improving municipal management for cities to succeed - the … · 2016-07-10 · the welfare impact...

Improving MunicipalManagement for Cities to SucceedAn IEG Special Study

Improving MunicipalManagement for Cities to SucceedAn IEG Special Study

49917P

ublic

Dis

clos

ure

Aut

horiz

edP

ublic

Dis

clos

ure

Aut

horiz

edP

ublic

Dis

clos

ure

Aut

horiz

edP

ublic

Dis

clos

ure

Aut

horiz

edP

ublic

Dis

clos

ure

Aut

horiz

edP

ublic

Dis

clos

ure

Aut

horiz

edP

ublic

Dis

clos

ure

Aut

horiz

edP

ublic

Dis

clos

ure

Aut

horiz

ed

Annual Review of Development Effectiveness 2008: Global Challenges

Addressing the Challenges of Globalization: An Independent Evaluation of the World Bank’s Approach to Global Programs

Assessing World Bank Support for Trade, 1987–2004: An IEG Evaluation

Books, Building, and Learning Outcomes: An Impact Evaluation of World Bank Support to Basic Education in Ghana

Bridging Troubled Waters: Assessing the World Bank Water Resources Strategy

China: An Evaluation of World Bank Assistance

The CGIAR at 31: An Independent Meta-Evaluation of the Consultative Group on International Agricultural Research

Debt Relief for the Poorest: An Evaluation Update of the HIPC Initiative

A Decade of Action in Transport: An Evaluation of World Bank Assistance to the Transport Sector, 1995–2005

The Development Potential of Regional Programs: An Evaluation of World Bank Support of Multicountry Operations

Development Results in Middle-Income Countries: An Evaluation of World Bank Support

Doing Business: An Independent Evaluation—Taking the Measure of the World Bank–IFC Doing Business Indicators

Engaging with Fragile States: An IEG Review of World Bank Support to Low-Income Countries Under Stress

Environmental Sustainability: An Evaluation of World Bank Group Support

Evaluating a Decade of World Bank Gender Policy: 1990–99

Evaluation of World Bank Assistance to Pacific Member Countries, 1992–2002

Extractive Industries and Sustainable Development: An Evaluation of World Bank Group Experience

Financial Sector Assessment Program: IEG Review of the Joint World Bank and IMF Initiative

From Schooling Access to Learning Outcomes: An Unfinished Agenda—An Evaluation of World Bank Support to PrimaryEducation

Hazards of Nature, Risks to Development: An IEG Evaluation of World Bank Assistance for Natural Disasters

How to Build M&E Systems to Support Better Government

IEG Review of World Bank Assistance for Financial Sector Reform

An Impact Evaluation of India’s Second and Third Andhra Pradesh Irrigation Projects: A Case of Poverty Reduction withLow Economic Returns

Improving Investment Climates: An Evaluation of World Bank Group Assistance

Improving the Lives of the Poor through Investment in Cities

Improving the World Bank’s Development Assistance: What Does Evaluation Show:

Maintaining Momentum to 2015: An Impact Evaluation of Interventions to Improve Maternal and Child Health andNutrition Outcomes in Bangladesh

New Renewable Energy: A Review of the World Bank’s Assistance

Pakistan: An Evaluation of the World Bank’s Assistance

Pension Reform and the Development of Pension Systems: An Evaluation of World Bank Assistance

Poland Country Assistance Review: Partnership in a Transition Economy

The Poverty Reduction Strategy Initiative: An Independent Evaluation of the World Bank’s Support Through 2003

The Poverty Reduction Strategy Initiative: Findings from 10 Country Case Studies of World Bank and IMF Support

Power for Development: A Review of the World Bank Group’s Experience with Private Participation in the ElectricitySector

Public Sector Reform: What Works and Why? An IEG Evaluation of World Bank Support

Small States: Making the Most of Development Assistance—A Synthesis of World Bank Findings

Social Funds: Assessing Effectiveness

Sourcebook for Evaluating Global and Regional Partnership Programs

Using Knowledge to Improve Development Effectiveness: An Evaluation of World Bank Economic and Sector Work andTechnical Assistance, 2000–2006

Using Training to Build Capacity for Development: An Evaluation of the World Bank’s Project-Based and WBI Training

Water Management in Agriculture: Ten Years of World Bank Assistance, 1994–2004

The Welfare Impact of Rural Electrification: A Reassessment of the Costs and Benefits—An IEG Impact Evaluation

World Bank Assistance to Agriculture in Sub-Saharan Africa: An IEG Review

World Bank Assistance to the Financial Sector: A Synthesis of IEG Evaluations

The World Bank in Turkey: 1993–2004—An IEG Country Assistance Evaluation

World Bank Lending for Lines of Credit: An IEG Evaluation

IEG PUBLICATIONS

All IEG evaluations are available, in whole or in part, in languages other than English. For our multilingual section, please visithttp://www.worldbank.org/ieg.

Improving Municipal

Management for Cities

to Succeed

An IEG Special Study

2009The World Bank

Washington, D.C.http://www.worldbank.org/ieg

©2009 The International Bank for Reconstruction and Development / The World Bank1818 H Street NWWashington, DC 20433Telephone: 202-473-1000Internet: www.worldbank.orgE-mail: [email protected]

All rights reserved

1 2 3 4 5 12 11 10 09

This volume is a product of the staff of the Independent Evaluation Group of the World Bank Group. The findings, interpretations, andconclusions expressed in this volume do not necessarily reflect the views of the Executive Directors of The World Bank or the governments theyrepresent. This volume does not support any general inferences beyond the scope of the evaluation, including any inferences about the WorldBank Group's past, current, or prospective overall performance.

The World Bank Group does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and otherinformation shown on any map in this work do not imply any judgement on the part of The World Bank Group concerning the legal status of anyterritory or the endorsement or acceptance of such boundaries.

Rights and PermissionsThe material in this publication is copyrighted. Copying and/or transmitting portions or all of this work without permission may be a violation ofapplicable law. The International Bank for Reconstruction and Development / The World Bank encourages dissemination of its work and willnormally grant permission to reproduce portions of the work promptly.

For permission to photocopy or reprint any part of this work, please send a request with complete information to the Copyright ClearanceCenter Inc., 222 Rosewood Drive, Danvers, MA 01923, USA; telephone: 978-750-8400; fax: 978-750-4470; Internet: www.copyright.com.

All other queries on rights and licenses, including subsidiary rights, should be addressed to the Office of the Publisher, The World Bank, 1818H Street NW, Washington, DC 20433, USA; fax: 202-522-2422; e- mail: [email protected].

Cover: Municipal service provision is imminent in Quixadá, Northeast Brazil. Photo courtesy of Roy Gilbert.

ISBN: 978-0-8213-8043-7e-ISBN-13: 978-0-8213-8044-4DOI: 10.1596/978-0-8213-8043-7

Library of Congress Cataloging-in-Publication Data

Improving municipal management for cities to succeed : an IEG special study.p. cm.

Includes bibliographical references.ISBN 978-0-8213-8043-7 (pbk.) - - ISBN 978-0-8213-8044-4 (e-book)1. City planning—Developing countries. 2. Municipal services—Developing countries. 3. Municipal government—Developing countries.

4. Municipal finance—Developing countries. 5. Economic assistance—Developing countries. I. World Bank. Independent Evaluation Group.HT169.5.I47 2009352.1609172'4—dc22

2009019206

Printed on Recycled Paper

World Bank InfoShop

E-mail: [email protected]

Telephone: 202-458-5454

Facsimile: 202-522-1500

Independent Evaluation Group

Knowledge Programs and Evaluation Capacity

Development (IEGKE)

E-mail: [email protected]

Telephone: 202-458-4497

Facsimile: 202-522-3125

v Abbreviations

vii Acknowledgments

ix Executive Summary

xv Management Comments

xvii Chairperson’s Summary: Committee on Development Effectiveness (CODE)

1 1 Managing Engines of Growth4 Evaluation of Good Municipal Management5 World Bank Policy Underpinning Support to Municipalities5 Findings of Earlier IEG Assessments6 This Study

7 2 Bank Support for Better Municipal Management9 Bank Assistance to 3,000 Municipalities10 MDP Approaches— Wholesale and Retail12 MDP Performance13 MDP Objectives— The Aims of Bank Assistance14 MDP Components— Instruments for Better Municipal Management15 Limited Attention to Poverty Reduction

17 3 Better Municipal Planning19 More Information for Planning21 M& E22 City Planning and City Development Strategies23 Investment Planning and Strategies

25 4 Stronger Municipal Finances27 Better Financial Management— Accounts and Audits 29 Mobilizing Own Revenues30 Municipal Creditworthiness and Debt Management32 Private Finance Participation

33 5 Managing Service Provision35 Elements of Service Provision38 Sectors Most Affected and Service Quality39 Income Levels of Beneficiaries— Poverty Reduction

41 6 Conclusions

45 Appendixes47 A: List of Bank-Financed MDPs, Fiscal 1998–200855 B: Banking on Municipalities: World Bank Support in Sub-Saharan

Africa

Contents

i i i

I M P R OV I N G M U N I C I PA L M A N AG E M E N T F O R C I T I E S TO S U C C E E D

i v

63 C: Banking on Municipalities: World Bank Support in East Asia and Pacific

71 D: Banking on Municipalities: World Bank Support in Europe and Central Asia

79 E: Banking on Municipalities: World Bank Support in Latin America andthe Caribbean

85 F: Banking on Municipalities: World Bank Support in the Middle East and North Africa

91 G: Banking on Municipalities: World Bank Support in South Asia97 H: Notes on Methodology of Evaluation

101 Endnotes

103 Bibliography

Box12 2.1 Defining Features of Different MDP Approaches

Figures9 2.1 MDP Trends, 1998–200810 2.2 MDPs across the Regions, 1998–2008

Tables11 2.1 Completed and Ongoing MDPs by Region, 1998–200811 2.2 Completed MDPs with More Municipal Clients Perform Better13 2.3 Completed MDPs: Performance by Region, 1998–200814 2.4 Focus of Objectives of Completed and Ongoing MDPs, 1998–200815 2.5 Focus of Components of Completed and Ongoing MDPs, 1998–200820 3.1 Summary of MDP Results in Municipal Planning28 4.1 Summary of MDP Results in Municipal Finance36 5.1 Summary of MDP Results in Service Provision

CDS City Development StrategyERR Economic rate of returnFINDETER Financiera de Desarrollo Territorial SA (Local Development Fund, Colombia)ICR Implementation Completion ReportIEG Independent Evaluation GroupM&E Monitoring and evaluationMDP Municipal development projectO&M Operations and maintenancePPAR Project Performance Assessment ReportSINIM Sistema Nacional de Informaciones Municipales (National System for Municipal

Information, Chile)TNUDF Tamil Nadu Urban Development Fund (India)WDR World Development Report

Abbreviations

v

Local residents Shaoxing, China, are keen in for municipal planning to help them relocate. Photo courtesy of Roy Gilbert.

Acknowledgments

This report was prepared by the team of RoyGilbert (task manager), Ramachandra Jammi,Kavita Mathur, and Heather Dittbrenner underthe supervision of Cheryl W. Gray (Director,Independent Evaluation Group [IEG]-WorldBank) and Monika Huppi (Manager, IEG SectorEvaluations). Peer review was by GeorgePeterson (expert in municipal governance andfinance, formerly of the Urban Institute) andEleoterio Codato (former sector manager for the

World Bank’s Urban Anchor). The study alsobenefited from valuable comments and inputs bythe following Bank staff: Patricia Annez, ArupBannerji, Alain Barbu, William Dillinger, PeterFreeman, and Christine Kessides. The study teamis grateful for all these valuable contributions, aswell as the positive collaboration of many urbansector staff and task managers during theplanning and realization of this study. The reportwas edited by William Hurlbut.

Director- General, Evaluation: Vinod ThomasDirector, Independent Evaluation Group–World Bank: Cheryl W. Gray

Manager, Independent Evaluation Group Sector Evaluations: Monika HuppiTask Manager: Roy Gilbert

v i i

Low-income housing has been upgraded with water supply and sanitation in Quixeramobim, Northeast Brazil. Photo courtesy of Roy Gilbert.

i x

Cities now host half the world’s population and provide 70 percent ofits gross domestic product, making them “engines of growth.” Managingthese economic centers well is essential for development. In nearly 3,000

municipalities worldwide, the Independent Evaluation Group (IEG) identified190 operations as municipal development projects (MDPs). These MDPs havebeen the World Bank’s principal instrument to help strengthen municipal man-agement over the past decade. Evidence from this experience can provide use-ful input to the design of future MDPs.

The best MDPs led to stronger own-revenueflows, better financial management, improvedmunicipal information systems, and localmanagement of procurement. Weaker resultswere common in monitoring and evaluation(M&E), operations and maintenance (O&M),private finance of municipal services, andpoverty focus. In these areas, MDPs can do moreand do it better. MDPs serving many municipali-ties—called wholesale MDPs—have had betteroutcomes than retail MDPs, which serve just afew, although more in-depth analysis of causalfactors is needed.

The purpose of this IEG special study is to illumi-nate the scale and scope of Bank support formunicipal development and to draw specificlessons from the achievements and failures of asample of individual projects. The findings of thestudy are based on a review of all 190 MDPscompleted or ongoing during the period1998–2008. In consultation with World Bankoperational staff, IEG identified MDPs as projectswith objectives and components focused onstrengthening municipal management in cities of12,500 inhabitants or more. Of these 190 MDPs,the 114 completed operations are the principalsource for the evaluation findings. Ninety MDPswere studied through IEG desk reviews ofImplementation Completion Reports, and 24

were the subject of detailed IEG field assess-ments, summarized in Project PerformanceAssessment Reports (henceforth called PPARMDPs).

The study focuses on three dimensions of munici-pal management—planning, finance, and serviceprovision—that figure repeatedly in Bank-financed MDPs. The planning dimension refers tothe capacity of a municipality to forecast andoversee its own progress. It includes informationsystems, M&E, city planning, and investmentstrategies. The finance dimension refers to how amunicipality manages the resources needed toprovide services to its constituents. It coversfinancial management, own-resource mobiliza-tion, access to credit, and private funding. Theservice provision dimension refers to the capacityof a municipality to manage the services requiredby city residents and business people through theeffective prioritization of investments, manage-ment of competitive procurement, and the abilityto sustain services through O&M.

Overview of Bank Support for MunicipalManagementFrom fiscal 1998 to 2008, the Bank committed $14.5billion, 3.4 percent of its total lending, to these 190MDPs. The projects have assisted nearly 3,000urban municipalities—about 15 percent of all those

Executive Summary

in developing countries, more than a third of whichare in the Latin America and the Caribbean Region.The level of MDP support to an individual munici-pality has varied enormously, from tailor-madetechnical assistance and significant investmentfunding to training just a few municipal staff. Up to345 million people—IEG’s estimate for the entirepopulation of the 3,000 participating municipali-ties—might have benefited.

The Bank has supported MDPs in all sixoperational Regions. The largest number hasbeen in Sub-Saharan Africa (27 percent of thetotal), and the largest lending commitment hasbeen in East Asia and Pacific (38 percent of thetotal). Seventy-four percent of the 114completed MDPs obtained satisfactory outcomesusing IEG criteria, compared with 77 percent forall Bank operations. The strongest Regional MDPperformers have been Latin America and theCaribbean and East Asia and Pacific, with 86 and80 percent satisfactory outcomes, respectively.

The number of municipal clients assisted by eachMDP has varied significantly. Wholesale MDPs—MDPs that serve seven or more municipalities—occupy the top 40 percent of this distribution.The average wholesale MDP covers 65 munici-palities. Wholesale MDPs have been strongperformers, with 85 percent having satisfactoryoutcomes. Retail MDPs, which serve six or fewermunicipalities, make up the bottom 60 percentof the distribution. The average retail MDP servesjust three municipalities. Only 67 percent ofthese MDPs have obtained satisfactoryoutcomes.

Although more analysis is needed, several factorsmay help explain the stronger performance ofwholesale MDPs. First, wholesale MDPs canspread the downside risk of failure broadlyacross many municipalities. Second, competitionamong municipalities, a feature of all thewholesale operations reviewed, means both thatmunicipalities that fail to meet MDP perform-ance criteria may no longer be entitled to projectsupport and that weak municipalities that do notqualify at the outset may become eligible forproject funding later if their performance

improves. Third, the study found that wholesaleMDPs allocate a significantly larger share ofproject spending to technical assistance andinstitutional development. Fourth, it is possiblethat municipality size is a factor—for example, ifwholesale MDPs deal more with smaller, less-complex municipalities, although this could notbe tested given the striking absence of popula-tion data for the municipalities they serve.

Each MDP in the portfolio of 190 has aimed tostrengthen municipal management in one ormore of its planning, finance, or serviceprovision dimensions. Surprisingly, given itspriority in the Bank’s urban strategy and theBank-supported Cities Alliance, better planninghas been an objective of just one-third of MDPs;that is, planning has received the least attentionamong the three dimensions. Finance has beenaddressed in MDP objectives more than half thetime. Service provision has featured in theobjectives of nearly all of them.

Only 27 percent of the 190 MDPs in the portfoliohave had project objectives focused on assistingthe poor or have indicated how the poor mightbenefit from stronger municipal management.Earlier IEG evaluations of urban lending foundtwice that share. The lack of MDP poverty focusis a serious shortcoming, especially given thepoverty emphasis in the Bank’s urban strategyand new estimates that put the number of poorpeople in cities at 746 million.

In-Depth Findings from ProjectAssessmentsIn addition to the broad portfolio reviewsummarized above, IEG undertook detailedfield-based assessments of the performance of24 MDPs. These assessments throw light on bothsuccessful practices and remaining challengesalong the three dimensions of planning, finance,and service delivery.

Better municipal planningThough planning is a priority in the Bank’s urbanstrategy and is widely used by municipalities formapping future city development, it was not aconsistent priority in the MDPs (17 of the 24 PPAR

x


MDPs focus on it). Six of those 24 PPAR MDPsobtained substantial or better results in enhancingmunicipal information systems, one dimension ofplanning. A notable success was the establishmentand consolidation of Chile’s Web-based NationalSystem for Municipal Information. In contrast,centralized municipal information systems in SriLanka and Mozambique failed, in part becausemunicipalities themselves had restricted access tothem. Clearly, municipal involvement in the use ofsuch systems is a factor of success.

M&E is another aspect of planning. When itworked well, which was rarely—only four of thePPAR MDPs had substantial results in this area—itwas a hands-on instrument for the day-to-daymanagement of project implementation and forevaluation. The weak performance of the majorityof the projects often reflected inadequateattention to project results themselves. Evenwhere MDP information systems were good—asthey were in Chile, China, and Indonesia—in mostcases project M&E measured only the delivery ofproject components and not the achievement ofan operation’s objectives (such as reaching thepoor in the MDP in Ceará, Brazil). M&E generallyworked better when countries used more widelyavailable municipal finance data, as in Tunisia andColombia. A very strong M&E system was builtinto the MDP in Kazan, Russian Federation, wheresome M&E performance indicators doubled astranche release conditions, enhancing the statusand importance of the M&E itself. Moreover, theKazan municipality saw the usefulness of M&E forits own planning, and not just for fulfilling a Bankproject requirement.

