcomes. Interviewees also acknowledged that ADB could provide good pressure on supra-local decision makers. 79 ADB. 1995. Report and Recommendation of the President to the Board of Directors on a Proposed Loan to the Republic of the Philippines for the Regional Municipal Development Project. Manila. 80 ADB. 1990. Report and Recommendation of the President to the Board of Directors on Proposed Loans to the Republic of the Philippines for the Metropolitan Cebu Water Supply Project. Manila. 81 ADB. 1993. Report and Recommendation of the President to the Board of Directors on a Proposed Loan and Technical Assistance Grant to the Local Water Utilities Administration in the Republic of the Philippines for the Municipal Water Supply Project. Manila. 82 ADB. 1991. Report and Recommendation of the President to the Board of Directors on a Proposed Loan to the Metropolitan Waterworks and Sewerage System in the Republic of the Philippines for the Manila South Water Distribution Project. Manila. 83 ADB. 1989. Report and Recommendation of the President to the Board of Directors on a Proposed Loan to the Metropolitan Waterworks and Sewerage System and a Technical Assistance Grant to the Republic of the Philippines for the Angat Water Supply Optimization Project. Manila. 84 ADB. 1997. Report and Recommendation of the President to the Board of Directors on a Proposed Loan to the Republic of the Philippines for the Subic Bay Area Municipal Development Project. Manila.

52 147. PMs, however, indicated the following problems with ADB involvement: (i) (ii) (iii) (iv) (v) (vi) (vii) ADB is prone to overestimating the government s capacity to apply a new, innovative approach. ADB has a propensity to make institutional arrangements complex. ADB has not been sufficiently careful in choosing or inviting financial intermediaries. ADB approval processes are too lengthy, have too stringent safeguard policies, and voluminous documentation requirements. ADB presence at project sites and at the local level is not sufficient. ADB has high commitment charges on OCR loans (local government projects often move slowly and are therefore prone to incur these charges). ADB s involvement in the selection of consultants funded from local sources is not understood, and the quality of the consultants and their sensitivity to local culture is often insufficient. 148. A project official of an LGU project observed that LGUs are a difficult market for ADB for the following reasons: (i) short term of office of mayors (3 years), (ii) lack of continuity of priorities and agreements from one local administration to another, (iii) weak capacity of staff which necessitates lengthy and costly capacity building and TA support, (iv) aversion of many small LGUs to borrowing because of the perceived political backlash it can cause, and (v) limited financial resources of most LGUs. 149. Some encouraging JFPR initiatives in the Philippines are summarized in Box 8. Box 8: New Urban Initiatives in the Philippines: A Fresh Approach to Poverty Reduction ADB is pilot testing new approaches for on-site urban upgrading and off-site relocation of informal settlers under the Japan Fund for Poverty Reduction (JFPR). Two innovative $1 million grant-financed initiatives are underway. The first is in Payatas, Quezon City, site of the tragic garbage slide in 1999, where 525 families are undertaking a community-led urban upgrading program assisted by the Vincentian Missionaries for Social Development. The second is in Muntinlupa, along the Philippine National Railroad right-of-way where there are many squatters. About 567 families are working with the Muntinlupa Development Foundation to voluntarily relocate to a safer, off-site location. Both efforts involve sustainable revolving funds, and will be replicated in surrounding communities over time. A third JFPR project aims to develop strategic private sector partnerships for urban poverty reduction in Metro Manila. It will leverage $6.9 million in assistance from JFPR and leading Philippine corporations for slum communities throughout the city, to include capacity building and the provision of revolving funds for financing home improvements, urban services and micro-enterprise development. Source: Special Evaluation Study on Urban Sector Strategy and Operations. D. Conclusions from the Country Case Studies 150. There appears to be a potentially strong demand from cities for ADB financing in India and the PRC. However, total borrowing from ADB must be consistent with overall macroeconomic and public debt considerations. The transaction costs associated with doing business with ADB and ADB s perceived limited value added in terms of knowledge and best international practice sometimes limits the demand for loans for urban projects. In the PRC, projects are mostly prepared by the EAs. PPTA is mainly used to check the quality of the local feasibility study and compliance with ADB s policies, particularly safeguard policies. In India, the central Government selects the states and cities eligible for borrowing, often without bankable

53 project proposals being available. ADB s PPTA plays an important role in bringing the project to a bankable level. PPTAs are often viewed as too limited for this important task, especially for complex multisector projects. In the Philippines, the national Government determines the borrowing capacity of cities. Within this framework, the projects are prepared by PPTA, usually address only a subset of urban sector problems, and are viewed as ADB-driven. The findings of the country case studies are reinforced by the questionnaire returns received from project directors and PIU managers in these three countries. They are elaborated in Appendix 8. E. Potential Effect of ADB Policy Initiatives 151. For most projects, ADB s involvement was regarded as having many advantages. The long-term financing of needed infrastructure that ADB can offer is a scarce good in many DMCs, especially for the type of investments cities make. ADB offers long grace and amortization periods, longer than are available in the local market. ADB also has a concessional lending window, and recently offers grant-based ADF funding for some low-income countries in the WSS sector. Many respondents felt that the provision of reliable, long-term financing was the key advantage of working with ADB. This is a somewhat different view than the one commonly held within ADB that ADB s contributions to knowledge transfer and policy reforms are at least as important as financial transfers. There was a greater appetite for ADB to introduce best practice and innovation in ADB s larger DMCs, which have some significant institutional strengths of their own (e.g., the PRC and India). 152. The country case studies illustrated that some of the problems related to urban sector lending are systemic to ADB rather than to country contexts. Such problems are not specific to the urban sector, but are present to varying degrees in all sectors and across all DMCs. Examples include long project preparation and approval periods, inflexible commitment charges, complicated procedures, and insufficient project administration by ADB. With the growing availability of project finance in some DMCs, the added value of ADB involvement in projects has come under more scrutiny and criticism in some DMCs. 153. Because of feedback from clients and the lack of growth of the ADB loan portfolio during the past decade, ADB is seeking solutions to become more responsive and relevant. In 2005, some ADB-wide initiatives to change its systems and processes were adopted that will have a major effect on ADB activities in the urban sector (footnote 6 [i]). Some recently approved policy changes are (i) for OCR countries, an increase in the maximum threshold for international competitive bidding from $3 million to $10 million on a case by case basis; (ii) poverty targeting at the CSP level instead of the project level; (iii) more flexible determination of cost sharing limits; (iv) expanding the list of items eligible for ADB financing to include land acquisition, taxes and duties, and (limited) recurrent costs; (v) allowing retroactive financing up to 20% of the project amount; and (vi) allowing greater flexibility for the size of imprest accounts. 154. Innovative lending modalities were approved in 2005 on a pilot basis for a period of 3 years (footnote 6 [ii]). These concern multitranche financing facilities, subsovereign and nonsovereign public sector financing, local currency financing, and refinancing. Local currency financing by ADB should be an interesting option for local governments, as such costs are usually considerable in urban sector projects, and have increased as a proportion over time. Before the USS, local costs accounted for, on average, 51% of the total project costs. Post USS, this proportion has risen to 58%. Local currency financing for urban projects would also be attractive because municipalities are generally not skilled in mitigating foreign exchange risk exposure. Subsovereign lending has the potential to be an important lending modality for the urban sector. However, ADB would need to considerably strengthen its credit risk assessment,

54 risk mitigation techniques, and ability to critically assess the financial position of municipalities before substantially increasing subsovereign lending. This would also require a better partnership between the public and private sides of ADB. A significant revision of ADB s policy on supplementary financing was approved November 2005, which will allow for more streamlined financing of additions to financing plans and even investment plans. 85 The new guidelines for procurement will give more flexibility and authority to Borrowers. Bidding periods will be shorter. On a case-to-case basis, depending on Borrower capacity, ADB may carry out post-review of procurement procedures. Also approved are new guidelines for the use of consultants. A review of ADB s system of commitment charges for OCR loans is ongoing. 155. ADB s further reform agenda is expected to cover a wider range of actions that have the potential to change operations in the urban sector as well as other sectors: (i) updating the sector lending, investment lending, and program lending modalities; (ii) introducing a regional cooperation facility, an emergency response facility, and a cluster TA facility; (iv) increasing the 20% cap on the proportion of total lending for programs; 86 (v) reviewing the graduation policy; and (vi) streamlining sector policies and strategies to transfer all mandatory elements to the ADB safeguard policies. 156. These initiatives are elaborated in Appendix 9, with their anticipated benefits for the urban sector and risks to DMCs and EAs. If the inevitable initial uncertainties regarding the implications and opportunities of the new business model can be overcome within ADB, the benefits have the potential to be substantial for the urban sector projects as well as other sectors. These measures, taken as a whole, would mean a radical departure from previous practices and may help to create new opportunities for increased and more effective lending in the urban sector. These may address some clients complaints, which have, when viewed from the perspective of DMCs, made ADB operations more restrictive such as the increased attention on fiduciary, resettlement, environmental, and indigenous people safeguards. With these new opportunities, ADB will then need to concern itself more with the possibility of ADB crowding out the involvement of local banks in the urban sector. 157. Even when all the changes are fully internalized, the challenge will remain to (i) find a good opportunity for development; (ii) create a consensus around it; (iii) expeditiously design good projects taking safeguards into account; (iv) find good EA staff, ADB staff, contractors, and consultants; (v) adopt an implementation model that is efficient and effective; and (vi) create sustainable results that address the MDGs, and that promote capacity and institutional development. Some of the substantive lessons learned from completed projects and TA of importance to the design of new projects, are summarized in Appendix 10. Appendix 11 summarizes some further lessons from PCRs. VII. EVALUATION AND RECOMMENDATIONS 158. This chapter draws together the findings of the study, and attempts to answer some key questions: (i) Does ADB need to address the issues raised by urbanization and if so, what is so special about them that ADB would need an urban sector strategy? (ii) What would be a sensible delimitation of the urban sector? and (iii) Why is ADB s urban sector program not much larger than it is today? This is followed by a rating of USS and urban sector operations, some suggestions for a new ADB USS, and a set of recommendations. 85 ADB. 2005. A Review of the Policy on Supplementary Financing. Manila. 86 This rule has not been rigidly enforced over the years.

55 A. Key Questions 159. Does ADB Need to Worry about Urbanization? Urbanization is one of the megatrends that is having a profound impact on the Asia and Pacific region. By 2020, the region will look very different than it does today. As described in Chapter 2, the region is in the midst of a major transition from a rural, agriculture dominated society to an urban, industrial/service sector society. Urbanization is progressing at a rapid rate in most DMCs. Urbanization has many positive features. Cities are the centers of economic growth and most of the jobs that are needed to win the war on poverty will be created in cities. However, urbanization presents many challenges that are evident in most of Asia s burgeoning cities, particularly megacities. For example, there are a set of interrelated environmental problems including water scarcity, water pollution, air pollution, and solid waste. The rapidly growing urban populations are placing increasing pressure on overextended infrastructure. This is exacerbated by lack of public and private investment, insufficient municipal finance options, deficient urban management, and corruption. If not managed well, urbanization can breed vast slums with grinding urban poverty, increased crime, and major social inequities. There is an interrelation between public health and public space problems. Water supply and its use produces wastewater, and dense human settlement produces special problems of drainage and solid waste management, which often need to be tackled jointly. Poorly managed solid waste often obstructs drainage and pipe networks, causing flooding and disruption and damaged infrastructure. Public health and public space are affected by squatters and growing slums. Because of the close interrelationship of urban issues, ADB should address the problems from an urban sector perspective rather than a sector-by-sector basis. Addressing urban problems requires a holistic perspective, although effective implementation of solutions simultaneously needs an approach that is as unbundled as possible, since complex, multicomponent projects are difficult to manage and can overburden the institutional and project management capacity of individual agencies. 160. What Could be a Meaningful Domain of the Urban Sector? Although the urban sector is not a typical sector in any functional classification system, it can be made to fit if it refers to the set of interrelated investments that are needed when populations are growing rapidly, and conglomeration results in deteriorating standards in public health, use of public space, and environmental quality of life. The investments would then cover one or more of the following subsectors: (i) water supply; (ii) sanitation (including public and private toilets, sewerage networks, sewage treatment systems, and solid waste management systems); (iii) storm water drainage and floodwater control; (iv) roads and transport and transit related systems; (v) environmentally oriented investments such as increased public space, green environment, and clean air; (vi) municipal facilities occupying key space such as markets, municipal offices, and bus or truck terminals; (vii) slum improvement; (viii) residential and industrial site and service development; (ix) public housing provision and private housing finance; (x) urban renewal, heritage, cultural, and tourism investments; (xi) land management; (xii) public environmental awareness; and (xiii) urban institutional or capacity development. 161. Why is ADB s Urban Sector Program Not Much Larger than it is Today? The answer needs to distinguish issues that are ADB-wide and issues which are specific to the urban sector. ADB-wide issues are summarized in Box 9: Box 9: ADB-Wide Issues Affecting Urban Sector Performance (i) Many DMCs do not wish to significantly expand their external borrowings, given their debt management policies and their fiscal position. As a result, total ADB lending, not that just for the urban sector, has remained constant for a decade.

56 (ii) (iii) (iv) (v) (vi) (vii) (viii) (ix) Unlike at the time ADB was established, many DMCs now have access to the international capital markets and do not need to borrow foreign exchange from ADB. The availability of alternative sources of project finance and foreign currency has grown for some of ADB s major clients (e.g., India, Indonesia, Pakistan, Philippines, and PRC). If ADB s clients feel that the transaction costs of working with ADB are too high, they will seek funds from other sources, even if the associated costs are higher than the financial cost of borrowing from ADB. Some DMCs do not want to increase their borrowings from ADB as they do not believe that ADB offers a suitable range of products or are not convinced of the value added of ADB s involvement. ADB has traditionally been more focused on rural than urban development. Since the bulk of Asia s poor live in rural areas, ADB s overarching objective of poverty reduction may result in focusing ADB s operations on rural areas rather than in the richer cities, even when there is a major investment gap in urban areas and there are growing urban problems. ADB does not have enough resources to take the time needed to identify a large program or to develop leadership in the complex urban sector field. ADB finds it difficult to manage and mitigate the potential risk to its balance sheet of lending without a sovereign guarantee for the urban sector (e.g., subsovereign loans to municipalities or private sector lending). ADB s organizational set-up does not have a clear urban focus, as there are no longer dedicated urban divisions, urban committees, or urban sector classifications. The professional urban sector staff, 16 in all, is scattered across six departments. There has been a turnover of urban specialists and there is limited urban sector expertise located in the resident missions. ADB s project preparation model is not conducive to preparing the type of small projects that individual cities may need. ADB = Asian Development Bank, DMC = developing member country, PRC = People s Republic of China. Source: Special Evaluation Study on Urban Sector Strategy and Operations. 162. Issues specific to the urban sector are summarized in Box 10: Box 10: Issues Specific to the Urban Sector Affecting Performance (i) (ii) (iii) (iv) (v) (vi) (vii) Some developing member countries have embarked on decentralization programs. In this context, some governments do not wish to incur national debt for problems specific to local governments; in a decentralized system, urban problems should be solved and paid for by local governments themselves. Many national governments have relatively weak urban strategies. Some of the main components of the urban sector (e.g., wastewater collection, urban roads, and solid waste collection) have traditionally been provided for free or at a subsidized cost to urban populations. Changing attitudes to support cost recovery is a major challenge. If the cost recovery issue cannot be addressed, the likelihood of major ADB lending is limited. Problems of full cost recovery also apply to water supply, although lifeline tariffs and the use of water meters should allow appropriate user cost systems to be developed. Some urban sector projects, especially water supply and sewage/waste management, require significant technical expertise in operation and maintenance. Problems may occur if there is insufficient technical expertise at the local level. Urban sector projects are often in need of special environmental and involuntary resettlement safeguard activities, which make the preparation and implementation process more challenging. Many urban sector projects use a multisector approach and include components in several subsectors. The resulting project complexity creates challenges for both the executing agencies and ADB, which are not always successfully overcome, resulting in less than optimal project outcomes. Urban sector projects, especially those covering many cities, are scattered geographically, are sometimes affected by local politics, tend to have hundreds of contracts, and require a greater than usual input from ADB in terms of their preparation and implementation.

