Report No. 25899 Improving the Lives of the Poor through Investment in Cities An update on the Performance of the World Bank’s urban portfolio March 23, 2004 Operations Evaluation Department Sector and Thematic Evaluation Group Document of the World Bank Abbreviations and Acronyms APPI Aggregate Project Performance Indicator CDF Comprehensive Development Framework CDS City Development Strategy CODE Committee on Development Effectiveness (of Bank Board of Directors) CPIA Country Policy and Institutional Assessment ED Executive Director (of the Board of the World Bank) ERR Economic Rate of Return ESSD Environmentally and Socially Sustainable Development (network of World Bank) ESw Economic and Sector Work FPSI Finance, Private Sector and Infrastructure (network of World Bank) HIV/AIDS Human Immunodeficiency Virus/Acquired Immune Deficiency Syndrome ID Institutional Development ICR Implementation Completion Report IFAD International Fund for Agricultural Development M&E Monitoring and Evaluation MDP Municipal Development Project MDG Millennium Development Goal NGO Non-government Organization OECD Organization for Economic Cooperation and Development OED Operations Evaluation Department OD (World Bank) Operational Directive OP (World Bank) Operational Policy PPAR (OED) Project Performance Assessment-formerly Audit-Report PRSP Poverty Reduction Strategy Paper PTI Poverty Targeted Operation QAG Quality Assurance Group RBM Results-based Management UNCHS UN Habitat (formerly United Nations Centre for Human Settlements) UNDP United Nations Development Programme UPP-9 1 Urban Policy Paper - 1991 Urban Policy and Economic Development USP Urban Strategy Paper Cities in Trhnsition WDI World Development Indicators WDR World Development Report Director-General, Operations Evaluation Mr. Gregory K. Ingram Director, Operations Evaluation Department Mr. Ajay Chhibber Manager, Sector and Thematic Evaluation Mr. Alain Barbu Task Manager Mr. Roy Gilbert Foreword This report i s based on a desk review o f the Bank's urban portfolio. I t focuses on the results o f the 99 operations completed in the past ten years. I t uses the four pillars o f livability, good governance, bankability and competitiveness of the Urban Strategy Paper as the evaluation framework. Over time, urban development projects have increasingly sought to improve the lives o f the poor but robust evidence o f impacts has been meager. Through the triangulation o f project performance ratings with changes in basic urban service coverage-reported by UN Habitat urban indicators data at the city level-this evaluation detects significantly greater improvements in cities that hosted Bank-financed projects than in others. A t the project level, the study identifies factors that help determine good outcomes, such as building upon previous operations, involving beneficiaries, and avoiding straining borrower resources and implementation capacity. At the strategic level, the study finds that the portfolio has concentrated on the livability pillar, through projects aiming to make the lives o f the urban poor more healthy and productive. Attention has also been paid to governance too, especially through operations that strengthen municipal administration. Bankability aspects received some attention while the competitiveness pillar-that seeks improvements to the workings o f urban markets-has proved the most elusive. OED recommends: 0 Systematic M&E and reporting o f results-ofpoverty alleviation especially-fiom the city to the sectorhtrategic levels. The Bank should go beyond the USP 's 'illustrative' indicators and put in place an M&E program to measure results o f Bank investments in cities and report on them on a regular basis. 0 Revision of the USP 's business strategy to ensure successful implementation, This would provide explicit targets and determine priorities that link the USP 's four key instruments- scaling-up services to the poor, city development strategies, national urban strategies, and local government capacity building-and four strategic pillars-livability, good governance, bankability and competitiveness-to urban poverty alleviation. 0 ClariJication of the concept and the operational consequences o f the competitiveness USP pillar for urban practitioners. One way of doing this could be through issuing region-speci$c guidelines explaining to Bank task managers, borrower project managers, city mayors and other oficials, how to get urban poverty alleviation results through the implementation of this pillar. Gregory K. Ingram Director-General Operations Evaluation 11 Table o f Contents Preface........................................................................................................................................... ... 111 Executive Summary....................................................................................................................... v 1. Context and Methods.............................................................................................................. 1 2. Evolution o f the Bank’s Urban Portfolio .............................................................................. 3 3. Better Projects in Cities, Better Lives for the Poor............................................................ 11 4. Improving Lives in Cities by Implementingthe Strategy ................................................. 18 5. Recommendations ................................................................................................................. 25 Annex 1 : Urban Output and Outcome Indicators from USP.................................................. 27 Annex 2: Selected Performance Data: Urban Portfolio 1993-2000......................................... 28 Annex 3: Urban Portfolio: Completed Projects 1993-2000 .................................................... 31 Annex 4: Urban Portfolio cont: Ongoing Projects* ................................................................. 35 Annex 5: Management Response ............................................................................................... 37 Annex 6: Chairman’s Summary: Committee On Development Effectiveness............ 39 Bibliography................................................................................................................................. 43 ... 111 Preface Cities are home to 525 million poor people, and the Bank makes substantial investments in developing country cities every year. This study by the Operations Evaluation Department (OED) reviews the performance o f 99 urban development operations completed since 1993 to see, in particular, how these interventions have improved the living conditions o f the urban poor, the primary goal o f the livability pillar o f the Bank’s current urban strategy. This i s OED’s first review o f the urban portfolio since its 1994 study o f the first 20 years of Bank urban lending. Since that study was completed, the Bank has invested US$6 billion more in assistance for urban development. From Regional and OED assessments o f project performance and other sources, the present study compiled a database o f more than 120 variables for each o f the 99 urban operations completed during 1993-2001. While the information in this database was the cornerstone o f the review, the evaluation also made use o f other data. These included country statistics from the Bank’s World Development Indicators and city statistics from UN Habitat Urban Indicators Program, as well as data from older projects covered by OED’s earlier Twenty Year review and newer operations that are s t i l l ongoing as o f September 2001, Designed primarily as a desk study, the present review also included interviews o f 45 borrower managers o f urban projects worldwide. Supporting all this effort was a review o f academic and professional literature relating to urban development, especially from non-Banksources. The core team for this work was made up o f Roy Gilbert, Anna Amato, and Romayne Pereira. Roy Gilbert wrote the report. The telephone survey o f project managers was designedjointly with OED, and conducted by the Gallup Organization under the supervision o f Ajay Bhardwaj, with the assistance o f Chris McComb, David Osborne, and Anna Maria Salih. For the extensive data analysis conducted, the study benefited from the help o f OED colleagues Sohail Malik and Anju Gupta Kapoor. The core team also benefited from advice and inputs at various stages o f this work from fellow OED Urban Cluster members Soniya Carvalho and Ron Parker. Bill Hurlbut edited the report. Finally, the team’s special thanks go to urban project managers in our client countries for their collaboration in our telephone interview and also to Bank urban task managers for facilitating these contacts. Written comments were gratefully received from: Robert Buckley, Tim Campbell, William Cobbett, Elizabeth Campbell-Page, Victoria Elliot, James Fitz Ford, N i l s Fostvedt, Jonathan Kamkwalala, Christine Kessides, Omar Razzaz, Gwen Swinburn, Gene Tidrick, and Tom Zearley. Christine Kessides was a peer reviewer o f the report. iv About the OED Rating System The time-tested evaluation methods used by OED are suited to the broad range of the World Bank's work. The methods offer both rigor and a necessary level of flexibility to adapt to lending instrument, project design, or sectoral approach. OED evaluators all apply the same basic method to arrive at their project ratings. Following is the definition and rating scale used for each evaluation criterion (more information is available on the OED website: http://worldbank.org/oed/eta-mainpage. html). Relevance of Objectives: The extent to which the project's objectives are consistent with the country's current development priorities and with current Bank country and sectoral assistance strategies and corporate goals (expressed in Poverty Reduction Strategy Papers, Country Assistance Strategies, Sector Strategy Papers, Operational Policies). Possible ratings: High, Substantial, Modest, Negligible. Efficacy: The extent to which the project's objectives were achieved, or expected to be achieved, taking into account their relative importance. Possible ratings: High, Substantial, Modest, Negligible. Efficiency: The extent to which the project achieved, or is expected to achieve, a return higher than the opportunity cost of capital and benefits at least cost compared to alternatives. Possible ratings: High, Substantial, Modest, Negligible. This rating is not generally applied to adjustment operations. Sustainability: The resilience to risk of net benefits flows over time. Possible rafings: Highly Likely, Likely, Unlikely, Highly Unlikely, Not Evaluable. lnstitutional Development Impact: The extent to which a project improves the ability of a country or region to make more efficient, equitable and sustainable use of its human, financial, and natural resources through: (a) better definition, stability, transparency, enforceability, and predictability of institutional arrangements and/or (b) better alignment of the mission and capacity of an organization with its mandate, which derives from these institutional arrangements. Institutional Development Impact includes both intended and unintended effects of a project. Possible ratings: High, Substantial, Modest, Negligible. Outcome: The extent to which the project's major relevant objectives were achieved, or are expected to be achieved, efficiently. Possible ratings: Highly Satisfactory, Satisfactory, Moderately Satisfactory, Moderately Unsatisfactory, Unsatisfactory, Highly Unsatisfactory. Bank Performance: The extent to which services provided by the Bank ensured quality at entry and supported implementation through appropriate supervision (including ensuring adequate transition arrangements for regular operation of the project). Possible ratings: Highly Satisfactory, Satisfactory, Unsatisfactory, Highly Unsatisfactory. Borrower Performance: The extent to which the Borrower assumed ownership and responsibility to ensure quality of preparation and implementation, and complied with covenants and agreements, towards the achievement of development objectives and sustainability. Possible ratings: Highly Satisfactory, Satisfactory, Unsatisfactory, Highly unsatisfactory. V Executive Summary CONTEXT AND METHODS Each year, the Bank invests US$6-7 billion in developing country cities where 525 million poor people reside. Some 30 per cent o f a l l the world’s poor live in cities, where a dollar-a-day does not go as far as it does in the countryside. Meeting the Millennium Development Goal (MDG) o f halving poverty by 2015 would imply raising 24 million people out of urban poverty every year for the next fifteen years, a daunting task. Has Bank investment in cities improved the lives o f the poor? This i s the central question addressed by this evaluation, a desk study o f the policy context and performance o f 99 urban projects completed during 1993-2000, and update o f OED’s 1994 review o f urban operations. The study assesses the implementation challenges o f Cities in Transition-the Bank’s new urban strategy paper (USP). For its analysis, the study built an evaluation database from existing Bank sources as well as others such as UN Habitat city-level urban indicators. The only primary data collected for this work was through a worldwide telephone survey o f borrower managers o f 45 of the 99 completed urban projects. EVOLUTIONOF THE BANK’S PORTFOLIO The first decade o f Bank lending-1972-82-pioneered Bank urban operations and set priorities, such as slum upgrading and sites and services focused upon the urban poor. Low-income urban households were target beneficiaries from the outset, a priority inspired by former World Bank President McNamara’s 1975 speech and clarified by subsequent Bank urban policy work. The second decade-1983-92-saw a rapid expansion in urban lending and made Latin America the main regional client o f the Bank’s urban program. Urban projects s t i l l concentrated upon improving basic infrastructure and shelter, while a new line o f urban business expanded in the form o f municipal development projects (MDPs). With 78 per cent o f project outcomes rated satisfactory, portfolio performance was good, according to OED’s 1994 review o f urban lending, which found that best practice projects had unequivocal borrower ownership and explicit poverty reduction aims, lessons absorbed by later operations. The Bank’s 1991 Urban Policy Paper - - UPP-9 l-focused upon three “P’S’’, (urban) productivity, poverty reduction, and pollution abatement. I t was most successful in maintaining the portfolio’s focus on poverty reduction. Urban practitioners rarely took up the urban productivity theme. Pollution was addressed, but urban environment project performance was weak at that time. The third decade-exit years 1993-2000-witnessed sustained portfolio activity. Africa hosted most projects, but Latin America s t i l l accounted for most lending. Overall, satisfactory projects f e l l to 71 per cent o f the total. The nadir came in 1995, the year o f the completion o f twelve projects prepared around the time o f the disruptive Bank reorganization o f 1987. Since 1995, there has been a strong rebound. Urban projects helped strengthen municipal management and the housing sub-sector. MDPs used financial intermediaries to on-lend Bank and other funds to local governments in exchange for structural reforms at the city level across all regions. The focus upon urban poverty reduction became even stronger, being addressed by more than half the completed portfolio at this time. This decade’s urban activities also show responsiveness to OED’s 20-Year Study recommendation to strengthen the congruence o f project objectives and design with better results. Almost all projects that aimed at improving livability and governance incorporated specific components designed to help achieve these goals. On the other hand, less vi was done by the Bank to develop guidelines for M&E and to implement project specific M&E systems that OED had recommended. The fourth decade-2001 and beyond-poses the challenge o f implementing the Bank’s new Urban Strategy Paper (USP) approved in 1999, with i t s primary focus upon improving livability- decent quality o f l i f e for all, including the poor-through the good governance, bankability and competitiveness dimensions o f sustainable cities (defined below). In i t s 1999 comments on the draft USP, OED welcomed the new strategy, especially i t s matrix o f performance indicators to monitor results under each o f these four strategic pillars, but indicated that more work was needed on implementation. Thus far, the prospects for implementing the livability agenda are promising, however. In the ongoing portfolio o f 90 urban projects worldwide, 68 per cent o f projects had objectives focused upon improving the living conditions o f the urban poor, the highest proportion to date. Self-evaluation by QAG and the latest supervision missions suggests that final outcome ratings o f these projects will be good. BETTERPROJECTS IN CITIES: BETTERLIVESFOR THE POOR Despite the desirable shift by urban projects toward a more concentrated poverty focus, robust evidence o f their actual impacts upon livability in cities has been meager, especially at the strategic or sector level. In the absence o f performance data, the present review turned to another source, city-level UN Habitat urban indicators. Through triangulation with urban project ratings, it found robust evidence o f positive project impacts upon livability. Between 1993 and 1998, water, sewerage and solid waste service coverage extensions were significantly greater in 24 Bank client cities hosting urban projects than in 37 otherwise similar comparator cities without the benefit o f such operations. This result confirmed that cities with urban projects designed to extend service coverage within their urban areas did perform better. The study also used a simple least-squares regression model to identify factors that determine project performance, and hence should be taken into account in project design. Project factors: High levels of demandingness-that strains borrower resources and capacity i s negatively correlated with project performance. The Bank’s overestimation o f borrower institutional capacity was typically detrimental to outcomes, based on OED evaluations. Building on p r i o r experience-urban projects that incorporate the lessons o f predecessor project experience (covering h a l f the portfolio in 3 1 countries) perform better. Involvement of benejciaries-especially during project identification, was positively correlated with good outcomes. Projects with substantial beneficiary involvement had satisfactory outcomes 89 per cent o f the time. Since fewer than h a l f o f completed operations had a substantial degree o f participation, however, designers should do more to incorporate it in the future. Borrower involvement-mainstreamed into project design t o a greater extent has a similarly positive impact upon performance. Excellence in Bank service-through good performance during identification, appraisal, and supervision, i s also key to satisfactory results. Countryfactors: Projects completed in more urbanized countries (with more than 58 per cent o f their populations living in cities-the lower limit o f the top tercile) were 81 per cent satisfactory, while in less urbanized countries (less than 34 per cent living in cities) only 59 per cent o f projects were satisfactory. GDP per capita i s correlated with urbanization and therefore has a similar relationship with project performance. More urbanized and higher income countries give borrowers greater urban development experience, resources (including counterpart funding), and institutional capacity to perform better. Urban project managers may not be able to change these conditions, but they certainly need to be aware o f them. vii This statistical analysis was unable to find significant relationships between three factors conventionally associated with project performance: (i) poverty-focus in objectives-often associated by task managers with poorer performance through weaker effective demand by poorer beneficiaries-did not appear to affect outcomes; (ii) partnerships with co-financiers-appeared to have no impact upon performance, despite expectations o f high ‘transaction costs’ o f such arrangements; and (iii) economic and sector work - ESW - preceded only a minority o f projects-1 8 out o f 99-whose performance was not significantly different from the performance o f the majority o f urban projects prepared without ESW. In some cases ESW may have been carried out for other purposes than to improve project performance, e.g. for country dialogue. IMPROVING LIVES IN CITIES BY IMPLEMENTING USP How much Bank urban assistance will continue to improve the lives o f the poor in cities in the future will depend upon how successfully the new urban strategy i s implemented and how well the results are monitored. The USP’s matrix o f performance indicators will help, but we s t i l l await the results of monitoring them at the strategic level. The results needed are outlined below for each strategic pillar. Livability - Decent quality o f life and opportunitiesfor all, especially the poor. As its beacon on the poor, livability i s deservedly the principal pillar o f the USP. Through it, the Bank seeks to help make the lives o f the poor in cities more