Relatively few MDPs attempted to strengthen cityplanning. Eight cases yielded substantial results,and two MDPs performed poorly. Among thesuccesses, retail MDPs in China helped the citiesof Ningbo and Tianjin develop city planning in away that served as a model for the whole country.Sri Lanka’s MDP enabled its capital Colombo toupdate its master plan, as Zimbabwe’s did for thesmall city of Victoria Falls. Wholesale MDPs inChile, Colombia, and Tunisia brought cityplanning to many smaller municipalities for thefirst time. Weaker results came in Indonesia,

where municipalities reacted coolly to thecomplex model of integrated planning proposedby one MDP and expressed little demand for cityplanning proposed in another. Notably absentwas the City Development Strategy, an instru-ment intensely promoted by the Cities Allianceyet rarely supported in MDPs.

Municipal (nonspatial) investment strategies madeheadway in five PPAR MDPs. Projects in Chile,China, India, Russia, and Tunisia enabled munici-palities to become more “business friendly,” andtwo MDP clients in China rose to the top of anationwide list of municipalities with the bestinvestment climate in the country.

Stronger municipal financesMost PPAR MDPs addressed the financialdimension of municipal management. Thisevaluation found more good results in thisdimension than in the planning or serviceprovision dimensions.

Half of these PPAR MDPs had substantial resultsin financial management. Good results camethrough project technical assistance and on-the-job learning that enabled many small municipali-ties in Chile, Georgia, The Gambia, India, andTanzania to adopt computerized accounting andfinancial systems for the first time. Larger munici-palities—such as Kazan, Maputo, and Tianjin—unified accounts and integrated financialmanagement across their large organizations.Among the less-successful MDPs, Georgia andUzbekistan were hindered by weak municipalcapacity before the project began.

Again, half the PPAR MDPs achieved substantialresults in enhancing revenue mobilization. Thesesuccessful MDPs updated tax records, expandedthe coverage of cadastres or land registers, andimproved collections. Municipalities receivingsuch support in Brazil and Colombia saw theirown revenues increase faster than fiscal transfers.Participating municipalities in Georgia saw signifi-cant growth of own revenues that had fallen fornonparticipants over the 2002–05 period, andown revenues of participating municipalities inThe Gambia grew 50 percent faster than expected.

E X E C UT I V E S U M M A RY

x i


x i i

Weaker results for eight MDPs in Brazil, Indonesia,Mozambique, and Zimbabwe arose from politicalreluctance by some municipalities to raise taxes.

Improvement of municipal access to credit was aninfrequent priority, with only six PPAR MDPsfocusing on it at all. Of these, five had substantialefficacy in helping to “bring municipalities tomarket.” MDPs in Colombia were particularlysuccessful in establishing a local credit market,complete with recognized credit ratings of activemunicipalities; some became able to issue munici-pal bonds for the first time. Municipalities learnedabout prudent debt management throughwholesale MDPs in Brazil, India, and—to a lesserdegree—Georgia.

Stimulating private finance of municipal serviceswas an objective in only five MDPs, and only one(in Colombia) yielded substantive results throughprivate funding of water, gas, and solid wasteservices in several municipalities. Many munici-palities lacked the expertise to staff the contractmanagement units needed to engage the privatesector. The less-successful MDPs promotedprivatization of solid waste operations in Sri Lankaand Uzbekistan; these did not go far, given poorfinancial performance and uncertain regulatoryenvironments. In Zimbabwe, funding of low-income municipal housing was not forthcomingfrom private building societies and their higher-income product lines. These weak results mighthave been averted with more accurate assess-ments of local financial markets and of thedemand for municipal services that arepotentially profitable.

Managing service provisionManagement of municipal service provision was apriority in all 24 PPAR MDPs. In prioritizing invest-ments in services, however, only seven MDPssuccessfully supported the clients’ application ofcost-benefit analysis with estimates of economicrates of return (ERRs). Simple yet robust estimatesof ERR were made for MDPs in China, Ghana,India, Indonesia, Tanzania, and Zimbabwe. Theyincluded accurate cost figures and realistic assess-ments of future benefits, often measured by thehigher value of urban land that has infrastructure

services. Good M&E systems helped producesome of the data needed for ERRs. In all cases,municipalities themselves were involved in theanalyses. Given its successful application of ERRsin cases such as these, why did MDPs use ERRestimates so infrequently? Among the reasonsgiven were high cost, lack of data, and externali-ties. But simple methods that make full use ofexisting data can help overcome these constraints.

Nine MDPs led to substantial strengthening ofprocurement management at the municipallevel; other MDPs dealt with municipalities thathad already handled their own procurement andneeded little project support. Where municipali-ties handled procurement, local beneficiarieswere better informed about the service improve-ments. Even larger municipalities, such as Kazan,Tashkent, and Tianjin, were introduced to morecomplex procurement packages, includinginternational competitive bidding, by theirrespective MDPs.

Few MDPs had substantial results in strengthen-ing the municipal management of O&M, which isnecessary to ensure ongoing service provision.The few successful cases were in Africa, whereMDPs helped computerize municipal mainte-nance in Tanzania and establish and fund munici-pal O&M accounts in The Gambia. Othersuccesses were evident in Ghana and Tunisia. Incontrast, lack of adequate O&M in MDPs led toservice failures in Georgia, Indonesia, andZimbabwe. These cases show that the risk todevelopment outcomes can increase signifi-cantly if O&M is neglected.

Only MDPs in Brazil, The Gambia, Ghana, andTanzania had objectives that explicitly addressedpoverty alleviation or service access by the poor.Visual inspections of these projects during fieldmissions confirmed that there were poor benefi-ciaries, although little data on specific povertyimpacts was available. Evidence elsewhere waseven thinner because of a lack of poverty focusand monitoring. The Bank still has much work todo to address its poverty reduction missionthrough partner municipalities. Being able todefine poverty-related objectives and measure

x i i i

E X E C UT I V E S U M M A RY

actual results of MDPs for the poor would makean important contribution.

LessonsSeveral forward-looking lessons from thefindings of this study are relevant for futureoperations and the broader municipal manage-ment agenda:

• Among the three dimensions of municipalmanagement—planning, finance, and serviceprovision—MDP support for strengtheningmunicipal finance most often yielded successfulresults, according to field assessments. TheBank should continue to support tightenedmunicipal financial management, own-revenueraising by municipalities, and municipalitiesbeing brought to local credit markets whenappropriate conditions are present.

• Project documentation that routinely reportsbasic data about each client (municipality name,population, and MDP investment) is vital to de-veloping a better understanding of the scopeof MDP results.

• Wholesale MDPs that have assisted many mu-nicipalities have yielded better outcomes thanretail MDPs over the past decade, but moreanalysis is needed to understand the precisereasons for the performance differentials. Re-tail MDPs might perform better if they incor-

porated more of the winning elements ofwholesale MDPs, such as performance-basedincentives and a focus on finance.

• More frequent use of cost-benefit or cost-ef-fectiveness analysis would help MDPs’ munic-ipal clients select the best investments andachieve outcomes efficiently. IEG found thatonly half of MDPs use such tools, with the bestcoverage in the Sub-Saharan Africa Region.

• For M&E to succeed in MDPs, it has to be use-ful and not unduly burdensome to municipal-ities themselves, and it must keep a focus onachieving results, particularly for the poor.Strong M&E can also help reduce the expenseof cost-benefit analysis by providing some of thedata needed to estimate ERRs. Few MDPs havesucceeded with this.

• Private financing of municipal services can beencouraged through better analysis of local fi-nancial markets and deeper understanding ofdemand to help municipalities gain the trustof private investors.

• Thus far, little evidence exists that strongermunicipal management has benefited thepoor. MDPs need to give much more attentionto poverty reduction in defining MDP objec-tives, showing how the poor would benefitfrom municipal investments and how serviceswould improve through stronger municipalmanagement.

A moderate-income housing development in a municipality on the periphery of Mexico City. Photo courtesy of Roy Gilbert.

More than 90 percent of world populationgrowth in the next decades will be in developingcountry cities, many of them secondary citiesand towns, whose systems are currently ill- prepared to provide services to all of theirpopulation. Strengthening management capacityis a thus a necessary condition for making citieslivable. Improved and sustainable access toservices is a key pillar for poverty reduction onthe urban agenda.

The main lessons from the study are useful forthe Urban Sector going forward: the successfulrole of municipal development projects’ supportfor strengthening municipal finance; continuedsupport for tightening of municipal financialmanagement, raising of municipal own revenues,and bringing of municipalities to local creditmarkets when appropriate; the importance ofproject documentation for measuring results;the relative success of wholesale versus retailapproaches; the need for use of cost- benefit or cost- effectiveness analysis to select the bestinvestments and achieve outcomes efficiently; aneed to strengthen monitoring and evaluationsystems; analysis of local financial markets anddemand to encourage private finance; and thepotential role that municipal developmentprojects (MDPs) can play in reaching the poor.

The study finding that less than one- third of theprojects reviewed cited poverty alleviation as aformal objective of the project is significant and

deserves further attention to understand thisbetter. In particular, it would be useful to explorethe extent to which the poverty focus in theseprojects may not be currently reflected in theformal development objectives of municipaldevelopment projects.1

Poverty reduction is at the core of the Bank’surban work and its forthcoming urban strategy.The Bank directly addresses poverty reduction incities through a variety of instruments that aredesigned to address immediate and basic needsof the poor while supporting institutional andmanagement capacity to improve and lay a solidfoundation for the sustainability of services.These include slum upgrading, or developmentpolicy loans targeting policy reforms to improveaccess to affordable housing. Typically low- income settlements are informal and thusbeyond the reach of formal service delivery. TheMDPs studied in this report focus on systemwideimprovements in planning, finance, and servicedelivery and are thus a complementary tool toensure sustainability and access to services forall, including the poor, over the longer term.

It is important to place the role of municipaldevelopment projects in context. MDPs representonly about 35 percent of Urban Developmentprojects prepared by the Urban Sector Board overthe same time period. Among the other sectorboards covered in the study, the sample includesonly 2 percent of the Environmental Sector Board

Management welcomes the Independent Evaluation Group’s (IEG)study on Bank experience in improving municipal management. TheWorld Bank recently did its own review of experience with urban in-

frastructure funds that serve smaller cities and towns, and one importantfinding was the need to understand better what works in building capacity atthe municipal level.

Management Comments

x v


x v i

projects, 7 percent of Water Sector Boardprojects, 3 percent of Transport Sector projects,and 1 percent of Public Governance projects.

It is also important to note that there may be areporting issue in capturing the poverty focus ofmunicipal development projects. The determi-nation of poverty focus in the study was basedonly on the project development objectives ofthe projects reviewed, not on the actual projectcontent or field review. Projects focusing onsystemwide improvements in accounting,planning, and tax collections are those least likelyto set specific poverty objectives as the projectdevelopment objective, because as explainedabove, impacts are more indirect and long term.Improvements in management municipal sys -tems will help the poor over the longer term asthe formal system expands its reach, but theseimpacts may extend beyond the period under evaluation.

Management looks forward to guidance from IEGregarding best practice on how clearer articula-tion of the poverty alleviation objectives and activi-ties in municipal development projects can becaptured, and on how to monitor the indirect and long- term impacts on poverty, including in smallercities and towns, which may have limited capacity.

A review of the Bank’s recent work indicates thatprojects with components specifically targetingthe urban poor are trending upward andaccounted for more than 40 percent of UrbanSector Board lending in fiscal 2008. A number ofrecent pieces of economic and sector work havealso been developed or approved in the UrbanSector, with a strong focus on urban poverty thatwill help to build the pipeline. That being said,management’s aim is to increase this further,reversing, for example, the decline in lending forslum upgrading over the previous two decades.

The release of the IEG report coincides with thelaunch of consultations on the new World BankUrban Strategy. This is an opportune time tobuild on the insights from the report as weengage with clients, development partners, andcivil society organizations, particularly in light ofthe report’s call to scale up urban services to thepoor. This is an agenda that calls for strengthen-ing our analytical base, mainstreaming urbanissues in Country Assistance Strategies and policydialogue, and expanding the Bank’s approachesfor reaching the urban poor. Scaling up programsfor delivery of services to the urban poor, innova-tive projects, and responsive instruments will allplay a role as the Bank seeks to respond in arapidly urbanizing world.

On March 16, 2009, the Informal Subcommittee of the Committee onDevelopment Effectiveness considered the Independent EvaluationGroup (IEG) special study Improving Municipal Management for Cities

to Succeed. The study covered a review of the entire portfolio of 190 munic-ipal development programs (MDPs) completed or ongoing during the 10-yearperiod 1998–2008.

Chairperson’s Summary: Committee on Development

Effectiveness (CODE)

Overall ConclusionsThe Committee welcomed the opportunity todiscuss the IEG study, taking note that almosthalf of the global population lives in cities. Thediscussion revolved around the main IEGfindings related to the poverty focus of MDPs andthe three dimensions of municipal management:planning, finance, and service provision.Members remarked that the study maycontribute to the overall update of the WorldBank’s urban sector strategy, for which theCommittee was expected to consider theconcept note in April. In this context, a memberremarked that the report could have clarified theimplications of the study findings for the Bank’surban sector strategy update. Speakers alsourged management to consider a significantfinding of the report, that the lack of MDPpoverty focus is a serious shortcoming, especiallygiven the poverty emphasis in the Bank’s urbanstrategy. Management noted that the IEG studyraised important questions that require furtherconsideration, such as addressing urban poverty,taking into account the complexity of trackingthis, and the reasons for wholesale MDPs yieldingbetter outcomes than retail MDPs.

Main Issues Raised

Poverty focusThe Committee noted management’s clarifica-tions that not all MDPs may be suitable tools foraddressing the needs of the poor (for example,municipal finance) and that other Bank urbanprojects are designed to focus on the urban poor,such as slum-upgrading projects and develop-ment policy loans that focus on policies to makehousing more affordable and to target subsidiesmore effectively. Mention was made of the CitiesAlliance global program, supported by the Bank,which has slum upgrading as a major focus.Members acknowledged the challenges oftracking urban poverty, given the shifting popula-tion and the need to consider national andmunicipal level linkages in addressing poverty.Nevertheless, speakers echoed IEG in urgingmore attention to poverty in MDP objectives,taking into consideration the distinct nature ofurban poverty. A member emphasized the needto expand economic opportunities for the poor,observing that improving services for the poor isnot sufficient to address urban poverty. Therewas a question about the decrease in poverty

x v i i


x v i i i

focus of MDPs over the years noted in the IEGstudy and a comment on the possible applica-tion of a Poverty and Social Impact Assessmentto measure the distributional impact for Bankoperations beyond policy reform programs.

Municipal servicesInterest was expressed in a broader review ofcausality between improvements in municipalplanning and finance and enhanced servicedelivery. As well, more information was soughton the extent to which Bank support led toincreased quality and access to services, includ-ing for the poor. Some members drew attentionto the importance of strengthening theoperations and maintenance focus in MDPs.Though appreciating the importance of cost-benefit and economic rates of return analyses, amember cautioned about giving emphasis tosuch analyses in crisis situations. IEG, however,stressed the importance of economic rates ofreturn, which can still be estimated at the latterpart of project implementation.

Municipal financeStrong financial management at the municipallevel was considered one of the prerequisites toenhanced service provision, but not an end initself. In response to a member’s comment,management provided assurances that althoughMDPs’ efforts have been successful in municipalfinance, the Bank will respond to countrydemand and not focus solely on this dimension.Some members noted the need to consider thecomplex interrelations with national policies(including decentralization) and their impact onmunicipal fiscal management, including revenuemobilization and expenditures, as well as politi-cal factors. A member remarked that manage-ment of foreign exchange and rollover risksshould be addressed as part of Bank support formunicipal access to credit.

Regarding private finance, there was generalagreement on the need for a good regulatorystance by municipalities that recognizes stability,effective demand, and potential for profitability.Yet a member also observed the need for moreanalysis of measures to increase tariffs forpromoting private finance. Interest was ex -pressed in the role of financial intermediariesand public-private partnerships at the municipallevel. In addition, some members commentedon innovative financing, such as the possibility ofsubsovereign lending without a sovereignguarantee and the use of performance-basedgrants. Management said that it will be holding atechnical briefing on subnational lending theweek of March 23, 2009.

Municipal planningEmphasis was put on ensuring that Bank supportis aligned with local city planning, including citydevelopment strategies (promoted by CitiesAlliance). Some members touched on theimportance of strengthening municipal institu-tions’ capacity, and in this regard a memberobserved that the IEG study could have providedadditional analysis on accountability andgovernance aspects. IEG responded that it isplanning to evaluate the implementation of theGovernance and Anticorruption Strategy in a fewyears; that evaluation can incorporate someissues encountered in the urban sector. As notedby IEG, a member underlined the need tostrengthen monitoring and evaluation.

DisseminationThe engagement of Regions in the IEG specialstudy and the strong learning element of theprocess were welcomed. There was a questionabout the presentation of the findings in otherforums in the future. IEG said that after theCommittee’s consideration of its study, thefindings would be disseminated.

Giovanni Majnoni, Chairman

Chapter 1

Evaluation Highlights• Good municipal management of

cities—which are important engines ofgrowth—is essential to development.

• Improved municipal managementhas become increasingly challeng-ing as cities grow, costs increase,and service expectations rise.

• This study reviews World Bank effortsto help strengthen three dimensions of municipal management: planning, finance, and service provision.

• This meta-evaluation assembles thefindings of existing IEG assessmentsof municipal development projectsduring the period 1998–2008.

Planning city growth and conserving historic assets side by side in Ningbo, China. Photo courtesy of Roy Gilbert.

3

Managing Engines of Growth

Cities now host more than half the world’s 6.6 billion people and pro-duce $42.4 trillion of gross domestic product—70 percent of theworld’s total. Hence, the management of these important development

centers is crucial.

Well- managed cities are “engines of growth,”offering people opportunities to build produc-tive lives, an idea articulated by the 1999/2000World Development Report (WDR) Entering the21st Century (World Bank 2000b, pp. 125–138).This idea was endorsed by the World Bank at theJune 2008 World Cities Summit in Singapore aswell as through work done for the Commissionon Growth and Development1 (Duranton 2008).It falls to local city government administrations,called municipalities in this report, to providegood management.2

Worldwide, some 31,000 municipalities— eachwith more than 12,500 inhabitants3—accommo-date the world’s urban population. Twentythousand of these are in developing countries,the client base for the World Bank.4 Each munici-pality typically manages a single city, and this isthe unitary model of municipal managementconsidered in this evaluation.5

This report reviews performance findings forthree dimensions of municipal management— planning, finance, and service provision— anddevotes a separate chapter to each. Thesedimensions accommodate the most commonobjectives of Bank operations to support munici-

pal management strengthening. Effective cityplanning can minimize spatial externalities, andrational investment planning can help allocatelimited resources efficiently to local strategicpriorities (chapter 3).

Proper financial management can help ensure thatcities have adequate revenues and that they spendthem well (chapter 4). Good preparation anddelivery of urban infrastructure and services bymunicipalities can enhance livability for residentsand productivity for businesses (chapter 5). MostBank assistance that has sought to strengthenmunicipal management has been deployed withinthis framework.

Other dimensions of municipal managementmay be important in other contexts, but they areaddressed by less than 5 percent of the projectscovered by this study and thus have not beenreviewed. They include, for example, the politi-cal dimension of strengthening local democracy,citizen participation and representation, thesecurity dimension of policing cities, the welfaredimension of a municipality’s role in providing asocial safety net, and the external relations ofmunicipalities in joining associations andforming twinning arrangements that encourage

4


the bilateral exchange of experiences betweenpairs of cities. Some of these dimensions arebeyond the mandate of the Bank. Otherdimensions, such as the environment andclimate change, housing, health, education, andculture in which municipal management isactive, have been covered by other evaluationsby the Independent Evaluation Group (IEG).