57 (viii) Many executing and implementing agencies in the urban sector report to both national bodies and locally elected politicians. This can lead to delays and a greater than normal need for coordination and consensus building. ADB = Asian Development Bank. Source: Special Evaluation Study on Urban Sector Strategy and Operations. 163. In spite of these issues, which have affected the portfolio growth in the urban sector that was expected in the USS, this study assesses that there is good potential to substantially increase ADB s involvement in the medium and long term with proactive urban sector planning, supported by a revised USS, changes in ADB s business model and products, and some organizational adjustments. The challenge is for ADB to successfully grasp this opportunity. In the MTS II, the urban sector is rated as a high priority sector in which ADB s operations are expected to grow. B. Rating of the Urban Sector Strategy and Urban Sector Operations 164. This section assesses the USS primarily in terms of its relevance and efficiency, and urban sector operations in terms of their effectiveness and sustainability. 165. Relevance of the USS. The USS was highly relevant in terms of highlighting urbanization in Asia as one of the mega-trends, and in identifying the need for ADB to focus on a response to the phenomenon as such, through a strategy on its own. All of the trends and issues identified by the USS are still relevant in 2005, although some could be prioritized and a few new issues could be added. The strategy appropriately anticipated ADB s increasing focus on poverty reduction. USS 1999 was consistent with ADB s subsequent 2001 Long-Term Strategic Framework (2001 2015), which reflected urbanization as a major challenge for ADB, and one which ADB was particularly suited to address. However, the USS had little effect on the 2001 Medium-Term Strategic Framework (2001 2005), which hardly mentioned the USS or urban sector priorities. The USS s objective of increasing lending to urban areas was highly relevant, given ADB s comparative strengths as an organization promoting infrastructure and capacity development. The strategy could have been more relevant as a guide to current operations if it had made more distinction between regions and countries within Asia in terms of urbanization, needs, and approaches; and if it had been more specific as to harmonization with those of other major funding agencies and development partners. Overall, the USS is assessed as relevant. 166. Efficiency of the USS in Directing Urban Operations. The number of goals and targets make the USS hard to pursue in practical terms and hard to evaluate. After ADB s reorganization, the USS was even harder to pursue; urban sector staff was scattered and the new RSDD, as a custodian of USS, was not staffed with sufficient specialists to play a leading role and to coordinate effectively. The USS was not successful in providing a statement that focused ADB as an institution on the urban sector. ADB s subsequent water policy captured some of the attention, but this had a much wider coverage than the urban sector, and the relation of the water policy with the urban sector was not fully addressed. Individual country urban sector profiles and strategies have generally not resulted in a clear urban focus in CSPs. ADB s TA for the urban sector diminished. ADB s 2002 reorganization and the accompanying new business processes led to much greater attention on country planning rather than sector planning. ADB s leadership and institutional focus on the urban sector, widely acknowledged in the 1990s, was due to the pre-2002 division-centered organization. Two divisions were explicitly focused on urban development. After the 2002 reorganization, ADB focus on the urban sector

58 dissipated. The urban sector does not perform more efficiently in terms of the preparation and implementing urban projects than before the USS; there are as many delays and loan cancellations as before, or as in many other projects. Borrowers perceive ADB s transaction costs as high, although many executing agencies believe that there is significant added value. Since ADB no longer has specialized urban divisions and has few urban sections in RMs, there is a lack of focal point for urban clients. This has led to a lack of lobby to promote ADB involvement in the urban sector. Overall, the USS is assessed as having been less efficient in directing a program of urban operations. 167. Effectiveness of Urban Sector Operations. Urban sector lending operations increased but not significantly. Intended expansions of operations into public private sector partnerships, urban (mass) transit, and land management were not achieved, although there was some expansion in the housing sector. The number of program loans and sector loans did not grow as planned. TA resources went down. Economic and sector work declined. No more large urban forums were organized by ADB. However, PCRs have generally rated urban sector projects higher than the average of the other projects in ADB, especially integrated urban development (multisector) projects. OED ratings of completed urban sector projects have been in the middle range. Some sectors have performed better (e.g., transport and energy) and some have performed worse (e.g., agriculture and lines of credit to government-owned financial institutions). Given a reduced TA program in the last years, and no growth in loan consulting services and loan components for institutional development or training, it is unlikely that the policy dialogue or capacity impact of current operations is higher than before the USS. Increased cofinancing sources may have partly offset this trend. The USS did not specify outcomes across Asia and these are hard to quantify. To give an indication of the coverage of ADB s lending operations: areas with ADB-supported urban sector projects in the last 12 years covered 12% of the DMCs urban population. Overall, ADB s operations in the post-uss period are assessed as effective. 168. Sustainability of Urban Sector Operations. The sustainability of the urban sector program varies with the subsector and country, in line with country contexts. OED evaluations have given lower ratings than PCRs to the sustainability of urban sector projects and impact on institutions. Some progress has been registered over the years with appropriate tariff setting and cost recovery in water supply and sewage systems, but practically all systems constructed still need government subsidies, and are therefore less sustainable. The housing finance subsector program, relying on financial intermediaries which generally significantly increase the cost of housing loans to final borrowers, does not seem to be sustainable given the increasing availability of competing private sector sources and lower commercial interest rates. The sustainability of the urban sector within ADB is low, given its incompatibility with the present organizational structure. As ADB no longer has urban divisions, there is a lack of a custodian for the strategy. Expansion in lending, TA, and knowledge products would normally need to be accompanied by appropriate levels of staffing. Given that ADB currently has a cap on increased levels of staffing, this would mean a redeployment of existing positions. From the side of ADB s clients, demand for urban sector projects continues, but an exponential rise is not foreseen, partly because of external debt considerations for many DMCs. ADB has yet to decide on a strategy for significantly increasing subsovereign lending. Overall, urban sector operations are assessed as less sustainable. 169. Overall, ADB has failed to seize the opportunity to significantly increase the size and quality of its urban portfolio. Although ADB operations are rated as broadly successful, the USS is rated as partly successful.

59 C. Suggestions for a New Urban Sector Strategy 170. Nature of the New Strategy. The new urban sector strategy will be more effective if it (i) is consistent with ADB s long-term or medium-term strategic frameworks; (ii) sets clear, monitorable targets by regional department that are consistent with reasonable assumptions about human resources and lending and TA volumes; (iii) has an overall custodian department to guide the process; (iv) is more integrated with (other) sectoral strategies, such as ADB s water policy; (v) identifies focal issues and opportunities for policy dialogue and capacity development; (vi) guides decision making on the type of projects to be pursued proactively; (vii) identifies the intended strategic partnerships with national and international agencies; and (viii) is reviewed every 5 years. If the new strategy is not consistent with ADB s strategic frameworks, does not have a custodian division, and does not set clear targets, monitorable for this division, then it is unlikely to be operationally useful within ADB. Its main use would likely be as a public relations document, much like the USS of 1999. Given that the MTS II identified urban sector operations as a priority, the challenge will be to align ADB s human and budgetary resources so that this vision is realized. 171. Substance of the New Strategy. In terms of its substance, many of the tenets of USS 1999 remain valid in OED s view, such as increased focus on urban transport, urban finance, and housing. The new strategy should identify ways to facilitate a significant growth in urban financing, given that there is in principle a vast financing gap in the urban sector in Asia, and that the MTS II identified the urban sector as a sector in which ADB s operations should grow. The strategy should specifically address new opportunities offered as a result of the new changed lending rules, modalities, and business processes. The new USS should identify ways to mobilize private sector investment and develop public private partnerships, and develop strong risk mitigation measures for subsovereign loans. The new urban sector strategy should address recent international concerns about the harmonization of the programs, policies, and procedures of various external agencies; government ownership of the projects; anticorruption measures; and capacity depletion as a result of substitution of regular functions of agencies by project staff. The new strategy may also need to pay more attention to disaster prevention and mitigation safeguards (e.g., earthquakes and flooding). 172. Need for Alignment of the TA Program with the Lending Operations. ADB has multiple objectives and areas of focus. Human resources and TA grant funds are limited. A previous OED recommendation, that the TA program should be aligned as much as possible with the ongoing and pipeline loan program, remains valid. The TA program should be focused on policy dialogue; demand surveys for project identification; project preparation; and, within the context of ongoing lending operations, on capacity development and policy reform. Optimal use should be made of urban research conducted by other organizations. RSDD and ADBI could play a lead role in channeling relevant research findings to operational divisions. 173. Urban Sector Projects Need to Respond to Demand. An integrated multisector approach should not be ruled out but, given the administrative challenges of complex projects, urban interventions may need to start small, develop city development strategies focusing on investment plans and pricing policies, and gradually expand in investments over the years. The urban multisector projects are good candidates for the new modality of the multitranche financing facility. 174. Single City or Multicity Projects. Policy dialogue with the DMCs should determine whether ADB should focus on megacities and city regions rather than the more scattered secondary cities in a particular country. Analysis of urbanization patterns should inform such

60 dialogue, and later be reflected in the country strategies. Multicity projects often require more transaction costs, both for ADB and for the cities themselves. The cities usually have to rely on a central borrower, which relends against more unfavorable conditions than when ADB directly lends to a large city or utility. PCRs and PPARs point to higher success rates for single city projects, whether multisector or not. Multicity projects, especially those with a sector modality and for smaller cities, have often been products of convenience given ADB s project preparation procedures. Now that the procedures have changed, more can be prepared under the multitranche financing facility. This should improve the quality of subproject appraisal and reduce transaction costs and commitment fees. 175. Need for Complementary Country-Specific Urban Sector Strategies. Now that ADB has chosen to focus on the urban sector, regularly updated country strategies based on policy dialogue and demand surveys will be needed. The urban sector operations should reflect country specific and localized social, institutional, and economic conditions. From ADB s perspective, the urban sector is labor intensive because investments are dispersed either across many components in few cities or few components in many cities. In such circumstances, it is usually efficient to delegate project administration to RMs. D. Recommendations for ADB 176. In light of the findings, the following recommendations are made for ADB consideration. Recommendation Responsibility 1. ADB Should Formulate an Urban Sector Strategy II. A greater focus on RSDD; with Asia s urban sector is appropriate, given the trends in urbanization in the inputs from region, the size and complexity of the sector s needs, its potential for operations significant investment, and ADB s role as a premier development departments; institution funding physical infrastructure and providing needed technical Human expertise and capacity development. The MTS II identified the urban Resources infrastructure as one of ADB s core operational sectors in which operations Division (BPHR) are expected to grow and ADB will build up a critical mass of expertise to for internal act as a leading provider of assistance, both financial and expertise. organization Preparation of the Urban Sector Strategy II, now deferred to the fourth quarter of 2007, should be accelerated. 2. The Urban Sector Strategy II Should Include a Road Map or Business RSDD to Plan for Implementation to Translate the Broad Goals and Objectives coordinate with of the Strategy into Reality, Given Various Constraints. This should set inputs of out the assumptions about the resources (staffing; TA and expected urban operations lending levels), the role of RSDD, sector divisions and resident missions, departments, and plans to address the organizational and staffing impediments that are including restricting the significant growth of ADB s urban portfolio. BPMSD 3. The Urban Sector Strategy II Should Also Include a Results-Based RSDD, with Framework with Clear, Monitorable Indicators that is consistent with inputs from assumptions made about resources, so that Management can monitor operations progress on its implementation and take corrective action as may be departments; required. Such frameworks have been absent from past ADB strategies. SPD for ADB s sector and thematic classifications should be reviewed with respect classification to their coverage of the urban sector. issue

61 Recommendation Responsibility 4. The New Strategy Needs to be Grounded in a Custodian Division in RSDD to RSDD, to Coordinate Across Departments, Monitor Targets and coordinate; Elaborate Common Themes. The custodian division needs to promote Operations learning across regional departments and prioritize the use of TA funds to Departments to address strategic issues that are common across a number of regions. It consider; may need to be assisted by some in-house consultants for a defined Human period of time. In those regional departments with significant urban sector Resources operations, the strategy should be guided by CSP priorities and Division (BPHR) implemented by divisions that are geared more to the urban sector than for internal currently is the case, with the exception of the newly created water and organization urban development division in the South Asia Regional Department. Given that ADB has a cap on increased levels of staffing, an expansion of staff in the urban sector would have to come from gradual redeployment of existing positions in MTS II s low priority sectors. Focal points should be established in RMs in those countries with a promising urban portfolio. More local staff needs to be employed in RMs for DMCs where the CSP indicates a significant portfolio, as urban sector projects are demanding in terms of processing and administration. The current organizational structure and resource commitment do not seem appropriate for the sector, given its classification in MTS II. 5. ADB Needs to Provide Some Guidance to Staff on How to Engage in Anticorruption Initiatives in the Urban Sector. Although some corruption risks are common across sectors (e.g., procurement), other types of corruption vary by sector. ADB should develop a standard check list of types of corruption that may occur at the various stages in the project cycle for urban projects, together with possible risk mitigation measures and entry points to raise and address the issue during policy dialogue, project processing, and project implementation. RSDD and the Integrity Division ADB = Asian Development Bank, BPMSD = Budget, Personnel, and Management Systems Department, CSP = country strategy and program, DMC = developing member country, MTS = Medium-Term Strategy, RSDD = Regional and Sustainable Development Department, SPD = Strategy and Policy Department, TA = technical assistance.

62 Appendix 1 METHODOLOGY OF THE QUESTIONNAIRE SURVEY OF ONGOING URBAN SECTOR PROJECTS 1. In May 2005, a list of all 54 approved loans pertaining to urban sector projects supported by the Asian Development Bank (ADB) loans was compiled. Project officers from ADB headquarters and resident missions were informed of the survey and were requested to provide their contact information. As some project officers believed that their respective projects had been very recently established, the total number of urban projects included was reduced to 51. Later on, one more project from India (Gujarat Earthquake Rehabilitation and Reconstruction Project) was included in the survey as it had a very sizeable urban sector component in it, leading to a total of 52 projects, from which a response was requested. 2. The questionnaire consisted of 19 questions divided into three parts: multiple choice questions on the project, its project office, and capacity development. The answers to most questions were suggested for consistency, and they could be ticked in boxes (multiple choice format). 3. Between June and July 2005, the questionnaires were sent by e-mail and fax, together with a brief explanatory letter about the survey background and objectives, including a request for the project directors to forward copies of the questionnaire to other project offices and/or project implementation units involved in the project. In some cases, questionnaires were compiled by the main project office before they were returned to the survey team. Follow-up calls and e-mails were made to project offices that did not respond. The total number of projects that responded with at least one questionnaire return was 44 (85%) while the total number of project offices responding was 88 (75%) out of a total of 117 questionnaires circulated to known project offices by either the study team or the main project office of their own accord. Details on the origin of the responses are shown in Table A1.1. A complete list of the projects surveyed is in Table A1.2. Table A1.1: Projects Ongoing in June 2005, and Response to Questionnaire Survey Country No. of Projects Projects with Response Total Respondents Azerbaijan 1 1 1 Bangladesh 1 1 1 Bhutan 1 1 1 Cambodia 2 2 3 Cook Islands 1 0 0 Fiji Islands 1 1 1 India 7 7 10 Indonesia 2 2 2 Kiribati 1 1 1 Kyrgyz Republic 1 0 0 Lao People s Dem. Rep. 4 3 17 Mongolia 3 3 3 Nepal 4 3 3 Pakistan 2 1 1 Papua New Guinea 1 1 1 Philippines 6 6 13 PRC 5 5 12 Samoa 1 0 0 Sri Lanka 3 2 5

Appendix 1 63 Country No. of Projects Projects with Response Total Respondents Uzbekistan 1 0 0 Viet Nam 4 4 13 Total 52 44 88 PRC = People s Republic of China. Source: Special Evaluation Study on Urban Sector Strategy and Operations. 4. Of the 21 countries covered by the survey, the highest number of responses came from the Lao People s Democratic Republic, where one project submitted 14 questionnaires from its project implementation units and other project offices. The good response rate from India, People s Republic of China, and Philippines was in part due to the follow-up and interviews by the study teams for the country case studies. Some projects and/or respondents had difficulty communicating with the survey team during distribution of the questionnaires. In the case of Viet Nam projects, questionnaires were coursed through the resident mission office for translation and facilitation of responses. Communication problems were also evident in the answers provided by some respondents. Some questions had to be left blank and some answers needed to be validated with information from the projects reports and recommendations of the President. This led to variable numbers of valid responses, which has been taken into account in the statistical analysis (averages are drawn only from valid responses). Aside from the language barrier, lack of internet access in the project offices prevented some of the respondents from answering completely as some questions from faxed questionnaires were no longer readable. 5. The analysis of the response was done in two ways: (i) by main offices, i.e., by project directors in the main executing agencies (44); and (ii) by all respondents, including main and subsidiary offices (88). The choice of the type of respondent depended on the question, but all questions reflecting perceptions were analyzed from the perspective of the replies by all respondents. Given the limited number of questionnaires, simple cross-tabulations and tables with means were used as the main sources for the statistical analysis.

64 Appendix 1 Table A1.2: List of Projects Surveyed Questionnaires Country/Loan No./Project Title Returned Azerbaijan 1. 2119/20 Urban Water Supply and Sanitation Project 1 Bangladesh 2. 1947 Urban Governance and Infrastructure Improvement Project 1 Bhutan 3. 1625 Urban Infrastructure Improvement Project 1 Cambodia 4. 1725/2013 Provincial Towns Improvement Project 2 5. 1969 GMS: Mekong Tourism Development Project 1 Cook Islands 6. 1832 Waste Management Project 0 Fiji Islands 7. 2055 Suva Nausori Water Supply & Sewerage Project 1 India 8. 1647 Rajasthan Urban Infrastructure Development Project 3 9. 1704 Karnataka Urban Development and Coastal Management 2 10. 1720 Urban and Environmental Infrastructure Facility 1 11. 1759/61 Housing Finance II Project 1 12. 1813 Calcutta Environmental Improvement Project 1 13. 1826 a Gujarat Earthquake Rehabilitation and Reconstruction Project 1 14. 2046 Urban Water Supply and Environmental Improvement in Madhya Pradesh 1 Indonesia 15. 1587 Metropolitan Medan Urban Development Project 1 16. 2072/73 Neighborhood Upgrading and Shelter Sector Project 1 Kiribati 17. 1648 Sanitation, Public Health, and Environmental Improvement Project 1 Kyrgyz Republic 18. 1742 Community-Based Infrastructure Services Sector Project 0 Lao People s Democratic Republic 19. 1525 Secondary Towns Urban Development 1 20. 1710 Water Supply and Sanitation Sector Project 14 21. 1834 Vientiane Urban Infrastructure Services Project 0 22. 1994 Small Towns Development Project 2 Mongolia 23. 1736 Cadastral Survey and Land Registration Project 1 24. 1847 Housing Finance (Sector) Project 1 25. 1907 Integrated Development of Basic Urban Services in Provincial Towns 1 Nepal 26. 1755 Small Towns Water Supply and Sanitation Sector Project 0 27. 1820 Melamchi Water Supply Project 1 28. 1966 Urban and Environmental Improvement Project 1 29. 2058/59 Integrated Development of Basic Urban Services in Provincial Towns 1 Pakistan 30. 1854 North-West Frontier Province Urban Development Sector Project 1 31. 2060/61 Southern Punjab Basic Urban Services Project 0 Papua New Guinea 32. 1812 Provincial Towns Water Supply and Sanitation Project 1 Philippines 33. 1658 Clark Area Municipal Development Project 5 34. 1665 Metro Manila Air Quality Improvement Project 3 64

Appendix 1 65 Questionnaires Country/ Loan No./ Project Title Filled Out 34. 1746 Pasig River Environmental Management and Rehabilitation Sector 1 Development Program 35. 1843 Mindanao Basic Urban Services Sector Project 2 36. 2012 MWSS New Water Source Development Project 1 37. 2063 Development of Poor Urban Communities Sector Project 1 PRC 39. 1692 Suzhou Creek Rehabilitation Project 1 40. 1797 Tianjin Wastewater Treatment and Water Resources Project 3 41. 1985 Hebei Province Wastewater Management Project 6 42. 1995 Harbin Water Supply Project 1 43. 1996 Wuhan Wastewater Management Project 1 Samoa 44. 2026 Sanitation and Drainage Project 0 Sri Lanka 45. 1575 Third Water Supply and Sanitation Project 1 46. 1632 Urban Development and Low Income Housing Sector Project 0 47. 1993 Secondary Towns and Rural Community-based Water Supply and Sanitation Project 4 Uzbekistan 48. 1842 Urban Water Supply Project 0 Viet Nam 49. 1514 Second Provincial Towns Water Supply and Sanitation Project 1 50. 1880 Third Provincial Towns Water Supply and Sanitation Project 1 51. 1702 Ho Chi Minh City Environmental Improvement Project 1 52. 2034 Central Region Urban Environmental Improvement Project 10 Total 88 GMS = Greater Mekong Subregion, MWSS = Metropolitan Waterworks and Sewerage System, PRC = People s Republic of China. Note: Loan 1990-VIE credit facility has not yet commenced; Loan 2117-BAN is in the process of setting up project offices. a Loan 1826-IND was previously not included in the list of ongoing urban sector loans. Source: Questionnaire survey of all urban sector projects ongoing by June 2005, for this study.