healthy and productive. Beyond its welfare content, it establishes conditions for the growth o f the urban economy that will benefit all citizens. New ongoing urban projects focus appropriately upon livability goals and dovetail neatly into the Bank’s own poverty reduction mandate (OD 4.15 -1990) that espouses the inclusion o f the poor through more healthy and productive lives. USP indicators for monitoring livability outcomes (e.g. urban poverty rates, infant or child mortality, rates o f waterborne diseases, and ambient air and water quality) and outputs (e.g. basic urban service coverage) are, for the most part, relevant to USP strategic goals and familiar to urban practitioners. Much more needs to be done, however, to monitor poverty results in practice. We can only claim good results o f Bank investments in cities on the lives o f the poor if we can base those claims on convincing evidence. Good Governance -Inclusion and accountability at the local level. For urban practitioners to assemble evidence o f governance results, they will need more guidance from the Urban Sector Board. In particular, they will need advice on quantitative measures o f performance. Meanwhile, anecdotal evidence suggests that by dealing directly with municipal governments, many completed urban projects have strengthened governance at the local level. If municipal governments are well managed, they are better poised to lead programs to improve livability in their cities. The most comprehensive instrument for improving local governance i s the City Development Strategy - (CDS). Projects also support governance through technical assistance, training and intensive borrower management o f procurement. Municipal governance can be enhanced by enabling private sector participation in service provision within a competitive environment. Bankability -financial soundness and creditworthiness: Proposed USP bankability performance indicators for local governments are among the strategy’s most robust and most readily quantifiable. In practice, municipal development projects have enabled the Bank to strengthen local government revenues. OED evaluations show local government fiscal gains attributable to Bank financed projects in Brazil, Ghana, the Philippines, Tunisia, and Venezuela. Progress with municipal creditworthiness, the second aspect o f the USP’s bankability pillar, has been much ... Vlll more limited. Even so, many urban practitioners f e e l the pursuit o f creditworthiness, insofar as it requires good governance, remains a worthy goal. Competitiveness - efficient markets in cities: The USP aims to improve the workings o f urban markets for land, labor, credit, and infrastructure and housing inputs using such instruments as city development strategies (CDS), and housing finance and infrastructure reforms. How to monitor these results will be a major challenge for urban practitioners, given that city-level data i s scarce for several proposed indicators. Currently, fewer than ten per cent o f urban projects address competitiveness, partly because urban practitioners are unsure what competitiveness means and how it should be fostered. More work i s needed to refine the term, clarify i t s underlying objectives and determine i t s practical application. In i t s Evaluative Note to CODE on the USP in 1999, OED raised concerns about the ambiguity o f the competitiveness concept and the uncertainty about how it would be addressed. It i s important for urban practitioners in the Bank to help mayors focus upon building good economic governance that emphasizes the comparative advantage o f their city economy. RECOMMENDATIONS To help the urban lending program focus more effectively upon getting urban poverty alleviation results within the current strategic framework, OED recommends: Systematic M&E and reporting o f results-of poverty alleviation especially-from the city to the sector/strategic levels. The Bank should go beyond the USP’s ‘illustrative’ indicators and put in place an M&E program to measure results o f Bank investments in cities and report on them on a regular basis. Revision o f the USP’s business strategy to ensure successful implementation. This would provide explicit targets and determine priorities that link the USP’s four key instruments- scaling-up services to the poor, city development strategies, national urban strategies, and local government capacity building-and four strategic pillars-livability, good governance, bankability and competitiveness-to urban poverty alleviation. Clarification o f the concept and the operational consequences o f the competitiveness USP pillar for urban practitioners. One way o f doing this could be through issuing region-specific guidelines explaining to Bank task managers, borrower project managers, city mayors and other officials, how to get urban poverty alleviation results through the implementation o f this pillar. 1 1. Context and Methods RATIONALE 1.1 The study deals with an important line o f business for the Bank which invests US$6-7 billion-some 30per cent‘ o f all its lending-in cities where some 525 million poor people reside: Five per cent o f all Bank lending i s for urban development projects reviewed in this report-infrastructure, housing, and municipal development operations-most o f which are focused upon the urban poor. Surprisingly for such an important parameter o f the Bank’s fight against poverty, a precise and broadly accepted measure o f how many people are living in urban poverty i s not readily available. This study’s estimate (details para. 1.6) o f 525 million urban poor currently living in cities corresponds to 25 per cent o f the 2.1 billion inhabitants o f developing country cities today. Worldwide, this would make cities home to some 30 per cent o f all the poor, against 70 per cent in rural areas.2 But the minority urban share understates the gravity o f the urban poverty problem. It gets bigger as cities themselves do. Moreover, poverty can be particularly degrading in cities, where a dollar-a-day goes less far than in the countryside. 1.2 The study updates progress since OED ’s I994 review: Since that report, 99 more urban development projects-with Bank investments o f US$6.6 billion-have been completed within a poverty-focused policy framework. How well they have done can point to the likely future success o f improving the lives o f the poor in cities. In addition, this update looks at progress in response to the earlier review’s recommendations (details Table 2.3), such as those that called for more coherence between urban project objectives and designs and a fuller exploitation o f the economic and fiscal linkages o f urban projects (World Bank 1994a). 1.3 The study assesses the implementation challenges o f the Bank’s new Urban Strategy - called USP in this report: Presented to the Committee on Development Effectiveness (CODE) in July 1999 and to the Board o f Directors in November 1999, the USP (Cities in Transition: World Bank Urban and Local Government Strategy) has guided Bank operations since i t s adoption. The present review examines the obstacles that stand in the way o f full implementation o f the strategy and at the likelihood o f achieving desired strategic outcomes o f improved urban lives, especially for the poor. Through pointing to what we have learned thus far fiom past experience, the review suggests where USP efforts are most likely to succeed and also points to where greater efforts are needed to overcome,strategic weaknesses. 1.4 Poverty alleviation goals in general and meeting the 2015 Millennium Development Goals (MDG) call for results which clearly show that Bank investment in cities improves the lives o f the urban poor. Poverty alleviation has been central to Bank urban development policy for decades. H a l f the 99 completed urban projects reviewed here had at least one objective trained upon poverty alleviation. The study asks if we are getting results. Can urban projects help halve poverty in cities by 2015 as the MDG imply? How can the Bank implement the current urban sector strategy to help achieve this? This study seeks answers to these and related questions by drawing upon the operational and policy lessons o f what now amounts to three decades o f urban lending by the Bank. DEVELOPINGCOUNTRY CITIES AND THE POOR 1.5 More and more of the developing world’s population lives in cities, even though urban population growth rates have slowed (Table 1.1). By 2015-the M D G target year-almost half the developing world will be urban, compared with today’s 40 per cent. Although there are many more large cities today, most urban residents live in cities o f less than one million people, the most com- 2 mon client o f Bank urban projects. Surprisingly given its centrality to the Bank’s mission statement, we do not have a precise and widely accepted measure o f how many o f the poor live in cities. 1.6 This study used a working estimate o f 525 million people living in cities below the resp- ective national poverty lines in 2000, using poverty shares estimated from the following: (i) 44 country reports to the Bank’s 2002 World Development Report; (ii) 132 developing country cities reporting to the UN Habitat Urban Indicators Program for 1998; and (iii) recent estimates at John Hopkins University (Bloomberg School 200 l).3 The estimate needs further refinement, o f course, not least o f all because it will serve as a baseline from which improvements can be observed. In the meantime, however, one thing i s certain. Urban poverty i s likely to increase, unless we can do more-in part through urban assistance-to spurn i t s encroachment among families in cities. Table 1.1. More People-Including the Poor-Living in Developing Country Cities 1970 1990 2000 2015 (projected) Urban population (total in millions) 654 1,320 2,100 2,849 Urban population (%of total population) 25% 34% 40% 48% Annual growth of urban population (% per annum) 3.6% 3.8% 2.6% 2.1% Large Cities >1 million inhabitants (number) 80 173 268 358 Urban poor (millions below national poverty line) 215 330 525 713 (356)’ Urban poor (% of urban population) 33% 25% 25% 25% (12.5%)* Sources and Notes: (i) Urban population andgrowth; 1970/1990 (World Bank 1991), 2000 (WDI 2001 CD-ROM), 2015 (UN Habitat 2001); (ii) Large cities (UN Habitat 2001a); (iii) Urban poor 1970-1990 (World Bank 1991), 2000 (World Bank 2002d - 44 countries), 201 5 share assumed unchanged; (iv) Developing countries include those defined as middle and low income by the World Bank for 1970-1990; (v) *MDG target figures in parentheses. 1.7 Despite prosperity in many o f these cities, poor people living in them often face squalid housing conditions, enjoy few urban services and lack security of tenure in illegal squatter settle- ments. Poor people in cities face constraints that their more numerous rural counterparts do not. Costs for most basic needs-notably food and shelter-are typically higher in a city, meaningthat a dollar-a-day there does not go as far as it does in the countryside. Nor do the urban poor have direct access to food as the rural poor do (Pernia 1994). Health deprivation in cities through inadequate water, sanitation, and drainage infrastructure-less necessary in rural areas-can be severe for the poor (McDade and Adair 2001). Relative deprivation i s worse as Gini coefficients estimated for 19 countries’ urban and rural areas show income inequality to be greater in cities (Bump and Hentschel 1998). In common with their rural counterparts, the urban poor are often without voice in political and bureaucratic processes, and rarely have the option to exit their deprivation. Women among them especially face discrimination in labor and housing markets, as well as difficulties in accessing property, credit, and urban services. Such exclusion poses not just a welfare problem for the urban poor themselves. It prevents their productive contribution to the urban economy. Thus, halving ur- ban poverty by 2015, as implied by the MDG, i s a worthy aim for the Bank’s current urban strategy (Tables 1.1 and 2.4). Achieving the M D G in cities, however, would imply taking 24 million4 people out o f poverty every year for the next 15 years, a daunting task indeed. 1.8 We now know that “watering and housing” the poor i s not by i t s e l f a solution to urban poverty (Moser 1997). But providing better sanitation and shelter through urban projects can enhance livability for the poor, affording them a decent quality o f l i f e and equitable opportunity, as defined in the Bank’s current urban strategy (Table 2.4). Support to the urban economy and i t s governance can help foster income earning opportunities, some o f which may benefit the poor. Thus, urban development assistance can be fully consistent with the Bank’s poverty reduction 3 policy which calls for projects to “raise the productivity o f the poor’s physical assets and increase their incomes, through the provision o f infrastructure, credit, technology and complementary inputs and by regularizing de facto land tenure rights” (OD 4.15 para. 27). METHOD EVALUATION 1.9 This review was based upon a desk study, conducted during the period July 2000- December 2001. It focused primarily upon the portfolio o f 99 urban projects completed between 1993 and 2000, how they performed and what their results were. Bank project investments in cities through other sectors-notably education, health, transport and power-are not covered here. They mostly lacked explicit urban (or rural) spatial references needed for a review like this. As their spatial dimension becomes clearer, their impact upon urban (and rural) poverty will also be amenable to assessment in the future. Meantime, this review’s assessment o f urban projects was complementedby reference to the very rich academic and professional literature on urban affairs in developing countries, and to Bank non-lending activities, particularly urban sector policy advice. 1.10 To take advantage o f the wide array o f sources of data related to urban development, this review relied upon six different instruments o f evaluation: (i) reassessment o f OED’s earlier review o f urban lending, called the 20-year Study in this report (World Bank 1994a); (ii) compilation and analysis o f a database of the 99 more recently completed urban operations, with data on their performance-assessed by OED through desk reviews o f Implementation Completion Reports (ICRs) and field assessments through Project Performance Assessment Reports (PPARs) and Impact Evaluations; (iii) intensive interaction with urban practitioners within the Bank through the Urban Sector Board, Urban Fora, OED ‘Urban Breakfasts’ and direct contacts with Anchor and Regional staff; (iv) a telephone survey worldwide o f borrower managers o f 45 of the 99 completed urban projects; (v) comparative analysis o f changes in livability in Bank client cities and others during 1993-98-61 cities worldwide; and (vi) review o f academic and other professional literature. 1.11 For the more data-rich livability aspect o f cities, this study tries to assess the results achieved by operations within the urban portfolio, within the results-based management (RBM) framework adopted by OED for its evaluations. The portfolio o f 99 completed urban projects seems able to fit that framework, since the objectives o f 92 per cent o f these operations were aimed explicitly at medium-term results (outcomes) and/or long-term results (impacts). This study examines whether the desired results were achieved and postulates how such achievements were made. There i s growing consensus among practitioners-which this review aims to bolster-of the need for greater emphasis upon results assessments in evaluation and the monitoring and evaluation (M&E) that this implies. 2. Evolution o f the Bank’s Urban Portfolio 2.1 The Bank urban portfolio has grown regularly since i t s inception in 1972. The volume o f lending for completed projects has increased substantially, the product o f more and, on average, larger projects (Figure 2.1). Portfolio performance ratings, measured by the share o f projects with satisfactory outcomes, were strongest in the earlier periods. The weaker performance for 1993- 2000 conceals a strong rebound in recent years (details: para. 3.7 and Fig. 3.1), as better performing operations approved since 1993 entered the portfolio o f completed projects. 4 2.2 Across regions, Africa now hosts the largest number o f urban projects, both ongoing and recently completed. Latin America had been the main urban borrower during 1983-92, but now hosts far fewer projects in the ongoing portfolio, as does South Asia. For ongoing projects East Asia i s responsible for the largest volume o f lending. Project performance during 1993-2000 was strongest in Middle East-North Africa (MNA), and had improved on the previous decade in Europe, Central Asia (ECA) and Latin America (Table 2.1). For a year-by-year portrayal o f project performance and associated events in the urban portfolio story, see Timeline (Figure 2.1). Table 2.1. The Bank's Urban Portfolio at a Glance : First decade: j Second decade:; Third decade: I Fourth decade j 1972-1982 1983-1992 ~ 1993-2000 j plans: 2007+ Projects completed (numbedyear) j 1.5 j 9.9 j 12.4 I 15.0(#) by exit Actual total lending (US$m/year) j 18 ; 460 I I 825 I, 1,033(#) year' Average loan size (US$m/project) j 15 46 67 69 - - - -with Percent - - - - - - - - - - outcomes - - - -satisfactory - - - - - - - !- - - - - - - - - - - I- - - - - - - - 88% 78% - p!- - - - - - 1. - - - - 9!?41% - - - - - - - - - I- - - - - - -71 --- -- ; ~ ~ ~ Projects approved (number/year) : 7.1 , 12.4 ! 12.6 ; bY entry Planned total lending (US$m/year) ; 245 980 ' 863 j year' Planned av. loan size (US$/project) / 35 79 I I 69 Percentwith satisfactory.............................................................................................. .................................................. outcomes I 81% I 71% I I loo%(*') I No.of I No.of I No.of I No.of !projects % sat lprojecfss % sat !projects % sat /projects % sat(') Africa I 5 100 j 21 86 ; 30 63 i 24 83 East Asia j 4 700 ~ 21 95 j 14 79 ; 18 89 re,$n: Europe, Central Asia I 1 700 j 2 j 17 0100 : 6 80 (exit year)Latin America, Caribbean ; 4 75 j 33 j 70 6489 24 79 Middle East, North Africa I 2 50 ; 11 100 ; 13 92 j 16 93 ....................................................................... ,------------------------------------------------------ South Asia I _ - I 11 73 j _____-___-___-_-___------- 12 42 ; 5 80 - low-income countries (all regions) 6 700 j 39 82 j 45 64 j 48 88 ALL URBAN PROJECTS I 16(") 88 1 99(") 78 I 99(") 71 / go(") 90 Source: OED Urban Database. Notes: (#) assumes 90 projects completed over six years; ("") refers to 12 completed out of 102 urban projects approved since 1993; (*) latest supervision self-evaluation of project achievement of development objectives; (") actual number of projects. Decades are "flexible". %Sat = percentage rated satisfactory. All years are calendar years. THE FIRST DECADE: 1972-1982' - INITIAL FOCUS ON POVERTY 2.3 Just 16 projects with loans o f US$188 million were completed in this first period- mostly in Africa, Latin America and East Asia-but further lending of US$1.9 billion was approved for 55 new projects. Inspiredby then World Bank President McNamara's 1975 Annual Meetings' speech-tliat had tackling urban poverty through service provision as its main theme- early urban projects targeted low-income beneficiaries. In particular, they called for upgrading existing squatter settlements notably through the large scale pioneering kampung improvement program in Jakarta, Indonesia (Lnl 040).6 The aim was to apply low physical standards to make solutions affordable to poor beneficiaries themselves and replicable on a large-scale. In the same spirit, early urban projects supported new housing through low-cost sites and services schemes across several regions as well as urban transport for the poor in a few key cities.' Items of urban ESW o c o w w h l o N T r r ~ - c o u l w N O 2000 Urban Strategy Paper - “Cities in Transition” 1999 1998 1997 1996 1995 Urban within FPSl network that includes 1994 Urban Sector Boardflhematic.Groups) 1993 Housing Policy Paper Urban within ESSD ~ 1992 1991 Urban Policy Paper 990 989 988 987 1987 Reorganization: Urban within 986 Sub-regional Infrastructure Divisions ~ 1985 1984 Urban Transport Paper ...-..-. .-..-..-..-..-..- ..-..-..-..-..-..-..-..-..-..-..-..-. __.. 1983 1982 “Learning by Doing” .. .-..-..-..-..-..-..-..-..-..-..-..-..- ..-..-..-..-..-..-..-..-..-.. 1981 1980 .. Regional Urban Projects Depts. .-. .-..-..-..-..-. .- . .-..-..-..-..-..-..- ..-..-..-. . 