Evaluation of Good Municipal ManagementAs part of what the Bank calls capacity building,assessing the effectiveness of Bank assistance tostrengthen municipal management requireslooking for evidence that explicit municipalmanagement objectives were achieved or thatmanagerial improvements contributed tomeeting other project objectives further downthe results chain. Within the municipal manage-ment framework, this evaluation highlightsnotably successful operations worthy ofemulation and others where shortcomings pointto important lessons for improvements. Thestudy reviewed evidence available on projectsthat supported the planning, finance, and serviceprovision dimensions of municipal management.

Within the planning dimension, this evidenceincluded information about the city, its localmarkets, and the local government itself, all asinputs into planning. Also, evidence of a substan-tial monitoring and evaluation (M&E) systemwould indicate a municipality ready for planningand able to manage and oversee its ownprogress. In addition, evidence of new andupdated city plans and investment strategieswould reflect a municipality able to diagnose andmanage future development expectations.

In reviewing the municipal finance dimension, thestudy focused on examples of financial manage-ment and reporting at the local level, looking forevidence of how effectively and transparently amunicipality is able to manage its resources. Thus,evidence of strong own- revenue flows would bean important indication of municipal autonomy inservice delivery. Access to credit and privatefinance would also point to municipalities beingable to provide more services.

For the dimension of managing serviceprovision, the study searched for evidence ofmunicipal management performance in threeareas related to better access and quality ofmunicipal services. First, it examined techniquesof choosing the best performing investments forservice delivery ex ante, as well as evaluatingperformance ex post— using cost- benefit anal -ysis, for instance. Second, it used municipalability to handle procurement for serviceprovision itself as one indicator of capablemunicipal management. Third, it assumed thatlocal attention to and funding for operations andmaintenance (O&M) indicated municipal man -agement that was able to sustain serviceprovision and ensure that services remainaccessible after the project assets are in place. Allthree areas, it should be emphasized, areconcerned with the management of serviceprovision, not actual service delivery itself.

Improving municipal management across thesedimensions has become increasingly challengingfor four reasons: ever- larger cities to administer,continuous rapid urban growth, rising costs ofurban investment, and citizens’ increasing expecta-tions of the level and quality of municipal services.Large municipalities, such as the 325 in thedeveloping world that serve more than 1 millionpeople each, require complex organizations. Inthis class, the large municipality of Montevideo,Uruguay (population 1.3 million), is staffed by8,500 people, who administer an annual budget of$261 million— which is equal in complexity to alarge corporation in many countries.

Rapid urban population growth, particularly inAsia and Africa, requires that municipal manage-ment respond to demands for services andinfrastructure that far exceed the capacity ofexisting resources and systems. Rising costs ofservice provision, driven by higher land prices inparticular, can outstrip even what the most agilemunicipal financial management can provide. Asexpectations of municipal services rise, municipalmanagement has to respond. Should it fail,residents and businesses might move to moresuccessful and better- served cities. For thesereasons alone, good municipal management is

essential, and Bank support to strengthen it is important.

World Bank Policy Underpinning Supportto MunicipalitiesMunicipal development has long beenmainstreamed in development thinking at theWorld Bank. Every WDR since 1990, for instance,refers to municipalities. Most recently, the 2009WDR Reshaping Economic Geography seesmunicipalities as key players in managing cityexpansion as countries urbanize (World Bank2008). A decade ago, the 1999/2000 WDREntering the 21st Century stressed that munici-palities have to raise substantial revenues to helpprovide the urban services and infrastructurethat cities need to succeed as engines of growth(World Bank 2000b, chaps. 5–7).

The 2001 WDR Attacking Poverty posited animportant municipal role in urban land tenurereform in favor of ownership by the poor (WorldBank 2001, p. 94). The 2002 WDR BuildingInstitutions for Markets saw municipalities playingan increasingly important role in contracting outand regulating private sector provision of services(World Bank 2002, p. 160). The 2004 WDR MakingServices Work for Poor People placed municipali-ties within a “policymakers- providers- poorpeople” triad as potentially efficient (local) serviceproviders to the poor when incentives are rightand institutional responsibilities clear (World Bank2004, pp. 49, 185).

Municipal management and the planning,finance, and service provision dimensionsreviewed in this study remain at the center of theBank’s current urban strategy, Cities in Transi-tion: World Bank Urban and Local GovernmentStrategy (World Bank 2000a). The strategy seesmunicipalities as key providers of local servicesfor improving livability for the poor in cities, thefirst pillar of the strategy.

The second pillar, good governance, is to beachieved through Bank support for capacitybuilding at the municipal level, especially munici-pal development itself, and city planning. Thethird pillar, bankability, calls for the Bank to help

municipalities achieve sound finances and credit-worthiness through better municipal financial management.

The fourth pillar, competitiveness, highlights CityDevelopment Strategies (CDSs) as a tool to helpurban markets work better. Thus, in relation tothis study, the strategy’s livability speaks to thisstudy’s service provision dimension, goodgovernance and competitiveness to the planningdimension, and bankability to the municipalfinance dimension.

Findings of Earlier IEG AssessmentsIEG’s review of the implementation of thisstrategy through the Bank’s urban portfolio, inits report Improving the Lives of the Poorthrough Investment in Cities (IEG 2004), foundthat urban projects did help improve livabilitybut recommended more explicit and operationaltargets to link municipal capacity building topoverty reduction. Furthermore, that reviewurged more systematic M&E of the resultsobtained (IEG 2004, p. 31). An earlier IEG studyof Bank support for municipalities, DevelopingTowns and Cities: Lessons from Brazil and thePhilippines (IEG 1999), found that theseoperations had helped municipal reform in thesetwo countries. A lesson of the study was thatreform should be broadened beyond focusingon just one municipal instrument, such asproperty tax, to enhance own revenues.

Municipalities were a focus of IEG’s GlobalProgram Review of the Cities Alliance—part of itsevaluation of global programs that are partneredby the Bank (IEG 2008). Tightly focused onupgrading low- income settlements in cities andCDSs, Cities Alliance is a multidonor partnershipthat aims to improve the quality of developmentcooperation and urban lending and strengthenthe impact of grant- funded programs. ThisGlobal Program Review found that the programwas well focused on its two main activities butneeded to do more to strengthen M&E and thedissemination of its work. Of particular interestto the present study (as well as to the Bank’sUrban Strategy) is the Cities Alliance champi-oning of CDSs, defined as action plans for

M A N AG I N G E N G I N E S O F G R OW T H

5

6


equitable growth in cities and surrounding regions.

This StudyThis study is a meta- evaluation of the findings ofIEG project assessments. Analogous to a litera-ture review, the study assembles evaluationfindings of what worked and what did not so thatproject task managers and others can findexamples of good municipal managementpractice to emulate and replicate. They can alsosee shortcomings that need to be avoided orovercome. It does not determine if municipalmanagement in general improved followingBank interventions.

The study constituted a portfolio of 190operations,6 called municipal developmentprojects (MDPs) in this report, either completedor ongoing during the period 1998–2008. Theprimary focus of these operations is on urbanmunicipalities with populations of 12,500 ormore, so it does not include projects aimed atrural development. Appendix H, on methodol-ogy, details how the study identified MDPsthrough their “municipal” Bank activity codes andthen verified with Bank Regional staff that projectobjectives and components really focused onstrengthening municipal management.

This portfolio review laid the groundwork for theevaluation by analyzing the scope of the MDPassistance provided and the design and

implementation approaches of the 190component projects. Of these, 114 completedMDPs were the principal source for the evalua-tion findings. Ninety MDPs were studied throughIEG desk reviews of Implementation CompletionReports, and 24 were the subject of detailed IEGfield assessments, summarized in ProjectPerformance Assessment Reports (PPARs).

The findings from these 24 PPAR MDPs, as theyare called in this report, were assembled andanalyzed to obtain a more in- depth view of theperformance of MDP assistance. Although thesePPAR MDPs do not cover every single aspect ofthe Bank’s broad work with municipalities, andalthough they were not randomly selected fromthe portfolio, they do share the main characteris-tics of the portfolio as a whole, notably projectsize, implementation period, and distribution.7

In one important respect, however, they aredifferent: 83 percent of the 24 PPAR MDPs hadsatisfactory outcomes, whereas across the wholeportfolio only 74 percent rated satisfactory. Inthis report, the results of the review of theportfolio of 190 MDPs are presented in chapter 2and the Regional appendixes (B–G); findingsdrawn from the 24 PPAR MDPs make up most ofthe material in chapters 3–5. The 76 ongoingMDPs are included in this report not to assessoutcomes and results that have yet to bereported ex post, but to carry forward the reviewof MDP designs and approaches.

Chapter 2

Evaluation Highlights• The Bank committed $14.5 billion to

190 MDPs in 76 countries over theperiod 1998–2008.

• Bank- financed MDPs have assisted3,000 municipalities, 15 percent ofthe total in the developing world.

• Wholesale MDPs, each servingmany municipalities, have beenstronger performers than retailMDPs, which serve just a few.

• Across the three dimensions of mu-nicipal management, MDPs focusmost on service provision and leaston planning.

• Few MDPs focus on assisting the poor.

MDP-funded municipal water treatment plant in Pereira, Colombia. Photo courtesy of Roy Gilbert.

9

Bank Support for Better Municipal Management

Bank Assistance to 3,000 Municipalities

Since 1998, the Bank has committed $14.5 billion, 3.4 percent of its totalfunding, to 190 completed and ongoing MDPs in 76 countries worldwide.The numbers of new MDPs approved and new Bank commitments

show a growth trend in the second half of the period after an uneven but gen-erally weaker first half (figure 2.1).

The portfolio of MDPs has assisted nearly 3,000municipalities,1 about 15 percent of the 20,0002

urban municipalities in the developing world.Across municipalities, the intensity of MDPsupport may vary considerably. It can range from tailor- made technical assistance and significant

investment funding at the high end, to trainingjust a few municipal staff, often remotely, to nophysical investment at the low end. Taken as awhole, however, this support has the potential tohelp many people benefit from improved munici-pal management.

0

500

1,000

1,500

2,000

1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008

US$

mill

ions

0

5

10

15

20

25

Num

ber o

f pro

ject

s

Bank MDP commitments (US$ m) Number of projects approved

Figure 2.1: MDP Trends, 1998–2008

Source: World Bank data.Note: MDP = municipal development project.

1 0


Exactly how many, however, is rarely computedaccurately in Bank project documentation atappraisal, during implementation, or at comple-tion. Of course, an accurate picture wouldemerge if MDPs routinely reported the numberof municipalities they served and providedsummary details of each one, including thename, population, and investment received.Assuming the population size distribution of thenearly 3,000 MDP beneficiary municipalities wassimilar to that of all urban municipalities, anestimated 345 million people would live inmunicipalities assisted by Bank- financed MDPs.Although not all inhabitants would be expectedto benefit directly from the operations, a parame-ter such as this one nevertheless points to apotentially significant and extensive impact ofmunicipal management improvements wroughtby Bank- financed MDPs.

MDPs have been implemented in all Regions,with the greatest number in Sub- Saharan Africaand the largest share of commitments in EastAsia and Pacific (figure 2.2). East Asia and SouthAsia have larger- than- average MDPs; Sub- SaharanAfrica, the Middle East and North Africa, and

Europe and Central Asia have smaller ones. Theaverage Bank commitment per project is nearlythree times larger in East Asia than in Africa (table2.1). The size of individual MDPs varies consider-ably, ranging from $300 millionplus megapro-jects in large countries— China, India, Mexico,and Turkey— to small $5 million- or- lessoperations with smaller clients such asHonduras, Kosovo, and Peru (details in appendixA). Also, apart from operations in Latin Americaand the Caribbean, ongoing MDPs have largerBank commitments than completed ones,pointing to increasing Bank support for these projects.

MDP Approaches— Wholesale and RetailThe design of most MDPs is quite simple. Themajority has just two basic components: (i) institu-tional development/policy reform through techni-cal assistance and training for municipalities (andtheir higher- level government minders) and (ii)infrastructure and service provision throughfunding physical investments. Physical invest-ments accounted for more than 85 percent of thetotal project costs of most of the 24 PPAR MDPs(with Chile I and Mozambique I as exceptions).

Sub-SaharanAfrica16%

East Asiaand

Pacific39%

n = US$14.5 billionn = 190

Europeand

Central Asia12%

Latin Americaand the

Caribbean18%

Middle Eastand

North Africa6%

Middle Eastand

North Africa9%

SouthAsia6%

SouthAsia9%

Latin Americaand the

Caribbean19%

Europeand

Central Asia15%

East Asiaand

Pacific23%

Sub-SaharanAfrica28%

Share of all projects Share of Bank commitments

Figure 2.2: MDPs across the Regions, 1998–2008

Source: IEG special study.Note: MDP = municipal development project.

Physical investments are popular with municipali-ties that wish to improve service provision in theirjurisdictions, something that can affect theoutcome of local elections. Consequently, mostPPAR MDPs offered to finance such investments,but only in those municipalities that were commit-ted to reform and institutional development. Tosupport such reform directly, MDPs themselvesfunded technical assistance and training formunicipalities, which accounted for up to 15percent of the total costs of these operations.Beyond technical assistance, better management

of service provision through physical investmentsin infrastructure also helped strengthen munici-pal management.

Although most MDPs embody this basic design,the number of municipal clients assisted by anindividual MDP varies widely, ranging from just 1to 257. Across all closed projects reviewed, 60percent served 6 or fewer municipalities. Theseare referred to as retail MDPs in this study. Thoseserving seven or more municipalities are calledwholesale MDPs (table 2.2).

B A N K S U P P O R T F O R B E T T E R M U N I C I PA L M A N AG E M E N T

1 1

Latin America Middle East Sub- Saharan East Asia Europe and and the and South All

MDPs Africa and Pacific Central Asia Caribbean North Africa Asia MDPs

All (number) 52 44 28 36 18 12 190

Completed (number) 32 30 16 21 8 7 114

Ongoing (number) 20 14 12 15 10 5 76

MDP client municipalitiesa

Completed + ongoing (number) 601 445 292 1,098 379 146 2, 961

Average per project (number) 12 10 10 31 21 12 16

Average Bank commitment per project

All (US$ millions) 47 129 61 74 53 112 79

Completed (US$ millions) 42 126 37 86 49 88 76

Ongoing (US$ millions) 56 136 94 53 56 146 82Source: IEG special study.Note: MDP = municipal development project.a. Includes all municipalities served by at least one MDP.

Table 2.1: Completed and Ongoing MDPs by Region, 1998–2008

Number of municipalities Percent with per project satisfactory

Quintile Range Mean Approach Number of projects outcomes

1 1–1 1 Retail 23 69

2 2–3 2 Retail 23 61

3 4–6 5 Retail 22 69

4 7–31 15 Wholesale 23 83

5 33–257 96 Wholesale 23 87

Overall 1–257 24 All MDPs 114 74Source: IEG database. Note: MDP = municipal development project.

Table 2.2: Completed MDPs with More Municipal Clients Perform Better

1 2


The defining features of wholesale and retailMDP designs are summarized in box 2.1. Themost significant features of wholesale MDPs,found in all those reviewed by PPARs, arecompetition among municipalities and the useof a special MDP agency as an intermediarybetween the Bank and individual municipalities.Such an agency provides (that is, retails) projectsupport to many municipal clients in the form offunding for municipal subprojects. This leavesthe Bank able to concentrate its support directlyon the special agency itself.3

Retail MDPs, in contrast, do not need such anagency. Existing government departments andthe Bank itself are able to interact directly withand provide tailor- made assistance to each of thefew municipalities involved. MDPs used thewholesale approach most intensively in LatinAmerica and the Caribbean. That Region has along history of municipal administration, wherewholesale MDPs outnumber retail MDPs. Forsimilar reasons, wholesale MDPs were alsocommon in the Middle East and North AfricaRegion. East Asia hosted the largest number ofretail MDPs, using this model to assist large citiesin China, which are managed by unitary mega- municipalities.4

MDP PerformanceOverall, 74 percent of the 114 completed MDPsin this study’s portfolio achieved satisfactoryoutcomes, slightly below the Bank- wide average

of 77 percent for completed projects over thesame period. During the first half of the period,MDP performance was weaker, with only 65percent of projects achieving satisfactoryoutcomes. Performance improved in the secondhalf, when 85 percent were rated satisfactory.Against the 74 percent satisfactory rate for theMDP portfolio as a whole, 85 percent ofcompleted wholesale MDPs achieved satisfactoryoutcomes, compared with 67 percent for retailMDPs (table 2.3). This difference between theaverage performance of wholesale and retailMDPs is statistically significant.5

Though more analysis is needed, several factorsmight help explain the stronger performance ofwholesale MDPs. First, wholesale MDPs can spreadthe downside risk of failure broadly. Second,competition among municipalities in wholesaleMDPs means that municipalities that fail to meetwholesale MDP performance criteria, for instance,may no longer be entitled to participate in theproject. Conversely, initially ineligible municipali-ties that have a subsequently stronger perform-ance can be brought on board. This giveswholesale MDPs the flexibility to allow changes intheir client profile during implementation, therebystimulating competition among municipalities.

Retail MDPs can choose their few municipal clientscarefully, too, but this can only be done at theoutset, and the selection criteria are not alwaystransparent. Once chosen, retail MDP clients

Wholesale (>6 municipalities)• Many municipal clients— average of 65 per project in this

study’s MDP portfolio— competing to participate• Bank wholesales project to special MDP agency that agrees

on policy with the Bank and retails project services andfunding to municipalities; little direct Bank contact with municipalities

• Rules of engagement generally the same for all municipalities• Participating municipalities and investment subprojects not

known up front

Retail (1–6 municipalities)• Few municipal clients— average of three per project in this

study’s MDP portfolio• Project agreements made directly with municipalities; di-

rect Bank contact with municipalities• Rules of engagement crafted for each municipality and may

vary within a project• Participating municipalities and investment subprojects usu-

ally part of project design

Box 2.1: Defining Features of Different MDP Approaches

Source: IEG.

remain the same during implementation. Policyimpact leading to improved municipal manage-ment is likely to be more widespread for wholesaleoperations, given the larger number of municipali-ties affected. In addition, at 10 percent of projectcosts, wholesale MDPs spent more on technicalassistance and institutional development thanretail MDPs did; they spent only 6.4 percent.6

Other possible explanations for strongerperformance by wholesale MDPs that might havebeen expected were not supported by evidence.Thus, location factors, such as wholesale MDPsbeing located in stronger performing Regions,did not come into play, as both types of MDP arewidely represented across all Regions (table 2.3).The level of Bank financing was also not a factor,being similar for both wholesale and retailoperations. An initial review did not point tosubstantive differences between the objectivesof wholesale and retail MDPs. However, thecomplexity of the issues addressed by each typeof MDP could affect performance and deservesfurther inquiry.

Another hypothesis is that wholesale MDPsperform better because they typically work with

less- complicated, smaller municipalities. However,this can only be tested when MDPs routinelyproduce population data of the municipalitiesassisted, something that is missing from most ofthe project documentation reviewed for this study.Because the study did not examine all the possiblefactors, there is still a need for more analysis toelicit the precise reasons for the performancedifferential observed.

MDP Objectives— The Aims of Bank AssistanceAll 190 MDPs reviewed here aim to strengthenmunicipal management in one or more of thethree dimensions of this study’s evaluationframework. Because project objective formula-tions do not always use the same language as thisframework does, this study used synonymkeyword searches of the formal objectivestatements of each MDP. This was done across all190 operations to identify which dimensions ofurban management were targeted for strength-ening (details in appendix H). The results aresummarized in table 2.4.