URBAN SECTOR PROJECTS APPROVED FROM JULY 1993 TO JUNE 2005 REPORTS AND RECOMMENDATIONS OF THE PRESIDENT FOR URBAN SECTOR PROJECTS BETWEEN MID-1993 AND MID-2005 No. Loan Number Country Project Title Project Cost Estimate ($ million) Loan ($ million) Date Approved Pre-USS (July 1993 to June 1999) Water Supply, Sanitation, and Solid Waste Management 1. 1250 RMI Majuro Water Supply Project 0.80 0.70 09-Sep-93 2. 1260 PAK Urban Water Supply and Sanitation Project 95.80 72.00 04-Nov-93 3. 1264 BAN Second Water Supply and Sanitation Project 43.00 31.00 16-Nov-93 4. 1267 LAO Northern Provincial Towns Water Supply and Sanitation Project 16.30 13.00 18-Nov-93 5. 1269 PHI Municipal Water Supply 72.00 43.20 25-Nov-93 6. 1273 VIE Ho Chi Minh City Water Supply and Sanitation Rehabilitation Project 81.00 65.00 29-Nov-93 7. 1313 PRC Dalian Water Supply Project 379.70 160.00 20-Sep-94 8. 1326 THA Chonburi Water Supply Project 65.00 38.50 18-Oct-94 9. 1361 VIE Provincial Towns Water Supply and Sanitation Project 82.50 66.00 17-Aug-95 10. 1389 RMI Majuro Water Supply and Sanitation Project 11.60 9.20 29-Sep-95 11. 1459 FSM Water Supply Sanitation Project 13.30 10.60 19-Sep-96 12. 1468 CAM Phnom Penh Water Supply and Drainage Project 35.80 20.00 26-Sep-96 13. 1472 PHI Small Towns Water Supply Sector Project 83.00 50.00 30-Sep-96 14. 1490/91 PRC Anhui Municipal Wastewater Treatment Project 336.00 140.00 26-Nov-96 15. 1514 VIE Second Provincial Towns Water Supply and Sanitation Project 92.00 69.00 27-Feb-97 16. 1527 INO Capacity Building of Water Supply Enterprise for Water Loss Reduction 119.40 66.00 17-Jul-97 17. 1528 THA Nong Khai-Udon Thani Water Supply Project 85.50 50.00 21-Aug-97 18. 1539 PAK Korangi Wastewater Management Project 101.00 70.00 18-Sep-97 19. 1544 PRC Zhejiang-Shanxi Water Supply Project (Phase I) 518.00 100.00 24-Sep-97 20. 1560 MON Provincial Towns Basic Urban Services Project 8.50 6.80 30-Sep-97 21. 1575 SRI Third Water Supply and Sanitation Sector Project 121.00 75.00 06-Nov-97 22. 1636 PRC Fuzhou Water Supply and Wastewater Treatment Project 192.20 102.00 30-Sep-98 23. 1640 NEP Melamchi Water Supply (Engineering) Project 6.75 5.00 10-Nov-98 24. 1648 KIR Sanitation, Public Health, and Environment Improvement 12.80 10.24 08-Dec-98 Subtotal 2,572.95 1,273.24 66 Appendix 2 66

No. Loan Number Country Project Title Multisector Project Cost Estimate ($ million) Loan ($ million) Date Approved 25. 1240 NEP Kathmandu Urban Development Project 0 12.00 29-Jun-93 26. 1292 INO Eastern Islands Urban Development Sector Project 142.00 85.00 21-Dec-93 27. 1362 LAO Vientiane Integrated Urban Development Project 27.70 20.00 17-Aug-95 28. 1367 PHI Regional Municipal Development Project 64.20 30.00 30-Aug-95 29. 1376 BAN Secondary Towns Infrastructure Development Project II 82.90 65.00 19-Sep-95 30. 1383/84 INO Sumatra and West Java Urban Development Sector Project 334.00 200.00 26-Sep-95 31. 1410 THA Samut Prakarn Wastewater Management Project 687.00 a 230.00 b 7-Dec-95 32. 1415/16 IND Karnataka Urban Infrastructure Development Project 132.00 105.00 14-Dec-95 33. 1448 VAN Urban Infrastructure Project 12.80 10.00 27-Jun-96 34. 1511 INO Metropolitan Bogor, Tangerang & Bekasi Urban Development (Sector) Project 228.00 80.00 19-Dec-96 35. 1525 LAO Secondary Towns Urban Development Project 38.50 27.00 26-Jun-97 36. 1587 INO Metropolitan Medan Urban Development Project 198.20 116.00 08-Dec-97 37. 1599 PHI Subic Bay Area Municipal Development Project 37.00 22.00 19-Dec-97 38. 1625 BHU Urban Infrastructure Improvement Project 8.20 5.70 30-Jul-98 39. 1632 SRI Urban Development and Low Income Housing (Section) Project 100.00 70.00 24-Sep-98 40. 1647 IND Rajasthan Urban Infrastructure Development Project 362.00 250.00 03-Dec-98 41. 1658 PHI Clark Area Municipal Development Project 41.80 24.30 15-Dec-98 Subtotal 1,809.30 1,352.00 Others 42. 1323 PAK Sukkur Bridge 45.00 29-Sep-94 43. 1549/1550/ IND Housing Finance Project (NHB/HUDCO/IDFC) 1,150.00 300.00 25-Sep-97 1551 44. 1663 PHI Metro Manila Air Quality Improvement 200.00 03-Dec-98 1664 PHI Metro Manila Air Quality Improvement 22.87 1665 PHI Metro Manila Air Quality Improvement 71.00 45. 1572 INO Capacity Building in Urban Infrastructure Management Project 70.00 42.00 04-Nov-97 Subtotal 1,220.00 680.87 Total for all Sectors Pre-USS 5,602.25 3,306.11 Appendix 2 67

No. Loan Number Country Project Title Post-USS (July 1999 to June 2005) Project Cost Estimate ($ million) Loan ($ million) Date Approved Water Supply, Sanitation, and Solid Waste Management 1. 1692 PRC Suzhou Creek Rehabilitation Project 876.00 300.00 29-Jun-99 2. 1710 LAO Water Supply and Sanitation Sector Project 25.00 20.00 16-Nov-99 3. 1725 CAM Provincial Towns Improvement Project 26.30 20.00 17-Dec-99 4. 1755 NEP Small Towns Water Supply Sanitation Sector Project 53.90 35.00 12-Sep-00 5. 1812 PNG Provincial Towns Water Supply and Sanitation Project 23.30 15.34 14-Dec-00 6. 1820 NEP Melamchi Water Supply Project 464.00 120.00 21-Dec-00 7. 1832 COO Waste Management Project 3.10 2.20 17-Jul-01 8. 1842 UZB Urban Water Supply Project 65.50 36.00 27-Sep-01 9. 1880 VIE Third Provincial Towns Water Supply and Sanitation Project 98.00 60.00 13-Dec-01 10. 1907 MON Integrated Development of Basic Urban Services In Provincial Terms 26.50 20.10 06-Aug-02 Project 11. 1985 PRC Hebei Province Wastewater Management Project 165.34 82.36 19-Dec-02 12. 1993 SRI Secondary Towns & Rural Community Based Water Supply and 100.00 60.29 16-Jan-03 Sanitation 13. 1995 PRC Harbin Water Supply 399.50 100.00 11-Mar-03 14. 1996 PRC Wuhan Wastewater Management Project 193.60 83.00 25-Apr-03 15. 2012 PHI MWSS New Water Source Development Project 4.97 3.26 14-Oct-03 16. 2013 CAM Provincial Towns Improvement Project Supply 6.26 28-Oct-03 17. 2034 VIE Central Region Urban Environmental Improvement Project 96.00 44.00 08-Dec-03 18. 2046 IND Urban Water Supply and Environmental Improvement in Madhya Pradesh 303.50 200.00 12-Dec-03 19. 2055 FIJ Suva-Nausori Water Supply and Sewerage Project 72.40 47.00 18-Dec-03 20. 2058/59 NEP Kathmandu Valley Water Services Sector Development Program/Project 15.00 15.00 18-Dec-03 21. 2119/20 AZE Urban Water Supply and Sanitation Project 39.90 30.00 07-Dec-04 Subtotal 3,051.81 1,299.82 Multisector 22. 1702 VIE Ho Chi Minh City Environmental Improvement Project 100.000 70.00 07-Oct-99 23. 1704 IND Karnataka Urban Development and Coastal Environmental Management 251.400 175.00 26-Oct-99 24. 1719/1720/ 1721 IND Urban and Environmental Infrastructure Facility Project (HUDCO/ICICI/IDFC) 200.00 17-Dec-99 68 Appendix 2

Loan Number Country Project Title Project Cost Estimate ($ million) Loan ($ million) Date Approved 25. 1742 KGZ Community-Based Infrastructure Services Sector Project 45.00 36.00 08-Jun-00 26. 1797 PRC Tianjin Wastewater Treatment and Water Resources Protection Project 340.70 130.00 11-Dec-00 27. 1813 IND Calcutta Environmental Improvement Project 360.00 250.00 19-Dec-00 28. 1834 LAO Vientiane Urban Infrastructure and Services Project 37.00 25.00 23-Aug-01 29. 1843 PHI Mindanao Basic Urban Services Sector 60.00 30.00 27-Sep-01 30. 1854 PAK North-West Frontier Province Urban Development Sector Project 30.00 20.80 08-Nov-01 31. 1947 BAN Urban Governance and Infrastructure Improvement (Sector) Project 87.00 60.00 28-Nov-02 32. 1966 NEP Urban and Environmental Improvement Project 37.50 30.00 10-Dec-02 33. 1994 LAO Small Towns Development Sector Project 20.00 16.00 28-Jan-03 34. 2026 SAM Sanitation and Drainage Project 10.00 8.00 27-Nov-03 35. 2060/61 PAK Southern Punjab Basic Urban Services Project 128.60 90.00 18-Dec-03 36. 2063 PHI Development of Poor Urban Communities Sector Project 48.80 30.50 18-Dec-03 37. 2072/73 INO Neighborhood Upgrading and Shelter Sector Project 126.50 88.60 19-Dec-03 38. 2117 BAN Secondary Towns Integrated Flood Protection Project Phase 2 128.88 80.00 02-Dec-04 39. 2151 IND Multisector Project for Infrastructure Rehabilitation in Jammu and Kashmir 358.00 250.00 21-Dec-04 Subtotal 2,169.38 1,589.90 Others 40. 1745/1746 PHI Pasig River Environmental Management and Rehabilitation Sector 150.00 175.000 20-Jul-00 Development Program 41. 1759/61 IND Housing Finance II Project 517.00 120.00 21-Sep-00 42. 1847 MON Housing Finance Sector Loan 24.00 15.00 18-Oct-01 43. 1990 VIE Housing Finance Project 51.80 30.00 20-Dec-02 44. 2024 PRC Xi'an Urban Transport 762.00 270.00 27-Nov-03 Subtotal 1,504.80 610.00 Total for all Sectors Post-USS 6,725.99 3,499.72 Total for all Sectors in the Study Period 12,328.24 6,805.83 AZE = Azerbaijan, BAN = Bangladesh, BHU = Bhutan, CAM = Cambodia, COO = Cook Islands, FIJ = Fiji Islands, FSM = Federated States of Micronesia, HUDCO = Housing and Urban Development Corporation Limited, ICICI = ICICI Limited, IDFC = Infrastructure Development Finance Company Limited, IND = India, INO = Indonesia, KGZ = Kyrgyz Republic, KIR = Kiribati, LAO = Lao People s Democratic Republic, MON = Mongolia, MWSS = Metropolitan Waterworks and Sewerage System, NEP = Nepal, NHB = National Housing Bank, No. = number, PAK = Pakistan, PHI = Philippines, PNG = Papua New Guinea, PRC = People s Republic of China, RMI = Republic of the Marshall Islands, SAM = Samoa, SRI = Sri Lanka, THA = Thailand, USS = Urban Sector Strategy, UZB = Uzbekistan, VAN = Vanuatu, VIE = Viet Nam. a At appraisal, total project cost was estimated at $507.00 million. At the time of the approval of the supplementary loan, estimated project cost had increased to $687.00 million). b Includes a supplementary loan (Loan 1646-THA) amounting to $80.00 million, approved 3 December 1998. Note: The actual period periods compared are from 28 June 1993 to 27 June 1999 (pre-uss) and 28 June 1999 to 27 June 2005 (post-uss). Source: Asian Development Bank s internal databases. Appendix 2 69

70 Appendix 3 LIST OF ALL URBAN SECTOR RELATED TECHNICAL ASSISTANCE GRANTS Table A3.1: Advisory Technical Assistance Grants for the Urban Sector From Mid-1993 to Mid-2005 Year No. Country Advisory Technical Assistance Total ($) 1993 1916 PRC Institutional Strengthening of the Environmental Protection Bureaus in Tangshan 450,000 and Chengde Municipalities 1993 1958 MAL Urban Transport Planning 600,000 1993 1980 BAN Institutional Strengthening of Pourashavas for Urban Water Supply and Sanitation 450,000 Services 1993 1999 VIE Institutional Strengthening of HCMC Water Supply Company 600,000 1993 2000 VIE HCMC Water Supply Master Plan 600,000 1993 2012 BAN Khulna-Jessore Drainage Rehabilitation 920,000 1993 2016 INO Private Sector Participation in Urban Development (Bandung and Semarang) 600,000 1993 2015 PRC Urban Environmental Improvement Planning 480,000 1993 2040 VIE Ho Chi Minh City Environmental Improvement Planning 600,000 1994 2098 IND Urban Sector Profile 400,000 1994 2148 VIE Urban Sector Strategy Study 300,000 1994 2198 KIR Integrated Urban Plans and Program Study 385,000 1994 2202 IND Capacity Building for Improved Infrastructure Dev. in Selected Municipality in 600,000 Karnataka State 1994 2212 PRC Beijing Urban Transport 715,000 1994 2211 PRC Capacity Building of the Beijing Industrial Hazardous Waste Management Center 425,000 1994 2210 PRC Capacity Building of the Beijing Municipal Environmental Protection Bureau and 600,000 Affiliated Agencies 1994 2246 BAN Study of Urban Poverty in Bangladesh 190,000 1994 2254 PHI MWSS Operational Strengthening Study 600,000 1995 2281 CAM Urban Development Strategy Study 500,000 1995 2302 PRC Symposium on Urban Transport 100,000 1995 2319 TUV Urban Planning and Environment Management 310,000 1995 2368 IND Institutional Strengthening of Karnataka Urban Infrastructure Finance Corporation 100,000 1995 2376 VIE Community Environmental Health Improvements for the Provincial Towns 500,000 1995 2401 PHI MWSS Privatization Support 582,000 1995 2456 PRC Pilot Environmental Plans for Selected Medium-Size Cities 537,000 1995 2471 IND Resource Mobilization Study for Local Governments in Karnataka 300,000 1995 2502 PHI Private Sector Participation in Urban Development 500,000 1996 2596 VAN Urban Growth Management Strategy for Port Vila 600,000 1996 2639 PAK Urban Institutional Strengthening 100,000 1996 2729 PRC Industrial Pollution Investigation and Assessment in Town and Village Enterprises 600,000 1997 2751 PRC Capacity Building of Wastewater Treatment Operations in Anhui Province 400,000 1997 2805 INO Strengthening of Urban Waste Management Policies and Strategies 600,000 1997 2804 PRC Build-Operate-Transfer (BOT) Chengdu Water Supply 600,000 1997 2837 INO Capacity Building for Private Sector Participation in Urban Development 850,000 1997 2881 MON Capacity Building for the Provision of Urban Services in Provincial Towns 825,000 1997 2906 PRC Leadership Training on Urban Environmental Management in Key Cities 600,000 1997 2959 VIE Preparation of a Development Strategy for the Central Region of Viet Nam 898,000 1997 2972 LAO Support for Urban Development Administration Authorities 500,000 1997 2893 PRC Hohhot Urban Development 430,000 1997 2833 IND Strengthening Housing Finance Institutions 600,000 1998 2893 PRC Hohhot Urban Development (Supplementary) 150,000 1998 2726 PRC Water Quality Management Planning for Suzhou Creek (Supplementary) 400,000 1998 2998 NEP Urban Water Supply Reforms in Kathmandu Valley 800,000 1998 3022 BHU Capacity Building in the Urban Sector 500,000 1998 3053 BAN Promoting Good Urban Governance in Dhaka 150,000