1979 Shelter Paper 1978 1977 1976 1975 Housing Paper 1974 1973 1972 Urbanization Paper Central Urban Projects Dept. (Bank urban milestones) 6 2.4 Bank policy and technical papers on urban development during this decade-Urbanization (World Bank 1972); Sites and Services (World Bank 1974); Housing (World Bank 1975a); Urban Transport (World Bank 1975b); and Shelter (World Bank 1980)-all emphasized what today we would call livability improvements, affordable to the poor. The first decade ended with a Bank self-evaluation-entitled Learning by Doing-of the incipient sector (World Bank 1983). I t concluded that affordable infrastructure and housing standards had been an important achievement, but to reach more people, the Bank had to wholesale benefits via local intermed- iaries rather than retail them individually to every single city. In today’s parlance, this was a call for scaling-up with more emphasis on strengthening what we now call governance. THE SECOND DECADE 1983-1992 - RAPID EXPANSION AFTER THE DEBTCRISIS 2.5 Urban projects were completed on a much larger scale during this period, which also saw a sharp acceleration o f new urban lending approved. The portfolio gave increasing attention to institutional development at this time, while 40 per cent o f projects remained focused upon the urban poor. With 78 per cent rated satisfactory, rapid expansion did not erode the quality o f the portfolio. In a decade overshadowed by the international debt crisis, Latin America became the Bank’s principal urban client region, hosting one third o f completed projects and nearly half the lending. Successful innovation came in the form o f well performing shelter and housing finance projects, such as those that successfully strengthened local housing finance agencies in India (Cr2929), Morocco (Ln2245), and Mexico (Ln26 12) and established wholesale municipal lending arrangements in Jordan (Ln1826), Paranti, Brazil (Ln2343), Morocco (Ln2272) and Nicaragua (Ln2086). This was also a time o f expansion o f Bank urban lending for reconstruction following natural disasters (see: Gilbert and Kreimer 1999 for details). Less successful innovation was through integrated urban development projects several o f which-including operations in Calcutta, India (Cr756), Recife, Brazil (Ln2170) and Guayaquil, Ecuador (Ln1776)-foundered under the weight of their multi-sectoral complexity. 2.6 The end o f this decade saw the publication of the Bank’s paper (details Table 2.2) Urban Policy and Economic Development, known hereafter as UPP-91. I t presented a policy framework o f the “three Ps”, namely productivity enhancement, poverty alleviation, and pollution abatement and supported a call for further research into them. For Bank urban work, productivity was the most innovative concept, although it had figured for a long time in the academic literature (Jacobs 1969, p. 18 and Richardson 1971, p. 47). While it was an important attempt to highlight links between cities and economic development, the urban productivity concept was not well understoodthen or later. As one writer recently questioned: “When we talk o f the need to improve urban productivity, are we talking o f increasing the productivity and efficiency o f cities in themselves or o f increasing the productivity and efficiency o f urban systems?” (Burgess 1997, p.21). Nor was the concept widely applied, being incorporated into only two urban operations during this period8and very few subsequently. More importantly, UPP-91 reaffirmed the priority o f tackling poverty as well as taking on board the growing environmental agenda. 2.7 Also at the end o f this decade, OED began i t s first assessment o f the Bank’s urban portfolio, whose report was issued in 1994 as Twenty Years o f Lendingfor Urban Development 1972-1992: an OED Review (World Bank 1994a). The “20-year Study” o f 115 urban projects completed between 1972 and 1992 found that physical output goals were often surpassed, but only one-third o f projects substantially met their institutional development (ID) objectives. Best practice urban projects enjoyed unequivocal borrower ownership and incorporated explicit poverty reduction aims. The present review’s finding o f substantial borrower involvement in project preparation and implementation and the resulting project success (Table 3.1 and para. 7 3.15) suggests that the lessons o f OED’s 20-year Study remain valid and have been absorbed. The poverty focus lesson has been similarly taken on board, with more projects addressing it. Recent findings show that such projects perform well although not significantly better than average (para. 3.21). The 20-year Study made four recommendations, adopted by the Bank to differing degrees during the following decade, as discussed below (Table 2.3 and para. 2.13) . Table 2.2. Urban Policy and Economic Development, UPP-91 Productivity Poverty Pollution Research lssues to Constraints upon the Unemployment and lack of Deterioration of the Serious gap in tackle: productivity of the urban infrastructure and urban environment understanding of urban economy services urban issues Strategic (i) Strengthen man- (i) Government-supported, lab- (i) More information (i) Assessment of instruments: agement of urban or-intensive productive activit- about the urban existing urban infrastructure; (ii) ies; (ii) regulatory reform to ease environmental crisis; research; (ii) make citywide regul- access by the poor to urban (ii) city-specific broad, long-term atory framework more services, credit, and markets; environmental urban research market efficient; (iii) (iii) facilitate women’s emp- strategies; (iii) strategy; (iii) improve municipal loyment; (iv) spend more on curative clean-up act- mobilizing financial, and basic services for urban poor; ions in cities; (iv) for- resources for technical capacity; (iv) (v) better access of poor to mation of national and urban research strengthen financial infrastructure and housing; (vi) urban policies on the especially in the services for urban recognize and support self-help environment. developing development. community and NGO efforts. countries. Expected (i) Faster growth of (i) Fewer urban families below (i) Lower levels of (i) Research results on the urban economy; the poverty-line; (ii) higher share pollution in urban results in the the ground: (ii) market-friendly of population with access to areas literature; (ii) behavior by national urban infrastructure and greater urban data and city authorities services available Related USP Competitiveness/ Livability Livability themes: Governance Source: World Bank 1991 1993-2000 - MANAGING THETHIRD DECADE CITIES AND MARKET REFORMS 2.8 Portfolio performance during this period-the busiest in the history o f Bank urban lending-is the principal focus o f the present review. The decade saw the largest volume o f lending for completed projects, although the volume for newly approved projects (Table 2.1) was down on the previous period. Africa overtook Latin America as the busiest region for Bank urban projects, but Latin America s t i l l hosted the largest volume o f Bank urban lending. With 71 per cent o f completed projects rated satisfactory, performance slipped to around the average for Bank projects as a whole. The performance o f projects approved during this period, however, has so far been much stronger, although the 100 per cent satisfactory rate applies only to the twelve operations completed by July 2001 (Table 2.1). 2.9 The portfolio o f completed projects during this period focused heavily upon the Urban Management and Urban Housing sub-sectors, strong performers that together accounted for two- thirds o f all operations (Fig 2.2). These sub-sectors incorporated the decade’s emphasis upon market reforms by trying to help bring local government and official housing agency actions and finances more into line with market behavior (Lee and Gilbert 1999). Even if they did not reach the lowest income groups, large and successful projects involving public sector housing finance were completed in Mexico (Ln2947 and Ln3497), as well as smaller operations in the Philippines (Ln2974) and Morocco (Ln3 122). Indonesia’s Housing Sector Loan (Ln2725) helped reform state 8 agencies, but with less Fig 2.2 Subsector Shares of Urban Projects and their Percent Satisfactory by Exit Year 1993-2000 private sector involvement than I Other Urban Development planned. Urban environment was the Urban Management weakest performing sub- 71% sector, due to the failure satisfactory o f three projects with large sanitation or solid waste component^.^ Urban development adjustment performed badly too, though this (Shaded areas are was a poorly defined proportional to sub- group with few common sectors’ shares of urban satisfactory satisfactory projects in portfolio) features among i t s nine component projects. 2.10 This period saw increased attention by completed urban projects to poverty alleviation; addressed by objectives o f 53 per cent o f a l l operations, up from 40 per cent in the previous dec- ade. They aimed to improve the lives o f the poor through slum upgrading, for example, that made a comeback during this decade in Venezuela, Ghana (Cr2157) and Indonesia (Ln3246 and Ln3304). The poor-those most vulnerable to the effects of natural disasters-were the focus o f emergency recovery operations, a few o f which included housing components (details: Gilbert 2001). Additionally, this decade saw an expansion o f completed Bank assistance to rebuilding lives and restarting economies in post conflict country cities in Bosnia-Herzegovinathrough emergency housing repair, and emergency reconstruction in Croatia and S r i Lanka (Cr 1883). 2.1 1 Municipal development projects (MDPs) took center stage among completed urban projects at this time. They greatly increased the number o f cities served by the Bark’s urban portfolio. The eight largest MDPs-in Brazil (Paranb, Rio Grande do Sul and Santa Catarina), Argentina, Ecuador, Tunisia, Morocco and India (Tamil Nadu)-directly helped more than 1,500 cities. This was through local financial intermediaries on-lending Bank (and other) funding to help municipalities finance priority investment programs in return for structural reforms at the city level. OED rated 89 per cent o f MDPs as having substantial or high borrower involvement in preparation, making them a particularly appropriate follow-up the 20-year Study recommendation to secure greater borrower ownership (Table 2.3). A later OED impact evaluation o f four MDPs in Brazil and the Philippines found that the projects facilitated reform, and enabled participant municipalities to outperform non-participants on the fiscal front (Lee and Gilbert 1999). Despite the decentralization implied, the Bank i t s e l f s t i l l prefers to lend directly to national governments (or their financial agents). As o f September 2001, the Bank made loans to the subnational level only three times, to states in Brazil and always with a federal government guarantee, as required by the Bank’s Articles o f Agreement (111, Section 4). Even in China with a program o f on-going projects aimed at particular cities, Bank loans were always to the national government. 2.12 Bank urban activities during this decade were strongly influenced by the UPP-91 (details Table 2.2)’ but did not implement all aspects o f the policy. UPP-91 did help keep the portfolio focused upon urban poverty and the urban environment, albeit with weak results for the seven operations officially classified as urban environment projects (Figure 2.2). There was little success in dealing with the constraints upon the productivity of the urban economy, however, a theme rarely taken up by urban practitioners (details: para. 2.6). As to the fourth aspect o f the UPP-91, urban 9 research, it appears that it became less active within the Bank itself, but important work was done elsewhere during the decade (see: bibliography). This study did not examine this aspect in detail. 2.13 This decade’s urban activities also show responsiveness to OED’s 20-Year Study recommendation (Table 2.3) to strengthen the congruence of project objectives and design focused upon achieving better results, particularly as far as livability issues (see: Table 2.4) are concerned. Without doubt, though, the most thoroughly adopted OED recommendationwas the one concerned with securing greater project ownership by both borrowers and beneficiaries. This led to good project performance (details: para. 3.15-3.16). On the other hand, OED had least success in convincing the Bank to develop guidelines for M&E and to implement project specific M&E systems, as recommended by both the 20-year Study (World Bank 1994a) and restated in i t s later Evaluative Note to CODE (World Bank 1999d). Insufficient attention to monitoring results o f Bank urban assistance continues to be an unfortunate feature o f urban lending, as discussed later in this report (Chapter 4). OED’s recommendation for urban projects to tap their economic and other linkages, while supported in principle by the Bank, led to few cases o f operations paying attention to the urban economy. Table 2.3. OED’s Review o f 20 Years o f Lending for Urban Development, 1972-92 Recommendation Response within Level of Current update through 1993-2000 Urban Bank Adoption Portfolio Enhance development impacts Agreement, but Low Limited attention to economic linkages still. through tapping the potential of limited action in Only 24% of projects had objectives explicitly economic, fiscal, and financial practice. focused upon enhancing economic develop- linkages of urban projects. ment. Strengthen the congruence of Agreement, but Medium 86% of projects with livability (definition: Table objectives and design. urban projects 2.4) objectives had livability components. 70% remain intrinsically of projects with governance objectives had complex. governance components. Secure project ownership by Pursuit of greater High Borrowers were highly or substantially involved borrowers and beneficiaries. involvement in in 85% of projects, according to OED. But design and imple- beneficiary participation was a major project mentation stages. emphasis for only 28% of completed projects. Accelerate the development of Agreement that this Low Only 34% of projects incorporated M&E guidelines for and implementa- should be a priority, indicators into design. Only one fifth of those tion of project-specific monito- but guidelines and projects incorporating M&E had their perform- ring and evaluation systems. adoption still weak. ance indicators substantially supervised. ~ Source: World Bank 1994a INTO THE FOURTH - PUTTING CITIES IN A 2001 AND BEYOND DECADE: GLOBAL CONTEXT 2.14 In Cities in Transition-called USP in this report-the Bank has a new urban strategy firmly harnessed to the earlier policy paper and the experience o f past lending. The rebound o f urban project performance in recent years bodes well for successfully implementing the strategy, but expectations are higher today and the Bank’s own poverty focus i s sharper. The top priority o f USP i s to improve livability using the three other strategic dimensions or pillars-good governance, bankability and competitiveness-to achieve this goal (see definitions in Table 2.4). How this can succeed-taking into account what has been learned thus far-is the focus o f more detailed discussion later in this report (Chapter 4). 10 Table 2.4. Cities in Transition: World Bank Urban & Local Government Strategy, USP Livability Good Governance Bankability Competitiveness ~~ Definition: Decent quality of life (a) Inclusion and rep- Financial soundness in Efficient markets in cities and equitable resentation of all in treatment of revenue for land, labor, credit and opportunities for all- urban society; sources and expend- for infrastructure and inc. the poorest-to (b) accountability, int- itures-and for some housing inputs, achieve a healthy and egrity, and transpar- cities, a level of permitting firms and dignified living ency of government in creditworthiness individuals to become standard. defining and pursuing permitting access to the more productive. shared goals capital market Issues to Urban poverty, (a) ExclusiorVIack of Cities lack of access to Inefficiencies in these tackle: inequality, unhealthy representation capital markets. markets. urban environment, (b) insufficient govern- Inequitable and and insecurity. accountability unsustainable local finance systems Strategic (i) Scaling-up services (i) Enhanced capacity (i) Public-private partn- (i) National urban instruments: to the poor; (ii) slum building; (ii) municipal ership in municipal ser- strategies; (ii) City upgrading development rooted in vices: (ii) financial in- Development Strategies market-based initiat- novations to bring cities (CDS) . . ives; (iii) City develop- to capital markets; (iii) ment strategies (CDS) risk-pooling Expected (i) Fewer urban famil- (i) Local government (i) More creditworthy (i) Well-functioning urb- results o n ies below the poverty more responsive and municipalities; (ii) more an markets; (ii) efficient the ground: line; (ii) higher share of inclusive toward the sound municipal use of urban inputs in population with access poor; (ii) more finances production and livability to urban infrastructure professional staffing and services; (iii) less of local government pollution ~ ~ ~ ~~~ ~~ ~ Selected (i) Child mortality rate; (i) Local government (i) Shares of local gov- (i) House price: income Indicators (ii) households below employees per 1,000 ernment income deriv- ratio; (ii) travel time to (complete list poverty line; (iii) house- people; (ii) wages as ing from: (a) taxes; (b) work in Annex 1 of holds with water and % of local government user charges; (c) tran- this report :(*) sewerage connections; budget; (iii) less cor- sfers; (d) debt; and (ii) (v) homicide rate ruption debt service charge Re/ated UPP- Poverty/pollution Productivity 97** themes: Source: World Bank 1997 and World Bank 2000a. (*) Taken from Annex D of USP . ** See Table 2.2 2.15 When the USP was first presented to CODE in July 1999, OED’s Evaluative Note to the Committee (World Bank 1999d) welcomed the draft o f the new strategic document. OED stated then that USP was better prepared for implementation than i t s predecessor 1991-UPP, especially as USP included a matrix o f urban performance indicators to monitor achievements under each o f the four strategic pillars (reproduced in Annex 1 to this report). For M&E to be able measure the results o f Bank supported interventions, however, OED suggested to CODE that additional work would be needed, such as: (i) guidance to help Bank task teams incorporate M&E into projects; (ii)clarification of baselines, or points of departure, from which changes indicators can be measured; (iii) suggestions for cost-effective methods o f monitoring; (iv) Bank partnership with others, particularly the UN Habitat. Having explicit targets to achieve also increases the urgency and relevance o f M&E to urban practitioners. At the same time, OED stated that M&E was most likely to succeed with the more familiar livability indicators, in contrast to indicators o f competitiveness, a concept that was not clearly understood. 2.16 Among the 90 ongoing urban projects, the focus upon the poor has never been stronger. The objectives of 69 per cent-compared to 53 per cent during the previous decade-aim at alleviating urban poverty, principally through improving livability. Across regions, Africa, East Asia and Europe-Central Asia give most attention to livability, which i s addressed by the 11 objectives o f more than 70 per cent o f all projects. As o f July 1,2001 they include three concurrent urban projects in Ghana, while six major operations presently address kampung improvement in Indonesia (see Annex 4 for complete l i s t o f projects). Good governance i s also a major target, being addressed by 62 per cent o f projects, with the most intensive coverage by urban projects in Africa, Latin America and South Asia. In contrast, competitiveness and banka- bility are each within the sights o f less than 10 per cent. Within this limited coverage, the Africa urban portfolio i s ahead o f other regions in addressing competitiveness through project object- ives. Bankability, on the other hand seems to have met with little attention in all regions. 2.17 The ongoing portfolio has incorporated the OED 20-year Study recommendation to strengthen the congruence between project objectives and design, at least as far as livability and governance are concerned (Table 2.3). O f the 61 ongoing projects with livability objectives, 97 per cent also had livability components (up from 86 per cent for operations completed during 1993-2000). A similarly high share o f projects aimed at better governance had relevant components. The story o f bankability and competitiveness-given much less attention by urban projects-is quite different, however. O f the five ongoing projects focused upon bankability, only 40 per cent had relevant components. O f the nine projects aimed at competitiveness, only 33 per cent had relevant components. Congruence between strategic objectives and project design i s key to the efficacy o f sector strategies and the Bank’s performance in this respect has been mixed. 2.18 Self-evaluation of on-going urban projects points to likely successful outcomes, however. Latest supervision reports rated 90 per cent o f them satisfactory in achieving their development objectives. This would imply an 80 per cent satisfactory OED rating for outcome at completion, if current ‘disconnect’ factors between self-evaluation and OED ratings o f ten percentage points hold. QAG 100 per cent satisfactory quality-at-entry ratings o f a sample o f 23 ongoing urban projects, and 92 per cent satisfactory supervision for another sample o f 48, support the positive trends observed. 