MDPs gave the most attention to service provisionand the least to planning. Some MDP objectives


1 3

Latin America Middle East Sub- Saharan East Asia Europe and and the and South

MDPs Africa and Pacific Central Asia Caribbean North Africa Asia Total

Completed wholesale MDPs (7–257 municipalities each)

Number of projects 11 8 6 13 5 3 46

Percent satisfactory 92 75 67 92 80 67 85

Number of municipalities 416 210 54 858 59 112 1, 709

Completed retail MDPs (1–6 municipalities each)



Number of municipalities 22 46 15 9 3 3 98

All completed MDPs



Number of municipalities 438 256 69 867 62 115 1, 807Source: IEG special study. Note: MDP = municipal development project.

Table 2.3: Completed MDPs: Performance by Region, 1998–2008

1 4


embraced broader goals, such as improving theurban environment or assisting with decentraliza-tion, issues reviewed by other recent IEG evalua-tions and not dealt with in detail here.

Municipal planning, as a dimension of manage-ment, received the least attention among theaims of Bank- financed MDPs in the portfolio.Only one- third of these MDPs included betterplanning as an objective. This poor showing issurprising in the context of Bank support forpublic sector capacity building in general andBank endorsement of CDSs in particular. Asnoted earlier, promoting good CDSs amongcities worldwide is one of the two aims and linesof business of the Cities Alliance program, whichthe Bank strongly supports. Good city planningallows for city (population and economic)growth and helps provide space for environmen-tal and other public goods, as well as locationsfor services and amenities that market mech -anisms by themselves cannot. For the Bank itself,a greater focus on planning would have made theMDP portfolio more relevant to the goals ofstronger local governance espoused by theBank’s 2000 urban strategy.

Despite the limited coverage, there have beenimportant examples of MDPs improving plan -ning. These are discussed in chapter 3.

Municipal finance improvement was moreprominent, addressed by the objectives of half theportfolio’s MDPs. Among the Regions, this receivedmost attention in Sub- Saharan Africa and South

Asia. A substantially larger share of wholesaleprojects—63 percent— included finance objectivesthan retail MDPs did—46 percent (table 2.4).Examples of actual MDP results in strengtheningmunicipal finances are presented in chapter 4.

Nearly all MDPs have explicit goals of strengthen-ing the ability of municipalities to better managethe provision of good- quality and accessibleservices. Among Regions, MDPs gave mostattention to this in South Asia, East Asia andPacific, and Sub- Saharan Africa. Wholesale andretail MDPs were equally attentive to the goal ofstrengthening the management of municipalservices (table 2.4). Service provision’s leadcomes from the Bank’s traditional support for thedelivery of urban infrastructure. MDP experienceillustrates several ways of strengthening munici-pal management of service provision. MDPresults with respect to municipal management ofservice provision are illustrated through detailedreferences to cases described in chapter 5.

MDP Components— Instruments forBetter Municipal ManagementAll MDPs in the portfolio chose projectcomponents to strengthen municipal manage-ment in at least one of the planning, finance, orservice provision dimensions of this study’sevaluation framework. Table 2.5 summarizes their distribution.

Planning, although the dimension least ad -dressed by MDP components, is covered bycomponents of more than half of the operations


MDPs Africa and Pacific Central Asia Caribbean North Africa Asia Total Wholesale Retail

Total (number) 52 44 28 36 18 12 190 72 118

Percent of all MDPs with OBJECTIVES focused on—

Municipal planning 23 50 29 39 22 17 33 31 34

Municipal finance 62 48 54 50 33 58 52 63 46

Service provision 92 93 86 78 83 100 88 86 90Sources: IEG and World Bank databases.Note: MDP = municipal development project.

Table 2.4: Focus of Objectives of Completed and Ongoing MDPs, 1998–2008

in the Europe and Central Asia and Latin Americaand the Caribbean Regions (table 2.5). The kindsof instruments considered here include performance- based municipal developmentagreements, training for municipal employees,city master plan updates, land mapping and landinformation systems, and preparation of strate-gic development programs for future invest-ments. This report reviews MDP results fromusing such components in chapter 3.

Finance components are found in slightly morethan half of MDPs overall, but in higher shares inthe Europe and Central Asia and Sub- SaharanAfrica Regions, where municipal finances havebeen particularly weak (table 2.5). For one- quarter of all MDPs, these components includethe establishment of urban development fundsto specifically finance municipal infrastructureand services, the subject of a recent Bank review(Annez, Huet, and Peterson 2008). Examples ofother finance components include municipalfinancial rehabilitation plans, improved cash flowmanagement, credit to fund municipal invest-ment needs, training in budgeting and revenuegeneration, and technical assistance for munici-palities to raise revenues from domestic capitalmarkets. Their efficacy in contributing to goodproject results is reviewed in chapter 4.

Service provision components are present innearly all MDPs (table 2.5). This predominancemakes sense. After all, the ultimate purpose ofstrengthening municipal management is to makemunicipalities more effective and efficient

providers of good- quality urban infrastructureand accessible services. It is through bettermanagement of services such as these thatmunicipalities, as local public sector entities, candirectly benefit users in cities.

Examples of such components in MDPs includecomputerized management information systems,model contracts for private sector operators, and on- the- job learning. In many MDPs this last itemcomes from implementing physical investmentsin roads; drainage; lighting; water supply(treatment plants and distribution networks);sanitation; solid waste collection and disposal;social, educational, health, and cultural facilities;urban transport (including mass transit); housingand communal services; neighborhood upgrad-ing; and environmental rehabilitation. These arediscussed in chapter 5.

Limited Attention to Poverty ReductionRelatively few MDPs—27 percent of the total portfolio— have objectives focused on assistingthe poor,7 such as aiming to improve their livingconditions through service provision to low- income areas. MDPs that omit poverty referencesin their objectives typically aim to improveservice delivery, too, but without specifying thepoor among the beneficiaries.

The sparse MDP attention to the poor is surpris-ing in view of the urban strategy focus on povertyand the Bank’s own top priority of reducing it.Municipalities have an important role in helping low- income beneficiaries, the starting point of


1 5


MDPs Africa and Pacific Central Asia Caribbean North Africa Asia Total Wholesale Retail

Total (number) 52 44 28 36 18 12 190 72 118

Percent of all MDPs with COMPONENTS focused on—

Municipal planning 48 48 61 50 33 25 47 38 53

Municipal finance 71 43 82 44 61 50 59 59 59

Service provision 99 95 86 92 89 100 94 92 95Sources: IEG and World Bank databases.Note: MDP = municipal development project.

Table 2.5: Focus of Components of Completed and Ongoing MDPs, 1998–2008

1 6


the 2004 WDR Making Services work for PoorPeople (World Bank 2004, pp. i, 75). The absolutenumber of urban poor, recently estimated atsome 756 million worldwide (Ravallion, Chen,and Sangraula 2007), is expected to remain high,

particularly if the faster urbanization of the poorthan of the population as a whole persists (WorldBank 2004). Clearly, MDPs need to strengthentheir poverty focus to ensure that they benefitthe poor more effectively.

Chapter 3

Evaluation Highlights• Only half the 190 MDPs in the port-

folio focused on planning; the sharewas higher among the 24 MDPs re-viewed by PPARs, where 7 projectsobtained good planning results.

• Field work showed that developingmunicipal information systems forplanning has had mixed results, withsuccess coming from more munici-pal involvement.

• M&E results of evaluated projectscommonly counted the delivery ofproject components rather thanmonitoring the achievement of outcomes.

• MDP coverage of city planning wasthin, pointing to the need for morework in this area.

“One-stop shop” for community participation in city planning and service provision in Novgorod, Russia. Photo courtesy of Roy Gilbert.

1 9

Better Municipal Planning

Planning is a priority in the Bank’s Urban Strategy and a tool widely usedby municipalities for mapping future city development. For instance, plan-ning applies to strategic spatial plans and land use. It is surprising,

therefore, that only half of the 190 MDPs in the portfolio focus on city plan-ning of any kind (tables 2.4 and 2.5).

Among the 24 MDPs reviewed by IEG’s PPARs,attention to the planning dimension of municipalmanagement was somewhat higher, with 17projects focused on planning through theirobjectives or components (table 3.1). This chapterassesses the effectiveness of MDP support withineach of four broad categories of planning: informa-tion management, M&E, city planning, and invest-ment strategies. The chapter highlights successfulprojects, as well as examples of performanceshortcomings, to inform MDP practitioners ofwhat has worked well and what has not.

More Information for PlanningGood information about a municipality and theeconomic, social, and financial challenges andpotential of its city are indispensable to soundplanning. Good information feeds directly intoM&E, enabling municipal authorities to knowmore about the problems they confront andprogress made in solving them. Altogether, 14PPAR MDPs specifically attempted to improveinformation systems. Of these, six obtainedsubstantial or better results and eight were lesssuccessful. The remainder of the PPAR MDPs didnot try to strengthen these systems (table 3.1).1

Perhaps the most significant MDP informationsuccess has been in Chile, where the centralgovernment’s National Information System on

Municipalities (SINIM, http://www.sinim.cl/) waslaunched (Chile I) and consolidated by systemimprovements (Chile II). Both operations werewholesale MDPs involving a strong higher- levelagency, the Regional Development Sub- Secretariat. At this writing, SINIM holds annualdata for the 2000–07 period on local finances,administration, education, and health services, aswell as social and geographical indicators—250variables altogether for every one of Chile’s 345municipalities. Being Web based and in the publicdomain, SINIM allows policy makers and citizensto know what their own— or any other— municipality has been doing and how well.Municipalities themselves upload the data intothe database. Sustainability has been good.

The information system, created more than eightyears ago, is still going strong without furtherBank assistance. Although SINIM data areavailable to anybody with access to the Internet,municipal information systems in othercountries have often been less accessible. Forexample, centralized information systemsintroduced by the Sri Lanka MDP in Colombo(population 2.3 million) and Mozambique II inMaputo (population 1.2 million) and Nampula(population 380,000) remained largely beyondthe control of or even access by municipalitiesthemselves and thus fell out of use. The clients of

2 0


the technical suppliers of such systems werecentral governments, meaning that the suppliersthemselves rarely responded to demands frommunicipalities. These shortcomings clearly pointto the need to involve municipalities actively up front.

A few MDPs also developed information systemsat the level of individual municipalities. Througha technical assistance contract under IndonesiaIX, a project that focused on municipal innova-tion, the municipality of Bogor (population769,000) developed a lively and informative Web

Overall Results in—outcome City Investment

Country MDP Project name of project Information M&E planning strategies

Sub- Saharan Africa

Gambia, The Pov. Alleviation & Capacity Building M Sat — — — —

Ghana I Second Urban Sat — — — —

Mozambique I Local Govt. Reform Unsat * — — —

Tanzania I Urban Sector Rehabilitation Sat — — — —

Zimbabwe Urban Sector & Regional Dev. M Sat * * √ —

East Asia and Pacific

China III Tianjin Urban Development Sat √ √ √ —

China IV Zhejiang Multicities Dev. Sat √ * √ √China VII Shanghai Environment Sat √ * — —

Indonesia II East Java/Bali Urban Dev. M Sat * * * —

Indonesia VI Second East Java Urban Dev. M Unsat * * — —

Indonesia IX Municipal Innovations Sat √ * * —

Europe and Central Asia

Georgia I Mun. Infrastructure Rehab. M Unsat — — — —

Georgia II Mun. Dev. & Decentralization M Sat — — — —

Georgia III Second Mun. Dev. & Decentral. Sat * — —

Russian Federation IV Kazan Municipal Dev. H Sat √ √ √ √Uzbekistan Tashkent Solid Waste Mgt. Sat — * — —

Latin America and the Caribbean

Brazil II Ceará Urban Dev/Water Res. M Sat * * — —

Chile II Second Municipal Dev. M Sat √ * √ √Colombia I Municipal Dev. Sat — √ √ —

Colombia IV Urban Infras. Services Dev. Sat — — — —

Middle East and North Africa

Tunisia I Municipal Sector Investment Sat * √ √ √South Asia

India I Tamil Nadu Urban Dev. M Sat — — — —

India II Tamil Nadu Second Urban Dev. Sat * * — √Sri Lanka Colombo Env Improvement Unsat * * √ —Sources: IEG PPARs.Note: Bold = MDPs focused on planning; √ = substantial or higher achievement of element; * = element tackled but with modest or lower achievement; — = element not attempted.(Roman numerals are attached to MDPs in their sequence in the portfolio of 190 operations, being dispensed with altogether when there is only one MDP in a particular country.) M&E= monitoring and evaluation; MDP = municipal development project. Ratings: H Sat = highly satisfactory; Sat = satisfactory; M Sat = moderately satisfactory; M Unsat = moderately un-satisfactory; Unsat = unsatisfactory.

Table 3.1: Summary of MDP Results in Municipal Planning

site to interact with and inform the public;basically, this was an online version of the munici-pality’s earlier and successful public informationbooths. IEG found it still functioning well fiveyears after completion.

In another innovation under the same project,the municipality of Surabaya (population 2.4million) set up a hotline for citizens’ inquiries orcomplaints about municipal services. Unfortu-nately, it had fallen out of use for lack of fundingto maintain it. In contrast, international consult-ing contracts under China III helped themegacity of Tianjin develop a computerizedtraffic information system. Set up more than 10years ago, the system is still operating withoutfurther assistance from the Bank.

Staying engaged for the long haul can helpdevelop municipal information systems even inthe most challenging circumstances. Thus,elements of a basic municipal finance informa-tion system are finally coming together afterthree successive MDPs in Georgia over 14 years. One- off efforts to improve information, as theBank tried through Zimbabwe’s MDP, had thinresults even before the country’s current crisis.

M& EWhen it worked well in MDPs, which was rarely,2

M&E was a hands- on instrument for the day- to- day management of project implementation andfor conducting evaluations. Of the 24 PPAR MDPsreviewed, only 4 obtained substantial results withM&E (table 3.1). The performance of the remain-der was weak, often reflecting inadequateattention to project results. Before it helps mapthe direction and paths a project should take,M&E begins as a planning tool; that is the reasonit is considered in this chapter. For M&E tosucceed, municipalities have to regard it as auseful tool for themselves, not just an instrumentfor government control or academic research byothers. This suggests that there should be moreinteraction between municipalities, govern-ments, and the Bank at the design phase of M& E.

A common weakness of M&E design for MDPswas excessive focus on monitoring the delivery

of a project’s components instead of measuringthe actual achievement of intended outcomes.Even an MDP that otherwise excelled in informa-tion management—M&E in Chile II— measuredonly the number of technical assistance assign-ments completed and their cost, but not indica-tors of municipal governance improvements orgreater efficiency in service provision, which waswhat the project intended.

M&E for Indonesia II and VI also ventured littlebeyond counting the number of subprojectcontracts awarded and the amount of disburse-ment, so M&E in those projects was able toprovide the exact number of community toiletsbuilt and their precise unit costs, for instance,but not how much they were used, which provedto be very little. During field visits, IEG sawcommunal toilet blocks designed for 15 familiesbeing used only by 1 or 2. Though China IV inthe Zhejiang province excelled in many otherrespects, its M&E did little more than count andcost the delivery of individual subprojects.

Another shortcoming often found in the M&E ofMDPs was the lack of baseline data and theabsence of explicit, preferably quantified targets.Brazil II, for instance, did not describe the livingconditions of the poor at the outset in munici-palities in the state of Ceará that were targetedfor project improvements. Endline conditionsmay be easy for an evaluator to observe atcompletion, but how much progress theyrepresent from the starting point can only beunderstood if there are also baseline data toshow clearly how far the endline is from it.

Even when a baseline is known, it must beexpressed in terms and units that make measure-ments of changes over time meaningful. Highinflation, for instance, can cloud the interpreta-tion of financial and economic indicatorsexpressed in current values, as they were in theM&E of Uzbekistan’s MDP. Sometimes, evengood quality and quantified indicators, such asthose measuring water quality of the environ-mentally stressed Huangpo River in Shanghaiunder China VII, are of limited use whenbaseline values are not comparable. In this case,

B E T T E R M U N I C I PA L P L A N N I N G

2 1

2 2


baseline and endline water samples were drawnfrom different parts of the river. Water qualitymeasurements that were taken from the pollutedBeira Lake in the city of Colombo under SriLanka’s MDP can reveal its condition today, butnot the reduction of pollution— because thereare no baseline data.

Experience shows that an initially weak M&Edesign can be enhanced during projectimplementation and later used effectively. Thishappened under China III, when the stronglocal planning team in Tianjin municipality, at itsown initiative, incorporated outcome indicatorsto measure municipal management effectivenessthat had been overlooked by the initial M&Edesign. Georgia I incorporated theoreticallysound indicators in its M&E design, such as thenumber of days lost through schools and healthcenters having to close for bad weather, but therewere no unit- level data available for them in thecountry at the time. M&E under Georgia II andIII progressed to incorporate municipal financeindicators, for which data became available aftermore than 10 years of consistent effort.

M&E of MDPs was strongest where its focusincluded municipal finances. This was the casewith Colombia I and Tunisia I, both of whichbuilt up extensive knowledge about municipalmanagement in their countries. Russia IV, aretail operation, also incorporated a strong M&Esystem, albeit for one city, Kazan. But its designincluded clear and easy- to- measure indicatorsthat addressed all project objectives. Theseincluded a declining municipal budget deficit,increasing targeted cash social assistance toeligible beneficiaries, and greater provision ofhousing and local infrastructure services by theprivate sector. This M&E was easy to implementand use, as the municipality of Kazan itselfwanted to know these results and had thecapability to collect and analyze all the necessary data.

This may not be the case in weaker municipalities.In Kazan, however, it is still in full use, two yearsafter project completion. Part of the M&E successin Kazan comes from the programmatic structural

adjustment loan design3 of the operation, whichrequired the MDP to meet specific outcome targets— some as tranche release conditions— over a period of just three years.

Kazan was willing and even enthusiastic aboutadopting M&E because the municipality foundthat the information generated was useful for itsown financial management purposes, not just formeeting the formal requirements of the project.During PPAR missions to review other MDPs,however, IEG heard several municipal officialssay that M&E information was being collected,often at significant expense to their municipali-ties, just to please the World Bank. It is thereforeimportant for project designers to ensure thatthe M&E will be useful for the municipalitiesthemselves and that it still fulfills the require-ments of due diligence and project evaluation.This can best be done by ensuring municipalinvolvement in M&E design from the outset.

City Planning and City Development StrategiesAlthough city planning has long been atraditional function of municipal administrationsworldwide, IEG found relatively few MDPsfocused on it. Of the 24 MDPs reviewed by PPARs,only 10 included a focus on city planning; mostof these achieved substantial results. The remain-ing MDPs achieved little in strengthening suchplanning, not because they tried and failed, butbecause city planning was not on their agenda(table 3.1).

Among the successes, China IV in ZhejiangProvince helped the municipality of Ningbo(population 720,000) update its city master plan;that plan enhanced long- term land use planningand firmly embedded the conservation ofhistoric and cultural assets into its city centerplanning. China III also strengthened cityplanning in Tianjin (population 11.2 million).Several official delegations of foreign plannershave visited the megacity to learn more about itssuccessful approach to planning, especially inupgrading transport corridors and providing forsolid waste and sewage disposal. As retail MDPs,these experiences point to how this approach

can bring substantial improvements in planningto a few cities.