Appendix 3 71 Year No. Country Advisory Technical Assistance Total ($) 1998 3096 PRC Urban Policy Workshop 150,000 1998 3124 PAK Urban Sector Strategy Study 150,000 1998 3130 PAK Water Resources Strategy Study 650,000 1998 3067 IND Restructuring State-Level Housing Institutions 500,000 1999 3185 NEP Institutional Strengthening of Kathmandu Metropolitan City 450,000 1999 3196 PHI National Urban Policy Framework 150,000 1999 3209 IND Strengthening Institutional Capacities for Urban Infrastructure Finance and 500,000 Development 1999 3226 BAN Urban Sector Strategy 150,000 1999 3243 FIJ Urban Sector Strategy Study 670,000 1999 3272 NEP Urban Sector Strategy 200,000 1999 3297 BAN Urban Transport and Environmental Improvement 645,000 1999 3324 IND Community Participation in Urban Environmental Improvement 150,000 1999 3326 INO Urban Sector Development in a Decentralizing Environment 600,000 1999 3331 LAO Capacity Building for Urban Development Administration Authorities 600,000 1999 3344 IND Strengthening Microfinance Institutions for Urban and Environmental Infrastructure 500,000 Finance 1999 3377 PRC Urban Poverty Study 410,000 1999 3379 IND Strengthening Disaster Mitigation and Management at the State Level 1,000,000 2000 3586 IND Building HUDCO's Capacity for Lending to Community-Based Finance Institutions 150,000 2000 3475 PHI Institutional Strengthening of Housing and Urban Development Sector 150,000 2000 3480 IND Reducing Poverty in Urban India 300,000 2000 3566 SAM Capacity Building for Urban Planning and Management 500,000 2000 3521 PRC Assessment of Small and Medium Cities Urban Infrastructure Development 150,000 2000 3487 VIE Low Income Housing and Secondary Towns Urban Development Needs 500,000 Assessment Study 2000 3447 PRC Strengthening Urban Solid Waste Management 600,000 2001 3673 PRC Pro-Poor Urban Heating Tariff Reforms 850,000 2001 3645 THA Planning for Sustainable Urbanization 500,000 2001 3703 PHI Capacity Building for the Regulatory Office of Metropolitan Waterworks and 800,000 Sewerage System 2001 3749 PRC National Guidelines for Urban Wastewater Tariffs and Management Study 700,000 2001 3729 UZB Capacity Building for Urban Water Supply 600,000 2001 3770 IND Support for Improvements in the Accounting System of the Calcutta Municipality 150,000 Corporation 2001 3784 IND Conservation and Livelihood Improvement in the Indian Sundarbans 450,000 2002 3860 SAM Implementation of the Urban Planning and Management Strategy 400,000 2002 3902 IND North Eastern Region Urban Sector Profile 150,000 2002 4003 BAN Supporting Urban Governance Reform 400,000 2002 3848 PHI Metro Manila Solid Waste Management 1,250,000 2003 4162 IND Urban Sector Review and Strategy 480,000 2003 4086 IND Capacity Building for Project Management and Comm. Mobilization in Madhya 520,000 Pradesh 2004 4402 PRC Heating Supply for Urban Poor in Liaoning Province 500,000 2004 4352 MON Developing an Urban Development and Housing Sector Strategy 350,000 2004 4335 PRC Town-Based Urbanization Strategy Study 750,000 2004 4415 AFG Kabul Air Quality Management 450,000 2004 4518 IND Capacity Building for Kerala Sustainable Urban Development 500,000 2004 4497 IND Capacity Building for Municipal Service Delivery in Kerala 800,000 2005 4604 PRC Nanjing Water Utility Long-Term Capital Finance in Commercial Markets 550,000 AFG = Afghanistan, BAN = Bangladesh, BHU = Bhutan, CAM = Cambodia, FIJ = Fiji Islands, HCMC = Ho Chi Minh City, HUDCO = Housing and Urban Development Corporation Limited, IND = India, INO = Indonesia, KIR = Kiribati, LAO = Lao People s Democratic Republic, MAL = Malaysia, MWSS = Metropolitan Waterworks and Sewerage System, MON = Mongolia, NEP = Nepal, PAK = Pakistan, PHI = Philippines, PRC = People s Republic of China, SAM = Samoa, THA = Thailand, TUV = Tuvalu, UZB = Uzbekistan, VAN = Vanuatu, VIE = Viet Nam. Source: Asian Development Bank s internal databases.

72 Appendix 3 Table A3.2: Project Preparatory Technical Assistance Grants Year No. Country Project Preparatory Technical Assistance Total ($) 1993 1911 LAO Vientiane Integrated Urban Development 600,000 1993 1335 NEP Bagmati Command Area Development (Supplementary) 16,000 1993 1513 PHI Manila North-East Water Supply (Supplementary) 48,000 1993 1917 PRC Beijing Environmental Improvement 600,000 1993 1952 VAN Urban Infrastructure 536,000 1993 1977 IND Urban Infrastructure Development 600,000 1993 2009 INO Industrial Pollution Control in Cimahi 590,000 1993 2006 PAK Second Urban Water Supply and Sanitation 500,000 1993 2014 THA Wastewater Management and Pollution Control in Samut Prakarn 600,000 1994 2070 COO Integrated Urban Infrastructure 455,000 1994 2074 PAK Sukkur Bridge 100,000 1994 2094 INO Second West Java and Sumatra Urban Development Sector 100,000 1994 2146 VIE Second Provincial Towns Water Supply and Sanitation 550,000 1994 2185 INO Java Flood Control 900,000 1994 2187 PRC Anhui Municipal Wastewater Treatment 283,000 1994 2262 KAZ Almaty Air Quality Study 100,000 1994 2263 PHI MWSS Water Supply Improvement Study 582,000 1994 2272 PHI Small Towns Water Supply and Sanitation Sector 100,000 1995 2280 CAM Urban Water Supply and Sanitation 600,000 1995 2292 THA Nongkai-Udonthani Water Supply and Sanitation 600,000 1995 2303 THA Bangkok Metropolitan Region Wastewater Management Action Plan and 600,000 Feasibility Study 1995 2327 PAK Bahawalpur Division Area Development 600,000 1995 2373 IND Rajasthan Urban Infrastructure Development 600,000 1995 2396 LAO Secondary Towns Integrated Urban Development 600,000 1995 2410 BAN Urban Poverty Reduction 600,000 1995 2412 INO Development of Madrasah Aliyahs 600,000 1995 2445 PRC Xian-Xianyang-Tongchuan Environment Improvement 500,000 1995 2480 SAM Integrated Urban Development 552,000 1995 2511 PRC Zhejiang-Shanxi Water Conservancy 1,000,000 1995 2515 INO Preparation of Metropolitan Medan Urban Development 100,000 1996 2581 PHI Infrastructure Improvement of Subic Bay Area Municipalities 800,000 1996 2582 MON Provincial Towns Basic Urban Services 600,000 1996 2592 CAM Stung Chinit Water Resource Development 800,000 1996 2655 INO Metropolitan Bandung Urban Development 600,000 1996 2666 BHU Urban Infrastructure Improvement 400,000 1996 2689 CAM Integrated Urban Development 800,000 1996 2711 LAO Small Towns Water Supply and Sanitation 500,000 1996 2666 BHU Urban Infrastructure Improvement 400,000 1996 2700 IND Housing Finance Facility 100,000 1997 2770 PRC Fuzhou Water Supply and Wastewater Treatment 598,000 1997 2790 VIE Ho Chi Minh City Environmental Improvement 600,000 1997 2791 SRI Urban Development and Low-Income Housing 1,100,000 1997 2803 PHI Pasig River Environmental Management and Rehabilitation 800,000 1997 2806 IND Karnataka Coastal Environmental Management and Urban Development 800,000 1997 2807 PHI Clark Area Municipal Development 600,000 1997 2816 BAN Third Urban Development 600,000

Appendix 3 73 Year No. Country Project Preparatory Technical Assistance Total ($) 1997 2666 BHU Urban Infrastructure Improvement (Supplementary) 100,000 1997 2835 PHI Metro Manila Air Quality Improvement 150,000 1997 2903 THA Border Towns Urban Development 800,000 1997 2928 PAK Quetta Water Supply and Environmental Improvement 900,000 1997 2936 IND Urban and Environmental Infrastructure Fund 400,000 1997 2940 PAK North West Frontier Province Urban Development 950,000 1998 3021 PHI Mindanao Urban Planning and Basic Services Sector 1,000,000 1998 3025 PRC Suzhou Creek Environmental Rehabilitation 965,000 1998 3049 PRC Zhejiang-Shanxi Water Supply (Phase II) 540,000 1998 3055 FIJ Suva-Nausori Water Supply and Sewerage 800,000 1998 3059 NEP Small Towns Water Supply and Sanitation 600,000 1998 3085 COO Urban Infrastructure 750,000 1998 3088 INO Development of Rural-Urban Linkages 890,000 1998 3089 IND Calcutta Environmental Improvement 1,000,000 1999 3291 PHI Development of Poor Urban Communities 850,000 1999 3333 LAO Vientiane Urban Infrastructure and Services 600,000 1999 3364 NEP Urban Environmental Improvement 750,000 2000 3440 UZB Urban Social Infrastructure Development 720,000 2001 3646 INO Urban Poverty Reduction 800,000 2001 3804 INO Private Sector Participation Development Facility for Urban Infrastructure 600,000 2001 3809 VIE Central Region Urban Development 1,000,000 2001 3690 BAN Urban Governance and Infrastructure Improvement 350,000 2001 3760 PHI Metro Manila Urban Services for the Poor 1,000,000 2001 3802 PAK Southern Punjab Basic Urban Services 800,000 2001 3685 MON Integrated Development of Basic Urban Services in Secondary Towns 700,000 2001 3759 IND Integrated Urban Development in Madhya Pradesh 1,000,000 2001 3774 AZE Urban Water Supply and Sanitation 740,000 2002 3646 INO Urban Poverty Reduction (Supplementary) 160,000 2002 3291 PHI Development of Poor Urban Communities (Supplementary) 150,000 2003 4182 IND Urban Clean Fuel 995,000 2003 4106 IND Kerala Sustainable Urban Development 1,000,000 2004 4515 IND Preparation of the Jammu and Kashmir Urban Infrastructure Development 500,000 Project 2004 4361 INO Urban Air Quality Improvement Sector Development Program 700,000 2004 4348 IND North Eastern Region Urban Development 1,000,000 2004 4385 PRC Guangxi Nanning Urban Infrastructure Development 560,000 2004 4377 LAO Northern and Central Region Water Supply and Urban Development 200,000 2004 4530 IND Karnataka Urban Infrastructure Development III 400,000 2004 4531 SRI Greater Colombo Wastewater 850,000 2004 4533 BHU Urban Infrastructure Development 600,000 2004 4534 PAK Sindh Basic Urban Services 795,000 2004 4535 BAN Secondary Towns Water Supply and Sanitation 800,000 2005 4578 PAK Mega City Development 150,000 AZE = Azerbaijan, BAN = Bangladesh, BHU = Bhutan, CAM = Cambodia, COO = Cook Islands, FIJ = Fiji Islands, IND = India, INO = Indonesia, KAZ = Kazakhstan, LAO = Lao People s Democratic Republic, MON = Mongolia, MWSS = Metropolitan Waterworks and Sewerage System, NEP = Nepal, PAK = Pakistan, PHI = Philippines, PRC = People s Republic of China, SAM = Samoa, SRI = Sri Lanka, THA = Thailand, UZB = Uzbekistan, VAN = Vanuatu, VIE = Viet Nam. Source: Asian Development Bank s internal databases.

74 Appendix 3 Table A3.3: Regional Technical Assistance Grants Year No. Country Regional Technical Assistance Total ($) 1993 None 1994 5611 Regional Megacities Management in Asia and the Pacific 550,000 1995 5662 Regional Impact Evaluation Study of Bank Assistance in the Urban Dev. and 400,000 Housing Sector 1995 5646 Regional Regional Study on Urban Infrastructure Finance 480,000 1996 None 1997 5764 Regional Enhancing Municipal Service Delivery Capability in Selected DMCs 550,000 1997 5728 Regional Chiang Rai-Kunming Road Improvement via Lao People s Democratic 600,000 Republic 1998 None 1999 5846 Regional Development of the Cities Data Book for the Asian and Pacific Region 400,000 2000 5926 Regional Public-Private Community Partnerships in Urban Services for the Poor 550,000 2001 6016 Regional Clean Air Initiative for Asian Cities 150,000 2002 6026 Regional Promoting Urban Poverty Reduction through Participation in the Cities 670,000 Alliance 2003 6016 Regional Clean Air Initiative for Asian Cities (Supplementary) 92,020 2003 6121 Regional Rural, Urban and Subregional Linkages in the Mekong Region 400,000 2003 6130 Regional Study of Urban Violence in Asia: Towards More Effective Urban 150,000 Upgrading 2003 6151 Regional ADB's Participation in the Cities Alliance 350,000 2004 6016 Regional Clean Air Initiative for Asian Cities (Supplementary) 466,500 2005 6240 Regional Special Evaluation Study on Urban Sector Strategy and Operations 150,000 ADB = Asian Development Bank, DMC = developing member country. Source: Asian Development Bank s internal databases. Table A3.4: Japan Fund for Poverty Reduction Grants Year No. Country Japan Fund for Poverty Reduction Total ($) 2000 9003 PHI On-Site Urban Upgrading for Vulnerable Slum Communities of Payatas 1,000,000 2000 9002 PNG Low-Cost Sanitation, Community Awareness and Health Education 1,740,000 Program 2000 9004 PHI Off-Site and Off-City Relocation of Vulnerable Slum Communities of 1,000,000 Muntinlupa 2002 9021 IND Rainwater Harvesting and Slum Development in Rajasthan 1,900,000 Strategic Private Sector Partnerships for Urban Poverty Reduction in 3,600,000 2002 9022 PHI Metro Manila 2002 9023 CAM Income for the Poor through Community-Based Environmental 1,000,000 Improvements in Phnom Penh 2003 9035 LAO Solid Waste Management and Income Generation for Vientiane's Poor 1,000,000 2004 9054 UZB Affordable Services and Water Conservation for the Urban Poor 1,500,000 2004 9058 VIE Expanding Benefits of the Poor through Urban Environmental 1,000,000 Improvements 2005 9065 INO Enriching Lives of the Urban Poor through Food Fortification 1,750,000 2005 9074 INO Seismically Upgraded Housing in Nanggroe Aceh Darussalam and North Sumatera 2,000,000 CAM = Cambodia, IND = India, INO = Indonesia, LAO = Lao People s Democratic Republic, PHI = Philippines, PNG = Papua New Guinea, UZB = Uzbekistan, VIE = Viet Nam. Source: Asian Development Bank s internal databases.

Appendix 4 75 URBAN SECTOR NEEDS 1. Well-managed urbanization taking place in well-regulated economies facilitates sustained economic growth and thereby promotes broad social welfare gains. The agglomeration of economic activities through the urbanization process contributes to growth through (i) raising the productivity of outputs and employment; (ii) mobilizing saving and allowing accumulation of wealth in the form of real estate; and (iii) increasing fiscal flows, with cities providing the largest share of tax revenues. However, policy and regulatory weaknesses can compromise the potential for beneficial development through urbanization, and poorly managed urbanization results in suboptimal use of public and private resources, failure of economic gains to be widely distributed, and significant social and environmental costs. Efficient urban growth is further compromised by inadequate, poorly distributed and poorly maintained infrastructure and services, and inadequate cost recovery for services provided. 2. The development needs of the urban sector in the Asia and Pacific region span the full range of activities, infrastructure, services, and facilities required to (i) improve the efficiency of cities as places to invest and work in, and (ii) improve the condition of cities as places in which to live and prosper. Improved governance and regulatory environments are required to iron out urban market distortions and attract capital. Administrative reform and capacity building is required to improve urban planning and management, and to ensure that cities develop efficiently and maximize their development potential. Improved and extended infrastructure, utilities, and services are required to support both economic development and improved quality of life. 3. In both low-income (Asian Development Fund) and medium-income (ordinary capital resources) countries, there is a need to address the requirements of the urban poor for decent living conditions and access to work so that they too can enjoy the benefits of economic growth. Poverty has many dimensions, in addition to material deprivation, Asia s urban poor suffer from the cumulative deprivation which results from squalid living conditions and poor access to services. Urban poverty is manifested not only in economic stress resulting from unemployment and low paying jobs, but in a poor quality of life as a result of inadequate and insecure access to shelter, education, health services, and infrastructure. The urban poor are exposed to environmental hazards and risks, have weak family support and coping mechanisms, and are highly vulnerable to economic and market changes. Poverty is compounded by the high levels of environmental stress in cities a function of rapid growth, industrialization, inadequate and ineffective waste management and disposal systems, and increasing vehicle densities. Notwithstanding the negative impact on quality of life, the economic impact of pollution in Asian cities, in terms of loss of productivity, has been estimated at between 1% and 5% of their city gross domestic product (GDP). 4. Addressing these needs requires intervention on a number of fronts. Firstly, national governments need to work with the newly empowered local governments in improving the policy and regulatory environment to (i) encourage efficient and effective urban development, and (ii) create the conditions for greater private sector investment in urban infrastructure and service provision. This requires policy reform and institutional change; the development and implementation of improved regulation; incentive systems; and strengthened relationships among local government, the private sector, and civil society. 5. Secondly, there is a need for the strengthening of local governments to better equip them to support efficient, effective, and sustainable urban development. This demands that local governments address issues such as promotion of the local economy, improved access to