3. Better Projects in Cities, Better Lives for the Poor 3.1 Despite the desirable shift by urban projects toward a more concentrated poverty focus, robust evidence o f their actual impacts upon livability in cities has been thin, especially at the strategic or sector level. But with the publication o f illustrative performance indicators in the USP (reproduced in Annex 1 to this report), more concrete results should be expected. A major challenge for M&E i s to produce these results soon, something that does not look very likely. The Urban Sector Board’s ‘stocktaking’ presentation o f USP progress to the Board o f Directors in October 2001 did not report results o f monitoring these indicators, nor describe M&E under way that might produce results in the near future. 3.2 In the absence o f project level data, the present review turned to another source, namely city-level UN Habitat Urban Indicators data for 1993 and 1998 to gather evidence o f such impacts, attributable to Bank-financed urban operations. The study first identified all city-level indicators related to livability for which there were consistent observations for both years. Then the changes in the values o f the indicators over the 1993-98 period were calculated and the results across cities were depicted on scatterplot charts. This helped identify a few outlier cities-with observations more than two standard deviations from the mean o f the group-that were excluded from the analysis, in view o f doubts about the quality o f their data. This would be the case, for instance o f a city reporting a sharp fall in water service coverage, when a modest increase was the expected result. After this ‘data cleaning’ the analysis used a good quality data set covering 61 cities-24 Bank clients and 37 comparators” -and seven indicators for the two years in question. 12 BANKSUPPORT MAKES A DIFFERENCE 3.3 The expected result o f this analysis was that Bank client city indicators would reveal greater improvements in livability than those o f otherwise similar comparator cities that did not benefit from assistance. Such a result was the explicit intent o f projects that sought to increase urban service coverage-particularly o f basic sanitation-in the client cities. Good quality data was most readily available for water, sewerage and solid waste disposal services. Being collected independently o f the projects themselves, the UN Habitat urban indicators data affords a valuable opportunity to veri@ project performance through the triangulation o f OED performance ratings at the project level with actual results reported at the city level. The results o f the analysis o f this small sample o f cities-and implicitly, the urban projects behind them-points to greater livability improvements in Bank client cities than in comparator cities with respect to:" Water: Bank client cities increased their service coverage during 1993-98 significantly more than comparator cities. This finding assumes that most higher-income households in these cities already have basic services-studies in Latin America show that 85-98 per cent are already served (PAHO 2001)-and most o f the additional coverage would go to the poor. The share o f households connected to water supply services rose, o n average, by 30.4 per cent in Bank client cities, but only by 4.9 per cent in comparator cities. Among others, urban projects in Armenia and Lesotho (Cr2400) contributed to this result. Sewerage: Similarly, Bank client cities fared better than others did. During 1993-98, the share o f households connected to proper sewerage rose on average by 86.6 per cent in Bank client cities, versus only 1.5 per cent in comparator cities (same distributional assumption as above). Bank-financed municipal development operations in Bolivia and Morocco were among those that helped client cities achieve better results. Solid Waste: The share o f all garbage poorly discarded in open dumpsfell by 6.7 per cent during 1993-98 in Bank client cities against an increase in comparator cities o f 285 per cent. Urban projects, such as those in Benin and Belize helped bring about improvements o f this kind. 3.4 Furthermore, the analysis found that the higher the project rating-measured by OED's Aggregate Project Performance Indicator (APPI)12-the greater the increase o f the shares o f all households connected to water and sewerage services in the host Bank client city. For both services the study found positive and significant correlations at the city level (coefficients o f 0.65 for water and 0.74 for sewerage). 3.5 In addition, the study pointed to other hypotheses about urban project impacts worthy o f further research, but which did not produce robust statistical results given high variances in the small pool o f observations available. Thus, the mean o f under-five child mortality rates for all cities reviewed here increased between 1993 and 1998-an average pulled up by the H I V / A I D S pandemic in Africa-the rate o f increase in Bank client cities was 45 per cent, while it was 153 per cent in comparator cities. The study also tried to test the hypothesis that poverty reduction was greater in Bank client cities, but the large variance o f the shares o f people living in poverty shrouded any significant differences between the t w o groups o f cities. In terms o f the efficiency o f the urban economy and markets, two more indicators provide (weak) evidence o f the favorable impacts o f Bank urban projects. First, the average travel time to work in Bank client cities fell by 7.9 per cent, while i t fell in comparator cities by only 2.4 per cent. Second, house prices as a multiple o f average annual incomes-an indicator o f affordability and the efficiency o f housing markets-increased in a l l cities, but by an average o f 13 per cent in Bank clients, against 71 per cent in comparator cities. The desirable result would be for this indicator to fall, o f course. 13 3.6 Clearly there i s a need for more data to test these and other hypotheses. Building upon the UN Habitat Urban Indicators-which s t i l l cover relatively few cities for both 1993 and 1998- would be a good place to start. In the UN Habitat database, fewer than half the cities and indicators are reported for both years, limiting the scope o f time series analysis. For these reasons, the Bank should support UN Habitat’s efforts to extend and improve the data set, starting with the next survey scheduled for 2003. UN Habitat’s work i s also currently under way-and should be encouraged-to consolidate a single City Development Index (CDI) measuring overall livability based upon: (i) city product; (ii)infrastructure; (iii) waste; (iv) health; and (v) education (UN Habitat 2001b). Broader and more consistent coverage would allow better urban livability impact analysis. I t would also be relevant to all Bank sectors that invest in cities-education, health, transport, energy, and water-that share the common aim o f improving the lives o f the poor. REBOUND OF PROJECT PERFORMANCE RATINGS 3.7 Figure 3.1 shows the positive trend o f urban project performance ratings, since their nadir in exit year 1995, when only 33 per cent o f (twelve) urban projects was rated satisfactory. That year saw the evaluation o f a number o f operations prepared around the time o f the Bank’s disruptive 1987 reorganization, which severely reduced funding for urban development. Though strong, the post-1995 rebound did not lift the average for the 1993-2000 period above 71 per cent satisfactory, a level significantly below the previous decade’s 78 per cent. This leaves the urban portfolio, which had been a good practice leader in the past, with ratings only slightly above the Bank-wide average o f 69 per cent satisfactory. Recovery looks set to continue, though, given the positive QAG ratings reported at the end o f the previous chapter. 3.8 As a component o f the overall outcome rating, OED evaluators also assess the r e l e ~ a n c eo ’~f a project’s objectives to country policies and Bank sector strategies (such as the CAS). O f 65 projects thus rated, 88 per cent were considered substantially or highly relevant. All but one o f the seven with the highest relevance rating-which included urban projects in Brazil- Salvador, Benin, and Ghana (Cr2157)-had project objectives and components focused upon the urban poor. The Ghana operation, for instance, sought to improve transport access to low-income was crucial for the country’s development. At the other Figure 3.1 Ratings have Rebounded Strongly since 1995 extreme, only one o f the seven 100% modestly relevant projects 90% 80% targeted the poor. 70% 60% 50% 3.9 Similarly as components 4ooh o f the final outcome rating, OED 30% 20% assessed urban project 10% performance in meeting 0% 1993 1994 1995 1996 1997 1998 1999 2000 objectives (eflcacy) and in the exit years -Outcome % Satisfactory efficient use o f resources in doing so (eflciency). Lower lnst Dev Impact % Substantial ratings for both undermined a I more positive assessment o f portfolio performance. For instance, only 58 per cent o f urban projects were rated highly or substantially eflcacious. Municipal development operations, implemented in Argentina, Brazil, India-Tamil Nadu, the Philippines, Morocco, Tunisia and Venezuela during 1993-2000, have high efficacy ratings, In Tamil Nadu, the project achieved more than originally intended through 14 the creation o f the privately managed urban development fund for the state, which helped ensure the remaining capacity building objectives and the provision of serviced land affordable to low- income households were fully met. Clear and realistic objectives o f projects contribute to project efficacy. OED evaluators rated the eficiency o f urban projects to be high or substantial only 5 1 per cent o f the time, a rating that could have been undermined by lack o f data. Only 28 project evaluations included estimates o f ex-post economic rates o f return (ERR), which averaged 26.6 per cent, against appraisal estimates o f 25.1 per cent. The highest ERRScame from urban projects with large housing components, such as Brazil-Rio Flood, and Albania Housing, which generated high returns in tight housing markets. Flood reconstructionprojects, such as those in Argentina and Yemen, also generated high ERRSas they helped kick-start disaster-hit local economies. 3.10 Urban project sustainability-understood as the resilience o f project benefits over time- has improved less dramatically, but more consistently since 1993 (Fig. 3.1). Still, only 38 per cent o f the 99 urban projects reviewed had likely sustainability, below 5 1 per cent for all Bank projects and down from 42 per cent for the urban portfolio during the previous decade. Inadequate maintenance and half-heartedcost recovery efforts are most frequently cited in OED PPARs as causes for unlikely sustainability. Maintenance was neglected altogether in urban projects in Ethiopia, Cameroon, and Mozambique (Cr 1949)-where five-year old street pavement was broken and water distribution networks not operating-and cost recovery failed in Nigeria (Ln2925) and Sri Lanka (Cr1697). Since larger projects perform better, 55 per cent o f all urban disbursement was sustainable, versus 60 per cent for Bank disbursement overall. Fewer urban projects with likely sustainability i s not the result o f more project sustainability being rated unlikely, however. In recent years, more urban project sustainability has been rated uncertain, a “don’t know” category used by evaluators when project data i s scarce. 3.1 1 ID impact ratings shadow outcome ratings, with a substantial improvement since 1995. Only 25 per cent o f the 99 urban projects had substantial or better ID impact, though, below the Bank-wide figure of 34 per cent and down from 29 per cent in the previous decade. For the recent period, evaluators gave more modest and fewer negligible ratings for ID impact. Nevertheless, the MDPs mentioned above (para. 3.9) and other follow-on projects such as in Ghana (Cr2 157) had important ID impacts through decentralizationthat helped cities conduct their urban development business more autonomously. K E Y DETERMINANTS OF PROJECT PERFORMANCE 3.12 Given that the impact o f an urban project upon livability depends upon the quality o f the operation-expressed through the performance ratings, urban project managers and others will want to know what factors help determine good project quality and performance. The present study designed a simple least-squares regression model to help identify such factors, some that managers can control and others that managers need to take into account as they design and implement urban projects. The model i s the outcome of: e Formulating hypotheses o f possible causes o f good project performance using factors related to project design and implementation, project management and country conditions, using data from a variety o f sources available. e Results o f simple bivariate correlation analyses among the 50 or so chosen variables, to select key independent variables to include in the regression equation (at the same time identifying their correlates to exclude multicolinearity from the model). e Estimatingthe results o f the regression model (yielding R-squared o f 0.48) (Table 3.1). 15 Table 3.1. Results o f Least-Squares Regression of Factors o f Urban Project Performance Dependent lndeper ent Variables: Variable: 1 2 3 4 (project factor b;) (pruject factor c:) (managemenb’counfry factors:) OED APPI (project factor a ;) : Beneficiary Poverty-focused Predicted overall borrower Demandingness involvement objective Derformance (1) Coefficient -0.900 0.601 0.028 1.714 ’ ....................................................................................... ;......................................................................................... .......................................................................... t statistic -3.451* I 2.660* 0.079 1.777* ............................................................................................ ................................................................................. ........................................................................... ~ + positive: I +positive: + positive: + positive: ~ Correlates - negative: Project complexity. Bank performance. Previous project. Monitoring & evaluation. I Borrower Borrower involvement. Previous project. I involvement. Country urbanization. I - Project quality at entry. Bank performance. Country GNPkapita. negative: Country interest rate. ......................................................................... ............................................................... j Urban pop. growth 1993-98. ...................................................................................... Interpretation The more a project Greater benefic- Incorporating Borrower performance, itself taxes government or , iary involvement poverty-focused the product of country implementation 1 during project objectives does condition, is a significant agency resources, preparation makes ~ not undermine driver of project outcomes. the weaker the for better project ~ project Bank performance is also a project performance. performance. performance. significant determinant of * Larger loan amount; ! * Greater govern- ~ A poverty-focused outcomes. better quality-at-ent- 1 ment involvement design is more More urbanized countries ry; being a follow-on 1 operation, all help good performance. 1 has a similar effect. I likely in follow-on projects. host better-performing projects. More rapid urbanization- HigherGNP per 1 with more unsatisfied de- capita facilitates 1 mand for services-under- performance. mines project performance. Regression R-squared = 0.48 F-value = 8.93 n = 54 (2) statistics: (significant at 99%) Source and notes: OED urban database. (*) Significant - t-test 90% or higher. (1) The model uses predicted values of Overall Borrower Performance estimated from a regression of Actual Overall Borrower Performance against three significantly correlated country factors: governance (CPIA), urbanization and urban growth - t-test 90% or higher. (2) All urban projects completed since fiscal 1995 for which OED has the most complete evaluation data. Difference of means tests show no significant differences between this subset and the complete set of 99 completed urban projects with respect to 18 variables, including OED APPI; country urbanization; country GNPkapifa; country GDP growth; poverty focus; and borrower and Bank performance. OF SUCCESS-THINGS THAT MANAGERS PROJECT FACTORS CAN WORK ON 3.13 Reduce “demandingness”. Urban projects should avoid straining borrower implementation capacity or resources, in other words not be too demanding. Only 42 per cent o f urban projects reckoned by OED evaluators to be highly demanding upon borrowers had outcomes that were rated satisfactory. A highly demanding project i s typically the product o f overestimating institutional capacities o f borrowers and what borrower agencies can be reasonably expected to achieve in the short term. Demandingness in urban projects can take several forms, among them: (i) complex designs involving myriad components and implementing agencies; (ii)assuming that a policy framework to be in place that does not actually exist, as occurred with urban projects in Mozambique (Cr1949) and Vanuatu; or (iii) project design based upon ideal institutional arrangements that in fact are not in place, as happened with housing in Argentina (Ln2997) that presumed a non-existent strong federal authority and Nepal’s municipal development that foundered without a local institutional base. These and many other experiences point to the need for careful appraisal o f borrower institutional capacity at the design stage. 16 experience has been Figure 3.2 Completed Urban Projects: Bank and Borrower broad enough for 50 o f Performance (YOsatisfactory by exit year) the 99 completed 1 100% projects in the 1993- 2000 period to have 90% predecessor urban 80% projects across 3 1 countries in all regions. 70% 60% 3.15 Involve 50% beneficiaries. 1995 1996 1997 1998 1999 2000 Beneficiary participation -Bank Performance Overall i s positively correlated a 3.16 Engage the borrower. Borrower involvement too i s very significant for project success, echoing the recommendation o f OED’s earlier 20-year Study to secure borrower ownership (details: Table 2.3). This lesson seems to have been well learned since borrowers were highly or substantially involved in the preparation o f the great majority o f urban projects-85 per cent o f the total. These projects were rated 82 per cent satisfactory, compared with the 38 per cent satisfactory rating o f operations lacking this involvement. 3.17 Excellence in Bank service. Good Bank performance was found t o be positively correlated with borrower performance and, therefore, with satisfactory project results (Figure 3.2). OED rated 79 per cent o f all completed projects as having satisfactory Bank performance (above the Bank-wide average o f 74 per cent). O f these, 82 per cent had satisfactory overall outcomes. There i s a positive payoff in terms o f project performance from good Bank performance, especially if it goes into preventing unduly demanding project designs and into recognizing and nurturingthe partnership with a well performing borrower. Bank performance has remained at high levels in recent years and future prospects look good as the high QAG ratings reported earlier (para. 2.18) would indicate. 17 3.18 Nurture good borrower management. The model found a positive and significant ’ ~ project outcomes. Nearly all o f the 74 per cent relationship between borrower p e r f ~ r m a n c eand o f completed projects with satisfactory borrower performance had satisfactory outcomes. The corollary i s that projects with poor borrower performance hardly ever make the grade o f a satisfactory outcome. Borrower performance for urban projects i s above the average for all Bank projects (69 per cent satisfactory), but has changed l i t t l e in recent years (Figure 3.2). COUNTRY FACTORS-THINGSTHAT NEED MANAGERS TO TAKEINTO ACCOUNT 3.19 Degree of urbanization. Urban projects appear to be more successful in more urbanized countries. Those completed in the most urbanized countries-the upper tercile with 58-89 per cent o f their populations living in cities-were 8 1 per cent satisfactory. The equivalent rating for the least urbanized countries-that made up the lower tercile with 6-34 per cent o f their people in cities-was only 59 per cent. More urbanized countries-which are also higher income-have more experience in managing urban development and more resources than those with rural-based economies. On the other hand, faster urban population growth was found to correlate negatively with project outcomes as resulting strains upon urban services and widening financial deficits undermines urban project management. 3.20 GNP per capita. The analysis for this review confirmed as earlier studies have long shown, that urbanization i t s e l f i s correlated with income per capita at the country level. There- fore, urban projects in upper-middle income countries-with 79 per cent satisfactory-performed significantly better than those in low-income countries, where 66 per cent were satisfactory.I6 Urban project managers can do little t o change income levels o f their client countries, but they need t o be aware o f them. Managers need to be sensitive to the capacity constraints posed by low- income countries, which account for a growing share o f Bank urban lending (Table 2.1). NOT FOUND FACTORS TO BE CORRELATED WITH PROJECT PERFORMANCE 3-21 The study model was unable to point to a statistically significant relationship between three factors conventionally associated with project performance and the quality o f project outcomes themselves. At this stage, we cannot affirm that no relationships exist; only that further work would be necessary to reveal them. 3.22 First, poverty focus in urban project objectives did not lead to weaker performance as many practitioners expect. Managers cite lack o f effective demand and financial sustainability coming from the poor that can undermine performance. The message i s that project managers need not fear fighting poverty through their projects. In fact, 75 per cent o f the 52 completed projects that targeted poverty in their objectives and components were rated satisfactory. While somewhat above the average for the urban portfolio as a whole, the difference was not found t o be statistically significant across the samples analyzed. 