Chile II provided support for 25 municipalitiesto prepare local city plans. MDPs also helpedmunicipalities update existing master plans, asthey did in Tianjin, China, and Colombo, SriLanka, or prepare one for the first time, as in thesmall town of Victoria Falls (population 36,000)through the Zimbabwe MDP. Indonesia IIsupported Integrated Urban InfrastructureDevelopment Planning, which the Bank andother donors hoped to establish as a modern-ized form of urban planning. But the approachwas not adopted successfully by smaller munici-palities in particular; they preferred to continuewith the traditional sectoral approach they knew.With its municipal demand- driven design,Indonesia IX also had little impact on strength-ening city planning, given the small demand forit among the project’s municipal clients.

Although CDSs are supported by the Bank andother donors worldwide and are one importantbusiness line for the Bank- supported CitiesAlliance,4 IEG assessments found that MDPs gaveCDSs very little attention. To strengthen cityplanning more, programming CDS work couldbe better coordinated with MDPs, and CDSsthemselves could be integrated into MDPoperations. CDSs tend to be quite diverse, and itwould be helpful if the Cities Alliance couldadvise the Bank, MDP agencies, and municipali-ties which model of CDS would be best suited toa particular MDP. The current absence of CDSswould appear to call for greater synergy on thisfront. In addition to their strengths as forward-

looking management tools, CDSs and cityplanning of all kinds do help municipalities knowthe baselines of the spatial configuration ofurban development better.

Investment Planning and StrategiesMunicipalities need to determine priorities fortheir investments, given the scarce resourcesavailable; this can be called investment planningor strategizing. Most MDPs of the 24 reviewed byPPARs have not systematically tried to do this.There were five projects that did so, and theyachieved substantial results.

India II in Tamil Nadu, for example, providedtechnical assistance to and facilitated exchangesamong 45 municipalities to help each oneprepare corporate plans. These plans set outtheir investment priorities for 10 years. UnderChile II, some smaller municipalities est -ablished Local Development Directorates tocoordinate municipal investment planningacross sectors. Exchanges of experiences amongmunicipalities in the Indian and Chilean casesprobably mean that the dissemination of MDPexperience is much wider than is obvious fromthe direct project results themselves.

The experience of China IV in Zhejiang Provincehelped make municipalities more businessfriendly. The cities of Hangzho (population 1.9million), Ningbo, and Shaoxing (population421,000), with planning and service provisiondirectly assisted by the project, were among thosewith the best investment climate in the country,according to a 2005 survey of 14,000 firms in 120cities throughout China (World Bank 2006).

B E T T E R M U N I C I PA L P L A N N I N G

2 3

Chapter 4

Evaluation Highlights• Across the three dimensions of mu-

nicipal management strengthening,MDP efforts in finance yielded suc-cessful results most often.

• Some small municipalities benefitedby adopting computerized account-ing systems for the first time.

• Some large municipalities consoli-dated and unified their accounts.

• Half of MDPs obtained substantialresults in mobilizing municipal revenues.

• Field work showed that the results ofprivate funding of municipal serviceswere thin and efforts inadequately prepared.

Oversight of municipal finances in Fortaleza, Northeast Brazil. Photo courtesy of Roy Gilbert.

2 7

Stronger Municipal Finances

The majority of MDPs addressed the financial dimension of municipal man-agement through their objectives, their components, or both (tables2.4 and 2.5). Among the 24 PPAR MDPs, there was a strong focus on fi-

nance. Altogether, 20 of the MDPs had objectives and components related tomunicipal finance (table 4.1). This study compiled PPAR findings of the per-formance of these projects within each of four categories: financial manage-ment, mobilizing revenues, access to credit, and private finance.

Better Financial Management— Accounts and Audits Financial management was the most frequentlysupported aspect among the projects reviewed,being addressed by 15 PPAR MDPs. Of these, theevaluation identified 11 that obtained substantialresults, and 4 had a weak performance. Theremaining nine did not address the issue.

Assistance with computerizing municipalaccounts was the tool most often used toimprove municipalities’ financial management.Smaller and more remote municipalities in many countries— Chile, The Gambia, Georgia,Tanzania, and Tamil Nadu State (India), for instance— adopted computerized accountingthanks to technical assistance and trainingpackages offered by their respective MDPs.Software and hardware packages were generallyprovided by local commercial firms that werefamiliar with local conditions and able to providethe service in the local language. In most cases,municipalities were able to meet nationalstandards of municipal financial reporting— requiring municipal financial information to beavailable in real time— promoted by theircountry’s Ministry of Finance.

Where training was more intense— for instance,with more than 35 courses in double- entryaccounting under India II in Tamil Nadu and withplenty of places for municipal applicants availablethrough The Gambia MDP— account modern-ization was successful. Through technicalassistance at the central and local levels, TanzaniaI ensured that all municipalities participating inthe project had— and still have today— up- to- dateand audited municipal accounts. This was not thecase before the project, or for most other munici-palities still, according to the government.

Some very large municipalities took well tomodernizing their accounting through MDPs. Forinstance, under China III, the mega- municipalityof Tianjin was able to integrate different financialinformation systems, using the project- supportedexpansion of its computer network. Russia IVhelped Kazan make municipal accounts moretransparent and strengthened municipal manage-ment. It did this by helping unify municipalaccounts, introducing standard financial indica-tors, and making projections to establish perform-ance targets. This was achieved not throughstandard contracts with consultant providers oftechnical assistance, but through intensive

2 8


interchanges between municipal officials andBank staff and consultants during the frequentBank project supervision missions to the city.

The municipality of Maputo in Mozambique wasable to unify its accounts in a similar way, thanksto later operations, Mozambique II and III.

Progress in systemizing municipal accounts wasslower under Georgia II, where computing skillswere in short supply among the weak municipaladministrations; this shortcoming was slowlyovercome by the follow- on Georgia III. Underthe Uzbekistan MDP, progress was slow inmodernizing the accounts and financial informa-

Overall Results in—outcome Financial Mobilizing Access to Private

Country MDP Project name of project management revenues credit finance

Sub-Saharan Africa

Gambia, The Poverty Allev. & Cap. Building M Sat √ √ — —

Ghana I Second Urban Sat √ * — —

Mozambique I Local Govt. Reform Unsat √ * — —

Tanzania I Urban Sector Rehabilitation Sat √ √ — —

Zimbabwe Urban Sector & Regional Dev. M Sat — * — *


China III Tianjin Urban Development Sat √ — — —

China IV Zhejiang Multicities Dev. Sat — — — —

China VII Shanghai Environment Sat — — — —

Indonesia II East Java/Bali Urban Dev. M Sat √ √ — —

Indonesia VI Second East Java Urban Dev. M Unsat √ * — —

Indonesia IX Municipal Innovations Sat * * — *


Georgia I Municipal Infrastructure Rehab. M Unsat — — — —

Georgia II Mun. Dev. & Decentralization M Sat * √ * —

Georgia III Second Mun. Dev. & Decentral. Sat √ √ √ —

Russian Federation IV Kazan Municipal Dev. H Sat √ √ — *

Uzbekistan Tashkent Solid Waste Mgt. Sat * * — *


Brazil II Ceará Urban Dev./Water Res. M Sat — * √ —

Chile II Second Municipal Dev. M Sat — * — —

Colombia I Municipal Dev. Sat — √ √ *

Colombia IV Urban Infras Services Dev. Sat — √ √ √Middle East and North Africa

Tunisia I Municipal Sector Investment Sat √ √ √ —

South Asia

India I Tamil Nadu Urban Dev. M Sat * √ * *

India II Tamil Nadu Second Urban Dev. Sat √ √ √ —

Sri Lanka Colombo Env. Improvement Unsat — — — —Sources: IEG PPARs.Note: Bold = MDPs focused on finance; √ = substantial or higher achievement of element; * = element tackled but with modest or lower achievement; — = element not attempted. (Romannumerals are attached to MDPs in their sequence in the portfolio of 190 operations, dispensed with altogether when there is only one MDP in a particular country.) MDP = municipal de-velopment project. Ratings: H Sat = highly satisfactory; Sat = satisfactory; M Sat = moderately satisfactory; M Unsat = moderately unsatisfactory; Unsat = unsatisfactory.

Table 4.1: Summary of MDP Results in Municipal Finance

tion systems of a financially stressed municipalsolid waste utility that had little experience withautonomous financial management. Theseresults show that a minimal existing capacity is aprecondition for MDPs to obtain substantialresults in strengthening financial management.

Mobilizing Own RevenuesMunicipalities’ own revenues, levied throughlocal taxes, user fees, and charges, have beenobserved to account for up to one- half of allmunicipal revenues (Shah 2006). Most of theremainder comes from transfers from higher- level governments.1 Because such fiscal transfersare usually beyond the control of a municipallevel recipient, the Bank generally supports theirreform through Development Policy Lending.MDP reform efforts instead have generallyfocused on strengthening own- revenue mobiliza-tion, which is more amenable to improvementthrough MDP project assistance (table 4.1).

Eleven of the 24 PPAR MDPs achieved substantialresults in revenue mobilization, eight MDPefforts yielded weaker results, and five did notfocus on this. Successful MDPs updated taxrecords, expanded the coverage of cadastres orland registers, and enhanced collections.2

Important factors contributing to the positiveresults were MDP incentives that requiredmunicipalities to raise revenues to remaineligible for project investment funding.

Factors leading to weaker results includedrevenues from other sources readily available tomunicipalities and lack of municipal control overparameters of local tax rates and collections.MDP technical assistance to the municipalities(with tight oversight by capable higher- levelauthorities for wholesale MDPs) helped themunicipalities both update their tax rolls andmonitor and follow up with delinquent taxpay-ers. Technical assistance worked well in thesecases when it was a condition of municipal accessto MDP credit for popular infrastructure.

Municipalities participating in Colombia I, forinstance, saw their own revenues increase morerapidly than revenues from fiscal transfers over

the 1991–2001 life of the project. More recentreviews of the follow- on projects underscore thatthe strong revenue performance is continuing.Large individual municipal clients of India I,such as Madurai (population 909,908) andTiruchirapalli (population 775,484), significantlyincreased their own revenues, enabling them tofinance more infrastructure investments fromtheir own resources.

Pointing to positive effects of project incentives andtechnical assistance, Georgia II participant munici-palities’ own revenues grew by 11 percent during1998–2002, but those of nonparticipants fell by 16percent. Under Georgia III, own revenues of the11 participant municipalities continued to grow by58 percent from 2002 to 2005; nonparticipants sawtheir revenues rise by only 35 percent.

Although such large figures may be the result ofincreases from a very low base, more importantfor an evaluation is the assessment of the differ-ences between the performance of municipalitiesthat participated in an MDP and the performanceof those that did not. Although details are scarce,Indonesia II did report enhanced revenuecollection among the 45 municipalities assisted bythe project in East Java and Bali. The Gambia’sMDP also reported good results, as own revenuesof participating municipalities grew by 9 percentannually, well above the 5–6 percent target theMDP set for them. Tanzania I enabled municipal-ities to more than double their own revenues overthe project period with the help of moderncomputer mapping and accounting.

Through Russia IV, Kazan municipality wasparticularly successful in increasing its ownrevenues. It had a particularly important andurgent reason to do so. Following the establish-ment of the new autonomous Kazan municipalauthority in 2004, existing large federal transferswere slated to be cut and had to be replacedquickly. The effect was similar to that of a fiscalshock, although not unexpected. The MDPhelped Kazan municipality find alternativesources of revenue, a task it did well, takingmunicipal finances into surplus by mid-2007 aftera succession of deficits.

S T R O N G E R M U N I C I PA L F I N A N C E S

2 9

3 0


Tunisia I produced excellent results as well thatcontinue, more than eight years after implemen-tation completion. Not only did MDP- parti -cipating municipalities there increase their ownrevenues more than other municipalities, but theparticipants produced surpluses that were twicethe size targeted. These results show thatimportant achievements in financial perform-ance can be obtained both through wholesaleand retail models of MDPs.

In two MDPs— Ghana I and Chile II— municipalclients increased their own revenues, but thegrowth could not be attributed to the projects.Other municipalities that did not participate in theprojects enjoyed similar increases, caused bybuoyant macroeconomic growth exogenous tothe operations. In such cases, it is important notto mistakenly read project success into the results.Indeed, questions still need to be asked aboutwhat more the MDPs could have done to helpclient municipalities achieve a better revenueperformance than nonparticipants. In the cases ofBrazil II, Indonesia VII, Mozambique I, andthe Zimbabwe MDP, results in raising more ownrevenues were weak where some municipalitieswere reluctant to raise taxes further.

Instead of looking broadly at improving munici-pal revenues overall, the Uzbekistan MDPconcentrated on stimulating direct cost recoveryfrom the project investments themselves; theresults were only modest. In this case, themunicipality of Tashkent allowed necessaryannual adjustments to the solid waste tariff tolapse for four years, undermining the sustainabil-ity of the service; a last- minute adjustment didthrow a financial lifeline to the system operator.In some other countries, notably China and SriLanka, increasing local revenues, whetherthrough MDPs or by other means, simply was nota priority for the national government.

Municipal Creditworthiness and Debt ManagementThrough offering lines of credit for financinginvestments in infrastructure and municipalservices, some wholesale MDPs were able tointroduce many municipalities to borrowing on a

significant scale. The metaphor of “bringingmunicipalities to market” through direct borrow-ing or issuing bonds appeals to the rigor suchcommitments impose on financial management,as municipalities service their debts and seek toremain creditworthy.3 Exposure to credit and itsaccompanying opportunities and risks forged anew approach to municipal finance in some cities.

Six of the eight MDPs reviewed by PPARs thatwere focused on strengthened municipal accessto credit yielded substantial results. An importantfactor in these positive outcomes was the MDPitself making resources available to municipali-ties that became creditworthy. One other MDPtried but failed to bring municipalities to marketbecause the availability of other grant fundingundermined their demand for credit. Theremaining 16 MDPs set no target for themselvesin this area (table 4.1). Retail MDPs had too fewmunicipal clients to start credit markets, whichgenerally require a large number of municipali-ties to be functional.

On the positive side, several small municipalitieslearned about servicing and paying off debt for thefirst time under Brazil II in Ceará State.Colombia I and IV, successive wholesale MDPs,both helped bring municipalities to market asintended by introducing 179 of them to creditoperations for the first time. These municipalitiesremain engaged today. Under these projectsColombia’s national Local Development Fund,FINDETER, currently with a AAA Fitch credit rating,stimulated the supply of private credit to munici-palities. It did this by rediscounting commercialbank loans to them for financing infrastructure andmunicipal services; that enabled the successfulconsolidation of a new local credit market. Further-more, the operations fostered a municipal cultureof creditworthiness that was publicly monitored byinternational credit- rating agencies that set uplocal operations in the country.

Eight municipalities were awarded a BBB orhigher credit rating, which in Colombia isregarded as being equivalent to investmentgrade. Through Colombia I, the municipality ofPereira (population 440,000), with an A+ Fitch

credit rating, went one step further: in 1998 itsuccessfully issued municipal bonds that wereheavily oversubscribed in the local market. Thisrich experience also revealed possible marketconstraints on the demand side, however. Evenwhen creditworthy, several municipalities inColombia were reluctant to assume debt formunicipal services. Their leaders were cautiousabout borrowing and their constituents seem -ingly satisfied with the existing levels of localservices provided.

Local credit markets in other countries have beenstimulated by other Bank- financed MDPs. India IIencouraged municipalities in Tamil Nadu tobecome creditworthy to have better access to loansawarded by the Tamil Nadu Urban DevelopmentFund. The state’s second largest city, Madurai, wentone credit step further. With the project’s technicalassistance, the municipality issued bonds to raisefunds to pay for the construction of an inner ringroad. But these successes were built on the back ofshortcomings of the earlier India I, which madelittle progress with municipal credit. Concessional state- level grant financing continued to be themajor source of revenue for municipalities, faroutstripping resources coming through the MDP’sline of credit.

Some progress, albeit modest and very protracted,has been made toward establishing the buildingblocks of municipal credit systems in Georgia.Georgia III helped make the country’s ninelargest municipalities creditworthy— at least in theeyes of the official Municipal Development Fund ofGeorgia. This was the first time this had beenaccomplished in that country’s history and was anecessary step toward a financial market recogni-tion of municipal creditworthiness. In bothcountries—Georgia and India—several municipal-ities took out loans for the first time to financeservices and infrastructure, learning quickly howto properly manage a debt portfolio.

Downside foreign exchange risks, especiallyvolatile at the end of 2008, have rarely beentreated openly in the designs of MDPs. Nor havethey been found to constrain municipal credit.But the consequences of sharp local currency

devaluations in several countries have laid barethe question of who should be responsible forthe additional local currency payment needed toamortize a foreign currency- denominated debt.Standard MDP practice has been for theborrower, usually the central governmentfinancial intermediary, to take this risk, part ofwhich it may be able to price into the creditoffered to municipalities. Municipalities, after all,are rarely foreign currency earners.

Under Colombia I and II, FINDETER itself wasable manage this risk professionally throughforward pricing of foreign currencies and also bymaintaining foreign account holdings, ensuringthat its own exposure did not exceed more than12.7 percent of its total liabilities. Meanwhile,Colombian municipalities were shielded fromthat risk. India II took a similar approach,expecting the state- level Tamil Nadu UrbanDevelopment Fund to manage the foreignexchange risks. It moved more cautiously,however, given that fewer hedging instrumentswere available in that country’s financial markets.

One MDP also provided temporary relief to amunicipality that was hard pressed by short- termdebt that an earlier municipal administration hadleft unpaid. Colombia IV, through advice onportfolio management, helped the large munici-pality of Barranquilla (population 1.4 million)pay off all its short- term debts, accumulatedduring the 1990s, by the year 2001.

Several questions about providing credit tomunicipalities remain, however. Should lendingconditions between public sector agencies try tosimulate exactly what the market would itselfprescribe? Do municipal credit markets indeveloping countries that are controlled ortightly regulated by government agencies provideaccurate pricing signals on interest rates, terms ofrepayment, and other lending conditions? TheLevy Report to the Bank (World Bank 1989) andthe subsequent 1998 Operational Policy 8.30 onfinancial intermediary lending tended towardnegative answers to these questions. This dis -couraged earmarked lending, such as credit tomunicipalities, which was thought to fragment

S T R O N G E R M U N I C I PA L F I N A N C E S

3 1

3 2


broader financial markets and undermine theirefficiency. A recent study by the Bank’s UrbanAnchor argued that this kind of lending tomunicipalities could effectively improve servicedelivery when the MDP is deliberately adapted tothe needs of its municipal clients (Annez, Huet,and Peterson 2008).

Private Finance ParticipationMDP incursions into enabling and stimulatingprivate finance of municipal services have yieldedfew positive results. The focus of the 24 MDPsreviewed by PPARs on this aspect of municipalmanagement was very weak. To seriously engagethe private sector, municipalities have to buildteams of experts to prepare and supervisecontract management units, something thatmany still lack.

Just seven projects addressed this, and only oneobtained substantial efficacy in achieving greaterprivate finance (table 4.1). This was underColombia IV, which helped municipalitiesincrease water, gas, and solid waste tariffs, therebymaking some services profitable for privateinvestors for the first time. At the same time itmade some services less affordable to the poorbecause it did not provide a corresponding safetynet for them. Average household expenditure onbasic sanitation in Colombia rose by 204 percentbetween 1997 and 2003. Thus, an important factorin this positive result was the ability and willing-ness of municipalities to increase local tariffs,albeit with some loss of affordability.