76 Appendix 4 private capital, reform in development planning and promoting effective real estate markets, and increased preparedness in disaster prevention and mitigation. To achieve this requires a range of interventions including institutional strengthening, support for improved urban governance, organizational reform, and human resource development. Improved urban management efficiency and sustainability also requires (i) municipal finance reform, (ii) enhanced cost recovery for services, (iii) improved urban planning and regulation, and (iv) the mobilization of private-sector resources for expanded and improved infrastructure and service provision. 6. Thirdly is the need for improved urban infrastructure and services. This requires both improved management and dramatic increases in investment. There are a variety of estimates for infrastructure financing requirements for Asia, the most recent of which were prepared for the Connecting East Asia: A New Framework for Infrastructure. 1 Such estimates are necessarily highly approximate since the relationship between the supply of infrastructure and services, its impact on growth, and the further demand this growth places on infrastructure provision, is complex. In 2004, conservative estimates suggested that in Asia and the Pacific, expenditures of about $400 billion per year on infrastructure (both capital requirements and operating costs) would be required for the next 25 years to sustain current levels of growth. 2 7. A model to predict future infrastructure demand in all sectors (net of requirements for rehabilitation) calculated that infrastructure investment needs for all developing countries (World Bank member developing countries) over the period 2005 2010 amounted to $465 billion per year or 5.5% of developing countries GDP (including both capital and maintenance requirements). 3 A similar methodology, adopted for East Asia 4 and extended to include all Asian Development Bank developing member countries (DMCs) over the same period, indicates a requirement of $230 billion per year, again including both capital and maintenance costs, and covering the following sectors: (i) electricity generation and supply, (ii) telephone landlines, (iii) telephone mobile, (iv) paved intercity roads, (v) rail lines, and (vi) access to improved water supply and sanitation. This methodology is based on infrastructure stock trends, using real demand data from eight Asian countries for which data is available, and adjusting for economic growth and geographic variables to extend this analysis to other countries within the same income group. Costs of increasing infrastructure stock, and for associated maintenance requirements, are based on best practice unit costs derived from the actual cost of infrastructure and service provision, and are reduced to an average cost per capita per year for middleincome and low-income economies to derive the total funding requirement. 5 8. To estimate approximate urban development financing requirements, a similar methodology was used excluding power, telecommunications, railways, and intercity roads; and including solid waste management and slum upgrading. The same best practice unit costs were used for water supply and sanitation, and similar figures were derived for solid waste management and slum upgrading based on recent field data 6 (Table A4.1). This analysis suggested that total financing requirements for just water supply, sanitation, solid waste 1 Asian Development Bank, Japan Bank for International Cooperation, World Bank. 2005. Connecting East Asia: A New Framework for Infrastructure. Washington DC. 2 World Bank. 2004. World Bank Development Report 2004. Washington DC. 3 Fay, Marianne and Tito Yepes. 2003. Investing in Infrastructure: What is Needed from 2000 2010? World Bank Working Paper No. 2545. Washington DC: World Bank. 4 People s Republic of China, Indonesia, Lao PDR, Malaysia, Mongolia, Papua New Guinea, Philippines, Thailand, Viet Nam, Cambodia, Fiji Islands, Kiribati, Marshall Islands, Federated States of Micronesia, Myanmar, Palau, Samoa, Solomon Islands, Timor-Leste, Tonga, and Vanuatu. 5 The figures for the People s Republic of China and Malaysia are adjusted slightly from the middle-income country figure to provide a better fit with observed data. 6 Figures taken from recent urban development projects with components in these sectors.

Appendix 4 77 management, and slum upgrading in urban areas will be $25 billion per annum (Table A4.2). Demand for additional urban roads and mass transit systems is difficult to estimate. However, based on past urban investment patterns on urban roads and highways, this number doubles to $50 billion per year if urban roads are included. Demand for mass transit systems is yet harder to assess. However, megacities and large cities across the region are going to find investment in such systems increasingly necessary as cities expand and traffic volumes exceed the carrying capacities of urban road networks. Even based on the conservative assumption that 100 kilometers of rail mass transit system are constructed annually across the region, this adds a further $10 billion per annum to the urban funding requirement. 9. Excluding electricity, intercity roads and rail, and telecommunications, it is estimated that the DMCs will need to spend about $60 billion per annum on the development of urban infrastructure and services between 2006 and 2010 in the infrastructure sectors of water supply, sanitation, solid waste management, slum upgrading, urban roads, and mass transit systems. Table A4.1: Best Practice Unit Costs Unit Cost ($) Sector Unit Capital Maintenance/Year as % of Stock Value Water Supply Per connected household 400 3 Sanitation Per connected household 700 3 Solid Waste Management Per household served 120 5 Slum Upgrading and/or Urban Renewal Per household upgraded 300 3 Urban Road Per km of 4 lane highway 650,000 2 Mass Transit Per km of underground or 100 million 5 elevated railway km = kilometer. Sources: Fay, Marianne and Tito Yepes. 2003. Investing in Infrastructure: What is needed from 2000 to 2010? World Bank Paper No. 3102. Washington DC: World Bank; Expenditures on Infrastructure in East Asia Region, 2006 2010 (for Asian Development Bank-Japan Bank for International Cooperation-World Bank East Asia Pacific Infrastructure Flagship Study, February 2005); This special evaluation study s estimates of urban roads and mass transit costs. 10. Overall, the funding requirements for infrastructure translate to 6.9% of GDP for the People s Republic of China, 3.6% for other middle-income countries, and 6.3% for low-income countries. While this analysis is useful in providing some indication of the likely funding requirements both for national and urban infrastructure and services these service costs can ultimately only be met either directly by consumers through service tariffs, or indirectly by taxpayers through subsidies. Financiers, whether the private sector or official lenders or donors, can only change the time profile of user charges or taxes through loans or equity which eventually must be repaid or remunerated.

Table A4.2: Expected Infrastructure Expenditure for Asia: Total and Urban Sector a Infrastructure Expenditures ($) per person per annum Urban Income b Total c Country Classification by 2003 Population (Million) Expected Annual Expenditure Needs Per Country ($ million) Urban Sector a Country Sector a Total d Urban e Total Afghanistan Low income 48.0 22.0 22.2 6.7 1,065.6 147.4 Azerbaijan Lower-middle income 90.0 21.0 8.2 4.3 741.1 90.3 Bangladesh Low income 48.0 22.0 133.4 37.0 6,403.2 814.0 Bhutan Low income 48.0 22.0 0.7 35.2 0.0 Cambodia Low income 48.0 22.0 13.3 2.5 637.8 55.0 China, People's Republic of Lower-middle income 90.0 21.0 1,292.3 498.0 116,307.0 10,458.0 Cook Islands 0.0 0.0 0.0 Fiji Islands Lower-middle income 90.0 21.0 0.8 74.8 0.0 Hong Kong, China High income 6.8 0.0 0.0 India Low income 48.0 22.0 1,068.0 301.3 51,264.0 6,628.6 Indonesia Lower-middle income 90.0 21.0 215.0 94.7 19,353.9 1,988.7 Kazakhstan Lower-middle income 90.0 21.0 14.9 8.3 1,341.8 174.3 Kiribati Lower-middle income 90.0 21.0 0.1 8.0 0.0 Republic of Korea High income 47.8 40.0 0.0 0.0 Kyrgyz Republic Low income 48.0 22.0 5.0 1.7 240.5 37.4 Lao PDR Low income 48.0 22.0 5.7 1.2 272.6 26.4 Malaysia Upper-middle income 97.0 19.0 25.1 14.7 2,429.9 279.3 Maldives Lower-middle income 90.0 21.0 0.3 25.7 0.0 Federated States of Micronesia, Lower-middle income 90.0 21.0 0.1 9.7 0.0 Mongolia Low income 48.0 22.0 2.5 1.4 120.0 30.8 Myanmar Low income 48.0 22.0 53.2 14.4 2,554.6 316.8 Nauru 0.0 0.0 Nepal Low income 48.0 22.0 24.2 3.2 1,161.6 70.4 Pakistan Low income 48.0 22.0 146.0 50.6 7,005.6 1,113.2 Palau Upper-middle income 97.0 19.0 0.0 2.0 0.0 Papua New Guinea Low income 48.0 22.0 5.6 1.0 269.8 22.0 Philippines Lower-middle income 90.0 21.0 81.1 49.7 7,299.0 1,043.7 Marshall Islands Lower-middle income 90.0 21.0 0.1 5.3 0.0 Samoa Lower-middle income 90.0 21.0 0.2 16.1 0.0 Singapore High income 4.2 4.3 0.0 0.0 Solomon Islands Low income 48.0 22.0 0.5 24.4 0.0 Sri Lanka Lower-middle income 90.0 21.0 19.3 4.6 1,732.7 96.6 Taipei,China 22.6 0.0 0.0 78 Appendix 4

Country Infrastructure Expenditures ($) per person per annum 2003 Population (Million) Expected Annual Expenditure Needs Per Country ($ million) Country Classification by Urban Urban Income b Total c Sector a Total d Urban e Total Sector a Tajikistan Low income 48.0 22.0 6.6 1.7 315.5 37.4 Thailand Lower-middle income 90.0 21.0 64.0 12.7 5,756.3 266.7 Timor-Leste Low income 48.0 22.0 0.8 37.3 0.0 Tonga Lower-middle income 90.0 21.0 0.1 9.1 0.0 Turkmenistan Lower-middle income 90.0 21.0 6.3 2.2 566.9 46.2 Tuvalu 0.0 0.0 0.0 Uzbekistan Low income 48.0 22.0 25.7 9.4 1,233.6 206.8 Vanuatu Lower-middle income 90.0 21.0 0.2 18.7 0.0 Viet Nam Low income 48.0 22.0 80.9 20.7 3,883.3 455.4 Total 3,397.0 1,193.1 232,222.5 24,405.4 a b c d e Urban sector expenditure in this table is confined to water supply, sanitation, solid waste management, and slum upgrading in urban areas. Country Classification, World Bank Data, and Statistics. Tito Yepes. 2005. Expenditure on Infrastructure in East Asia Region, 2006 2010. ADB-JBIC-World Bank East Asia Pacific Infrastructure Flagship Study. Asian Development Bank Statistical Database System. World Bank. 2005. 2005 World Development Indicators. Washington DC. Appendix 4 79

ASIAN DEVELOPMENT BANK S LENDING TO THE URBAN SECTOR Table A5.1: Urban Sector Loans by Country, 1993 2004 ($ million) 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 Total Country No. Amt. No. Amt. No. Amt. No. Amt. No. Amt. No. Amt. No. Amt. No. Amt. No. Amt. No. Amt. No. Amt. No. Amt. No. Amt. AZE 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 2 30.0 2 30.0 BAN 1 31.0 0 0.0 1 65.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 1 60.0 0 0.0 1 80.0 4 236.0 BHU 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 1 5.7 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 1 5.7 CAM 0 0.0 0 0.0 0 0.0 1 20.0 0 0.0 0 0.0 1 20.0 0 0.0 0 0.0 0 0.0 1 6.3 0 0.0 3 46.3 COO 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 1 2.2 0 0.0 0 0.0 0 0.0 1 2.2 FIJ 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 1 47.0 0 0.0 1 47.0 FSM 0 0.0 0 0.0 0 0.0 1 10.6 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 1 10.6 HKG 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 IND 0 0.0 0 0.0 2 105.0 0 0.0 3 300.0 1 250.0 4 375.0 3 370.0 0 0.0 0 0.0 1 200.0 1 250.0 15 1,850.0 INO 1 85.0 0 0.0 2 200.0 1 80.0 3 224.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 2 88.6 0 0.0 9 677.6 KGZ 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 1 36.0 0 0.0 0 0.0 0 0.0 0 0.0 1 36.0 KIR 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 1 10.2 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 1 10.2 KOR 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 LAO 1 13.0 0 0.0 1 20.0 0 0.0 1 27.0 0 0.0 1 20.0 0 0.0 1 25.0 0 0.0 1 16.0 0 0.0 6 121.0 MAL 0 0.0 0 0.0 0 0.0 0 0.0 0 0/0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 MON 0 0.0 0 0.0 0 0.0 0 0.0 1 6.8 0 0.0 0 0.0 0 0.0 1 15.0 1 20.1 0 0.0 0 0.0 3 41.9 MYA 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 NEP 1 12.0 0 0.0 0 0.0 0 0.0 0 0.0 1 5.0 0 0.0 2 155.0 0 0.0 1 30.0 2 15.0 0 0.0 7 217.0 PAK 1 72.0 1 45.0 0 0.0 0 0.0 1 70.0 0 0.0 0 0.0 0 0.0 1 20.8 0 0.0 2 90.0 0 0.0 6 297.8 PHI 1 43.2 0 0.0 1 30.0 1 50.0 1 22.0 4 318.2 0 0.0 2 175.0 1 30.0 0 0.0 2 33.8 0 0.0 13 702.1 PNG 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 1 15.3 0 0.0 0 0.0 0 0.0 0 0.0 1 15.3 PRC 0 0.0 1 160.0 0 0.0 1 28.0 1 100.0 1 102.0 1 300.0 1 130.0 0 0.0 1 82.4 3 453.0 0 0.0 10 1,355.4 RMI 1 0.7 0 0.0 1 9.2 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 2 9.9 SAM 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 1 8.0 0 0.0 1 8.0 SIN 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 SOL 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 SRI 1 40.0 0 0.0 0 0.0 0 0.0 1 75.0 1 70.0 0 0.0 0 0.0 0 0.0 0 0.0 1 60.3 0 0.0 4 245.3 THA 0 0.0 1 38.5 1 150.0 0 0.0 1 50.0 1 80.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 4 318.5 UZB 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 1 36.0 0 0.0 0 0.0 0 0.0 1 36.0 VAN 0 0.0 0 0.0 0 0.0 1 10.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 0 0.0 1 10.0 VIE 1 65.0 0 0.0 1 66.0 0 0.0 1 69.0 0 0.0 1 70.0 0 0.0 1 60.0 1 30.0 1 44.0 0 0.0 7 404.0 Total 9 361.9 3 243.5 10 645.2 6 198.6 14 943.8 11 841.1 8 785.0 10 881.3 7 189.0 5 222.5 18 1,061.9 4 360.0 105 6,733.8 Amt = amount; AZE = Azerbaijan; BAN = Bangladesh; BHU = Bhutan; CAM = Cambodia; COO = Cook Islands; FIJ = Fiji Islands; FSM = Federated States of Micronesia; HKG = Hong Kong, China; IND = India; INO = Indonesia; KGZ = Kyrgyz Republic; KIR = Kiribati; KOR = Republic of Korea; LAO = Lao People s Democratic Republic; MAL = Malaysia; MON = Mongolia; MYA = Myanmar; NEP = Nepal; PAK = Pakistan; PHI = Philippines; PNG = Papua New Guinea; PRC = People s Republic of China; RMI = Republic of the Marshall Islands; SAM = Samoa; SIN = Singapore; SOL = Solomon Islands; SRI = Sri Lanka; THA = Thailand; UZB = Uzbekistan; VAN = Vanuatu; VIE = Viet Nam. Source: Asian Development Bank s internal databases. 80 Appendix 5

Table A5.2: Annual Urban Lending by Country ($ million) Year AZE BAN BHU CAM COO FIJ FSM HKG IND INO KIR KOR KGZ LAO MAL MON MYA NEP PAK 1968 7.2 1969 1970 1971 8.8 1972 21.5 31.3 6.1 1973 13.0 1974 11.5 6.0 1975 20.0 1976 1.15 22.0 1977 20.50 39.5 1978 16.0 7.96 1979 40.3 47.1 1980 20.0 32.0 27.9 1981 74.8 102.2 1982 14.4 30.2 15.0 1983 36.7 122.8 1984 94.0 1985 172.6 35.0 45.0 1986 55.2 1987 3.3 33.2 1988 175.0 20.6 1989 24.2 9.6 120.0 151.0 1990 43.0 39.0 1991 91.5 253.0 9.6 1992 55.0 158.4 9.5 1993 31.0 85.0 13.0 12.0 72.0 1994 45.0 1995 65.0 105.0 200.0 20.0 1996 20.0 10.6 80.0 1997 300.0 224.0 27.0 6.8 70.0 1998 5.7 250.0 10.2 5.0 1999 20.0 375.0 20.0 2000 370.0 36.0 155.0 2001 2.2 25.0 15.0 20.8 2002 60.0 20.1 30.0 2003 6.3 47.0 200.0 88.6 16.0 15.0 90.0 2004 30.0 80.0 250.0 Grand total 30.0 464.1 9.0 46.3 2.2 56.6 10.6 82.0 1,850.0 1,792.0 10.2 522.9 36.0 146.1 104.5 41.9 36.0 217.0 565.5 Appendix 5 81

Year PHI PNG PRC RMI SAM SIN 1968 7.2 1969 1970 8.3 8.3 1971 8.8 1972 58.9 1973 19.6 4.6 37.2 1974 68.8 1975 36.8 1976 13.5 23.6 60.3 1977 60.0 1978 49.0 5.4 78.4 1979 15.1 68.0 170.5 1980 42.8 1.65 20.0 144.4 1981 46.0 223.0 1982 59.6 1983 39.3 198.8 1984 94.0 1985 168.9 421.5 1986 55.2 1987 36.5 1988 195.6 1989 156.4 461.2 1990 22.0 104.0 1991 31.4 70.0 20.0 475.5 1992 11.3 85.0 27.0 70.3 416.5 1993 43.2 0.7 40.0 65.0 361.9 1994 160.0 38.5 243.5 1995 30.0 9.2 150.0 66.0 645.2 1996 50.0 28.0 10.0 198.6 1997 22.0 100.0 75.0 50.0 69.0 943.8 1998 318.8 102.0 70.0 80.0 841.1 1999 175.0 15.4 300.0 70.0 785.0 2000 130.0 881.4 2001 30.0 36.0 60.0 189.0 2002 82.36 30.0 222.5 2003 33.8 453.0 8.0 60.3 44.0 1,061.9 2004 360.0 Grand Total 1,157.1 45.6 1,510.0 9.9 8.0 47.0 1.65 292.3 665.3 36.0 10.0 408.6 10,214.7 AZE = Azerbaijan; BAN = Bangladesh; BHU = Bhutan; CAM = Cambodia; COO = Cook Islands; FIJ = Fiji Islands; FSM = Federated States of Micronesia; HKG = Hong Kong, China; IND = India; INO = Indonesia; KIR = Kiribati; KOR = Republic of Korea; KGZ = Kyrgyz Republic; LAO = Lao People s Democratic Republic; MAL = Malaysia, MON = Mongolia; MYA = Myanmar; NEP = Nepal; PAK = Pakistan; PHI = Philippines; PNG = Papua New Guinea; PRC = People s Republic of China; RMI = Republic of the Marshall Islands; SAM = Samoa; SIN = Singapore; SOL = Solomon Islands; SRI = Sri Lanka; THA = Thailand; UZB = Uzbekistan; VAN = Vanuatu; VIE = Viet Nam. Source: Asian Development Bank s internal databases. SOL SRI THA UZB VAN VIE Grand Total 82 Appendix 5