3.23 Second, the performance o f the 37 completed urban projects that featured partnerships with co-financiers on the one hand was also not significantly different from that o f stand-alone Bank operations on the other. In part, this result reflects the fact that such partnerships are forged within a broad array o f urban projects across all regions and subsectors. I t does not point to evidence of these operations’ much talked about “transaction costs” undermining outcomes. The model used here does not capture other valuable benefits o f partnerships, such as learning and dissemination for instance. 18 3.24 Finally, the model did not point to any significant difference in the performance o f 18 projects prepared with the benefit o f recent-up to three years before project appraisal-urban Economic and Sector Work (ESW) and the 71 other urban projects prepared without it. Other aims o f ESW, such as establishing a policy dialogue with potential borrowers-as pursued in 12 countries with ESW but no ensuing projects-were not examined by the present study. While the results thus far are inconclusive, they do beg questions on the purpose o f ESW, where to focus it, and whether it should underpin more innovative approaches to urban lending in the future. WHERE TO FOCUS ACTION? 3.25 The evidence assembled in this chapter points to a valuable role for Bank-financedurban projects in improving the living conditions o f the poor in cities. Cities that hosted urban projects saw improvements that others did not. Across cities, the higher the project ratings, the greater the improvements. Furthermore, there are a number o f identifiable “handles” o f project performance that managers can grasp in order to improve project and hence.portfolio quality. To arrive at recommendations for action, however, we first need to identify the strategic areas o f urban action that hold the greatest promise o f success in Bank assistance to improve the lives o f the poor in cities. To do this, the discussion now turns in the following chapter to a more detailed review o f each o f the strategic pillars o f the current urban strategy. 4. Improving Lives in Cities by Implementing the Strategy 4.1 How much Bank urban assistance will continue to improve the lives o f the poor in cities in the future will depend upon how successfully the new urban strategy i s implemented and how well the results are monitored. These twin aspects o f implementationand monitoring are examined here in relation to each o f the pillars o f the strategy, livability, good governance, bankability and competitiveness. Successful strategy implementationrequires that these pillars constitute a framework for applying what we have learned thus far, and provide a road map showing where Bank urban assistance i s headed, what it aims to achieve, and how we will monitor the results obtained. 4.2 For monitoring results, a USP attachment provides a matrix of ‘illustrative indicators for monitoringurban and local government performance under the new strategy’. The matrix- reproduced in Annex 1 to this report-is one step in adopting the OED 20-year Study recommendation to improve M&E o f urban projects (Table 2.3). Having explicit targets to achieve can also increase the urgency and relevance of M&E in the eyes o f urban practitioners. As mentioned earlier in this report (para 3.1) we still await the results o f the monitoring o f these indicators at the strategic level o f the USP. The need and potential for getting these results in discussed below, pillar by strategic pillar. LIVABILITY-DECENT QUALITY OF LIFEAND OPPORTUNITY FOR ALL, ESPECIALLY THE POOR 4.3 Livability i s deservedly the principal pillar o f the USP. By continuing to focus upon it, as under different labels in the past, the Bank can help improve the lives o f the urban poor by making them more healthy and productive. Better health can come from a safer urban environment and productivity can come from better access to job opportunities in the urban economy. In addition to being a welfare issue, this i s a question of setting up conditions for the growth o f the urban economy that will benefit a l l citizens. Through helping provide a safe water supply, sewers that work and drainage and paving that permit access to neighborhoods however inclement the weather, urban projects are instrumental in such livability improvements. They 19 have been in the past, most notably through 12 successive operations in Indonesia that brought the benefits o f low-income kampung improvement to nearly 15 million people (World Bank 2000a). Newer ongoing urban projects focus correctly upon livability goals, that dovetail into the Bank’s own poverty reduction mandate (OD 4.15 - 1990 and i t s successor) that espouses the inclusion o f the poor through more healthy and productive lives. 4.4 USP indicators for monitoring livability outcomes” (e.g. urban poverty rates, infant or child mortality, rates of waterborne diseases, and ambient air and water quality) and outputs (e.g. basic urban service coverage) are, for the most part, relevant to USP strategic goals and familiar to urban practitioners (see: Annex 1 for complete list). Results measured by them are s t i l l not systematically reported at the sector or strategic level in the Bank, even though they are necessary for the Bank to demonstrate the effectiveness o f i t s contribution to the livability agenda, and eventually to the achievement o f the 20 15 MDGs. Long before then, however, the Board and external reviewers will want to see systematic evidence o f urban livability improvements for the poor, especially those attributable to Bank supported interventions. 4.5 More could be done to monitor poverty results in practice at the project level too. O f a l l urban operations completed during 1993-2000, 53 per cent focused at least one of their objectives on the poor, but ICR and OED evaluations drew poverty-related lessons for only 10 per cent o f projects. Even fewer reported impact indicators related to urban poverty. Typically at the closing o f an urban project, we s t i l l do not know who the poor beneficiaries were or how the operation made their lives better. Even the present study’s own findings o f the poverty impact o f livability improvements (para 3.3) are premised upon (albeit reasonable) assumptions about the distribution o f project benefits within the cities themselves. A critical (and self-critical) OED finding sums up the problem well: “A history o f inadequate project monitoring data and weak follow-up on poverty-related project objectives in ICRs and ICR reviews, has resulted in a serious gap in the Bank’s knowledge about the effectiveness of i t s lending in reaching poverty goals. The lack o f a clear poverty focus in OED’s work until recently has only served to perpetuate this gap” (Evans 2000, p. 46). We can only claim to have achieved positive impacts on the lives o f the poor if we have clearly seen the results, shown them to others and also known how they were obtained. 4.6 Despite these shortcomings, the USP and future urban portfolio are poised to support several key initiatives in the Bank’s poverty reduction arsenal, including: 0 Millennium Development Goals (MDG) especially reducing poverty and infant mortality in cities and improving the lives o f urban slum dwellers. (see Chapter 2). 0 Priorities of the WDR 2000/2001--Attacking Poverty: (i) opportunity, through improving livability o f the poor in cities; (ii) empowerment, through substantial participation o f poor beneficiaries in project design and implementation; and (iii)security, through legal tenure and mitigating the risks from natural disasters. 0 Poverty Reduction Strategy Papers (PRSP): As o f December 200 1, 42 have been prepared, 24 o f which by countries that have borrowed from the Bank for urban projects. Altogether 18 PRSPs” address urban poverty directly and explicitly as recommendedby the PRSP Sourcebook. 0 Priorities of the WDR 2001-Institutions: The important role assigned to institutions at the local level can draw upon urban portfolio’s rich experience o f dealing with municipal governments for more than two decades. The principal challenge will be to demonstrate convincingly the results achieved and that they can be attributed to the Bank interventions. 20 4.7 T o successfully raise the USP livability pillar, urban teams at the Bank and among borrowers can draw upon a r i c h body o f experience. First and foremost, are the three decades o f experience in urban lending, which had the urban poor in i t s sights from the outset. Second, borrower project managers interviewed for this study said that their project experiences had taught them a lot more about livability than any other pillar o f the USP. Altogether 37 o f the 45 interviewed reported substantial learning about alleviating poverty through targeted infrastructure provision especially. Third, other studies have reported that city mayors especially have learned considerably about improving livability through contacts with their peers from other cities (Campbell 2001, p. 229). Very direct learning also comes from incorporating the experience o f prior urban projects, a key factor in improving the performance o f the successor project, as noted earlier (para. 3.14). Among strongly performing follow-on operations was the MDP in R i o Grande do SUI, Brazil that benefited from a similar predecessor operation in the neighboring state o f Santa Catarina (Ln2623). Tunisia’s M D P (Ln3507) was the first o f i t s kind in i t s country, but could draw on a long and solid experience o f city management elsewhere. GOOD LEVEL GOVERNANCE-INCLUSIONAND ACCOUNTABILITY, ESPECIALLY AT THE LOCAL 4.8 This study’s findings about the importance for project success o f good borrower performance and o f operations that do not strain borrower capacity together point to the relevance o f the good governance pillar o f the USP. Although the urban strategy highlights its political inclusion and government accountability aspects (see definitions in Table 2.4), the term governance itself has recently gained wide currency in the Bank to describe more broadly: “(i) the process by which governments are selected, monitored and replaced; (ii) the capacity o f the government to effectively formulate and implement sound policies, and (iii) the respect o f citizens and the state for the institutions that govern economic and social interactions among them.” (Kaufmann, et al. 1999a, p. 1). An environment o f good governance-without corruption, with well-functioning bureaucracies, contract enforcement and protection o f property rights-is regarded as synonymous with a sound investment climate (Collier and Dollar 2001). At the city level, these conditions would enable the urban economy to prosper. 4.9 Indicators recommended by the USP for monitoring governance outcomes (e.g. local government responsiveness to the poor, citizen trust in local government, and service efficiency) cover relevant topics, but do not lend themselves readily to quantitative or systematic measurement. T o a lesser extent, this i s also true o f the governance output indicators (e.g. partnerships with local governments, public consultation in budgeting and investment decisions, share local government staff with professional qualifications), some o f which are really input indicators, as OED informed C O D E previously (World Bank 1999d). For urban practitioners to assemble evidence o f governance results that are meaningful at the sector or strategic level, the practitioners themselves will need more guidance from the Urban Sector Board. This would include baselines for these indicators from which improvements can be measured, proxy quantitative measures for the less robust indicators proposed, and targets t o be achieved. 4.10 Meantime, anecdotal evidence suggests that through dealing directly w i t h municipal governments, many completed urban projects have succeeded in strengthening governance at the local level. Bank-financed urban projects introduced municipalities to rigorous debt financing in Tunisia (Ln3 507) and Venezuela. In return for tighter financial management through municipal financial action plans monitored by a central agency, such projects typically gave better performing local governments access to funding for their priority infrastructure projects. With such incentives, well-managed municipalities are better poised t o lead programs to improve livability in their cities. 21 4.1 1 Urban projects, especially MDPs, have supported decentralization programs in many countries throughout the world. In federated systems, such as in Argentina and Brazil, such projects have consolidated existing decentralized arrangements through local capacity building. Some operations have contributed to fiscal decentralization too, by helping municipalities to become financially more self sufficient through bolstering their o w n revenues (Lee and Gilbert 1999). In more unitary or centralized regimes-in Chile and Indonesia, for instance-urban projects have helped central governments take early steps in the decentralization process. More generally, they also helped city mayors deal with globalization and “adopt a more sophisticated approach to world events that might affect them” (Harris and Fabricius 1996, p. 8) as well as enabling them to “adapt global technologies to local needs”(UNDP 2001, p. 5). Perhaps the most comprehensive instrument for improving governance at the local level i s the City Development Strategy - CDS (see Box 1). Box 1: City Development Strategies A key instrument for implementing the urban strategy, a CDS i s defined as “both a process and a product that identify ways o f creating conditions for urban sustainability along the dimensions o f livability, competitiveness, good governance and bankability” (World Bank 2000a, p. 64). To date, more than 80 city development strategies have been prepared or are under preparation in more than 20 countries worldwide. First prepared in 1997, a CDS i s a city-driven initiative that relies upon Bank and other donor assistance. Since the CDS process i s defined by the city itself, it shares the philosophy o f the Comprehensive Development Framework (CDF) o f being a partnership in which the local partner retains control. As with the CDF, a CDS i s also driven by participation o f stakeholders and inspired by a holistic view o f the challenges o f local development. In fact, some claim that a CDS provides a city-level CDF. According to the Cities Alliance, CDSs have recently been completed successfully for Kathmandu, Johannesburg, Kigali, Sofia and four cities in China-Changsha, Zhuzhu, Xiangtan and Guiyang. At i t s own initiative, the Municipal League o f the Philippines has planned to extend CDS coverage in the country to many more cities than initially included. Under the direct command o f the city mayor, a CDS tries to aid understanding o f the urban economy by asking the simple question, “how are people in the city, especially the poor, going to make a living?” The answer can come from CDS diagnosis o f how urban constraints prevent local industries from competing in broader markets. A CDS can also lead focus studies o f the poor that concentrate upon the inefficiencies behind the poverty and how they can be overcome. A CDS can be an important lens through which the c i t y executive can better understand the economic development o f the city and remove constraints to i t s efficient functioning in the hope o f improving the lives o f its people. This can involve strengthening local governance to remove potential obstacles to urban economic development, such as those that prevent access to housing, land, credit, and infrastructure. A CDS i s also intended to provide a framework for investment by a city in similar way that Bank Country Assistance Strategies guide Bank investment in a particular client country. The “bottom-line” o f a CDS exercise, however, i s to enable the c i t y administration and stakeholders to make sustainable improvements in their city’s livability, especially for the poor. For a summary o f an early CDS experience, for the city o f Haiphong, Vietnam, see (Campbell 1999). 4.12 Urban project support for local governance also came through conventional channels o f technical assistance (TA) and training to strengthen municipalities’ project appraisal capacities and financial accountability. Some US$658 million-12.6 per cent o f expenditures on physical investments-was spent on TA by 65 urban projects completed since 1995. Training programs-for borrower agency staff in particular-featured in one-third o f completed projects. Urban projects also enhanced borrower management o f procurement, a day-to-day aspect o f governance typically involving the most intense BanWborrower interaction. Borrower managers interviewed for this study reported learning more about procurement than any other aspect o f project management. The achievement o f efficient procurement processes means greater transparency and accountability, and 22 can contribute to anti-corruption efforts (Klitgaard 2000, p. 46). Although not necessarily the result o f any specific instrument, leadershipnow recognized as an important feature o f good governance (DiGaetano and Klemanski 1999, p. 250)--was often fostered by urban projects. The operations gave many individuals opportunities to champion reform initiatives, reflecting a new style o f leadership emerging in Latin America, for instance (Campbell 2001, p. 227). 4.13 The urban portfolio has provided little support to the institutional framework for private sector participation in urban services. Given the public good nature o f many services and government responsibility for addressing market failures in their delivery, the public sector i s likely to remain a key player in poverty-oriented urban development projects for the foreseeable future (Kessides 1997, p. 23). Publidprivate sector cooperation, nevertheless prospered under the first urban project in Zimbabwe (Ln2445) when city governments providing serviced land, while private (mortgage-lending) building societies financed the construction and/or sale o f the house, although macro-economic crises undermined building societies’ contribution to the follow-on project (Ln3079). The private sector has also participated in solid waste collection supported by Bank urban projects in Indonesia (Ln3246) and Brazil-Salvador, and Guinea-Conakry. Informal private sector activity, by small and medium enterprises in productive activities and through community-based organizations in neighborhood upgrading, has been supported by the urban portfolio. Indeed, successful experiences with beneficiary participation (para. 3.13), that strengthen the inclusion aspect o f governance, do engage the informal private sector through self- help efforts by communities themselves. While the cases reported here are important, they represent only small steps toward private sector participation in urban development programs, suggesting that opportunities for private sector participation may not be fully exhausted. Municipal governance can be enhanced by enabling private sector participation in service provision within a competitive environment. BANKABILITY-FINANCIAL SOUNDNESS AND CREDITWORTHINESS 4.14 Proposed USP bankability performance indicators-exclusively concerned with local government-are among the strategy’s most robust. Both outcome indicators (e.g. balanced budgets, investment capacity, creditworthiness) and output indicators (e.g. tax effort, debt service ratio, cost recovery on services) are readily quantifiable. With municipal finance data more readily available, the performance o f these indicators has been more widely reported, by OED i t s e l f in Brazil, the Philippines and Tunisia. In the case o f these indicators, the Urban Sector Board could do more to disseminate these practices across regions to increase M&E coverage. 4.15 In practice, municipal development projects have enabled the Bank to build up important experience with assistance aimed at strengthening local government revenues. With a stronger revenue base, municipalities can o f course make greater inroads into improving the lives o f the urban poor within their jurisdictions. OED evaluations have garnered evidence o f fiscal improvements attributable to Bank financed projects at the local government level in Brazil, Philippines, Tunisia and Venezuela. Borrower project managers interviewed for this study reported substantial learning about fiscal improvements through projects such as these. Such improvements typically come from better financial management systems, more complete property lists (cadastres), effective tax collection and accurate and transparent accounting, actions typically associated with good governance. 