Factors undermining successful private financingare mostly related to private investors hesitatingto participate in the face of pricing or regulatoryuncertainties. Thus, land development for low- income families by private developers underIndia I did not go far, as the developers showedlittle interest in the scheme, again for pricing

reasons. Privatizing the loss- making municipalsolid waste management operation in theUzbekistan MDP, a specific aim of the EuropeanBank for Reconstruction and Development, theBank’s partner in the project, did not proceedbecause of lack of interest by private investors.

The planned privatization of the municipal waterutility in Kazan, supported by the Bank underRussia IV, did not go far because the EuropeanBank for Reconstruction and Development (notpartnering with the Bank in this operation)sought to strengthen the existing public sectoroperator instead. The Zimbabwe MDP’s hopeof stimulating private finance of municipalhousing was frustrated by private buildingsocieties’ own financial weaknesses, which arosefrom their inability to compete with postalsavings and their inability to provide an afford-able product for the needs of lower- and middle- income households. In hindsight, a morethorough analysis of Zimbabwe’s housing marketwould not have recommended pursuing thisobjective. Under Indonesia IX, private financ-ing of local municipal markets or otherinfrastructure did not appear on the scale hopedfor, as small municipalities especially wereunable to articulate a regulatory framework forthe local investors.

The limited results in securing private finance forlocal services point to several areas in which MDPdesign should be strengthened: accurate analysisof local financial markets, a realistic assessmentof the demand for such services at the prices theywill be offered, and, most important, anassurance of a reasonable chance of some profit.In addition, it is important for an MDP appraisalto affirm that municipalities understand theirresponsibilities in relation to the regulatoryframework, especially for pricing that wouldgovern such private finance.

Chapter 5

Evaluation Highlights• Although all 190 MDPs focused on

municipal service provision, lessthan one- third used cost- benefitanalysis for evaluating or prioritizing investments.

• More than one- third of the evalu-ated MDPs were able to substan-tially help municipalities manageprocurement for the first time.

• Most MDPs paid little attention toO&M, significantly increasing therisk to development outcomes.

• Few MDPs focused explicitly on im-proving the lives of the poor, and ev-idence of actual results of betteraccess to services obtained is thin.

Municipal solid waste disposal in Tashkent, Uzbekistan. Photo courtesy of Roy Gilbert.

3 5

Managing Service Provision

This chapter reports the findings of IEG assessments of the efficacy ofthe following elements of managing service provision: prioritizing in-vestments through cost- benefit analysis, conducting procurement for

service investments, and handling the O&M of ongoing services.

Elements of Service Provision

Prioritizing investments in service provisionWhen resources are scarce, municipal managersneed to know, at the outset, the best options forinvestment and how well those investmentsperform at completion. To help them, the Bankrequires MDPs, like other investment operationsit finances, to conduct ex ante and ex posteconomic evaluations of investment perform-ance. The present study found, however, thatonly 7 of the 24 PPAR MDPs did this effectively.Those seven applied the instrument of choice, a cost- benefit analysis that yielded an economicrate of return (ERR) or an estimate of the netpresent value, at appraisal and/or completion(table 5.1).

Project documentation of the remaining MDPsgave diverse reasons for the lack of economicanalysis. The reasons included the high cost ofestimating ERRs, the complexities of measuringexternal costs and benefits, and the exempt statusof some MDPs because of their quasi- emergencystatus. Nevertheless, seven MDPs facing chal -lenges such as these were able to successfullyassess their efficiency through ERR estimates.

Most of these MDPs applied simple models of cost- benefit analysis that used available data,

notably data produced by project M&E. Ghana Imade simple but methodologically robustestimates that did not require a lot of data. Theproject’s municipal slum upgrading componentsin Tamale (population 360,579) yielded an ERRof 29 percent, using benefits derived from the(realistic) increases of land values followingproject improvements.

Tanzania I applied a similar method, and theZimbabwe MDP incorporated shadow pricinginto the assessment. Beyond Africa, China III,Indonesia II, Georgia III, and India I all sawmunicipalities themselves directly involved inthe ERR work, under the guidance of the MDPs.

Making the investments: ProcurementThrough MDPs, many municipalities engaged incompetitive tendering for works and supplies forthe first time; traditionally that procurement hadbeen in the hands of the central government.From its PPAR assessments in the field, IEG foundthat local MDP beneficiaries were better in -formed about service improvements to theirneighborhoods when municipalities themselveshad carried out the procurement. Nine MDPsreviewed by PPARs across most Regions obtainedsubstantial results in this area mainly by introduc-ing municipalities to procurement managementfor the first time.

3 6


In other cases, particularly in East Asia and Pacificand South Asia, the impact of MDPs on procure-ment practice was less. But this does not meanthat they all failed. Several larger municipalities

already had experience with procurement andcontinued to handle it proficiently, as they haddone before an MDP whose design rightly didnot focus unduly on this issue (table 5.1).

Overalloutcome Results in—

Country MDP Project name of project ERRs Procurement O&M

Sub- Saharan Africa

Gambia, The Pov. Allev. & Capacity Building M Sat * * √Ghana I Second Urban Project Sat √ * √Mozambique I Local Govt. Reform Unsat — * —

Tanzania I Urban Sector Rehabilitation Sat √ √ √Zimbabwe Urban Sector & Regional Dev. M Sat √ * *


China III Tianjin Urban Development Sat √ √ *

China IV Zhejiang Multicities Dev. Sat * * —

China VII Shanghai Environment Sat — * —

Indonesia II East Java/Bali Urban Dev. M Sat √ * *

Indonesia VI Second East Java Urban Dev. M Unsat * * *

Indonesia IX Municipal Innovations Sat — * —


Georgia I Mun. Infrastructure Rehab. M Unsat — — *

Georgia II Mun. Dev. & Decentralization M Sat — — *

Georgia III Second Mun. Dev. & Decentral. Sat √ √ *

Russian Federation IV Kazan Municipal Dev. H Sat * √ *

Uzbekistan Tashkent Solid Waste Mgt. Sat * √ *


Brazil II Ceará Urban Dev. & Water Res. M Sat * √ —

Chile II Second Municipal Dev. M Sat — — *

Colombia I Municipal Dev. Sat * √ *

Colombia IV Urban Infras. Services Dev. Sat — √ *


Tunisia I Municipal Sector Investment Sat — * √South Asia

India I Tamil Nadu Urban Dev. M Sat √ * —

India II Tamil Nadu Second Urban Dev. Sat * * *

Sri Lanka Colombo Env. Improvement Unsat — √ —Sources: IEG PPARs.Note: Bold = MDPs focused on service provision; √ = substantial or higher achievement of element; * = element tackled but with modest or lower achievement; — = element not at-tempted. In the case of ERRs, achievements are rated according to the use made of the tool itself, not according to the value of the estimated rate of return. (Roman numerals are at-tached to MDPs in their sequence in the portfolio of 190 operations, dispensed with altogether when there is only one MDP in a particular country.) ERR = economic rate of return; MDP= municipal development project; O&M = operations and maintenance. Ratings: H Sat = highly satisfactory; Sat = satisfactory; M Sat = moderately satisfactory; M Unsat = moderatelyunsatisfactory; Unsat = unsatisfactory.

Table 5.1: Summary of MDP Results in Service Provision

For example, Brazil II enabled 49 mostly smallmunicipalities such as Quixadá (population49,328) in Ceará State to oversee competitiveprocurement of works for upgrading low- incomeneighborhoods for the first time. At the oppositeend of the scale, a very large municipality, such asTianjin under China III, became quite expert inconducting complex international competitivebidding after its purchase of sophisticated trafficmonitoring equipment for the city and state- of- the- art equipment for the solid waste disposalsite at Shuangkou. These became a model for allof China.

The experience of Russia IV convinced Kazanmunicipality that Bank- standard local competi-tive procurement procedures helped it obtainlower prices. Kazan thus decided to apply theseprocedures voluntarily to urban street upgradesfinanced from its own budget. This practicecontinues today, though the project has closed.Previously, Kazan had relied on sole- sourceacquisitions, typical of former Soviet practices,which the MDP helped reform. Although munici-palities themselves drive such results, for themost part they have to be consistent withnational or higher- level legislation.

O&M of servicesFor their own investments and for those made byothers within their jurisdictions, municipalitiesare generally responsible for the use and upkeepof the assets provided by MDPs. Past urbandevelopment projects in which central govern-ment authorities simply delivered infrastructureassets to a city without involving or sometimeseven consulting the municipality about O&Mgenerally failed for lack of municipal ownership(IEG 2004, p. 18).

By putting municipalities at center stage, MDPsmake municipal O&M responsibilities clearer.They also provide both challenges and opportu-nities for local administrations to ensure thaturban infrastructure continues to provide good- quality services and benefits to users. It is clearthat many municipalities have yet to rise to thischallenge. This study found only four MDPs thatachieved substantial results in strengthening the

municipal management of O&M. In the remain-der of the cases, O&M was either disregarded byMDPs that focused primarily on supporting theinitial service investment, or it did not succeedfor lack of funding (table 5.1).

On the positive side, Tanzania I successfullyintroduced computerized maintenance systemsto 10 municipalities that used them to track andplan the maintenance of urban streets. Althoughbudget shortfalls are often given as a reason forthe lack of O&M, The Gambia MDP helpednine municipalities, home to half the country’surban population, to establish O&M accountsthat were adequately funded by the municipali-ties themselves. Although quite an achievementin itself, local administrations still lacked thenecessary equipment and technical and manage-rial capacity to carry out all the maintenance needed.

Better results were obtained through Ghana I,which even introduced parking controls in thecentral areas of the cities of Accra (population2.0 million) and Sekondi- Takoradi (population371,791) to facilitate street cleaning and accessby maintenance vehicles. Under Tunisia I, theremote municipality of Kasserine (population82,000) upgraded its Ezzouhour district, keepingit in good condition through careful mainte-nance, sometimes with the help of the local residents.

Inadequate attention to O&M led to negativeproject results for MDPs. Insufficient O&Mfunding meant, for instance, that the municipalityof Bulawayo (population 699,000) under theZimbabwe MDP could not pay for the pumpingneeded to keep the project’s Nkulumane sewageplant operating. Similarly, under Indonesia II, awater supply system for the Kintamani district ofDenpasar (population 405,923), Bali, fell intodisuse because the municipality could afford tooperate the pumps for only a fraction of the timeneeded each day. Other municipalities in East Javawere able to conduct everyday maintenance suchas patching minor pavement failures, but notheavier repairs. Even prosperous municipalitiesmay feel that incentives are not right to encourage

M A N AG I N G S E R V I C E P R OV I S I O N

3 7

3 8


O&M. Under Tunisia I, for example, officials ofAriana (population 237,395) felt that it was betterto neglect routine maintenance in upgraded areasand allow the infrastructure to fail, because thatincreased the chance that the government wouldfinance a complete replacement.

Sectors Most Affected and Service QualityIEG PPARs show that MDPs cover a widespectrum of municipal services in their efforts tostrengthen municipal management: upgradingexisting infrastructure, providing new assets,and improving the operation of existinginfrastructure. MDPs typically supportedimprovements to urban road and street paving,drainage and lighting, basic sanitation, solidwaste and slum upgrading, environmentalimprovements, transport, and others. By con -vention, Bank support for im portant municipaleducation and health services has beenprovided through dedicated sectoral projects,not through MDPs. By mandate, Bank assistancehas generally not been involved with theimportant political and security work thatmunicipalities in many countries carry out.

Street paving and drainage were the mostpopular municipal services supported by MDPmanagement strengthening in Colombia, Ghana,India, and Georgia. Colombia IV helpedimprove the urban environment of low- incomeperipheral areas of several cities, improvementsthat are still being maintained. Paving streetsreduced dust during dry seasons, and betterdrainage avoided repeat flooding and impassablestreets during rainy seasons.

Ghana I radically transformed the central areasof Accra and Sekondi- Takoradi through streetpaving and drainage, especially in places wherestreet markets were held daily. This improvedaccess to markets, whose stallholders reportedincreased business. The upgraded locationsmade the cleanup after the markets easier, too,thereby improving the urban environment.Georgia’s three MDPs improved residentialneighborhoods by paving streets.

Neighborhood upgrading and basic sanitationtook MDP support for services one step furtherby introducing water and sewer services intopoor neighborhoods. The large scale of serviceprovision across Tamil Nadu by India I is evidentfrom the upgrading of 489 slums. Through theupgrade, 76,000 people gained better access totheir homes and businesses through pavedfootpaths and proper drainage. These improve-ments were implemented across 35 municipali-ties that participated directly for the first time inproviding such better- quality services. India IIcontinued to provide more of the same in 102additional municipalities, where sample benefici-ary assessments point to better- quality basicsanitation services.

Tanzania I helped the smaller municipalities ofMororgoro (population 251,000) and Tabora(population 145,000) significantly reduce unaccounted- for water by helping to repair leakymain lines. Consumers reported that waterbecame available for more hours per day thanbefore the project. Unaccounted- for watercontinued to be a problem in some Colombia Imunicipalities, however, where network cov -erage was extended without achieving overallsystem improvements.

When trying to offer large- scale sewagetreatment plants through MDPs, municipalitieshad more limited success. Investment costoverruns and high operating costs put them outof the reach of municipal finances and resultedin incomplete and nonoperational plants, as inthe municipality of Tema (population 155,782)under Ghana I and in the Nkulumane district ofBulawayo under Zimbabwe’s MDP. The innova-tive introduction of a low- cost, small- scalemodular approach to sewage treatment by themunicipality of Malang (population 747,000)through Indonesia VI also has met with littlesuccess. Beneficiaries continue to dischargesewage into storm drains, rather than paying the(modest) fee imposed by the new system.

MDPs provided basic sanitation through solidwaste management, too, with generally positive

results. China III led to the building andoperation of the country’s first sanitary landfill atShuangkou near Tianjin. Similar solid wastedisposal solutions were offered through MDPs inSri Lanka, Uzbekistan, and Tanzania.Although they did not fully meet all the ambitiousperformance standards set for them, the MDPsolutions do represent significant advances overprevious practices of uncontrolled dumping.Solid waste collection equipment and technicalassistance provided through Tanzania I, forinstance, raised the share of garbage producedthat was actually collected from 40 percent acrossthe 9 cities to 55 percent, a significant improve-ment in service quality and benefit for the urbanenvironment. In the Uzbekistan MDP, thequality of solid waste collection improvedthrough more regular pickups at controlledcollection points rather than through an increasein the quantity collected.

In one case, a large number of municipalitiesbenefited. Colombia I’s help to 179 municipali-ties had an impact on service levels that made itsmark on indicators at the national level. Between1993 and 2003, when the project was imple -mented, basic sanitation coverage of the lowestquintile of Colombia’s population in incomedistribution rose from 77 percent to 83 percent.

Environmental improvements by MDPs beyondthose resulting from street paving and basicsanitation included both long- term and short- term provision of other services. Through SriLanka’s MDP, the municipality of Colombo(population 2.3 million) sought to reduce thepollution of the city’s Beira Lake, but with limiteddemonstrable results. A significant short- termenvironmental gain was made by China VII,however. The project enabled the Shanghaimunicipality (population 14.6 million) to build alarge water catchment plant on the upper, less- polluted reaches of the environmentally stressedHuangpo River. Today, more than five years aftercompletion, the plant continues to provide safedrinking water to more than 8 million con -sumers, whose health had been seriously at riskfrom the poor water quality of the old intake.

Other municipal services were mostly in the areaof urban transport. China III introduced bettertraffic surveillance and monitoring to the megacityof Tianjin, although without giving much higherpriority to public transport. India I in Tamil Nadu,however, brought considerable improvements tothe quality of the bus service in the state capitalChennai (population 4.3 million) through thepurchase of new bus chassis. Bus transport was akey feature, too, of Tanzania I, which upgradedmunicipal bus terminals in eight municipalities,providing paved areas for buses and coveredshelters for passengers. The terminals becamehives of commercial activity and are still boomingtoday, in addition to handling 50 percent morebuses than before the project. A special MDPtransport improvement came through Georgia I,which provided spare parts to enable the Metrosystem of the capital Tbilisi to continue operationsafter supplies had been interrupted following thecollapse of the Soviet Union.

Income Levels of Beneficiaries— Poverty ReductionBetter municipal management of service deliverycould benefit all income groups of the popula-tion living and working within the jurisdiction ofa municipality. But does it bring benefits to thepoor and reduce poverty?

Only 27 percent of the 190 MDPs in the portfolioexplicitly aimed to bring municipal services tothe poor. Evidence of actual results achieved bythe 114 completed MDPs is patchy, at best. Thereis some evidence that MDPs benefitted the poor,but it is thin, which is to be expected from anMDP portfolio so little focused on poverty. Only4 of the 24 PPAR MDPs had project objectivesexplicitly aimed at improving the lives of thepoor: Brazil II, Ghana I, The Gambia, andTanzania I. A few others, such as Colombia Iand II and Indonesia VI, introduced basicsanitation to poorer districts of client municipal-ities, even though their formal objectivestatements did not specify a focus on the poor.

IEG field inspections confirmed that basicmunicipal services provided through Brazil II in

M A N AG I N G S E R V I C E P R OV I S I O N

3 9

4 0


Ceará State had indeed benefited poor districts ofwhat were mostly low- income municipalities inthis poor northeastern region of the country. TheGambia MDP, with poverty alleviation in theproject title, benefited more poor, unskilledconstruction workers than targeted, albeit onlytemporarily through construction work arisingdirectly from the project implementation.Tanzania I brought urgent relief to some 13,600poor residents of Dar es Salaam (population 2.8million) through drilling 34 emergency boreholes(serving 400 people each) during a droughtemergency. Ghana I successfully introducedbasic sanitation, street paving, and lighting to24,000 poor people living in the Ashaimansquatter district of Tema, in the port city of Accra.

The poverty results of these operations arethemselves worth further scrutiny by those whowish to emulate such results in futureoperations. Indeed, the evidence showing thelack of poverty focus of most MDPs points to aclear need to report the results of such experi-ences much more thoroughly.

Implementation Completion Reports rarelyprovide information about the income levels ofMDP beneficiaries. If future MDPs are to benefitthe poor on a larger scale and are to be seen doingso, they will have to do two things. First, they needto sharpen the focus of their objectives on helpingthe poor, making it explicit in project objectivestatements and in project design. A more poverty- oriented approach would explain, for instance,how the poor would benefit from municipalinvestments and services rendered better throughMDP municipal management strengthening.

Second, they need to harness M&E so that thedesign of these systems helps establish clear goalsfor poverty alleviation and so that M&Eimplementation shows exactly what has beenachieved. As noted earlier, there is much workstill to do in this area for the Bank to fully deployits poverty reduction mission through its partnermunicipalities. Being able to assess poverty- related objectives and project design, as well asactual results of MDPs, would be an important contribution.

Chapter 6

Community life resumes for newly resettled residents of Ningbo, China. Photo courtesy of Roy Gilbert.

4 3

Conclusions

W ith cities now home to more than half the world’s population andproviding the majority of world gross domestic product, improvingmunicipal management is crucial for development. The Bank’s cho-

sen instrument to support this has been the MDP, 190 of which have assistednearly 3,000 municipalities in 76 countries worldwide over the past decade.

IEG field assessments of 24 completed MDPs,desk reviews of 90 more, and an overview of 76ongoing operations have highlighted thestrengths and weaknesses of these operations,which are aimed at helping municipalitiesstrengthen their management in the planning,finance, and service provision dimensions. Thisstudy has found positive MDP experiences worthemulation, as well as weaker results that point toareas in need of improvement. The following forward- looking lessons may help strengthenmunicipal management.