Appendix 6 83 FINDINGS FROM ALL 51 PROJECT COMPLETION REPORTS FOR URBAN SECTOR PROJECTS, ISSUED BETWEEN MID-1998 AND MID-2005 1. Success Ratings. Thailand and the People s Republic of China (PRC) have had the highest success rates for completed projects. Comparing urban sector projects with and without project preparatory technical assistance (PPTA), projects preceded by a PPTA were not significantly more successful than projects not preceded by a PPTA (81% versus 79% of all projects with project completion reports [PCRs] were rated successful). 1 So far, urban sector projects administered by resident missions have not been ranked more successful than projects administered by staff in headquarters. 2. Project Support by the Asian Development Bank. Comparing the average number of mission days needed for urban sector projects and all Asian Development Bank (ADB) projects (according to the Special Evaluation Study on Project Cost Estimates 2 ), there was little difference: fact-finding took around 50 days and appraisal lasted around 70 days in both cases. Project administration missions took around 150 days. 3 3. Disbursement Ratios. Amounts disbursed were on average 84% of the loans for the 51 completed projects with PCRs, somewhat similar to the average for all projects (footnote 2). Large amounts of ADB loans were cancelled or lapsed when the loan was closed, both in the urban sector and in all other sectors. Other Operations Evaluation Department studies have pointed out the problems with estimating project cost accurately (footnote 2), and the effects of certain contingent problems, such as local currency devaluations against the US dollar in the 1990s, which have contributed to loan cancellations. The ADB-wide phenomenon of loan cancellations also points to certain systemic problems that ADB loans face in general, that are unrelated to the specific features of the urban sector. However, some sector influence was evident, since water supply, sanitation, and waste management (WSSWM) projects fared worse than the other categories, at 76% disbursements; multisector projects (88%) and other urban sector projects did relatively better (86%). 4. Scope Changes. Multisector projects had an average of 9.2 scope changes per project, water supply 5.5, and other types of projects 2.0. Projects had more major scope reductions than scope additions. From the special evaluation study on project cost estimates (footnote 2), it is clear that the number of scope changes officially sanctioned by ADB is usually between one and two per project, so that it can be concluded that when the project ends, there is usually a different set of outputs from that which was set as the target at the project approval stage. This is especially the case for the multisector projects. 5. Time Overruns. Forty-five of the 51 projects had time overruns, and the average duration of an urban sector project was 6.9 years compared to an intended duration of 4.8 years. This was not very different from the average for all ADB supported projects: 4.8 years versus 6.6 years (footnote 2). WSSWM projects had especially long delays between appraisal and actual loan effectiveness. The reasons are unclear. Comparing the relative prevalence of 1 The reasons were not investigated. More projects without a project preparatory TA may have been follow-up projects; projects without a project preparatory TA may have had a better standard of preparation by the Executing Agency. 2 ADB. 2004. Special Evaluation Study on Project Cost Estimates. Manila. 3 Table 37 of the Annual Report on Loan and Technical Assistance Portfolio Performance for the Year Ending 31 December 2004 shows that from 1998 2004, project administration missions averaged 19.4 person-days per project. With an average actual project duration of 7 years (footnote 2), the total number of project administration mission days per project is approximately 136 days on the average.

84 Appendix 6 reasons for time overruns, delays caused by land acquisition were much more prevalent in urban sector projects than in other types of projects, as well as delays caused by slow progress with works, slow start-up, and government inaction. Other project delays, such as those resulting from procurement activities, policy approvals, consultant progress, contractor progress, ADB inaction, and calamities, were similar to those in other sectors. Thirty-five percent of the PCRs mentioned that there were problems with project design, which reduced the success of the project; this was mostly the case with multisector projects (47%) but also WSSWM projects (27%). Twenty percent of the projects had problems caused by unanticipated decentralization processes in various developing member countries. 6. Land Acquisition and Resettlement. Land acquisition led to problems in two thirds of the multisector projects where land was to be acquired, and half of the WSSWM projects where it was acquired. Issues with land acquisition were often underestimated, partly because of unforeseen project delays which set in motion a vicious cycle. Sixty-one percent of projects had a land acquisition cost component, a percentage much higher than for other types of projects. Resettlement was needed in 14 projects (27%), and in about half of the cases, it proved to be a much larger issue than foreseen again mainly in multisector projects and much less so in WSSWM projects. Resettlement, particularly in the PRC, was implemented successfully, involving almost 40,000 people in four projects. Outside the PRC, the experiences were more varied, mostly because of delays and, in two cases, court action (Loan 942-BAN[SF] 4 and Loan 1154-PNG). 5 7. Loan Covenants. The average number of covenants in the urban sector projects was 28, with little difference between the averages for multisector, WSSWM, and other urban sector projects. Compliance with loan covenants was on average 80%, with no differences between the averages for the different types of projects. Partly complied with covenants were generally between 5% and 8%; noncompliance was around 6% 7%. Compliance with loan covenants was overall highest in the PRC at 88%. Commonly, the covenants not complied with pertained to nonrevenue water reduction, financial targets, and benefit and monitoring evaluation report preparation and submission. 4 Loan 942-BAN(SF): Dhaka Urban Infrastructure Improvement Project, for $24.2 million, approved 12 January 1989. 5 Loan 1154-PNG: Transport Infrastructure Development Project, for $30.0 million, approved 14 January 1992.

Appendix 7 85 COUNTRY CASE STUDIES: PEOPLE S REPUBLIC OF CHINA, INDIA, AND PHILIPPINES A. People s Republic of China (PRC) 1. The Government of the People s Republic of China (PRC) does not see a need for borrowings from the Asian Development Bank (ADB) higher than $1.5 billion $1.7 billion per year, and transport and energy have traditionally been the major ADB-supported sectors. A senior official at the Ministry of Finance stated to the study team that the PRC did not regard itself as dependent on ADB financing for the urban sector; sufficient project finance was currently available from state and commercial banks in the country. The ministry and municipal borrowers were looking to ADB for significant added value in support of knowledge generation and transfer to address the PRC s formidable urbanization challenges. ADB s added value currently left a lot to be desired and could not always justify the transaction costs that PRC governments incurred when dealing with ADB loans. Right or wrong and it was corroborated by subsequent interviews with stakeholders and a questionnaire survey that other stakeholders held a more positive view of ADB s added value the perception at the center may limit the potential for expansion of ADB s lending to the urban sector. Nevertheless, the urban sector has been one of the four sectors for which the PRC borrows, as it evolved from its early focus on mitigating severe environmental pollution in cities and addressing pressing problems, such as restoring city or district heating systems. 2. Urbanization in the PRC. The PRC s degree of urbanization has doubled in the last 25 years. In 2005, it reached 42% of the population. With 536 million urbanites, the number is larger than the combined urban population of the next four largest urban developing member countries (DMCs): India, Indonesia, Pakistan and the Philippines. By 2015, the expectation is that urbanization will have crossed the internationally still modest 50% line, and that 700 million Chinese will then live in cities. 1 The PRC has 663 statutory cities, 60 of which have populations above 1 million, and over 20,000 statutory towns. The 60 largest cities, with 12% of the population, produced 34% of non-farming gross domestic product (GDP) in 2005. 3. Government Response. The PRC started paying more attention to the urbanization challenge by the end of the 1990s. The PRC s Tenth Five-Year Plan (2001 2005) diverted from earlier policies, which promoted efforts to reduce cities attraction to rural migrants. Promoting small town-based urbanization became one of the five key policy thrusts. Urbanization in the PRC follows different patterns from many other countries. Rural migration to cities has been more controlled because of mandatory urban registration and restricted access to social safety nets in places other than where rural people have been registered. The dominant enterprise structure of vertically integrated state-owned enterprises has historically localized supply chains within cities. The policy of allocation of inner urban land to state-controlled enterprises at no cost meant that, up until very recently, there had been little economic incentive for these firms to relocate to lower-cost suburban sites. This hampered suburbanization and caused a high degree of congestion and pollution in inner cities. 4. Problems with Urbanization. Apart from air and water pollution, which continue to be serious urban problems in the PRC, traffic congestion and water shortages are serious negative externalities of PRC urbanization. The Ministry of Water Resources has reported that 420 out of 658 cities had water shortages, 110 of these severe. Fifty-three percent of monitored sections of the seven key river systems in the PRC had water quality unsuitable for normal water treatment 1 United Nations. 2004. World Urbanization Prospects: The 2003 Revision. New York: Department of Economic and Social Affairs, Population Division.

86 Appendix 7 processes or irrigation. Despite the huge investments in domestic wastewater infrastructure in urban areas directed by the state council at the start of the millennium, it is likely that surface and groundwater quality will deteriorate further in many cities because of nonpoint pollution in areas outside the purview of municipal monitoring and control. Land consumption has also become a more serious issue. Concerns over rampant conversion of agricultural land have triggered periodic inspections and clampdowns by the Ministry of Natural Resources and Lands. However, the basic problem is that, under the current fiscal system in the PRC, leasing of landuse rights accounts for a large proportion of local government financing. Municipal solid waste will increasingly contribute to consumption of suburban land, and to air and water pollution. The World Bank estimates that waste generation in PRC cities will grow from 195 million tons in 2005 to 306 million tons in 2015. These estimates do not include waste generated in the expanding suburban areas. 5. Land Markets in the PRC. The PRC is different from non-transition DMCs in that it owns all city land. In large cities such as Shanghai, the model of auctioning the long-term leasing of land to companies has faciliated the financing of much of the modern inner city infrastructure and the road corridors. Unfortunately, in many other cities, the absence of fullyfunctioning land markets typical of transition economies still constrains the availability of serviced land, drives up the price of available land, and hence limits the location choices of enterprises and households. In core cities, the old practice of designating land-use rights to state-owned enterprises is changing, but in many places this does not happen fast enough to respond to new demands for land. Maximizing agglomeration benefits to metropolitan regions in the PRC will require both the removal of constraints to town-based urbanization and better management of the metropolises. There is considerable scope for policy development and the associated capacity and institution building. 6. Capital Investment Requirements for the PRC s Urban Regions. An ADB advisory technical assistance (TA) on public infrastructure investment policy in the PRC 2 has estimated capital investment requirements, based on data available for 1999. The TA estimated that capital investment in urban infrastructure would climb from just under 2% of GDP in 1999 to 2.3% in 2010, and 2.7% in 2020. Approximately $50 billion of new annual capital investment was projected in 2010 for water supply, wastewater treatment, solid waste, power, and roads in PRC s urban areas; rising to $100 billion per year by 2020. Recurrent investment in operations and maintenance would incur an additional $25 billion per year. 7. ADB s Lending Operations. Compared to other DMCs, the PRC started borrowing from ADB for the cofinancing of urban operations relatively late. The first loan in the PRC was for a large bridge in Shanghai in 1991. Since then, urban sector lending for the PRC accounts for about a quarter of all ADB s urban sector lending to DMCs a figure that may seem large but is still lower than the PRC s share of total urban population in ADB s DMCs, which is 42%. On a per capita basis, India has borrowed about 26% more than the PRC for the urban sector. 3 So far, 12 projects (with loans totaling $1,513 million) have been completed, 6 of which in water supply, sanitation, and waste management (WSSWM). Nine projects are ongoing (including 6 in WSSWM) with loans totaling $994 million, and 11 are under preparation with indicative loan amounts totaling $1,221 million, 9 of which will be WSSWM or multisector. The location of the projects is indicated in Figure A7.1. 2 ADB. 2002. Strengthening Public Infrastructure Investment Policy in PRC: Strategic Options for Central, Provincial and Local Governments. Final Report by Chreod Ltd. on TA 3253-PRC. Manila (also published by China Finance and Economic Publishing House, Beijing. 2004). 3 India: $6.34 per capita, PRC: $5.02 per 2005 urban resident, over the period 1985 2005.

Appendix 7 87 Figure A7.1: Urban Lending in the People s Republic of China Notes: 1990 (year of approval) completed projects: Loan 1082-PRC: Shanghai Nanpu Bridge Project, for $70.0 million, approved 28 May 1991; Loan 1188-PRC: Shanghai Yangpu Bridge Project, for $85.0 million, approved 17 November 1992; Loan 1313-PRC: Dalian Water Supply Project, for $160.0 million, approved 20 September 1994; Loan 1490-PRC: Anhui Environmental Improvement Project for Municipal Wastewater Treatment, for $28.0 million, approved 26 November 1996); Loan 1636-PRC: Fuzhou Water Supply and Wastewater Treatment Project, for $102.0 million, approved 30 September 1998; Loan 1205-PRC: Qingdao Environmental Improvement Project, for $103.0 million, approved 10 December 1992; Loan 1491-PRC: Anhui Environmental Improvement Project for Industrial Pollution Abatement, for $112.0 million, approved 26 November 1996; Loan 1328-PRC: and Qitaihe Thermal Energy and Environmental Improvement Project, for $165.0 million, dated 27 October 1994; 1990 (year of approval) multi-location projects (completed): Loan 1544-PRC: Zheijiang-Shanxi Water Supply Project (Phase I), for $100.0 million, approved 24 September 1997; Loan 1270-PRC: and Tangshan and Chengde Environmental Improvement Project, for $140.0 million, approved 25 November 1993; 1990 (year of approval) ongoing projects: Loan 1692-PRC: Suzhou Creek Rehabilitation Project, for $300.0 million, approved 29 June 1999; 2000 (year of approval) ongoing projects: Loan 1797-PRC: Tianjin Wastewater Treatment & Water Resources Protection Project, for $130.0 million, approved 11 December 2000; Loan 1985-PRC: $82.36 million, approved 19 December 2002; Loan 1985-PRC: Hebei Province Wastewater Management Project, for $82.36 million, approved 19 December 2002; Loan 1995-PRC: Harbin Water Supply Project, for $100.0 million, approved 11 March 2003; Loan 1996-PRC: Wuhan Wastewater Management Project, for $83.0 million, approved 25 April 2003; and Loan 2024-PRC: Xi An Urban Transport Project, for $270.0 million, approved 27 November 2003. 8. The indications are that ADB is reducing its urban orientation in favor of other sectors, contrary to the World Bank which is increasingly concentrating on urban sector lending. The

88 Appendix 7 PRC/ADB Poverty Reduction Partnership Agreement in 2003 targeted two thirds of lending volume to central and western provinces by 2005, while recognizing that urban projects could continue in coastal regions if directed towards the urban poor. ADB s support is also focusing more on medium and small metropolitan regions, as opposed to large metropolitan regions. Most likely, ADB s first urban sector loan in the PRC will be for water and wastewater investments in small cities in Henan Province. 9. Success of ADB s Lending. Completed projects have invariably been rated successful or highly successful by operating department s project completion reports (PCRs) and by the Operations Evaluation Department s evaluations. The PRC s overall good performance with respect to ADB loans extends to the urban sector. In the project locations, ADB has contributed to significant results, and scores on the sustainability of the investments have been higher than in many other DMCs. However, not everything has gone well. Considerable transaction costs are incurred both by ADB and the PRC. Loan cancellations for the completed projects have been extraordinarily high, amounting to 25% of the loan amounts. This is 5-8% higher than the (high) ADB average for loan cancellations. Project implementation units (PIUs) consulted by this study reported that they were dissuaded from seeking the reallocation of savings because the process required approval by the headquarters in Manila, and were tied to pre-set reimbursement ceilings that seemed difficult to change. The more competitive financing from domestic sources on some projects, although not financially competitive with ADB loans, required far less administration costs and could finance directly, rather than reimburse, such as is the case with some of ADB s financing. It was easier to get replacement financing from local banks. The complete dropping of components from the ADB loan agreement, while occasionally unavoidable, suggests that initial project preparation may have been less than adequate. Special circumstances also played a role, i.e., the slump in equipment prices and heavy competition for civil works contracts in the 1990s. 10. Comparison with the World Bank. Project administration by ADB was often compared with that by the World Bank. Decision making in the World Bank was viewed as more decentralized. When a World Bank loan encountered savings, the task team leader had the authority to approve reallocation to other components. If a component was dropped, the task team leader could appraise an additional component or reallocate to existing components that might need more financing. As long as the original intention of the loan remained consistent, the only internal approval was generally a sign-off and amendment to the Loan Agreement by legal staff. Senior management was consulted depending on the scale of a loan amendment, but there was more flexibility and approval authority vested by the World Bank in its task team leaders than by ADB in its project officers. 11. In spite of the drawbacks, the ongoing ADB urban sector project portfolio is widely acknowledged to perform well, as was confirmed during the field visits (Box A7.1). Box A7.1: Excellent Progress with ADB-supported Projects in the People s Republic of China Harbin Water Supply Project, a with an estimated cost of $400 million, is supported by a loan of $100 million. It is 1.5 years into its implementation period, and scheduled to be completed a year ahead of schedule, by the end of 2007. The project is already regarded as very successful. It is constructing the Mopanshan dam, resettling about 10 villages (1,871 families) in the reservoir basin, laying a 176 kilometer (km) 2.3 meter (m) diameter pipeline from the dam to Harbin city (including a number of river crossings), and constructing a large water treatment plant and pumping stations, and associated reservoir storage, with a capacity to provide around 450,000 cubic meters (m 3 ) of water per day to the