4.16 Progress with municipal creditworthiness, the second aspects o f the USP’s bankability pil- lar, has been much more limited, though. Bank urban lending has been in countries that themselves are not considered creditworthy-taking Moody’s sovereign rating o f Ba or below as the bench- mark. In these countries, the prospects for creditworthiness at subnational levels o f government 23 must be clouded by the market perceptions at the national level. In addition, market misgiving about local governments, especially if they lack autonomy to raise revenues or reduce expenditures, i s likely to continue to bar many from the capital markets (Dillinger et al. 1999, p. 58). Even after many years o f successful municipal development work in Paranh, Brazil municipalities there are s t i l l not “going t o market” on a large scale. Still, 52 municipalities in Latin America did issue bonds during 1991-99 and Asia’s local bond market was worth US$477 million in 1999, but this is less than 0.2 per cent of worldwide municipal revenues o f US$347 billionlg per annum and less than 0.01 per cent o f the US$7.4 trillion debt outstanding in North America’s local bond markets. In spite o f the limited results thus far, many urban practitioners feel the pursuit o f creditworthiness, insofar as it requires good governance, remains a worthy goal. Creditworthiness helps the growth o f domestic bank borrowing by municipalities, an aspect that is also relevant to the Bankability pillar. COMPETITIVENESS-EFFICIENT IN CITIES MARKETS 4.17 Through i t s competitiveness pillar, the USP seeks to improve the workings o f urban markets for land, labor, credit and infrastructure and housing inputs in order to enhance livability. Strategic instruments for achieving this include city development strategies CDS (which double up as instruments o f governance too - see B o x l), and housing finance and infrastructure reforms.20Beyond competitiveness lies a worthy vision o f livable cities with buoyant, broad- based growth of employment, incomes and investment (World Bank 2000b p. 48). The USP proposes to monitor this through several outcome indicators (e.g. urban employment, urban incomes, foreign direct investment in cities, satisfaction with business climate) for which data i s scarcely available at the city level. The same i s true o f output indicators (e.g. functioning land markets, availability o f microcredit, access to information technology). Guidance i s particularly needed here for practitioners to know what proxy measures to use for indicators such as these. 4.18 Emphasizing the removal o f constraints to the efficient working o f markets, the competitiveness concept bears more than a family resemblance t o its urban productivity predecessor in the UPP-91 (see: Table 2.2). It shares too with i t s earlier counterpart a general neglect by urban practitioners. Fewer than ten per cent o f completed and ongoing urban operations addressed market issues o f competitiveness in their objectives or components.21 Thus far, ICRs have yet to report concrete results in this area. Borrower urban project managers interviewed for this study reported competitiveness to be the least relevant strategic pillar, a similar sentiment expressed through a p o l l taken o f participants o f the Bank’s Urban Forum in 2000. Why i s this sector strategy objective neglected? 4.19 Part o f the reason may be that urban development practitioners are unsure about what competitiveness in cities means and how it should be fostered. In its July 1999 Evaluation Note to CODE, OED raised its concerns about the ambiguity o f the competitiveness concept, and uncertainty about whom it applies to, whether to local governments, firms, urban households or other economic players. Lively discussions within the Bank provoked by earlier drafts o f the present paper brought up disparate interpretations o f competitiveness such as: business environment; j o b creation; backward/forward linkages; and locally grown businesses. Each idea deserves scrutiny, since each one’s links to better working markets are unclear and their implications for action are most diverse. D o we risk losing sight o f the original meaning-which has rivalry and contest at its h e a r t - o f competition, a word used in city halls throughout the world? 4.20 The academic literature-which the USP does not claim to represent-does not help clarify this, even though competition among firms and households i s mainstreamed in microeconomics. Economists argue that cities-which are merely the physical locus o f economic activity-cannot compete as firms do (Krugman 1996b). In most cities, the greater part o f the 24 urban economy involves non-tradables, and yet competition theory applies only to the tradable part o f that economy (Begg 1999 and Ingram 1998, p. 1032). For cities, therefore, we are l e f t with “precious little agreement either on what the term ‘competitiveness’ means or how policy should aim to enhance it” (Begg 1999, p. 795), and a concept that i s “notoriously difficult to measure” (Freire and Stren 2001, p. 47). Clearly, more work i s needed to refine the term, clarify the underlying objective and determine i t s practical application. 4.21 Urban practitioners are no doubt confounded further by the questions in the academic and business literature on competition about whether developing country cities can really succeed in this game. The USP itself does not take an explicit position in this debate, but the strategic prescriptions are inevitably influenced by it. The most influential text The Competitive Advantage o f Nations describes how spatial clusters o f high-tech innovative and risk-taking companies in cities are the primary source o f economic growth (Porter 1990). But the analysis i s o f clusters in rich O E C D countries, only. B y 1990 according to Porter, only Korea and perhaps Brazil among developing countries had competed their ways into this game (Ibid. pp. 544-556). Developing country cities are s t i l l excluded because they are peripheralized (Sassen 1998) and a globalized economy needs very few nodal city regions in any case (Friedmann 2001 p.120). N o major innovative and risk taking clusters o f the Porter type have been found in Africa (Rakodi 1998), and very few even in Brazil (Henderson 1988).22 With their technology, infrastructure, knowledge workers, high-tech industry, e-government, e-commerce and e-communities, they should be easy to spot in poor countries (Cal- dow 1999, p. 145-7). O f course, f i r m s and other services in close physical proximity in developing country city neighborhoods are often called clusters, and can promote efficiency through agglom- eration economies. But they lack the innovative and risk taking features o f the dynamic clusters that are behind competitive advantage that drives economic growth in rich OECD economies. Uncer- tainties about the applicability o f competitiveness, combined with very modest results from past urban project efforts to jump-start urban economies (World Bank 1994a pp. 27-28) may help ex- plain practitioners’ reluctance to bear the competitiveness standard. 4.22 Past experience has shown that city mayors themselves often see competition as a zero-sum game that benefits the winner only at the expense o f the loser (or losers). Apart from the few trained as economists, mayors are likely to see competition as defined in a dictionary, in English as ‘the activity o f striving to gain something by establishing superiority over others engaged in the same attempt.’ In practice, this has translated into mayors giving costly tax-breaks and free land and in- frastructure in attempts to attract outside enterprises and to nurture local ones. A recent extreme case o f this led to the overproduction and collapse o f the market, as municipalities in China counted the losses o f their competitive involvement in assembling TV sets.23Several other cases o f what has been called a ‘race to the bottom’ and ‘predatory competition’ are documented for both developed and developing countries.24The Bank’s recent keynote paper Globalization, Growth and Poverty puts this risk into context: “A sound investment climate i s not one full o f tax breaks and subsidies for f m s . It i s rather an environment for good economic governance-control o f corrup- tion, well functioning bureaucracies and regulation, contract enforcement, and protection o f prop- erty rights” (Collier and Dollar 2001). It i s important for urban practitioners in the Bank t o help mayors avoid the downside risks o f competition and help them focus upon building this good eco- nomic governance. 4.23 This focus can come from a good understanding o f their local economy, particularly i t s comparative advantage. Every city has some kind o f comparative advantage. A poor city can benefit when a very efficient and rich city i s willing to forgo its o w n efficient production o f some good (allowing a poor city to produce) so that the efficient city can focus upon those activities where it i s comparatively most efficient. City mayors themselves will not make such decisions; they are not CEOs o f business corporations. It i s the j o b o f businesses and households t o exploit 25 these comparative advantages. As leaders o f local public administrations, city mayors can help by posing questions like “what can we do to encourage what i s different and advantageous about our city”25 and “how can we help innovations prosper here?” Most likely, answers will highlight tradable activities that w i l l deepen a city’s involvement in intercity (and international) trade, which urban economists have long regarded as the main drivers o f city economic expansion (Jacobs 1984 p. 42). Through helping their cities to prosper, city mayors will not only be able to improve livability, but they w i l l help build national prosperity. After all, cities account for more than 50 per cent o f GDP (Begg 1996; World Bank 1996). LOOKING AHEAD 4.24 The focus on poverty and results o f the Bank’s past urban portfolio work confirms the relevance o f the USP’s primary goal o f improving livability. Action needed to support it can draw upon the most effective side o f Bank assistance to the urban sector to date, namely improving the lives o f the poor in cities. Thus, the Bank i s well poised for the successful implementation o f the livability dimension o f the strategy. Governance too i s a highly relevant strategic pillar, especially insofar as it makes achieving the livability goal more efficient. I t too can draw upon a successful line o f Bank urban business in the form o f assistance to local government through municipal development projects (MDPs). MDPs have also advanced the fiscal side o f the USP’s bankability agenda-which some argue i s a part o f governance-but progress with municipal creditworthiness has remained slow to date. Competitiveness, like its urban productivity predecessor o f the UPP-91, looks set to be the strategic pillar least adopted by urban practitioners. In the short to medium term this may not be a bad thing. It w i l l give the Bank and borrowers more time to refine the concept and clarify how it can be translated into practical actions that improve livability. It i s important to clarify i t s meaning to help dispel doubts and risks that, thus far, have discouraged i t s adoption. Looking at the strategy as a whole, we should expect good results from urban projects under this strategy in the years ahead. Specific recommendations to help achieve them are presented below. 5. Recommendations To help the urban lending program focus more effectively upon getting urban poverty alleviation results within the current strategic framework, OED recommends: Systematic M&E and reporting o f results-of poverty alleviation especially-from the city to the sector/strategic levels. The Bank should go beyond the USP’s ‘illustrative’ indicators and put in place an M&E program to measure results o f Bank investments in cities, and report on them on a regular basis. Revision o f the USP’s business strategy to ensure successful implementation. This would provide explicit targets and determine priorities that link the USP’s four key instruments-scaling-up services to the poor, city development strategies, national urban strategies, and local government capacity building-and four strategic pillars-livability, good governance, bankability and competitiveness-to urban poverty alleviation. Clarification of the concept and the operational consequences o f the competitiveness USP pillar for urban practitioners. One way o f doing this could be through issuing region-specific guidelines explaining to Bank task managers, borrower project managers, city mayors and other officials, how to get urban poverty alleviation results through the implementation o f this pillar. 27 Annex 1 Annex 1: Urban Output and Outcome Indicators from USP Output indicators Outcome indicators Percentage o f households with access to piped water, Urban poverty rates, especiallv for female-headed households sanitation, power, heating, and social services Percentage of income spent by low-income households on housing, water, energy, transport, food and social services ._________________---~---~-------------~---------------. Percentage of households with secure tenure Crowding (floor space per person) Percentage of households in informal housing or slum neighborhoods ___________________-------------~------~---------------. Housing affordability (ratio of housing prices to incomes) .________________-_-~-----~----------------------------. Rental market turnover ._________-____-__--------------------------------~----. Availability and use of public transport Percentage of solid waste adequately disposed. Percentage o f wastewater treated School dropout rates Response time to disasters Extent of strategic intent or vision developed in partnership Accountable and honest local govemment that i s responsive to with stakeholders that guides local govemment activities the needs o f the poor ........................................................... Extent of regular public consultation in local Efficiency and competency o f local government in fulfilling government’s budgeting and investment selection process essential responsibilities ........................................................... Percentageof local government staff with professional Extent of trust and satisfactionwith local governmentperformance qualification expressedby citizens and other stakeholda inrepresentativesurveys Percentage of local government services that are subject to competition with the private sector to assure efficient and effective service delivery Public access to information about local govt. decisions (e.g. policy and regulatory contract awards, procurement service delivery and budgetary performance, etc.) Tax collection rates (or tax effort relative to revenue base) _ _ _ _ _ _o _ _ _ _ _ _ _ _ _Balance _f_local ____ govemment budget ___________________--------------~---- Development o f municipal credit market (percentage Local government capital investment as share o f its budget ________________________________________------------------- o f banking system lending to municipalities, Local government creditworthinessratings (actual or proxy) percentage o f bank assets for municipal credit, ___-____________________________________------------------- Integrationof municipal finance and mortgage finance into the municipal bond issues) country’s overall financial systems Repayment record o f municipal credit funds Local government debt service ratio Improved cost recovery on revenue-earning services Trends in infrastructureservice quality and efficiency (telephone call completion rates, power and water service interruptions) ________----------------------------------------------. Functioning land markets (with land use reflecting market value) ___________________-____________________--~-----------. Mean travel time to work _______.___________-------------------------------- Regulatory delays for real estate transactions and for approval of business licenses. ....................................................... Availability o f microcredit ....................................................... Access o f firms to information technology and financial services Source: USP (Attachment 2 - Illustrative indicators for monitor - >anand local government nerformance under the new 1 “ strategy) World Bank 2000a pp. 23-25, Note: Being side-by-sid Y this table does not necessarily mean that there is a direct relationship between the proximate output and outcome indicators. 28 Annex 2 : Selected Performance Data: U r b a n Portfolio 1993-2000 Table A.1. Project Factors of Urban Project Performance (total) (per cent of row total) Number of Satisfactory Likely Substantial projects outcome sustainability ID impact Poverty reduction In objectives 52 75 42 25 In components 42 67 38 17 In lessons 9 78 67 22 PTI (poverty targeted 9 89 33 44 intervention) Loan size Large - .(66-450) 33 85 58 39 (millions Of us Medium (23-65) 33 64 39 18 dollars) Small 11.4-23) 33 66 19 19 Size of Project Large (>500,000) 41 68 39 24 target city Medium (50,000-500,000) 40 77 46 28 (population) Small (<50,000) 17 65 29 18 Results-based Long term (>5 years) 41 63 22 17 focus Medium-term (1-5 years) 50 78 52 28 Short-term (immediate effect) 8 63 38 38 Partnerships with Two or more 15 73 33 33 cofinancers One 32 72 44 28 None 35 66 46 37 Other factors Highlylsubstantially relevant 36 80 49 39 Strong borrower ownership 59 83 52 38 Beneficiary participation 26 89 65 50 Highly/substantially complex 42 67 40 24 Source: OED urban review database. Table A.2 Management Factors of U r b a n Project Performance (total) (percent of row total) Number of Satisfactory Likely Substantial projects outcome sustainability ID impact Borrower Highly satisfactory Performance Satisfactory 44 93 59 43 Unsatisfactory 17 24 6 0 Highly unsatisfactory Bank Highly satisfactory 2 100 100 100 Performance Satisfactory 46 85 54 37 Unsatisfactory 12 33 0 0 Highly unsatisfactory 1 0 0 0 Source: OED urban review database. 29 Table A.3 Country Factors o f Urban Project Performance (total) (per cent of row total) Number of Satisfactory Likely Substantial projects outcome sustainability ID impact Country income Upper middle 19 79 68 42 category Lower middle 35 74 40 20 Low income 44 66 25 21 Least Developed Least developed countries 22 59 18 18 Country Country Credit AI-Baa2 (investment grade) 19 74 63 16 Rating Baa3-61 (speculative grade) 24 42 88 54 (Moody's)* 62-Caa3 (junk grade) 14 71 21 21 Not rated 42 60 26 21 Country urbaniz- High (58%-89%) 33 81 50 34 ation (% Of total Medium (35%-58y0) 33 73 36 24 population in urban areas) LOW (6%-34%) 33 59 31 19 Country High (3.5-4.7) 30 70 39 27 governance Medium (3.1-3.5) 31 84 48 33 category(*) (av. CPlA indicator) !-OW (2.7-3.1) 31 55 28 16 Source: OED urban review database, WDR and Moody's Note: Standard errors associated with estimates of governance can be large and recommend the use of broad intervals of cases according to levels of govemance, as we do here (Kaufmann et al. 1999b) * Moody's ratings refer to sovereign debt issued in 2001, which, while rated, is not formally graded as corporate debt is. Nevertheless, "investment grade" etc. is noted here as a metaphor for equivalent qualify commercial paper in the financial markets. 30 41 0 a, g E 3 3 a, Y 2 e 3 + n W . n v1 a, * .3 .3 C 0 m 8 c) 0 -I .e 3 m m m m m m 3 m 41 0 9 0 n ?5 h #, h + n 9 3 E? .. 0 0 n t .3 .. .- c * v1 0 .? a, % * a, .- (d 0 a .- c S 0 M 2 . . 2 . 