• Among the three dimensions of municipal management— planning, finance, and service provision— MDP support for strengtheningmunicipal finance most often yielded successfulresults. The Bank should continue to supporttightened municipal financial management,municipalities raising their own revenues, andmunicipalities being brought to local creditmarkets when conditions are appropriate.

• Project documentation that routinely reportsbasic data about each client (municipality name,population, and MDP investment) is vital to de-veloping a better understanding of the scopeof MDP results.

• Wholesale MDPs have yielded better outcomesthan retail MDPs over the past decade, butmore analysis is needed to understand theprecise reasons for the performance differen-

tial. Retail MDPs might perform better if theyincorporated more of the winning elements ofwholesale MDPs, such as performance- basedincentives and a focus on finance.

• More frequent use of cost- benefit or cost- effectiveness analysis would help MDPs’ mu-nicipal clients select the best investments andachieve better outcomes. IEG found that onlyhalf of the 114 completed MDPs did this, withthe best coverage in the Sub- Saharan Africa Region.

• For M&E to succeed in MDPs, it has to be use-ful and not unduly burdensome to municipal-ities themselves; it must also keep a focus onachieving results, particularly for the poor.Strong M&E can also help reduce the expenseof cost- benefit analyses by providing some ofthe data needed to estimate ERRs. Few MDPshave succeeded with this.

• Private finance of municipal services can be en-couraged through better analysis of demandand of local financial markets. Stable regula-tions also help municipalities gain the trust ofprivate investors.

• Thus far, little evidence exists that strongermunicipal management has benefited the poor.MDPs need to give much more attention topoverty reduction in defining MDP objectives,showing how the poor would benefit frommunicipal investments and services improvedthrough stronger municipal management.

Appendixes

47

APPEN

DIX

A: LIST

OF B

AN

K-FIN

AN

CED

MD

PS, FISC

AL 1998–2008

Commitment Bank Borrower Project ID MDP Type Project name Perioda US$ millions Outcome performance performance

Sub-Saharan Africa

P000097 Benin I R Urban Rehabilitation & Management 1992–98 23 H Sat H Sat Sat

P035648 Benin II R First Decentralized City Management 1999–2005 26 Sat Sat H Sat

P082725 Benin III R Second Decentralized City Management 2006– 35 — — —

P000297 Burkina Faso I R Urban Environment 1995–2005 37 Sat Sat Sat

P084027 Burkina Faso II R Decentralized Urban Capacity Building 2007– 10 — — —

P064961 Burundi W Public Works & Employment Creation 2001– 40 — — —

P084002 Cameroon R Urban & Water Development Support 2007– 80 — — —

P072030 Chad R Urban Development 2007– 15 — — —

P037575 Côte d’Ivoire W Municipal Support 1995–2004 40 Unsat Unsat Unsat

P000712 Ethiopia I R Second Addis Urban Development 1990–99 35 Unsat Sat Unsat

P050938 Ethiopia II W Capacity Building for Decentralized Service Delivery 2003– 26 — — —

P074020 Ethiopia III R Public Sector Capacity Building 2004– 100 — — —

P101473 Ethiopia IV R Urban Water Supply & Sanitation 2007– 100 — — —

P057997 Gambia, The R Poverty Alleviation & Capacity Building 1999–2007 15 M Sat Sat Sat

P000910 Ghana I R Second Urban 1990–99 70 Sat Sat Sat

P000936 Ghana II W Local Govt. Development 1994–2003 39 Sat Sat Sat

P000973 Ghana III R Urban Environment & Sanitation 1996–2004 71 M Sat Unsat Unsat

P050624 Ghana IV W Fifth Urban 2000–04 11 Sat Sat Sat

P001074 Guinea I R Third Urban Development (APL) 1999–2005 18 Sat Sat Sat

P091297 Guinea II R Third Urban (Phase 2) 2008– 15 — — —

P001319 Kenya R Urban Transport 1996–2005 115 M Unsat Unsat Unsat

P001512 Madagascar I R Antananarivo Plain Development 1990–2000 31 M Unsat Unsat Unsat

P001583 Madagascar II R Antananarivo Urban Works Pilot 1994–99 18 Sat Sat Sat

P048697 Madagascar III R Urban Infrastructure 1997–2005 35 Sat Sat Sat

P001636 Malawi R Local Govt. Development 1992–2001 24 M Unsat Unsat Unsat

P001750 Mali I R Urban Development & Decentralization 1997–2005 80 M Unsat Sat Unsat

P090075 Mali II R Second Transport Sector 2007– 90 — — —

P034106 Mauritania I W Urban Infrastructure & Pilot Decentralization 1996–2002 14 Sat Sat Sat

P069095 Mauritania II R Urban Development Program 2002– 70 — — —

P001791 Mozambique I R Local Govt. Reform & Engineering 1993–99 23 Unsat H Unsat H Unsat

(Continues on the next page.)

IMP

RO

VIN

G M

UN

ICIP

AL

MA

NA

GE

ME

NT

FO

R C

ITIE

S T

O S

UC

CE

ED

48

Commitment Bank Borrower

Project ID MDP Type Project name Perioda US$ millions Outcome performance performance

Sub-Saharan Africa (continued)

P001806 Mozambique II W Municipal Development 2002–07 34 M Sat M Sat M Sat

P096332 Mozambique III R Maputo Municipal Development Program 2007– 30 — — —

P049691 Niger I W Urban Infrastructure Rehabilitation 1997–2003 20 Sat Sat Sat

P095949 Niger II R Local Urban Infrastructure Development 2008– — — —

P002074 Nigeria I R Oyo State Urban 1990–99 50 Unsat Sat Unsat

P071340 Nigeria II R Lagos Metropolitan Development & Governance 2007– 200 — — —

P060005 Rwanda R Urban Infrastructure & City Management (APL) 2006– 20 — — —

P002365 Senegal I W Urban Development & Decentralization Program 1998–2005 75 H Sat H Sat Sat

P084022 Senegal II R Local Authorities Development Program 2007– 80 — — —

P076901 South Africa W Municipal Financial Management (TAL) 2003– 15 — — —

P002669 Swaziland I R Urban Development 1995–2005 29 Sat Sat Sat

P095232 Swaziland II R Local Government 2008– — — —

P002758 Tanzania I R Urban Sector Rehabilitation 1996–2005 105 H Sat Sat Sat

P070736 Tanzania II R Local Govt. Support 2005– 52 — — —

P002865 Togo R Lome Urban Development 1994–2003 26 Sat Sat Unsat

P002933 Uganda I R First Urban 1991–2000 29 Sat Sat Sat

P059223 Uganda II R Nakivubo Channel Rehabilitation 1999–2004 22 Sat Sat Sat

P002992 Uganda III R Local Govt. Development Program 2000–04 81 Sat Sat Sat

P044679 Uganda IV W Economic & Financial Management 2000–07 34 Sat M Sat M Sat

P078382 Uganda V R Kampala Institutional & Infrastructure Development 2008– 34 — — —

P003241 Zambia R Urban Restructuring 1995–2002 33 M Sat Sat Sat

P003294 Zimbabwe W Urban Sector & Regional Development 1989–2000 80 M Sat Sat Sat


P003564 China I R Beijing Environment 1992–99 125 M Sat Sat Sat

P003565 China II R Shanghai Metropolitan Transport 1992–99 60 Sat Sat Sat

P003568 China III R Tianjin Urban Development 1992–2001 100 Sat Sat H Sat

P003473 China IV R Zhejiang Multicities Development 1993–2003 110 Sat Sat Sat

P003580 China V R Southern Jiangsu Environmental Protection 1993–2001 250 M Unsat Unsat Unsat

P003622 China VI R Second Shanghai Metropolitan Transport 1994–2001 150 Sat Sat Sat

P003586 China VII R Shanghai Environment 1994–2003 160 Sat Sat Sat

AP

PE

ND

IX A

: LIS

T O

F B

AN

K-F

INA

NC

ED

MD

Ps

, FIS

CA

L 1

99

8–

20

08

49 (Continues on the next page.)

P003598 China VIII R Liaoning Environment 1995–2004 110 Sat Sat Sat

P003603 China IX R Enterprise Housing & Social Security Reform 1995–2005 350 Sat Sat Sat

P003602 China X R Hubei Urban Environment 1996–2005 150 M Sat Sat Sat

P003646 China XI R Chongqing Industrial Pollution Control & Reform 1996–2003 170 Unsat Sat Sat

P003599 China XII R Yunnan Environment 1996–2005 150 M Sat Sat Sat

P040185 China XIII R Shandong Environment 1998–2006 95 Sat Sat Sat

P041890 China XIV R Liaoning Urban Transport 1999–2006 150 M Sat Sat Sat

P043933 China XV R Sichuan Urban Environment 1999–2007 152 M Sat Sat M Sat

P049436 China XVI R Chongqing Urban Environment 2000– 200 — — —

P045915 China XVII R Urumqi Urban Transport 2001– 100 — — —

P056596 China XVIII R Shijiazhuang Urban Transport 2001– 100 — — —

P040599 China XIX R Tianjin Second Urban Development 2003– 150 — — —

P069852 China XX R Wuhan Urban Transport 2004– 200 — — —

P081346 China XXI R Liuzhou Environment Management 2005– 100 — — —

P081161 China XXII W Chongqing Small Cities Infrastructure Improvement 2005– 180 — — —

P075732 China XXIII R Second Shanghai Urban Environment (APL) 2006– 180 — — —

P003922 Indonesia I R Sulawesi—Irian Jaya Urban Development 1991–99 100 M Sat Sat Sat

P003943 Indonesia II W East Java/Bali Urban Development 1991–98 180 M Sat Sat Sat

P003998 Indonesia III R Surabaya Urban Development 1994–2001 175 H Unsat Unsat Unsat

P003890 Indonesia IV R Semarang Surakarta Urban Development 1994–2002 174 M Unsat Sat Unsat

P003951 Indonesia V R Kalimantan Urban Development 1995–2003 136 Sat Sat Sat

P039312 Indonesia VI W Second East Java Urban Development 1996–2002 117 M Unsat Sat Sat

P036053 Indonesia VII W Second Sulawesi Urban Development 1997–2003 155 Unsat Unsat Unsat

P055821 Indonesia VIII W Urban Poverty 1999–2004 100 Sat Sat Sat

P056074 Indonesia IX W Municipal Innovations 1999–2003 5 Sat Sat Sat

P040528 Indonesia X R Western Java Environmental Management 2001–06 17 M Sat Sat Sat

P072852 Indonesia XI W Second Urban Poverty 2002– 100 — — —

P071296 Indonesia XII W Urban Sector Development & Reform 2005– 45 — — —

P004175 Korea, Rep. of R Pusan Urban Transport 1995–2002 100 M Sat Sat Sat

P036052 Mongolia R Urban Services Improvement 1998–2004 17 Sat Sat Sat

P004592 Philippines I W Third Municipal Development 1992–2001 68 Sat Sat Sat

P048588 Philippines II W Local Govt. Unit Finance & Development 1999– 100 — — —

P064925 Philippines III W Support for Strategic Local Development & Investment 2006– 100 — — —

IMP

RO

VIN

G M

UN

ICIP

AL

MA

NA

GE

ME

NT

FO

R C

ITIE

S T

O S

UC

CE

ED

50


Project ID MDP Type Project name Perioda US$ millions Outcome performance performance

East Asia and Pacific (continued)

P004830 Vietnam I R Water Supply 1997–2005 99 Sat Sat Sat

P004833 Vietnam II R Urban Transport Improvement 1999–2006 43 M Sat Sat Unsat

P070197 Vietnam III R Urban Upgrading 2004– 222 — — —

P082295 Vietnam IV R Coastal Cities Environmental Sanitation 2007– 125 — — —


P094225 Armenia R Third Social Investment Fund 2007– 25 — — —

P056192 Bosnia & Herzegovina I W Local Development 1999–2005 15 Sat Sat Sat

P070995 Bosnia & Herzegovina II W Community Development 2001– 15 — — —

P057950 Bosnia & Herzegovina III R Solid Waste Management 2002– 18 — — —

P083353 Bosnia & Herzegovina IV W Urban Infrastructure & Service 2005– 20 — — —

P065416 Croatia W Coastal Cities Pollution Control 2004– 48 — — —

P008417 Georgia I w Municipal Infrastructure Rehabilitation 1995–2000 18 M Unsat M Unsat M Unsat

P050910 Georgia II W Municipal Development & Decentralization 1998–2003 21 M Sat M Sat M Sat

P077368 Georgia III W Second Municipal Development & Decentralization 2003– 19 Sat Sat Sat

P008506 Kazakhstan I R Social Protection 1995–2002 41 Unsat Sat Unsat

P008500 Kazakhstan II R Atyrau Pilot Water 1999–2005 17 Sat Sat Sat

P079259 Kosovo W Second Community Development Fund 2004–07 4 M Sat Sat Sat

P050719 Kyrgyz Republic I R Urban Transport 2001–05 22 Sat Sat Sat

P083377 Kyrgyz Republic II W Small Towns Infrastructure & Capacity Building 2005– 15 — — —

P034584 Latvia R Municipal Services Development 1996–2002 27 M Sat Sat Sat

P035802 Lithuania R Municipal Development 1999–2005 20 Unsat Unsat Sat

P035082 Poland W Municipal Finance 1998–2002 22 Unsat Unsat Unsat

P042720 Russian Federation I R St. Petersburg Center City Rehabilitation 1997–2002 31 M Unsat Unsat Unsat

P064238 Russian Federation II R Northern Restructuring 2001– 80 — — —

P069063 Russian Federation III R St. Petersburg Economic Development 2003– 161 — — —

P082018 Russian Federation IV R Kazan Municipal Development 2005–07 125 H Sat Sat H Sat

P079027 Tajikistan R Municipal Infrastructure 2006– 15 — — —

P009065 Turkey I R Bursa Water & Sanitation 1993–2001 130 Sat Sat Sat

P081880 Turkey II W Municipal Services 2005– 275 — — —

P100383 Turkey III R Istanbul Municipal Infrastructure 2007– 322 — — —

P034083 Turkmenistan R Urban Transport 1997–2001 34 M Unsat Unsat Unsat

P095337 Ukraine R Urban Infrastructure 2008– 140 — — —

P049582 Uzbekistan R Tashkent Solid Waste Management 1998–2006 24 Sat Sat Sat

AP

PE

ND

IX A

: LIS

T O

F B

AN

K-F

INA

NC

ED

MD

Ps

, FIS

CA

L 1

99

8–

20

08

51

(Continues on the next page.)


P006060 Argentina I W Second Municipal Development 1995–2005 210 Sat Sat Sat

P060484 Argentina II R Basic Municipal Services 2006– 110 — — —

P070448 Argentina III W Subnational Govt. Public Sector Modernization 2006– 40 — — —

P006104 Belize R Belize City Infrastructure 1994–98 20 M Sat Unsat Unsat

P006190 Bolivia I W Municipal Development 1994–2000 42 Sat Sat Sat

P083979 Bolivia II R Urban Infrastructure 2007– 30 — — —

P006524 Brazil I W Minas Municipal Development 1994–2002 150 Sat Sat Sat

P006436 Brazil II W Ceará Urban Development & Water Resource 1995–2004 140 M Sat Sat Sat

P006562 Brazil III W Bahia Municipal Infrast. Dev. & Mgmt. 1997–2005 100 Sat Sat Sat

P081436 Brazil IV R Bahia Poor Urban Areas Integrated Development 2006– 49 — — —

P089013 Brazil V R Municipal APL: Recife 2008– 33 — — —

P006677 Chile I W Municipal Development 1994–98 10 Sat Sat Sat

P055480 Chile II W Second Municipal Development 1999–2005 10 M Sat M Sat M Sat

P006852 Colombia I W Municipal Development 1991–2001 60 Sat Sat Sat

P006872 Colombia II R Bogota Urban Transport 1996–2001 65 H Sat H Sat H Sat

P039291 Colombia III R Urban Environment (TAL) 1996–2003 20 Sat Sat Sat

P006861 Colombia IV W Urban Infrastructure Services Development 1998–2004 75 Sat Sat Sat

P074726 Colombia V R Bogota Urban Services Project 2003– 100 — — —

P082466 Colombia VI R Integrated Mass Transit Systems 2004– 250 — — —

P085727 Colombia VII R Disaster Vulnerability Reduction Project (APL 2) 2006– 80 — — —

P007123 Ecuador I W First Municipal Development 1991–99 104 Sat Sat Sat

P007128 Ecuador II R Environmental Management 1996–2002 15 M Sat Unsat Unsat

P007292 Haiti R Port-Au-Prince Water Supply 1989–99 20 Unsat Unsat Unsat

P064913 Honduras I W Natural Disaster Mitigation 2000– 11 — — —

P057859 Honduras II R Sustainable Coastal Tourism Project 2002–06 5 M Sat Sat Sat

P088319 Honduras III W Barrio Ciudad 2006– 15 — — —

P103881 Honduras IV R Water & Sanitation Program 2007– 30 — — —

P007710 Mexico I R Northern Border Environment 1994–2004 368 M Unsat Unsat Unsat

P007612 Mexico II R Solid Waste 1994–2001 200 M Unsat Unsat Unsat

P064916 Nicaragua W Natural Disaster Vulnerability Reduction 2001– 14 — — —

IMP

RO

VIN

G M

UN

ICIP

AL

MA

NA

GE

ME

NT

FO

R C

ITIE

S T

O S

UC

CE

ED

52


Project ID MDP Type Project name Perioda US$ millions Outcome performance performanceLatin America and the Caribbean (continued)

P035740 Peru I R Lima Transport 2004– 45 — — —

P082625 Peru II R Vilcanota Valley Rehabilitation & Management 2005– 5 — — —

P078894 Peru III W Second Real Property Rights 2006– 25 — — —

P008212 Venezuela I W Low-Income Barrios Improvement 1992–99 40 Sat H Sat Sat

P008210 Venezuela II W Urban Transport 1994–2002 100 Sat Sat Sat

P040174 Venezuela III R Caracas Slum Upgrading 1999–2006 61 M Sat M Sat M Sat


P094229 Egypt, Arab Rep. of R Alexandria Development 2008– 100 — — —

P073433 Iran, Islamic Rep. of R Urban Upgrading & Housing Reform 2004– 80 — — —

P070958 Jordan I W Regional & Local Development 2007– 20 — — —

P081823 Jordan II R Cultural Heritage, Tourism & Urban Development 2007– 56 — — —

P050544 Lebanon I W First Municipal Infrastructure 2000– 80 — — —

P050529 Lebanon II R Cultural Heritage & Urban Development 2003– 32 — — —

P005524 Morocco R Fes Medina Rehabilitation 1999–2006 14 Unsat Unsat Unsat

P005687 Tunisia I W Municipal Sector Investment 1993–99 75 Sat Sat Sat

P046832 Tunisia II W Second Municipal Development 1997–2003 80 M Sat Unsat Sat

P064082 Tunisia III R Transport Sector Investment 2001– 38 — — —

P074398 Tunisia IV W Third Municipal Development 2003– 78 — — —

P043339 West Bank & Gaza I R Municipal Development 1996–2003 40 Unsat Sat Sat

P053985 West Bank & Gaza II R Bethlehem 2000 1999–2004 25 M Unsat Sat Sat

P058683 West Bank & Gaza III R Second Municipal Infrastructure Development 2000–05 14 Sat Sat Sat

P078212 West Bank & Gaza IV W Emergency Municipal Services Rehabilitation 2003–06 20 M Sat Sat M Sat

P005907 Yemen, Republic of I R Sana’a Water Supply & Sanitation 1999–2003 25 M Sat Unsat Sat

P070092 Yemen, Republic of II R Taiz Municipal Development & Flood Protection 2002– 45 — — —

P065111 Yemen, Republic of III R Port Cities Development Program 2003– 23 — — —

South Asia

P083919 Afghanistan R Kabul Urban Reconstruction 2005– 25 — — —

P009467 Bangladesh I R Urban Development 1988–98 48 Unsat Unsat Unsat

P041887 Bangladesh II W Municipal Services 1999– 139 — — —

P057570 Bhutan R Urban Development 2000–06 11 M Unsat M Sat M Sat

P009872 India I W Tamil Nadu Urban Development 1988–98 300 M Sat Sat Sat

AP

PE

ND

IX A

: LIS

T O

F B

AN

K-F

INA

NC

ED

MD

Ps

, FIS

CA

L 1

99

8–

20

08

53

P050637 India II W Tamil Nadu Second Urban Development 1999–2005 105 Sat Sat Sat

P083780 India III R Third Tamil Nadu Urban Development 2006– 300 — — —

P079675 India IV R Karnataka Municipal Reform 2006– 216 — — —

P010305 Pakistan I R Punjab Urban Development 1988–98 90 M Sat Sat Sat

P010478 Pakistan II R North West Frontier Province Community Infrastructure1996–2003 22 M Unsat Unsat Unsat

P083929 Pakistan III W Punjab Municipal Services Improvement 2006– 50 — — —

P010467 Sri Lanka R Colombo Environment Improvement 1995–2001 39 Unsat Unsat UnsatSource: World Bank data.Note: MDP = municipal development project; R = retail; W = wholesale; — = not rated, as project not closed yet. Ratings: H Sat = highly satisfactory; M Sat = moderately satisfactory; Sat = satisfactory; M Unsat = moderately unsatisfactory;Unsat = unsatisfactory; H Unsat = highy unsatisfactory.a. Entry amd exit years; projects not yet closed just show entry year.