Appendix 7 89 city. The project subcomponent, which involves laying around 110 km of pipeline within the city to improve the overall distribution network, has not yet started. Suzhou Creek Rehabilitation Project, b at an estimated cost of $876 million, included an Asian Development Bank (ADB) loan of $300 million, approved in 1999. The project had 10 components, including wastewater interception in various areas, low flow augmentation structures, wastewater treatment, relocation of night soil and garbage wharves, a flow control structure, sediment dredging and disposal, re-aeration, and embankment construction. The project had top political support and a high public profile, won international environmental awards, and created much goodwill for ADB. ADB s recent project completion report rated it highly successful. However, the Government cancelled $135 million out of the loan, due to cost savings and ADB s adherence to the agreed (low) reimbursement percentages on the civil works portions. It was easier to get replacement financing from local banks. ADB s reimbursement inflexibility has not encouraged the Shanghai local government to request the Central Government for ADB involvement in follow-up projects. Tianjin Wastewater Treatment and Water Resources Protection Project c is 2.5 years into its implementation period. The project is on schedule, and close to completion. The total project cost is $340.7 million and ADB supports the project with a $130 million loan (38%). There are two components: (i) wastewater collection and treatment, and (ii) water resources protection. The Government is satisfied with the smooth implementation of the project. Hebei Province Wastewater Management Project, d at $165.34 million, includes financing of $82.36 million by an ADB loan. It has five components, covering improved wastewater collection and treatment systems in four project cities. Hebei project management office manages both ADB and World Bank projects. The project is ahead of schedule; most components do well. The wastewater project in Zhangjiakou has a very high profile because of its effect on the water for Beijing, which is downstream. The sewage company consulted was content with ADB s support to the project. The Wuhan Wastewater Management Project e consists of (i) three wastewater treatment plants for the central city of Wuhan, with a joint capacity of 520,000 m 3 /day; (ii) 105 km of sewers; and (iii) capacity building (including project management), expanded environmental monitoring, development of a water quality model, and establishment of a wastewater management training center. The total cost of the project is $193.6 million, including $83 million of an ADB loan. Counterpart funds are provided through revenues ($86.8 million) and local bank loans ($5.7 million). This is Wuhan s first project in the urban sector. The project management office commended ADB s involvement. The successful cooperation has already led to the preparation of another wastewater project with a total investment of $250 million, with $100 million ADB cofinancing. a ADB. 2003. Report and Recommendation of the President to the Board of Directors on a Proposed Loan to the People s Republic of China for the Harbin Water Supply Project. Manila. b ADB. 1999. Report and Recommendation of the President to the Board of Directors on a Proposed Loan and Technical Assistance Grant to the People s Republic of China for the Suzhou Creek Rehabilitation Project. Manila. c ADB. 2000. Report and Recommendation of the President to the Board of Directors on a Proposed Loan to the People s Republic of China for the Tianjin Wastewater Treatment and Water Resources Protection Project. Manila. d ADB. 2002. Report and Recommendation of the President to the Board of Directors on a Proposed Loan to the People s Republic of China for the Hebei Province Wastewater Management Project. Manila. e ADB. 2002. Report and Recommendation of the President to the Board of Directors on a Proposed Loan to the People s Republic of China for the Wuhan Wastewater Management Project. Manila. Source: Special evaluation study field visits to the People s Republic of China, August September 2005. 12. The group of ongoing projects is almost exclusively in the WSSWM sector, and their straightforward nature might be one reason for their success. However, there is strong demand from cities for multisector projects, in order to deal with complex and pressing issues in a more comprehensive way. Demand from smaller cities has led to the preparation of the first sector loan. The new demands find growing support for more sector loans and multisector projects within the Ministry of Finance and the National Development and Reform Commission. The

90 Appendix 7 trend is, therefore, towards greater complexity but also greater responsiveness to city needs, capacity building, and larger potential impact. 13. Perceptions Regarding ADB s Added Value and its Transaction Costs. These were gauged during the interviews. The response can be summarized as follows: (i) (ii) (iii) (iv) (v) (vi) (vii) (viii) ADB s involvement leads to better preparation of the projects. As a result, implementation is more smooth and successful than that of projects funded exclusively by domestic commercial banks. The good reputation of ADB helps the executing agencies (EAs) to raise additional loan financing from domestic commercial banks at lower costs. ADB projects often receive more support from local government in land acquisition, resettlement, staffing, and tariff adjustment; and counterpart funds are more easily secured. ADB helps EAs develop capacity; projects are mainly implemented through internally staffed PIUs and specially created companies that later operate the project outputs. There are fewer influences from local government during implementation regarding issues that have been decided already before (requests to change the schedules or the scope of projects, etc.). Contractors in the PRC are more interested in bidding for ADB project contracts due to ADB s good reputation with respect to transparent procedures and secure payment. Contractors put in lower bids as the risks are perceived as lower. ADB s interest rates are competitive, and long-term loans on similar conditions can not be obtained from banks in the PRC. ADB s environmental safeguards have generally been appreciated and are acknowledged to have improved the quality of the projects. ADB s resettlement policy has been criticized on the ground that the policy of negotiating compensation directly with households is unworkable in the PRC, where land is collectively owned. The country s own system provides for negotiation and payment to village and town governments. 14. Key Recommendations. ADB s clients in the PRC gave the following recommendations: (i) (ii) (iii) (iv) More responsibility should go to ADB mission leaders in order to improve efficiency and effectiveness, particularly at the preparation stage, and to minimize approval procedures by invisible departments based in ADB headquarters. Responsibilities for procurement and payment control should be transferred to the ADB resident mission in Beijing, and ideally to Chinese speaking officers and/or staff. Problems due to the language barrier and the needs of back and forth language translation and transmission should be minimized. ADB s support in the form of capacity building TA, and with policy and strategy studies for local governments, is highly appreciated and should be strengthened further. This could then be followed by loans targeting the priorities and challenges identified. The strict reimbursement mechanism should be replaced by a more flexible mechanism that would allow the reallocation of loan funds among projects major expenditures (civil works, equipment, training, etc.) so that any loan savings are easily utilized.

Appendix 7 91 (v) (vi) (vii) The size and performance of imprest accounts should be increased, in order to minimize the need for a resorting to local banks in order to obtain bridging funds. Project preparation procedures should be simplified and shortened. It should be possible to make more use of retroactive financing in ADB projects. 15. ADB s TA Program in the PRC. ADB s support of institutional and capacity development was substantial but has been on the decline in the period after the urban sector strategy (USS). The same has been the case for ADB s project preparatory TA resources for the PRC. Loan consulting services for similar objectives has been at around 1.8% of the loans for the PRC before the USS; this was below the ADB average and declined to 0.8% after the USS. Much of the loan component is in practice directed towards management improvement of the PIUs. The World Bank, while not having devoted similar resources to grant-based TA, has often succeeded in securing bilateral grant funds for loan preparation; its loan consulting services have been between 3% and 5% of project cost, and have been directed more at local policy reforms. A close look at ADB s advisory TA program shows that they have performed important reform-oriented functions at the national level, and often kick-started capacity and institutions at the local level. B. India 16. India, like the PRC, has only had access to ordinary capital resources (OCR) loans and has not borrowed for the agriculture and natural resources sector (barring one exception), the education and health sectors, and some other sectors. Apart from the urban sector, the mainstay has been borrowing for the energy sector, the finance and governance sectors, and the transport and communications sector. ADB lending for the urban sector in India began in 1995 with the approval of two loans of $85 million and $20 million respectively, for the Karnataka Urban Infrastructure Project (KUIDP). 4 This was regarded as somewhat of a breakthrough, as the urban sector was then regarded as part of the social infrastructure macro sector, for which India had not previously wished to borrow. Fifteen loan approvals for the urban sector followed for another nine projects, for a total of $2,071.2 million until 2005 (Table A7.1 and Figure A7.2). Two approved housing finance loans were declined before being signed (not shown below). 5 Two projects have been completed so far. ADB has a considerable pipeline of new urban sector projects for India, which is envisaged to lead to loans totaling $1,250 million in 2005 and 2006. Between 1993 and 2004, ADB approved 31 grant-based TAs, for $17 million: 52% were for urban sector profile studies and project preparation activities, 22% for institutional strengthening of urban local bodies and housing finance institutions, 16% for capacity building of project implementation agencies, and 10% for poverty reduction and community development objectives. 4 ADB. 1995. Report and Recommendation of the President to the Board of Directors on Two Proposed Loans and Technical Assistance to the Republic of India and the Housing Development Finance Corporation Limited for the Karnataka Urban Infrastructure Development Project. Manila. 5 ADB. 2000. Report and Recommendation of the President to the Board of Directors on Four Proposed Loans to the Housing and Urban Development Corporation, National Housing Bank, Housing Development Finance Corporation, and ICICI for the Housing Finance II Project in India. Manila.

92 Appendix 7 Table A7.1: Projects and Asian Development Bank Loans for the Urban Sector in India $ million Year of Loan No. Project Title loan size Approval 1415/16 Karnataka Urban Infrastructure Development Project 105.0 1995 1549/50/51 Housing Finance Project 300.0 1997 1647 Rajasthan Urban Infrastructure Development Project 250.0 1998 1704 Karnataka Urban Development and Coastal Environmental 175.0 1999 Management Project 1719/20/21 Urban and Environmental Infrastructure Facility Project 200.0 1999 1759/61 Housing Finance II Project 120.0 2000 1813 Calcutta Environment Improvement Project 250.0 2000 2046 Urban Water Supply and Environmental Improvement in Madhya 200.0 2003 Pradesh 2151 Multisector Project for Infrastructure Rehabilitation Project in Jammu 250.0 2004 and Kashmir 2226 Kerala Sustainable Urban Development Project 221.2 2005 Source: Asian Development Bank databases. Figure A7.2: Urban Sector Lending in India Notes: 1990 (year of approval) completed projects: Loan 1415-IND: Karnataka Urban Infrastructure Development Project, for $85.0 million, approved 14 December 1995; and Loan 1416-IND: Karnataka Urban Infrastructure Development Project, for $20.0 million, approved 14 December 1995; 1990 (year of approval) ongoing project: Loan 1647-IND: Rajasthan Urban Infrastructure Development Project, for $250.0 million, approved 3 December 1998; 1990 (year of approval) completed multi-location projects: Loan 1759-IND: Housing Finance II Project (National Housing Bank), for $40.0 million, approved

Appendix 7 93 21 September 2000; Loan 1758-IND: Housing Finance II Project (Housing and Urban Development Corporation [HUDCO]),for $100.0 million, approved 21 September 2000 (The approved amount was terminated 7 September 2001); Housing Finance II Project (Housing Development Finance Corporation [HDFC]), for $80.0 million, approved 21 September 2000 (the approved amount was terminated 8 October 2001); 2000 (year of approval) ongoing projects: Loan 1704-IND: Karnataka Urban Development and Coastal Environmental Management Project, for $175.0 million, approved 26 October 1999; Loan 1813-IND: Calcutta Environmental Improvement Project, for $250.0 million, approved 19 December 2000; Loan 2151-IND: Multisector Project for Infrastructure Rehabilitation in Jammu and Kashmir, for $250.0 million, approved 21 December 2004; Loan 2046-IND: Urban Water Supply and Environmental Improvement in Madhya Pradesh, for $200.0 million, approved 12 December 2003; 2000 (year of approval) completed multi-location projects: Loan 1719-IND: Urban and Environmental Infrastructure Facility Project (HUDCO), for $90.0 million, approved 17 December 1999; Loan 1721-IND: Urban and Environmental Infrastructure Facility Project (Infrastructure Development Finance Corporation), for $30.0 million, approved 17 December 1999; 2000 (year of approval) ongoing multi-location projects: Loan 1720-IND: Urban and Environmental Infrastructure Facility Project (ICICI), for $80.0 million, approved 17 December 1999; Loan 1759-IND, and Loan 1761-IND: Housing Finance II Project (ICICI), for $80.0 million, approved 21 September 2000. 17. Urbanization. Although the share of the urban sector in India s GDP has increased from 60% to nearly 70% in 2001, urban population growth has strained the level and quality of urban services. While 46% of urban households have safe sanitary facilities, only 28% are connected to public sewerage systems. Of the 89% who nominally have access to piped drinking water, most receive water of poor quality, and very irregularly. India s Planning Commission has estimated that 21% of the urban population is living in squatter settlements. The urban population living below the income poverty line is smaller than the rural population, and has shown a gradual reduction over the years. However, there has been an increase in the actual number of urban poor living in squalid social, health, and housing conditions. Overall, Indian cities house 315 million people, which represents 25% of the population. 18. Government Response. Since the formulation of the ADB USS in 1999, India has witnessed rapid economic growth and the national urbanization approach has gone through important revisions. The core elements of the new national strategy in India are (i) administrative decentralization and devolution of fiscal authority to urban bodies as a result of the 74 th constitutional amendments in 1992; (ii) increased emphasis on governance and business process reforms in urban management; (iii) growing awareness of the need to introduce market disciplines in urban infrastructure financing and service provisions; (iv) emergence of urban investment policies by the Government of India, based on state and municipal reforms in fiscal management; and (v) increased reliance on market loans for financing urban infrastructure investments. The Indian financial market witnessed significant structural changes with the introduction of innovative municipal and urban infrastructure development bonds. The rules of the game changed for external assistance in urban development, with the emergence of private financial institutions to support urban infrastructure investments, larger foreign direct investment flows in infrastructure and housing sectors, growing demand-based infrastructure financing structures, and increasing community participation in municipal service management. 19. Public sector investment in India has traditionally been oriented more to the rural than the urban sector. This started changing at the time of the 8 th Five Year Plan (FYP) in 1992. India s 10 th FYP (2002 2007) estimated the investment requirements of urban water supply and sanitation of $2.5 billion per year. 6 In comparison, the combined planned allocation for the Ministry of Urban Development (MUD) and Ministry of Employment and Poverty Alleviation 6 This estimate is based on the assumptions of 100% service coverage for water supply and 75% coverage for sanitation by 2007.

94 Appendix 7 (MEPA) for urban development projects 7 is $2 billion for the plan period, i.e., $400 million per year, which is only 16% of the investment needs. This means that nearly 84% of the needed urban investment would have to come from other sources such as external assistance, the private sector, and institutional financial markets. 20. As part of the new reform-based urban investment strategy, MUD and MEPA formulated the Urban Reform Incentive Fund, City Challenge Fund, and Pooled Financing Facility for Small Urban Local Bodies. 8 These schemes were subsequently merged with the much larger National Urban Renewal Mission (NURM) declared in 15 August 2005. The main objectives are (i) planned development of urban centers, (ii) integrated development of urban infrastructure, (iii) provision of adequate urban services with a primary focus on the urban poor, (iv) effective linkage between asset creation and asset management, and (v) reform-linked provision of investment funds to urban local bodies. Under the NURM, the Government proposes to provide Rs55 billion ($1.24 billion as of 3 April 2006) during 2005 2006 through a combination of sources such as plan allocation, external assistance, and market borrowing to urban local bodies as loans and grants. This represents a tripling of investment compared to previous years. The investment support under NURM is conditioned on a large number of fiscal and service management reforms at the level of urban local bodies, and institutional reforms at the level of state governments. 9 Although NURM is a quantum leap for the urban sector, it is only one of the special programs launched by the new Government in 2005, and by no means the largest. 10 21. Support of External Agencies. During 1993 2004, external agencies approved $6.78 billion to support the urban sector, 90% of which was loans. ADB was the largest external contributor during this period but World Bank funding is currently growing faster. After 1999, the World Bank approved urban projects worth $2.12 billion in loans. Japan Bank for International Cooperation funding was smaller, and was mainly for the urban water supply and transportation sectors. Of the grant donors, the United States Agency for International Development, United Nations Development Programme, United Kingdom Department for International Development, and the Netherlands have been particularly active. Since 2003, the Government of India has limited the number of bilateral agencies in India. 22. External assistance has not kept pace with growing urban investment needs. The Government s strategy to preselect states for external loan-based urban investments within the framework of capital expenditure plan allocations, its loan guarantee policy, and the limited number of states and urban local bodies that are allowed to absorb reform-linked loans limits 7 Major urban sector projects covered in the 10 th FYP are the Mega City Project, Integrated Development of Small and Medium Towns, Accelerated Urban Water Supply Program, Local Sanitation Program, Solid Waste Management Projects, Urban Transportation Projects, and Poverty Alleviation Programs. 8 Some of these initiatives are supported by the United States Agency for International Development-assisted Financial Institutional Restructuring Project. 9 Urban Local Body level reforms are: implementation of 74 th Constitutional Amendment Act, accrual and fund based municipal accounting, e-governance, property tax reforms, reasonable cost recovery, and earmarking internal resources for service provision for the urban poor. State level reforms are related to repeal of the urban land ceiling act, rationalization of stamp duties, reforms in the rent control act, and the introduction of independent regulators for urban services. 10 The total outlay for the Jawaharlal Nehru National Urban Renewal Mission for urban infrastructure development is expected to be Rs100,000 crores ($22 billion) in 63 of the largest cities in India. ADB's plan is to dovetail with this new initiative. While the funds are grants, ranging from 30% for larger cities to 70% for smaller cities, subject to certain commitments by the cities, ADB loans could provide the remaining funds. ADB is providing a TA to assist the Government in initiating the Jawaharlal Nehru National Urban Renewal Mission. This may lead to subsovereign lending, partnerships with private sector banks, public private partnerships, and bond issues.