2 . w m m m m m m m 2 0 - 0 0 c c c c w 2 w w Y C m '+ . w * h h s 4 cc 0 + h W m M h ?5 W w h h t T t h ru 0 - a - a + n W d- m ru 0 6 n ?5 + h V c h 3 W a e, 35 Annex 4 Annex 4: Urban Portfolio cont: Ongoing Projects* Region Africa Country Project Title Loan# Date Appr. Loan Amt. Benin First Decentralized.City Mgmt. C3234 6/3/1999 25.5 Burkina Faso Urban Environment C2728 5/23/1995 37.0 Burundi Public Works and Employment Creation C3460 1/23/2001 40.0 Cote dIvoire Urban Land Mgmt CN036 11/5/1997 10.0 Cote dIvoire Municipal Support C2704 4/13/1995 40.0 Gabon Pilot Com. Infrastructure L4387 8/24/1998 5.0 Gambia Poverty Alleviation & MunicipaLDev C3176 3/16/1999 15 Ghana Local Govt Dev. C2568 211711994 38.5 Ghana Urban 5 C3330 3/30/2000 10.8 Ghana Urban Environmental Sanitation C2836 3/26/1996 71.0 Guinea Urban I11 C3196 412011999 18.0 Guinea-Bissau Transport And Urban Infrastructure C2748 612211995 0.0 Kenya Emergency 1nfrastructure.Rehab C3120 711611998 40.0 Madagascar Urban Infrastructure C2968 6/25/1997 35.0 Malawi Local Govt. C2379 61411992 23.3 Mali Urban Devt & Decentralization CN004 12/13/1996 80.0 Mauritania Infrast & Pilot Decentralization. C2835 312611996 14.0 Niger Urban Infrastructure. Rehab. C2957 5/29/1997 20.0 Senegal Urban Devt & Decentralization. Project C3006 11/20/1997 75.0 Sierra Leone Freetown Infrastructure C25 11 611011993 26.0 Swaziland Urban Development C3807 11/15/1994 29.0 Tanzania Urban Sector Rehab C2867 512311996 105.0 Togo Lome Urban Development C2620 5/31/1994 26.2 Uganda Nakivubo Channel Rehab C3203 5/6/1999 22.4 Region East Asia Pacific Country Project Title Loan# Date Appr. Loan Amt. China Shanghai Sewerage Project I1 L3987 3/21/1996 250.0 China Enterprise Housing & Soc Sec Reform L3773 71511994 330.0 China Huai River Pollution Control L4597 3/22/2001 105.5 China Hubei Urban Environment L3966 1211911995 121.7 China Zhejiang Multicities Development L2475 312511993 110.0 China Liao River Basin L46 17 61191200 1 100.0 Indonesia Kalimantan Urban Dev L3854 3/21/1995 118.0 Indonesia East Java Urban Dev I1 L40 17 5/16/1996 82.2 Indonesia Sulawesi Urban Dev I1 L4105 11/21/1996 88.0 Indonesia Bali Urban Infrastructure Project L4155 51611997 79.9 Indonesia Surabaya Urban Dev L3726 411211994 0.0 Indonesia Urban Poverty Project L3210 511811999 100.0 Indonesia Semarang Surakarta Urban Dev L3749 61711994 132.6 Indonesia Gef-West Java Environment Mgmt 6/12/2001 0.0 Indonesia Western Java Environment Management L4612 6/12/200 1 17.4 Indonesia Municipal Innovations Project L4440 2/9/1999 5.0 Philippines Lgu Finance And Development Project L4446 3/23/1999 100.0 Vietnam H o Chi Min City Env Sanitation. c3475 3/20/2001 166.3 * (As of July 1, 2001) 36 Annex 4 Region Europe and Central Asia Country Project Title Loan# Date Appr. Loan Amt. Albania Land Dev C3066 5/12/1998 10.0 Azerbaijan Pilot Reconstruction C3 100 7/21] 998 20.0 Bosnia-Herzegovi Local Dev C3191 4/13/1999 15.0 Bosnia-Herzegovi Cultural Heritage Pilot C3269 6/28/1999 4.0 Croatia East Slavonia Reconstruction L4351 611811998 40.6 Georgia Municipal Dev C2976 7/15/1997 20.9 Latvia Municipal Services Dev L3964 1211411995 27.3 Lithuania Municipal Dev L448 1 5/27/1999 30.1 Lithuania Energy Efficient Housing L4064 711111996 10.0 Moldova First Cadastre C3 06 1 4/23/1998 15.9 Poland Flood Emergency L4264 1211811997 200.0 Russia St Petersburg Rehab L4144 3/27/1997 31.0 Russia Housing L3850 31711995 249.3 Russia Municipal Heating L460 1 3/27/200 1 85.0 Tajikistan Emergency Flood Asst C3 123 812711998 7.0 Turkey Emergency Flood Recovery L4388 9/10/1998 369.0 Turkey Marmara Earthquake Emergency L4517 11/16/1999 505.0 Region Latin America and theCaribbean Country Project Title Loan# Date Appr. Loan Amt. Argentina Munic Devt I1 L3860 3/23/1995 210.0 Brazil Bahia MunicipaLDev L 4 140 31411997 100.0 Brazil Ceara Urban Development & Water L3789 9/6/1994 140.0 Brazil Minas Municipal Dev L363 9 712011993 140.3 Colombia Urban Infrastructure L4345 6/11/1998 75.0 Colombia Urban Environment TA L3973 1/16/1996 20.0 Honduras Emergency Disaster Mgmt C3361 5/30/2000 10.8 Nicaragua Natural Disaster Vulnerability Reduction C3487 4/3/200 1 13.5 OECS Countries OECS Ship Waste Mgmt C2859 5/411995 12.5 Venezuela Caracas Slum Upgrading L4400 1012211998 60.7 Region MNA Country Project Title Loan# Date Appr. Loan Amt. Algeria Low-Income Housing L4361 6/25/1998 150.0 Egypt Private Sector Tourism Infras & Env C3605 5/18/1993 70.5 Jordan Community 1nfrastructure.Dev. L42 15 8/2 1/1997 30.0 Jordan Second Tourism Dev. L4214 713 1/1997 32.0 Lebanon Emergency Recovery L3562 31411993 225.0 Lebanon First Municipal Infrastructure L7026 6/22/2000 80.0 Morocco Municipal Finance I1 L423 1 911 111997 70.0 Morocco Sustainable Coastal Tourism L4573 6/3 0/2000 2.2 Morocco FesMedina Rehab. L4402 10/29/1998 14.0 Tunisia Cultural Heritage Project L7059 6/12/2001 17.0 Tunisia Municipal Dev. I1 L4202 612411997 80.0 West Bank-Gaza MIDPII 6/20/2000 7.5 West Bank-Gaza Housing C2605 4/8/1997 25.0 West Bank-Gaza Municipal Development C2605 513011996 40.0 West Bank-Gaza Bethlehem 2000 C2604 912211998 25.0 Yemen Public Works I1 C3 168 1/28/1999 50.0 Region SAR Country Project Title Loan# Date Appr. Loan Amt. Bangladesh Air Quality Management Project C3404 7/25/2000 4.7 Bhutan Urban Development Project C33 10 121214999 10.8 India Urban Dev I1 C4478 5/27/1999 105.0 Pakistan N. West Frontier Prov Community Infra C2829 3/14/1996 21.5 Sri Lanka Colombo Env. Improvement C2757 6/29/1995 31.2 37 Annex 5 Annex 5: Management Response I.INTRODUCTION 1. Management commends the OED study team for this thorough review and for their credible effort to take account o f previous staff comments and concerns. For the most part, the Review provides valuable analysis and stocktaking with which Management agrees. The Review also includes some valid criticisms (including o f certain aspects o f portfolio performance up to FYOO), and three important and challenging recommendations. VIEWS 11. MANAGEMENT’S 2. Management agrees with and supports the recommendations o f the OED review. Management has recognized the issues that prompt the recommendations and, as indicated in the matrix below, had already begun to formulate responses along those lines. This OED report will give additional impetus to the implementation o f Management’s plans. 3. Management has noted that our earlier comments on the draft report have largely been taken into account. However, Management would like to clarify i t s views on a few items without detracting from our general appreciation o f the present report. 4. Management agrees that the recommendation that the Bank support urban data indicators collection i s generally useful. Although collecting urban indicator data for international comparison i s valid, Management notes that the Bank and other donors have directed most o f their recent efforts toward strengthening locally-defined indicators relevant to projects and strategies, and to promoting the use o f these locally-defined indicators in operations. B o t h objectives are justifiable and needed, but from our perspective, there i s less international momentum behind data collection for international comparison. Since there are limited resources devoted to either, Management would choose to emphasize efforts to ward strengthening more locally-defined indicators. 5. Management was not convinced by OED’s findings that E S W does not affect outcomes. W e submit that E S W I A A A that Urban Poverty Analyses and Strategies (see matrix below) provide a reliable basis for the orderly application o f Bank instruments resulting in the successful development o f urban areas. In fact, Bank experience indicates that for multi-sectoral themes such as urban, several types of E S W are necessary because o f the inter-relatedness and interactions among sectors that affect the urban environment. The Bank has found it essential that the design o f individual interventions be based on an understanding o f these critical cross-sectoral relationships, and Management believes that carefully designed, multisectoral E S W i s indispensable to reach that level o f understanding across sectors. [corit: 38 Annex 5 111. SUMMARY OF OED RECOMMENDATIONSWITH MANAGEMENT RESPONSE 1 OED RECOMMENDATIONS MANAGEMENT RESPONSE The Bank should go beyond the USP’s Management endorses this recommendation. Working ‘illustrative’ indicators and put in place an with other Bankwide institutional efforts to improve M & M&E program to measure results o f Bank E in Bank operations, Management i s developing a investments in cities and report them on a program to provide advice, training, and support to Urban regular basis, Sector staff, specifically to help fill this important gap. Management i s also working with other international agencies to develop approaches to monitor national and global progress in urban poverty reduction, 2. Revision o f the USP’s business strategy to I lsure successful implementation This would provide explicit targets and Management accepts this recommendation. The Urban determine piiorities that link the USP’s four Anchor’s business plan reflects ongoing efforts to instruments-scaling up services to the poor, promote the effective implementation o f the Urban city development strategies, national urban Strategy Paper and the linkage of its key components. strategies, and local capacity building-and Urban staff are collaborating with cross-sectoral four strategic pillars-livability, good colleagues in a new non-lendingproduct currently governance, bankability and competitiveness- designated as Urban Poverty Analysis and Strategy to poverty alleviation (UPAS) to bring together multi-sectoral, poverty- focused analyses and a consultative process modeled on City Development Strategy experience. The aim i s to obtain locally-ownedstrategies and multi-year interventions for targeted, manageable support to reduce poverty in cities. This approach i s being tested in several pilots initiated in FY03. 3. Clarification o f the concept and Operations consequences o f the competitiveness USP pillar for urban practitioners One way o f doing this could be through Management agrees with the thrust o f this issuing region-specific guidelines explaining recommendation. The urban anchor has been working to Bank task managers, borrowing project for almost two years on a local economic development managers, city mayors and other officials, how (LED) knowledge and capacity-buildingprogram, to get urban poverty alleviation results through which includes methodologies for strategic planning o f the implementation o f this pillar LED based on international review of experience, good practice, and testing in client countries. These applications are underway, mainly in ECA, but considerable interest i s emerging in other Regions. A draft manual has been prepared and i s being presented in courses and workshops for clients and staff. The manual will be further refined in FY04 based on experience with its use. Substantial material on LED has been developed on the urban KM website and may be accessed under the LED link. The LED program i s being externally peer reviewed in FY03 and further refinements in content and approach will be identified and pursued accordingly. Issues of mainstreaming LED into Bank operations are also being addressed through practitioner workshops and internal peer reviews during FY03. 39 Annex 6 Annex 6: Chairman’s Summary: Committee On Development Effectiveness Report Findings. This review i s a sequel to OED’s first review Twenty Years o f Lending for Urban Development 1972-1992. The current urban strategy paper (USP), Cities in Transition, was discussed by CODE in July 1999 and endorsed by the Board in November o f the same year. The review uses the four pillars of the USP -- livability, good governance, bankability, and competitiveness -- as the framework for evaluating project performance. The results are promising. There were significantly greater improvements in the lives o f the poor in cities that benefited from Bank-financed urban projects than in similar cities that did not have such interventions. Projects undertaken in higher income and more urbanized countries also performed better. However, evidence o f project impact on the poor has not been systematically monitored, and improved monitoring and evaluation o f results at the project and sectorhtrategic levels are necessary. Management i s in broad agreement with OED’s recommendations and with the areas that OED concludes need priority action. Management has already initiated a number o f actions along the lines proposed by OED. 2. Main Conclusions and Next Steps. The committee welcomed the positive findings o f the OED study, but raised the question of attribution. CODE endorsed the findings o f the review and its recommendations but believed it could have spelled out more filly the operational lessons for successful strategy implementation, particularly those relevant for improving policy and project outcomes. The committee broadly endorsed the management response, but noted that it would have liked more information on how management planned to make progress in M&E o f urban projects. The findings and recommendations of the OED review will inform management’s on-going activities and implementationo f the FY03 Urban business plan. 3. Attribution of Outcomes to Bank Interventions. The committee noted that the correlation between the Bank’s investment and improvements in the cities was encouraging but it urged further analysis o f this relationship in order to attribute the causes o f the improvements more precisely to Bank activities. OED emphasized that the results showed a strong association and not necessarily direct causality and that the challenge would be to provide more evidence to inform future policy and project design. Management agreed that it would be important to be able to assess the impact of the Bank’s urban projects but also cautioned that it would be difficult to measure how many people actually benefit from a particular urban investment, given the nature o f urban dynamics. 4. Links to the Macro-economic Policy Framework and Other Sectorflhematic Strategies. The committee noted that the urban strategy needs to be set within the country context and be closely linked to macro-economic policy. OED and management agreed that it was important to assess the urban strategy within country and sub-regional contexts. Members stressed the importance o f a cross-sectoral approach and asked how the priority issues from related sectors (rural, transport, water) are being integrated into the urban strategy and the urban work program. Management clarified that the OED review was evaluating urban projects carried out between 1990 and 1999-2000. Following the new USP the urban program today i s being executed with cross-sectoral teams and consultations with all the networks. 40 Annex 6 5 . Links Between Urban Strategy and Poverty. The committee believed that it would be important to increase the operational focus o f the strategy on “economic growth” and “improving the investment climate”. In this regard, it would have been useful for the OED review to say more about the trade-offs between focusing resources on improving the living conditions o f the poor and directing resources towards promoting growth. The DGO noted that there may be no tradeoff, since improving the conditions o f the urban poor can help productivity and hence growth. A member noted the widening gap between the supply and demand for urban services and said it would be important for the review to highlight the need for more investments in the urban sector, particularly in light o f the daunting task o f achieving the MDGs. 6. Operationalizing the “competitivenessIf Pillar. The committee was concerned that the concept o f “competitiveness o f cities” i s relatively unknown to staff and that the operational implications o f this pillar are unclear. It agreed with OED on the need to clarify the concept further and hoped future work will be linked with on-going work on the investment climate and the private sector development strategy. The DGO suggested that an approach focusing on the comparative advantage o f cities and gains o f trade would be more useful and would avoid cities engaging in potential zero-sum games. Management said that the main message and operational focus o f this pillar now emphasized the role local governments could play in shaping the conditions o f the local economy, identifying constraints to improving the local investment climate, and developing good communications with the private sector. 7. Monitoring and Evaluation. The committee agreed that there was a clear need for data collection and for monitoring results at the project, at the country, and at the international level. Members, however, cautioned that such efforts should not over-burden clients. Management noted that harmonization of data collection and M&E at the country level was key. OED and management also supported collecting indicators at all levels, but noted the difficulty in balancing the need for indicators to assess project level impacts and the need to be able to track progress at the global level. The committee stressed the importance o f self-evaluation as a vehicle for focusing on results and asked when management anticipated having such a system in place and what role the sector board would play. Management informed the committee that it was moving ahead on monitoring and evaluation and was preparing a work program on the results agenda. Speakers also discussed the appropriateness o f some o f the indicators to monitor strategy implementation (education, health). OED noted that many of the social indicators are linked directly to the MDGs. 8. Bankability. The committee noted the need for the strategy to focus on municipal creditworthiness and fiscal sustainability at the national and sub-national levels. The DGO noted that there were examples o f cities that have successfully borrowed funds, some without the benefit o f central government guarantees. Management emphasized the importance o f strengthening the capacity o f cities to raise local currency, particularly for capital improvements, and noted that work was underway with partners to explore how to provide partial guarantees to public sector authorities or cities that are creditworthy and can issue bonds. Other issues raised by members were the need to consider the role o f secondary cities in achieving the strategy’s goals, including addressing issues o f economies o f scale. OED noted that the review focused on an array o f cities of various sizes, but omitted mega cities. Management confirmed that a minority share o f urban lending i s going to mega cities. OED also noted that there i s a significant inverse relationship between urban service costs and population density. The committee observed 41 Annex 6 that the terminology o f an urban PRSP was confusing and asked that it be revisited. Management agreed and w i l l make the change. 9. The Role o f ESW in Urban Sector Work. The committee recognized management’s view that the issue o f weak correlation between ESW and the performance o f urban projects was more one o f measurement than fact. Members also recognized that the link between ESW and the quality o f urban projects may require further analysis and asked management to take a more critical look at ESW. Members noted that strong and relevant urban ESW was critical to generating solid urban operational programs. They felt that Bank ESW could play an advocacy role in informing policymakers about the contribution o f the urban sector to development. Finn Jonck, Chairman 43 Bibliography Bibliography Abers, Rebecca. 200 1. “Learning Democratic Practice: Distributing Government Resources through Popular Participation in Porto Alegre, Brazil” in: Freire and Stren (2001). Bartone, Carl R. 2001. “The Role o f the Private Sector in Municipal Solid Waste Service Delivery in Developing Countries: Keys to Success” in: Freire and Stren (2001). Begg, Iain. 1999. “Cities and Competitiveness’’ in: Urban Studies Vol. 36, nos. 5-6. Bergstrom, Fredrick 1999. JThy do Local Governments Privatize Paper presented to the International Society for New Institutional Economics 1999 Conference held in Washington D.C. Bloomberg School o f Public Health. 2001. Basic Health Needs in Developing Country Cities - 2000 Baltimore, John Hopkins University. Boddy, Martin 2000. Cities and Competition: Definitions and Issues - A UK Outlook in; World Markets Research Center (2000). Buckley, Robert and Federico Mini 2000. From Commissars to Mayors: Cities in the Transition Economies Report prepared for World BanWIMF Annual Meetings in Prague in September 2000. Bump, Jesse and Jesko Hentschel. 1998. Urban Income Poverty-Some Cross-country Comparisons: A Background Note Washington D.C., World Bank PREM Network draft. Burgess, Rod; Marisa Carmona and The0 Kolstee. 1997. The Challenge o f Sustainable Cities: Neoliberalism and urban strategies in developing countries London, Zed Books. Caine, Christopher G. 2000. Readiness for the Networked World: Competitive Cities in the Digital Economy in: World Markets Research Center (2000). Caldow, Janet 1999. Competitive Cities in a Digital Society in: World Markets Research Center (1 999). Campbell, Tim 1999. The Changing Prospectsfor Cities in Development: The Case of Vietnam in: World Markets Research Center (1 999). Campbell, Tim 2001. ‘Innovation and Risk-Taking in Urban Governance in Latin America’ in: Scott (2001). Cohen, Michael 2001a. “The Impact o f the Global Economy on Cities” in: Freire and Stren (2001) Cohen, Michael 2001b. “Urban Assistance and the Material World: Learning by Doing at the World Bank” in: Environment and Urbanization Vol. 13. No. 1 April 200 1. Collier, Paul and David Dollar. 2001 Globalization, Growth and Poverty Washington DC, World Bank Policy Research Report. Costa, Rene 1998. Urban Development Projects: Lessonsfrom OED’s Performance Evaluation Results Urban Development Division, Washington D.C., The World Bank. Diaz, Andres Cabanas; Emma Grant; Paula Irene del Cid Vargas; and Ver6nica Sajbin Velasquez The Role of External Agencies in the Development of El Mezquital in Guatemala City in: Environment and Urbanization Vol. 13. No.1 April 2001. DiGaetano, Alan and John Klemanski 1999. Power and City Governance: Comparative Perspectives on Urban Development University o f Minneapolis Press. 44 Bibliography Dillinger, William and Shahid Yussuf 1999. Financing Cities: The View from the WDR 2000 in: World Markets Research Center (1999). Duffy, Hazel 1995. Competitive Cities: Succeeding in the Global Economy London, E & F N Spon. Ebel, Robert D. and Franqois Vaillancourt. 2001. “Fiscal Decentralization and Financing Urban Governments: Framing the Problem” in: Freire and Stren (2001). Evans, Alison 2000. Poverty Reduction in the 1990s: An Evaluation of Strategy and Performance Washington D.C., World Bank Operations Evaluation Department. Fernandes, EdBsio and Ann Varley. 1998. Illegal Cities: Law and Urban Change in Developing Countries London, Zed Books. Freire, M i l a and Richard Stren. 