Appendix 7 95 the flow of external finance to the urban sector. 11 ADB projects are in Jammu, Karnataka, Kashmir, Kerala, Madhya Pradesh, North Eastern States, Rajasthan and West Bengal. The World Bank has a similarly limited number of focal states. 23. Urban Sector Strategic Guidelines. ADB s urban strategic guidelines for India revolved around the India Urban Sector Strategy of August 1998, the USS of 1999, and the Country Strategy and Program (CSP) 2003 2006. 12 The India Urban Sector Strategy was based on the outcome of the TA on the Urban Sector Profile of 1994. 13 The strategy remained a staff paper and, therefore, had no official status in ADB. 24. The strategy reflected a 1996 expert group s estimate that the annual investment requirement for water supply, sanitation, and roads (excluding urban mass transit) was equivalent to $8.1 billion for the period 1996 2001 and $7.9 billion for 2001 2006. Under the 8th FYP, the total annual outlay for urban infrastructure averaged only $1.1 billion, which is less than half the annual operations and maintenance requirement for core services. Factoring in investments from the private sector and other sources, the expert group estimated the funds available for urban infrastructure as $1.4 billion per annum. This implied that there was a gap of at least $200 million per year to clear the backlog over the period 1995 2005. If urban transport requirements were added, then the gap would rise another $150 million a year. 25. The strategy indicated that ADB s principal activity should be to finance multisector urban projects. In line with ADB s USS, it stressed that such projects should be demand-driven, improve urban environmental and living conditions (especially of the urban poor), and strengthen the capacity of local governments. The strategy outlined various ongoing initiatives, notably in the area of housing finance, but did not project the size of the ADB investment required, the commitment made, and the ADB staff input required for implementing the strategy. 26. ADB set out to revise the India strategy of 1998 in 2003 through a TA. 14 Like the 1998 strategy, the TA recommended three focus areas for ADB s urban intervention: (i) infrastructure and services, (ii) transportation, and (iii) housing. Contrary to the above recommendations of the TA, the urban sector strategy formulated by MUD and MEPA focused on the reform-linked urban infrastructure investment approach, as well as increasing the provision of basic urban services, particularly for the urban poor. 27. The CSP 2003 2006 was the first operational strategy proposed for India after the adoption of ADB s Poverty Reduction Strategy in 1999, and it aimed to help achieve the Millennium Development Goals (MDGs) by 2015. The strategic link of urban development with poverty reduction was established through social development, which included social infrastructure projects and state level interventions to strengthen the delivery of pro-poor social services. The CSP shared the Government s vision of sustainable urban development but acknowledged two constraints to ADB urban sector operations in India; (i) ADB was anticipated to provide one urban project at most a year given the immense resource constraint; and (ii) lack of Asian Development Fund 15 resources would not allow ADB to extend support to direct poverty reduction projects. Financially viable, revenue-generating projects at the state level appeared to be best suited for ADB financing considering the relatively high cost of ADB funds 11 The World Bank has confined its urban operations to Andhra Pradesh, Gujarat, (recently) Karnataka and Tamil Nadu. 12 ADB. 2003. Country Strategy and Program (2003 2006): India. Manila. 13 ADB. 1994. Technical Assistance to India for the Urban Sector Profile. Manila. 14 ADB. 2003. India Urban Sector Review and Strategy. Manila. 15 Concessional loan funds that ADB makes available to certain low income countries.

96 Appendix 7 (relative to cheaper World Bank concessional funds) and the opportunity to promote urban governance reform at the level of local government. The strategy was to help develop decentralized, professionally managed, and financially sound urban local governments that could be engines of growth and provide essential social services in a cost-effective manner (footnote 12). 28. Multisector Projects. ADB s urban sector project interventions in India, starting with the KUIDP in 1995, have been generally in line with the multisector framework used in the urban development planning practice in India. Except for the housing and urban finance credit lines, all ADB s loans for the urban sector have been for multisector projects. The Government s policy to assign focal states for external assistance caused state governments to view externally aided projects as a one-time opportunity, demanding a wide range of investments in many sectors at once, which has increased the complexity of the urban projects. Project preparatory TA consultants were often pressed to incorporate components such as urban water supply, drainage and sanitation, roads, traffic management, low-income housing, slum improvements, residential and industrial sites and services, truck and bus terminals, microcredit mechanisms, and community development. The approach delivered sometimes impressive results in high profile activities such as water supply, slum improvement, and community development. However, it also caused the delivery of large sewage treatment plants, which have remained insufficiently utilized due to lack of connections with households (as in the PRC). It is early to assess the extent to which the multisector project modality works better than the single sector modality, as KUIDP is the only completed multisector project in India so far. It was rated successful by ADB s PCR, and OED s Project Performance Audit Report has confirmed this rating, mainly because it generated significant positive externalities in Karnataka State, and significantly informed the national debate on urbanization. It was the first major integrated urban investment project in several mid-sized cities in India in many years, not only for ADB but also for the Government of India. As such, it generated substantial experience with the modality, and led to further statewide and India-wide policy initiatives. It also helped establish the Karnataka (KUIDFC) as a professional urban investment center. After the start of KUIDP, the World Bank developed similar multisector projects, and other agencies followed with related types of projects for the urban sector. The Rajasthan Urban Infrastructure Development Project (RUIDP) also looks successful, due to the highly appropriate inclusion of the construction of a 2.3 m diameter pipeline over 72 km to a reservoir, which will save Jaipur s water supply. Projects in Madhya Pradesh, Coastal Karnataka, and Kolkata appear to be very appropriate in addressing major problem areas, but are at relatively early stages and are facing implementation delays. Some key experiences with ongoing multisector projects are summarized in Box A7.2. Box A7.2: Uneven Progress in India with Multisector Projects Karnataka Urban Development and Coastal Environment Management Project, supported by an Asian Development Bank (ADB) loan of $145 million, has 60 subcomponents in six sectors across 10 towns. Project performance has been satisfactory except in Mangalore, which receives around half of the project investments. Mangalore subcomponents have experienced a 2-year delay, mainly due to poor project ownership, lack of agreement on the size of loan, inability to channel the planned community participation, and disagreement over project-related reform initiatives. In addition, there were internal problems between the project management unit (PMU), project implementation unit (PIU), project management consultant, design and supervision engineers, and quality control consultants. Rajasthan Urban Infrastructure Development Project (RUIDP), supported by a $250 million loan, is helping six towns in Rajasthan, including Jaipur. In the absence of a dedicated finance corporation, as in Karnataka, the RUIDP has been designed on the traditional model of a PMU set up within line department

Appendix 7 97 structures. Project implementation is carried out by eight PIUs established in the six project towns. Project performance is rated satisfactory in spite of a slippage of about 2 years. ADB is regarded as having taken long to decide on a major scope change, in order for the project to build a necessary pipeline to a remote water source. Kolkata Environmental Improvement Project (KEIP), supported by a loan of $220 million, is to improve sewerage, drainage, solid waste, canals, slums, and helps with resettlement. The Department for International Development (DFID) supports the project with a grant equal to $42 million for capacity building of the Kolkata Municipal Corporation (KMC). The project is already delayed by 18 months, due to reasons such as (i) lack of project ownership by the state government and KMC, (ii) sketchy feasibility studies, (iii) poor staffing of the PMU, (iv) lack of incentives to promote speedy implementation (the Government pays the commitment charges), and (v) a weak setup for community consultation and public relations. Due to slow progress and high commitment charges, the Government recently canceled $44 million of the ADB loan. This negative experience has led to a revitalization of the project in 2005 and has significantly improved project performance. For instance, contract awards have speeded up significantly, and coordination of KMC s poverty reduction initiatives with those under KEIP and DFID is very successful. Urban Water Supply and Environmental Improvement Project of Madhya Pradesh, supported by a $181 million loan, is helping six urban centers with water supply, environmental improvements, and capacity building. The project is cofinanced by DFID with a grant equivalent to $30 million. After the project preparatory technical assistance in 2003, the loan took nearly 20 months to become effective (May 2005). Four public elections, the code of conduct laws, and political discussion of the cost of ADB loans, contributed to the delay. One of the present difficulties is the lack of funds for establishing PMUs and PIUs. The absence of retroactive financing by ADB delayed the establishment of the project management consultant and Design and Supervision Consultant teams. Although consultants funded by a bridging technical assistance designed fast-track subprojects, their mainstreaming to satisfy community expectations could not be achieved. Given the need for and the suitability of the subprojects planned, the expectations are still high. Source: Special evaluation study field visits to India, July September 2005. 29. Credit Lines for Urban Sector Projects. Housing finance loans were an important component of ADB s urban sector programs during the late 1990s, the main objective of which was to provide housing financial assistance to low-income households. The results varied significantly, from highly successful projects to unsuccessful projects and cancelled loans. Although the first loans provided to the National Housing Bank (NHB) and Housing and Urban Development Corporation (HUDCO) were successful, only NHB has participated in a second (smaller) housing finance loan. With more attractive finance gradually becoming widely available in India, HUDCO did not use the second loan, mainly because of the high cost of the necessary foreign exchange swap and a project design which was perceived as poor. HUDCO was also discouraged by observations of ADB auditors that HUDCO (as NHB later) did not fully comply with loan eligibility and end-user guidelines. ICICI Limited (ICICI), a private banking and financial institution, also participated in the second housing loan with a net loan of $80 million (footnote 5). 30. NHB utilized $20 million from its $40 million loan for refinancing of loans of housing finance institutions. Following problems with the provision of end-user details to ADB, NHB cancelled $20 million and further refunded $13 million from the amount withdrawn earlier, reducing the net loan to $7 million. In contrast, ICICI has withdrawn and disbursed $70 million from the second housing loan, and is likely to withdraw the remaining part before loan completion in 2007. ADB loan review missions assess the ICICI housing loan as successful.

98 Appendix 7 31. Determination of the eligibility of low-income households for utilizing housing loans was one of the main difficulties experienced in these loans. ICICI resolved this by an agreement with ADB to use a household monthly income threshold of Rs7,000 ($165) taking into account the income of the head of the household and spouse (and not any other earning members of the family) and linking this with the general consumer price index. While the loan has been very successful in the retail segment, its performance in housing loans to community-based financial institutions and employer assisted housing schemes, was less than satisfactory. Based on its poor disbursement through community/group-based lending mechanisms, the loan was later reoriented to meet growing demand from the retail segment. 32. ADB supported an urban infrastructure financing facility with an $80 million loan to ICICI for supporting urban environmental projects on a commercial format. The loan became effective in 2000 and is expected to close in 2006. Due to poor disbursement, ICICI has cancelled $38 million from the agreed loan of $80 million. The facility s performance is less than satisfactory, mainly because of the lack of commercially viable water supply and sanitation projects. This has prompted both ADB and ICICI to agree upon funding industrial projects, mostly in pollution mitigation, treatment of industrial waste, and other related projects. 33. With the exception of the ICICI component of the second housing finance loan, the performance of ongoing housing and urban infrastructure finance loans is less than satisfactory. Difficulties in determining low-income households, monitoring the end-use of funds, the high cost of ADB funds, and the Reserve Bank of India s guidelines on priority lending are some of the reasons for the lack of demand for ADB housing loans. 34. Views on Project Problems. A discussion with a senior official of the Ministry of Finance, as the Borrower, pointed to (i) dissatisfaction with ADB s commitment fees, which are charged rigidly when there are project delays, even when project schedules are legitimately different from the standard commitment fee schedules; and (ii) high transaction costs to meet ADB safeguards. 35. Interviews held with senior project staff pointed to the following main reasons for the generally slow rate of fund absorption and ADB urban sector lending: (i) institutional complexity involved in designing and implementing integrated urban projects; (ii) inadequate local government ownership in many cases, projects were driven by the central Government and state governments; (iii) weak eligibility criteria for the inclusion of urban local bodies in the project; (iv) lack of market discipline by local governments in formulating revenue-generating subprojects; (v) poor local capacity to design and implement projects; (vi) lack of advance actions to sustain investments such as institutional reforms particularly those related to cost recovery, acquisition of land, and procurement of technical expertise; (vii) inadequate community participation; and (viii) weak communication and media relation strategy of projects. 36. Performance of Nationally Funded Projects. In spite of all the drawbacks to the success of the completed and ongoing ADB supported projects, their performance may not be worse than that of projects exclusively supported by the Government of India. Supplementary Appendix I contains performance assessments of two of the three main centrally sponsored schemes by the Comptroller and Auditor General and the National Institute of Urban Affairs: (i) the Accelerated Urban Water Supply Program, and (ii) the Mega City Scheme. 37. One argument often brought forward was the need for clear added value beyond the financial value of ADB loans. Project financing within India by commercial banks and semi government financing organizations has recently become much more competitive; the difference

Appendix 7 99 between ADB and commercial and parastatal financing organizations is sometimes summarized as 7.0% real interest rate (including swap) for OCR loans and 7.5% to 9.0% for other project financiers. State governments in particular are sensitive to this comparison. Views from various stakeholders on value created and value lost as a result of ADB involvement in urban sector projects were summarized in Supplementary Appendix J. There are some considerable criticisms of ADB s practices. 38. ADB s urban sector program is clearly evolving, and experimenting with a variety of project modalities and implementation structures: KUIDP was decentralized, Karnataka Urban Development and Coastal Environment Management Project and RUIDP were centralized, the Kolkata Environmental Improvement Project (KEIP) is driven by a PMU based in a local government, and the Urban Water Supply and Environmental Improvement Project of Madhya Pradesh is driven by PIUs similarly based in local governments. The Northern Karnataka Urban Infrastructure Development Project will most likely have much more involvement from the private sector. Some of the lessons learned from the experience with these projects include reliance of projects on (i) involvement of state-based finance corporations to identify and manage urban sector investments; (ii) e-procurement and open tenders for contracts rather than on prequalification, which can be manipulated by contractor coteries; (iii) bridging TA consultants for initiation of advance actions; and (iv) preparation of management action plans for urban local bodies. C. The Philippines 39. ADB in the Sector. The Philippines is ADB s fourth largest borrower in the urban sector and ADB s lending program has historically been concentrated on water supply. The first urban sector project in the Philippines was approved in 1976. Consequently, it has a much older urban sector program than India and the PRC. ADB s involvement consisted of 2 water supply projects in the second half of the 1970s, 4 in the 1980s (and a first sewerage project), and 6 more water supply projects in the period 1990 1995. 40. In the last 10 years, the program has diversified significantly. ADB s first multisector project was approved in 1995: the Regional Municipal Project. 16 Since then, 4 more water supply projects were started, 3 integrated urban development projects, 2 environmental management projects, 1 project for shelter and poverty reduction, and 1 project for private sector finance (Figure A7.3). 16 Loan 1367-PHI, for P30.0 million, approved 30 August 1995.

100 Appendix 7 Figure A7.3: Urban Sector Lending in the Philippines Note: 1970 (year of approval) completed projects: Loan 190-PHI: Manila Water Supply Project, for $51.3 million, approved 28 August 1974; Loan 351-PHI: Second Manila Water Supply Project, for $49.0 million, approved 7 September 1978; 1970 (year of approval) completed multi-location project: Loan 251-PHI: Provincial Cities Water Supply Project, for $16.8 million, approved 16 December 1975; 1980 (year of approval) completed multi-location project: Loan 545-PHI: Water Supply Sector Project, for $46.0 million, approved 25 November 1981; 1980 (year of approval) completed projects: Loan 457-PHI: Manila Sewerage Project, for $42.8 million, approved 24 June 1980; Loan 645-PHI: Manila Water Supply Rehabilitation Project, $39.3 million, approved 27 October 1983; Loan 947-PHI: Second Manila Water Supply Rehabilitation Project, for $26.4 million, approved 24 January 1989; Loan 986-PHI: Angat Water Supply Optimization Project, for $130.0 million, approved 14 November 1989; 1990 (year of approval) completed projects: Loans 1056-PHI and 1057-PHI: Metropolitan Cebu Water Supply Project, for $16.0 million and $6.0 million, respectively, approved 29 November 1990; Loan 1150-PHI: Manila South Water Distribution Project, for $31.4 million, approved 19 December 1991; Loan 1599-PHI: Subic Bay Area Municipal Development Project, for $22.0 million, approved 9 December 1997; Loan 1663-PHI: Metro Manila Air Quality Improvement Sector Development Program (Policy Loan), for $200.0 million, approved 16 December 1998; Loan 1664-PHI: Metro Manila Air Quality Improvement Sector Development Program (Air Pollution Control Facility), for $25.0 million, approved 16 December 1998; 1990 (year of approval) completed multi-location projects: Loan 1269-PHI: Municipal Water Supply Project, for $43.2 million, approved 25 November 1993; Loan 1367-PHI: Regional Municipal Development Project, for $30.0 million, approved 30 August 1995; Loan 1472-PHI: Small Towns Water Supply Sector Project, for $50.0 million, approved 30 September 1996; 1990 (year of approval) ongoing projects: Loan 1658-PHI: Clark Area Municipal Development Project, for $24.3 million, approved 15 December 1998; Loan 1665-PHI: Metro Manila Air Quality Improvement Sector Development Program (Investment Loan), for $71.0 million, approved 16 December 1998; 2000 (year of approval) ongoing projects: Loan 1745-PHI: Pasig River Environmental Management and Rehabilitation Sector Development Program, for $100.0 million, approved 20 July 2000; Loan 1843- PHI: Mindanao Basic Urban Services Sector Project, for $30.0 million, approved 27 September 2001; Loan 2012-PHI: Metropolitan Waterworks and Sewerage System (MWSS) New Water Source Development Project, for $3.26 million, approved 14 October 2003; 2000 (year of approval) completed project: Loan 1746-PHI: Pasig River Environmental Management and Rehabilitation Sector Development Program, for $75.0 million, approved 20 July 2000; 2000 (year of approval) ongoing multi-location project: Loan 2063- PHI: Development of Poor Urban Communities Sector Project, for $30.5 million, approved 18 December 2003.