2001. The Challenge o f Urban Government: Policies and Practices Washington D.C., World Bank Institute Development Studies. Friedmann, John. 2001. “Intercity Networks in a Globalizing Era” in Scott (2001). Fujita, Masahisa; Paul Krugman and Anthony J. Venables. 1999. The Spatial Economy: Cities, Regions and International Trade Cambridge, Massachusetts, MIT Press. Gilbert, Roy. 1995. “Rio de Janeiro: the Making o f a Modern Megacity” in Habitat International Vol. 19, no. 1. Gilbert, Roy and Alcira Kreimer 1999. Learningfrom the World Bank’s Experience o f Natural Disaster Related Assistance Washington D.C., Urban Development Division, the World Bank. Gilbert, Roy. 200 1. Doing More for Those Made Homeless by Natural Disasters Washington D.C., World Bank Disaster Risk Management Working Paper No.1. Gordon, Ian. 1999. “Internationalism and Urban Competition” in: Urban Studies Vol. 36, nos.5-6. Hall, Peter. 1966. The World Cities London. Weidenfeld and Nicolson. Hall, Peter. 1998a. Cities in Civilization New York, Pantheon Books. Hall, Peter. 1998. “Globalization and the World Cities” in: L o and Leung (1 998) Harris, Nigel and Ida Fabricius 1996. Cities and Structural Adjustment London, U C L Press. Henderson, J. Vernon. 1988. Urban Development: Theory, Fact and Illusion New York, Oxford University Press. Henderson, J. Vernon; Zmarak Shalizi and Anthony Venables. 2000. Geography and Development Washington DC, World Bank Policy Research Paper. Herr, Harvey and Guenter Karl. 2002. Estimating Global Slum Dwellers: Monitoring the Millenium Development Goal 7, Target 11 Nairobi, Consultant paper for UN Habitat. Hirschman, Albert 0. 1992. Rival Views of Market Society Cambridge Massachusetts, Harvard University Press Hirschman, Albert 0.. 1998. Crossing Borders: Selected Writings New York, Zone Books. Huriot, Jean-Marie and Jacques-Franqois Thisse 2000. Economics of Cities : Theoretical Perspectives Cambridge University Press. IFAD. 2002. Enabling the Rural Poor to Overcome their Poverty: Strategic Framework for IFAD 2002-2006 Rome, IFAD. 45 Bibliography IFAD. 2002. IFAD Strategy for Rural Poverty Reduction in Latin America and the Caribbean Rome, IFAD Regional Strategy Paper. Ingram, Gregory K. 1998. “Patterns o f Metropolitan Development: What Have We Learned?” in: Urban Studies, Vol. 3, No. 7. Jacobs, Jane 1969. The Economy of Cities Penguin Books. Jacobs, Jane 1984. Cities and the Wealth of Nations New York, Random House. Jones, Gavin W. and Pravin Visaria 1997. Urbanization and Large Developing Countries: China, Indonesia, Brazil and India Oxford, Clarendon Press. Kaufmann, Daniel; Aart Kraay and Pablo Zoido-Lobatbn 1999a. Governance Matters Washington D.C., The World Bank. Kaufmann, Daniel; Aart Kraay and Pablo Zoido-Lobatbn 1999b. Aggregating Governance Indicators Washington D.C., World Bank Policy Research Paper #2195. Kessides, Christine 1993a. Institutional Optionsfor the Provision of Infrastructure Washington D.C., World Bank Discussion Paper no. 212. Kessides, Christine 1993b. The Contributions of Infrastructure to Economic Development: A Review o f Experience and Policy Implications Washington D.C., World Bank Discussion Paper no. 213. Kessides, Christine 1997. World Bank Experience with the Provision of Infrastructure Services for the Urban Poor: Preliminary IdentiJcation and Review o f Best Practices Washington D.C., Transportation, Water and Urban Development Department, The World Bank. Klitgaard, Robert; Ronald Maclean-Araroa and H. Lindsey Parris 2000. Corrupt Cities: A Practical Guide to Cure and Prevention Oakland, Institute o f Contemporary Studies. Krugman, Paul. 1996a. “Making Sense o f the Competitiveness Debate” in: Oxford Review of Economic Policy Vol. 12 no.3. Krugman, Paul. 1996b. Pop Internationalism Cambridge Massachusetts, The M I T Press. Krugman, Paul. 1998. “What’s New about New Economic Geography” in: Oxford Review of Economic Policy Vo1.14, No. 2, Summer 1998. Leautier, Frannie A. 200 1. Facing the Poverty Challenge in: World Markets Research Center (200 1) Lee, K y u Sik and Roy Gilbert 1999. Developing Towns and Cities: Lessons from Brazil and the Philippines Washington D.C., World Bank Operations Evaluation Department. Lo, Fu-Chen, and Yue-Man Yeung (editors). 1998. Globalization and the World o f Large Cities Tokyo, United Nations University Press. McDade, Thomas W. and Linda S. Adair 200 1. “Defining the “urban” in urbanization and health: a factor analysis approach” in Social Science and Medicine 53 (2001) 55-70. Mills, Edwin S. and Ernest0 M. Pernia 1994. ‘Introduction and Overview in: Pernia (1994) Moser, Caroline 1997. ‘Urban Social Policy and Poverty Reduction’ in Burgess (1997). Ohmae, Kenichi. 2001. ‘How to invite prosperity from the Global Economy into a Region’ in: Scott (2001). Organizacibn Panamericana de la Salud OPS. 200 1. Desigualdades en el Acceso, Us0 y Gasto con el Agua Potable en Ame‘rica Latina y el Caribe Washington D.C, February 2001 OPS 46 Bibliography Perdomo, Rogelio PBrez and Teolinda Bolivar. 1997. Legal Pluralism in Caracas, Venezuela in: Fernandes and Varley (1997). Pernia, Ernest0 M. 1994. Urban Poverty in Asia: A Survey o f Critical Issues Hong Kong, Asian Development Bank, Oxford University Press. Piermay, J-L. 1997. “Kinshasa: A Reprieved Megacity?” in Rakodi, Carole (1997). Porter, Michael E. 1990. The Competitive Advantage of Nations New York, The Free Press. Porter, Michael E. 1996. ‘What i s Strategy’ in: Harvard Business Review November-December 1996. Porter, Michael E. 1998a. ‘The Adam Smith Address; Location, Clusters and the “New” Microeconomics o f Competition’ in: Business Economics Vol. XXXIII, Number 1, January 1998. Porter, Michael E. 1998b. ‘Clusters and the New Economics o f Competition’ in: Harvard Business Review November-December 1998. Porter, Michael E. 2000. ‘Regions and the New Economics o f Competition’ in: World Markets Research Center (2000). Porter, Michael E. 2001. ‘Regions and the New Economics o f Competition’ in: Scott (2001) - expanded version. Rakodi, Carole. 1997. The Urban Challenge in Africa: Growth and Management of its Largest Cities Tokyo, United Nations Press. Rakodi, Carole. 1998. “Globalization Trends and sub-Saharan African Cities” in: L o and Yeung (1998) Richardson, Harry W. 1971. Urban Economics Harmondsworth, England, Penguin Books. Sattherwaite, David. 2001. “Reducing Urban Poverty: Constraints on the Effectiveness o f Aid Agencies and Development Banks and Some Suggestions for Change” in: Environment and Urbanization Vol. 13. No.1 April 2001. Sassen, Saskia. 1991. The Global City: New York, London, Tokyo Princeton, Princeton University Press. Sassen, Saskia. 1998. “The impact o f the new technologies and globalization on cities” in: Lo and Yeung (1998) Sassen, Saskia 2000. Cities in a World Economy 2nd.Edition, London, Pink Forge Press. Scott, Allen. 2001. Global City Regions Oxford, Oxford University Press. Simmonds, Roger and Gary Hack (editors). 2000. Global City Regions: Their Emerging Forms London, Spon Press. Stren, Richard. 200 1. ‘Local Governance and Social Diversity in the Developing World: New Challenges for Globalizing City Regions’ in: Scott (2001). Tolosa, Hamilton. 1998. “Rio de Janeiro as a World City” in: L o and Yeung (1998). United Nations Centre for Human Settlements (UN Habitat). 1987. Global Report on Human Settlements 2001 Oxford, Oxford University Press. United Nations Centre for Human Settlements (UN Habitat). 2001a. Cities in a Globalizing World: Global Report on Human Settlements 2001 London, Earthscan Publications Ltd. 47 Bibliography United Nations Centre for Human Settlements (UN Habitat). 2001b. The State of the World’s Cities2001 Nairobi, UNCHS. United Nations Development Programme (UNDP). 2001. Human Development Report 2001 New York, Oxford University Press. Vanderschueren, Franz. 2001. “From Violence to Justice and Security in Cities” in: Freire and Stren (2001). Venables, Anthony J. 1998, “The Assessment: Trade and Location” in: Oxford Review o f Economic Policy Vo1.14, No. 2, Summer 1998. Wolfensohn, James. 2001. “The World Bank and Global City Regions: Reaching the Poor” in: Scott (200 1). World Bank. 1972. Urbanization Washington D.C., Sector Working Paper. World Bank. 1974. Sites and Services Projects Washington D.C., World Bank. World Bank. 1975a. Housing: Sector Policy Paper Washington D.C., World Bank. World Bank. 1975b. Urban Transport: Sector Policy Paper Washington D.C., World Bank World Bank. 1980. Shelter: Poverty and Basic Needs Washington D.C., World Bank. World Bank. 1983. Learning by Doing: World Bank Lendingfor Urban Development 1972-1982 Washington D.C., World Bank. World Bank. 1990. Issues of i‘Governance’’ in borrowing members: the extent of their relevance f Agreement Memorandum o f the Vice-president and General under the Bank’s Articles o Counsel, December 21, 1990. Washington DC, World Bank. World Bank. 1991. Urban Policy and Economic Development: an Agenda for the 1990s Washington D.C., A World Bank policy paper. World Bank. 1993. Housing: Enabling Markets to Work Washington D.C., World Bank Policy Paper. World Bank. 1994a. Twenty Years of Lendingfor Urban Development 1972-1992: An OED Review Washington D.C., World Bank Operations Evaluation Department. World Bank. 1994b. World Development Report 1994: Inpastructure for Development Washington D.C., The World Bank. World Bank. 1995. Better Urban Services: Finding the Right Incentives Washington D.C., The World Bank. World Bank. 1996. Livable Cities for the 21” Century published for the U.N. City Summit, Istanbul, Turkey 1996. World Bank. 1997. Brazil: Learning from Best Practices in Five Urban Projects Washington D.C., World Bank Operations Evaluation Department. World Bank. 1999a 1998 Annual Review of Development Effectiveness Washington D.C., World Bank Operations Evaluation Department. World Bank. 1999b. Rural Development: from Vision to Action? Washington D.C., World Bank Operations Evaluation Department, World Bank . 1999c. Cities without Slums: Action Plan for moving slum upgrading to scale.. Washington D.C., World Bank. 48 Bibliography World Bank. 1999d. Evaluative Note on the Draft “A Strategic View of Urban and Local Government Issues: Implications for the Bank” Washington DC, World Bank Committee on Development Effectiveness. World Bank. 2000a. Cities in Transition: World Bank Urban and Local Government Strategy Washington D.C., World Bank Infrastructure Group. World Bank. 2000b. World Development Report 2000/2001: Attacking Poverty Washington D.C., Oxford University Press. World Bank. 2001a. Toward Sharpening the Focus on Rural Poverty: A Review of World Bank Experience Washington D.C., Operations Evaluation Department, (September 28,2001). World Bank. 200 1b. Cultural Properties in Policy and Practice: A Review of World Bank Experience Washington D.C. Operations Evaluation Department (June 28,200 1 draft). World Bank. 200 1c. World Development Report 2002: Building Institutions for Markets Washington D.C. (draft discussion proof). World Bank. 2002. Social Funds: Assessing Effectiveness Washington D.C., Operations Evaluation Department World Markets Research Center 1999. World Urban Economic Development Official business and technology briefing for the World Competitive Cities Congress 19-21 May 1999, Washington D.C. World Markets Research Center 2000. World Urban Economic Development Official business and technology briefing for the World Competitive Cities Congress 17-19 May 2000, Washington D.C. World Markets Research Center 200 1. World Infrastructure and Development: Infrastructure and Poverty Briefing for the World Bank Infrastructure Forum May 2-1 1, 200 1 Washington D.C. 49 Endnotes: 1. Estimated as follows: Of Bank investment in education, health, transport, water and energy 40 per cent goes to cities, their share o f the total population. For agriculture, environment, finance, mining, oil and gas, and also for non-spatial lending, zero per cent goes to cities. For urban development itself, the share i s 100 percent. 2. The rural poor exceed the urban poor in most regions except Europe & Central Asia (ECA), where the urban poor are twice as numerous as the rural poor (Buckley and Mini 2000) and Latin America and the Caribbean (LCR), where the urban poor outnumber the rural poor 142 million/82 million (IFAD 2002). 3. This study’s estimate implies a total o f 1.85 million poor living below national poverty lines. An alternative estimate recently made by the International Fund for Agricultural Development (IFAD) focuses upon the 1.2 billion people living on less than a-dollar-a-day. According to IFAD’s estimate, urban areas account for 25 per cent of this group, or 300 million (extreme) urban poor altogether (IFAD 2002). Another recent related estimate made for UN Habitat using a straight line urban population projection from 1993 points to 837 million people living in slums globally in 2001. This i s equivalent to 38 percent o f the urban population o f developing countries and 4 per cent o f the urban population o f developed countries, but this estimate does not explicitly affirm that all slum dwellers are poor (Herr 2002). 4. The number would fall to 14 million people per year, if only those living on less than a-dollar-a-day were included as per the IFAD estimate (see: endnote 3). 5. For a detailed assessment o f performance during this period, see OED’s earlier review (World Bank 1994a) 6. There was also a very earlier upgrading project in Africa, namely the Zambia: Lusaka Upgrading and Sites and Services (Ln1057). 7. Sites and service projects included: Senegal (Cr336), Tanzania (Cr495), Zambia (Ln1057), Indonesia (LnlO40), El Salvador (Lnl OSO), and Peru (Ln1283). Early urban transport projects included: Malaysia (Ln851), Iran (Ln952) and Tunisia (Ln937). 8. Brazil: Salvador Metropolitan Development (Ln268 1) and in Argentina (Ln2992). 9. The operations were: India; Uttar Pradesh Urban (Cr1780); Bangladesh: Urban I(Cr1930); and Mexico: Solid Waste Management (Ln2669). 10. Bank client cities were those that hosted Bank financed urban projects-11 completed and 13 ongoing-approved during 1992-96. The 24 urban projects would likely impact the indicator values for 1998, but not 1993. This sample o f 24 projects was representative o f the urban portfolio as a whole insofar as key sample mean values-loan size, performance rating etc.-were not significantly different (t-test at 90 per cent, assuming equal variances) from those o f the portfolio as a whole. T h e 24 B a n k client cities were: Accra, Antananarivo, Belize City, Bobo Dioulasso, Colombo, Cotonou, Kumasi, Lilongwe, Lome, Maputo, Maseru, Nizhny Novgorod, Nouakchott, Ouagadougou, Rabat, Riga, Santa Cruz de la Sierra, Santiago, Semarang, Surabaya, Tirana, Tunis, Vilnius, Yerevan. Cities with more than five million inhabitants were not included, as a single project was not expected to have a perceptible citywide impact at this scale. Comparator cities were chosen as those that best matched the Bank client cities by regional location, city population size, city population growth, country GNP per capita, country GDP growth, and country urbanization. The means o f these variables were not significantly different between the two groups o f cities (t-test at 90 per cent level, assuming equal variances). T h e 37 Comparator cities were: Addis Ababa, Amman, Asuncion, Bamako, Bandung, Bulawayo, Cajamarca, Cebu, Chittagong, Conakry, Cuenca, Douala, Gabarone, Guayaquil, Harare, Ibadan, Jinja, Kigali, Kingston, Kisumu, Kostromo, Minsk, Mombasa, Mysore, N’Djamena, Niamey, Quito, Recife, San Salvador, Sana’a, Tallin, Tbilisi, Ulaanbaatar, Vilnius, Windhoek, Yaoude, Zagreb. 11. Results reported here are of analyses o f differences o f means (assuming equal variances) between Bank client and comparator cities, with t-test statistics significant at the 90 per cent level or higher. 12. APPI simply consolidates OED’s three principal ratings by the following formulae into a continuous numerical index with values form 2 to 10 that i s amendable to quantitative analysis : (a) Outcome: highly satisfactory = 7.75; satisfactory = 6.00; moderately satisfactory = 5.25; moderately unsatisfactory = 4.50; 50 unsatisfactory = 3.75; highly unsatisfactory = 2.00; (b) Sustainability: likely = 0.75; uncertain = 0.25; unlikely = 0.0; (c) Institutional Development Impact: substantial = 1.50; modest = 0.50; negligible = 0.00. (World Bank 1999a, p. 39) 13. This i s not the same concept as the USP’sprinciple of relevance which i s about: “mobilizing skills and knowledge across the widening range o f urban issues and offering assistance with flexible designs, realistic time frames, and appropriate forms o f financing to meet varied client needs” (World Bank 2000a p. 61). 14. Participation does not preclude the need for project staff to develop a good understanding o f the demand for project services, however. OED’s review o f social funds found a community’s first choice o f investment can be overridden by the preferences o f a community’s “prime mover”, such as a teacher preferring a school or a doctor preferring a health post a school, for instance (World Bank 2002 p. xxvii). 15. In OED evaluations this refers to performance o f the government and implementing agency in preparing and implementing a project. For the preparation phase, OED evaluators take account o f how well the governmentfimplementation agency took account o f economic, financial, technical, policy, and resource considerations, and ensured participation o f major stakeholders in preparing the project. For the implementation phase, OED considers the extent to which the government supported the project through sectoral policies, commitment to the operation itself and counterpart funding, and the extent to which the implementing agency supported the project through good management, adequate staffing, cost controls and the involvement o f beneficiaries. 16. Urban client countries were: Upper-middle income: Argentina, Brazil, Chile, Croatia, Korea, Mexico, St. Lucia, and Venezuela; Lower-middle income: Albania, Algeria, Belize, Bolivia, Bosnia and Herzegovina, China, Djibouti, Ecuador, E l Salvador, Fiji, Guatemala, Iran, Jordan, Morocco, Papua New Guinea, Philippines, Sri Lanka, Thailand, Tunisia, and Vanuatu; Low-income: CBte d’Ivoire, Ethiopia, Gambia, Ghana, Guinea, India, Indonesia, Lesotho, Madagascar, Malawi, Mali, Mozambique, Nepal, Nigeria, Pakistan, Rwanda, Senegal, Tanzania, Uganda, Ukraine, Yemen, and Zimbabwe. 17. The USP lists both output and outcome indicators. In results-basedmanagement (RBM), outputs are generally regarded as short-term results o f a project, often directly produced by a project within five years o f its completion. Outcomes, on the other hand, consist o f medium-term results, often indirectly achieved by a project five years after completion. The Canadian International Development Agency (CIDA) has done a lot o f work on RBM methodology and terminology. For more details see (CIDA 2000). 18. In detail: Algeria, Bolivia, Cameroon, Chad, Georgia, Honduras, Mauritania, Vietnam and Zambia; less so: Benin, Burkina Faso, Cambodia, Gambia, Guinea, Guinea Bissau, Kenya, Mongolia, Mozambique, Nicaragua and Niger. 19. Average 1998 per capita own revenues o f US$165 reported for 192 developing country cities by the Urban Indicators Program, multiplied by the 2.1 billion urban inhabitants o f the developing world. 20. A key Bank instrument for stimulating competitiveness over the past two years in the ECA region especially has been the Local Economic Development a process by which actors within cities and towns work collectively with public, business and non-governmental sector partners to create better conditions for economic growth and employment creation. 21, The ECA region o f transition economies, as to be expected, i s where the urban portfolio gives most attention to strengthening markets. It has been reported that prosperous urban economies help build markets in these economies (Buckley & Mini 2000). 22. I t i s important to be able to identify existing clusters. New ones cannot be created from scratch since they depend upon a pre-existing critical mass o f enterprises and skills (Schmitz 2001, p. 403). Historical chance has been an important factor in the creation o f clusters (Porter 1998b, p. 84 and Krugman 1996a p. 22). For existing clusters, the literature prescribes how they can be upgraded through: improved education, skill levels, technological capacity, opening access to capital markets, improving institutions, and market- friendly regulatory standards (Porter 2001, p. 153), actions that are on USP’s governance agenda. 23. As reported in the FinancialTimes October 19,2001, p. 13. 24. See, for instance: Duffy 1995 pp. 4-5; Scott 2001 p. 22; Schmitz 2001, p. 408; Begg 1999, p. 805; Boddy 2000; and Scott 200 1, p. 22. 51 25. Among other things, they might want to consider the following: “The attractiveness o f urban areas for the location o f industry and services stems from scale economies in production (efficient plant sizes are large), lower transport costs (reduced by clustering activities together), the modest use o f land by industry and services as an input to production (allowing higher densities), externalities among firms (sharing o f information), linkages across firms (providing intermediate inputs to each other) and potential agglomeration economies (because large clusters o f activities use specialized inputs more efficiently)” (Ingram 1998 p.1020).