5 5

APPENDIX B: BANKING ON MUNICIPALITIES: WORLD BANK SUPPORT IN SUB-SAHARAN AFRICA

Bank SupportWith 280 million people—36 percent of the total population— living in cities, the Sub- SaharanAfrica Region is experiencing rapid urbanpopulation growth of 3.9 percent per annum.The World Bank financed 52 municipal develop-ment projects (MDPs) in this Region that wereactive during 1998–2008, with loan commit-ments of $2.4 billion. The portfolio aimed tostrengthen the management of 656 municipali-ties in 27 countries.

By the number of MDPs, the most active borrow-ers were Uganda (5 projects), Ethiopia (4),Ghana (4), Madagascar (3), Benin (3), andMozambique (3). The following countries hostedtwo MDPs each: Burkina Faso, Guinea, Mali,Mauritania, Niger, Nigeria, Senegal, Swaziland,and Tanzania. Another 12 countries had just 1each: Burundi, Cameroon, Chad, Côte d’Ivoire,The Gambia, Kenya, Malawi, Rwanda, SouthAfrica, Togo, Zambia, and Zimbabwe. Nearly allthe MDPs— more than 90 percent of the total— were thus implemented in low- incomecountries. The MDP portfolio covered allcountries in the Region with large urban popula-tions (15 million plus) except for the DemocraticRepublic of Congo and Sudan.

Portfolio PerformanceWith 75 percent of projects achieving satisfactoryoutcomes, the Region’s MDP performance wassimilar to that of the worldwide portfolio. Interms of Bank performance, Sub- Saharan Africa’sMDPs did better, with 81 percent satisfactory,against a Bank- wide average of 78 percentsatisfactory. For borrower performance, theRegion lagged behind, with 69 percent satisfac-tory against 75 percent satisfactory Bank- wide.

Of particular note were two MDPs that achievedhighly satisfactory outcomes. These can serve asmodel operations for others to emulate.

The first, Benin I, helped improve urbanservices in the country’s two largest cities,Cotonou (population 690,584) and Porto Novo(population 234,168). This result was confirmedby beneficiary assessments at completion andwas helped by the introduction of delegatedcontract management practices. These enabledrapid processing and execution of servicecontracts with local small and medium- sizeenterprises that provided higher- quality, lower- cost urban infrastructure services and leftmunicipal administrations more time to concen-trate on their planning and programming tasks.

The second highly satisfactory MDP, Senegal I,was a wholesale operation that helped 67 munici-palities throughout the country strengthen their

Completed (number) 32

Completed MDPs (% satisfactory) 75

Ongoing MDPs (number) 20

IBRD commitments (US$ million) 174

IDA commitments (US$ million) 2, 179

Bank commitments per completed MDP (US$ million) 42

Commitments per ongoing MDP (US$ million) 56

Wholesale MDPs (number) 12

Retail MDPs (number) 40

Countries served (number) 27

Municipalities served (number) 656Source: World Bank data.Note: IBRD = International Bank for Reconstruction and Development; IDA = International Develop-ment Association; MDP = municipal development project.

Table B.1: Summary of MDP Portfolio, 1998–2008


5 6

75 8169

74 78 75

0

20

40

60

80

100

Outcome Bank performance Borrower performance

Perc

ent s

atis

fact

ory

Sub-Saharan Africa Region All Bank

Figure B.1: MDP Portfolio Performance, Fiscal 1998–2008


financial and organizational management andimprove programming of investments in urbaninfrastructure and services. The project achievedthis through what were called “municipalcontracts,” participating agreements betweencentral and individual local governments withbenchmarks for municipal reform. Currently,more than 170 municipalities across French- speaking West Africa are implementing suchcontracts. The short- term results of the reformswere an increased municipal capacity to invest.Over 2001–03, for instance, municipal capitalinvestment as a share of current revenues rosefrom 10 to 17 percent.

Other operations with good outcomes in Sub- Saharan Africa included all five MDPs in Uganda.Uganda I introduced private sector participationin the municipal services of the capital Kampala(population 1.4 million). Flooding in that city wascurbed thanks to Uganda II’s rehabilitation ofthe Nakivubo Channel. In providing investmentfunds and technical assistance to other munici-palities, Uganda III strengthened municipalmanagement across the country, using awholesale approach. Uganda IV consolidatedthis approach by helping 30 municipalitiestighten their management controls, deployingnovel distance learning techniques to this end.

Ghana also had a string of successful MDPs.Ghana I brought significant service improve-ments, notably solid waste and stronger manage-

ment, to six municipalities. This success wasextended to 11 more municipalities by GhanaII. Ghana IV took the wholesale model furtherby investing intensely in financial and technicaltraining for the staff of 23 municipalities throughthe national Institute of Local GovernmentStudies, which itself came out of the projectconsiderably strengthened.

Madagascar II and III helped municipalitiesimprove municipal services through localagreements with an AGETIP- style provider. Byinvesting in sewerage and solid waste in particular,Tanzania I improved environmental managementin eight municipalities that a later governmentassessment found be the best in the country.Initially conceived as a retail operation only forNiamey (population 774,237) and Dogondoutchi(population 31,767), Niger I was successfullybroadened as a wholesale operation that, amongother things, introduced new digital cartographyskills into urban planning in 19 other municipalities.

Some MDPs performed poorly, however. CôteD’Ivoire’s MDP delivered fewer than half theservice improvements planned as the countrysituation became more volatile and centralgovernment support for the operation waned.Ethiopia I failed to address the major issues ofbudgeting, accounting, and financial managementof the client municipality of Addis Ababa (popula-tion 2.8 million), and infrastructure investmentsfor which demand was weak yielded inadequate

A P P E N D I X B : WO R L D B A N K S U P P O R T I N S U B - S A H A R A N A F R I C A

5 7

returns. Nigeria I failed to build municipalmanagement capabilities in Oyo State as intended,because of unresolved conflicts among the partiesthat led to implementation delays and cancella-tion of key components. Mozambique I did notimplement the majority of the project’s physicalcomponents because of procurement problems,inadequate Bank- borrower communication, andpoor performance by consultants.

Better City Planning

More informationParticularly notable have been MDP efforts andresults of obtaining better information aboutspatial configuration of cities. Burkina Faso I,for instance, helped create an informationsystem and database for urban managementbased on street address mapping in the munici-palities of Ougadougou (population 1.1 million)and Bobo Dioulasso (population 360,106).Niger I introduced a set of simple planning andprogramming tools covering digital cartographyand the production of an atlas, initially for 2 citiesbut extended through a wholesale approach to21 cities in all. Under Ghana II, maps wereproduced for 11 municipalities, leading toproperty valuations for twice as many propertiesas planned; only two municipalities ultimatelydeveloped land use structure plans on this basis.

Monitoring and evaluationSwaziland I incorporated a good monitoring andevaluation (M&E) framework with clearly designedperformance indicators that enabled a clearcomparison of targeted and actual results, as wellas of how far the municipalities had come from thebaseline. Mali I also had good M&E, although itsbaseline references were less clear. Uganda IIIimproved the evaluative capacity of the Ministry ofLocal Government, and Uganda IV helped theproject implementation unit consolidate project- specific information that made it possible toundertake an evaluation of project outcomes.

Mozambique II and Burkina Faso I providedresults frameworks with specific outputs andoutcomes, but the latter were not always measur-able, and baseline information was generally

missing. M&E systems of several MDPs wereweakened by a focus on project outputs; this andother factors resulted in the inadequate perform-ance of the operations. IEG found this to be casefor Guinea I, the Kenya MDP, Tanzania I, andthe Togo MDP.

Urban and spatial planningSeveral MDPs made important contributions to theurban planning capabilities of client municipalities.Under guidance and technical assistance throughMali I, five municipalities prepared strategic long- term physical and spatial plans that particularlyhelped them better understand the workings ofthe land markets in their cities. Through awholesale approach, Mauritania I introducedseveral management instruments, including urbanplans, inventories of assets, and priority invest-ment plans that were widely accepted by the localelected officials and staff of 13 municipalities.

Through Uganda II and III, the municipality ofKampala (population 1.4 million) prepared itsDrainage Master Plan and Urban TransportImprovement Strategy. Moving toward thewholesale model of MDP, Uganda IV helped 30municipalities make their existing municipalplanning committees more functional, resultingin 90 percent of them preparing 3- year develop-ment plans. In Benin I, in addition to the

Share of all MDPs with a project design focus on: Completed Ongoing

City planningIn objectives (%) 28 15

In components (%) 44 55

Municipal finance In objectives (%) 75 35


Service deliveryIn objectives (%) 94 85


Number of all MDPs 32 20Source: IEG special study.Note: MDP = municipal development project.

Table B.2: Municipal Management Focus of Region Portfolio


5 8

successful project efforts to strengthen urbanplanning at the municipal level, the centralgovernment itself adopted a declaration of urbanpolicy, through which it elaborated a coherent long- term strategy for urban planning, includingits environmental and sanitary aspects.

Investment planning and strategiesMDP results were thin in a Region where develop-ment projects often propose ad hoc planning andimplementation arrangements beyond the formalframework of central and local government. Thus,under Mozambique I, the five client municipali-ties that had prepared urban land- use andstructure plans saw none of them result in theintended municipal investment strategies.


Better financial managementThere is considerable evidence of MDP technicalassistance and support helping municipalitiesimprove their financial reporting and manage-ment. Under Mali I, for instance, three of theproject’s five targeted municipalities set upcomputerized accounting. More would have beenachieved had there not been a shortage of munici-pal staff qualified in finance and programming.This points to the need for more basic training inmunicipal financial management in the future.

Uganda I and II enhanced the Kampala munici-pality’s ability to plan, manage, and executecomplex investment decisions and report onthem, but roles and responsibilities for executingthem remained ill defined at the project’s end.This retail MDP focused on one municipality thatwas particularly prone to political interference in day- to- day operations. Uganda III and IV, incontrast, helped a much larger number of munici-palities strengthen and harmonize their planningand budgeting processes. The project- createdAudit Compliance Unit in the central governmentprovided a strong incentive for municipalities tobring their financial records up to date and makethem audit compliant.

Under Senegal I, a wholesale operation aimedat 67 municipalities throughout the country,

project audits confirmed that municipal financialbudgeting was placed on a sound footing in all ofthem for the first time. This MDP also preparedthe municipalities for incurring debt andmanaging debt service. In actual practice, theirfinancial management was solid, as reflected intheir being up to date in 95 percent of their loanrepayments. Tanzania I and Zimbabwe’s MDPhelped the targeted municipalities—10 and 21, respectively— routinely prepare and deliver up- to- date and audited accounts.

Mobilizing own revenuesEvidence is beginning to emerge that municipali-ties assisted by MDP operations strengthenedtheir revenue mobilization, perhaps more so thanunassisted municipalities, although robustevidence of control group performance remainsthin. Thus, under Uganda III and IV, some 30municipalities saw own revenues increase by 40percent in real terms, mainly because of computerprocessing of financial reports that highlightedarrears and areas of lax tax effort more promptly.Under Swaziland I, the 4 client municipalitiesextended property tax collection from 65 to 80percent of eligible properties following training oftax collectors, supervisors, and billing and financestaff. Tanzania I enabled municipalities to morethan double their own revenues, though from alow base, over the project period with the help ofmodern computer mapping and accounting. TheTogo MDP introduced improved collectionprocedures and penalties for nonpayment, acomputerized tax registry, and transparentmethods for assessing property values, but theimpact of all these measures on revenue collec-tion is not known. Technical assistance throughBenin II enabled its three municipal clients toexceed targeted own- revenue growth. BurkinaFaso I saw revenue collection nearly double inthe country’s two largest municipalities after theydeve loped a residential tax database with the helpof project technical assistance.

In contrast, elections dampened the political willto pursue energetic local tax collection inKampala under Uganda I and II. Similarconstraints prevented the five client municipali-ties of Mali I from actually collecting revenues


5 9

through a new urban tax that had been formallyinstituted but not acted on, leaving the munici-pal finances of Bamako (population 1.3 million)and Mopti (population 108,456) in deficit.

Aside from taxation itself, the muni cipality ofLomé (population 718,797) was able to create anew municipal agency with the help of the TogoMDP; this agency managed thriving local marketsand collected user fees. Similar user chargesenhan ced the revenues of six municipalities underGhana I. Progress with revenue mobilizationcontinued among municipali ties participating inGhana II and III but slowed somewhat in theface of municipal re sistance to setting usercharges high enough to cover costs. Revenue per -formance was weak under Nigeria I and leftmunicipal billing and collection machinery at thelocal level weak. Though Ethiopia I did adoptcost recovery for housing and some municipalservices, it made them less affordable to the poorin the short run.

Municipal creditworthiness and debt managementTo date, MDPs have done little to involve munici-palities in local credit markets, although morecan be expected in the future as incipientmarkets develop.

Private finance participationOnly modest results have been reported.Swaziland MDP’s attempts to bring togetherlocal governments and private sector financialinstitutions did lead to some private financing ofmunicipal services on a small scale in the capitalMbabane (population 76, 218). Under Maurita-nia I, better municipal management gave moreconfidence for private suppliers to work with the13 MDP municipalities in that country.

Improved Service Provision

Investment prioritiesOf 32 completed operations, more than halfreported economic rates of return (ERRs) at bothappraisal and completion, the highest propor-tion for any Region. ERR estimates at completionranged from 7 to 84 percent, exceeding appraisal

ERRs in half the cases. Estimates of internal ratesof return were more readily available for road-and water- related components. The main issuesrelating to ERRs were partial coverage of theinvestment, lack of credible data, and, still in afew cases, an apparent lack of appreciation forthe importance of conducting an economic cost- benefit analysis.

Ghana I reported the highest ERR of 82 percentafter detailed assessments showed the highimpact of below- cost road improvements in themunicipality of Accra (population 2.0 million)that accounted for one- third of the MDP projectcosts. Efficient upgrading of existing urban roadsin 11 other municipalities also explained the 68percent ERR reported for Ghana II, although inthis case the estimate covered only 13 percent ofthe total project costs.

Seven more successful MDPs reported high ERRsin the 25–40 percent range. Benin II also yieldeda 73 percent ERR, nearly five times the appraisalestimate, owing to much higher traffic volumesthan expected on the urban road improvementsthat accounted for most of the project costs. TheGambia MDP reported a 38 percent ERR forroad paving and drainage works in two munici-palities; again, these components accounted foronly about one- fifth of the total project cost.Burkina Faso I yielded a 33 percent ERR, thanksto the rehabilitation of urban roads and streets inthe country’s two main cities, components thataccounted for 40 percent of project costs.

Interestingly, Niger I found ERRs in the 20–41percent range for other components, such as busstations, slaughterhouses, public latrines, andleisure parks, using a beneficiary willingness- to- pay concept. A good cost- benefit analysisthrough Uganda II that carefully identified thecounterfactual and the flood- protection benefitsto the municipality of Kampala of reducingbuilding and infrastructure damage and of timesavings yielded a 25 percent ERR at completion.Tanzania I also yielded an ERR of 25 percent,based on the benefits of road and water supplyimprovements across 10 municipalities, account-ing for 55 percent of project costs.


6 0

Even though the Region has done the most cost-benefit analysis of all Regions, several MDPs did notinclude ERR estimates at completion even ofmunicipal infrastructure and service provisioncomponents that are amenable to cost-benefitanalysis. Lack of adequate data was most commonlygiven as the reason for not estimating ERR.

ProcurementMunicipalities’ experience in taking charge ofprocurement has been mixed, but there has beenprogress. Implementation of Mozambique I wasstalled by municipalities’ difficulties in meetingprocurement norms. In contrast, more recentlyTanzania I gave 10 municipalities the opportu-nity to manage procurement effectively for thefirst time.

Operations and maintenanceIEG found municipal clients of MDPs werebeginning to give more attention to operationsand maintenance (O&M). Under the Kenya MDP,for instance, four municipalities did more O&Mbecause of the incentive of special fundingprovided by the National Road Fund. Nearly allmunicipal water utilities funded O&M from theirown budgets under Tanzania I for the first time.Although Uganda II and III raised the profile ofO&M in the eyes of municipalities in Kampala, themunicipality was not able to raise all the revenueneeded to sustain this over the long run. Thesustained impact of improved municipal servicesunder Benin I and II is only ensured if the threeclient municipalities continue to mobilize thenecessary revenues on their own. However,municipal O&M continues to be weak in manymunicipalities, as the results of the Côte d’IvoireMDP, Swaziland I, and the Zambia MDP showed.

Services— Most affected sectorsMDPs provided a wide range of municipalservices, from low- income area upgrading; tourban road and street paving and drainageworks; to water and basic sanitation, otherenvironmental improvements, and urbantransport. In upgrading existing low- incomeareas, Burkina Faso I was particularly effective,using beneficiary participation. Under Benin I

and II, drainage works helped protect 403,000people in 6 municipalities against flooding.Ghana I and IV had an impact on 11 municipal-ities by rehabilitating urban roads and marketswhile increasing opportunities for smallbusinesses in vehicle repair and commerce.

Senegal I financed 421 basic infrastructuresubprojects across Senegal’s 67 urban municipal-ities and introduced systematic street addresssystems for the first time in 11 of them. TheZimbabwe MDP enabled 21 municipalities torehabilitate urban roads and extend water supplyand basic sanitation. Similar improvements tobasic sanitation under Guinea I and Mali I mayhave contributed to a steep decline in deathsfrom cholera. Water metering and networkimprovements enabled Swaziland I to reduce unaccounted- for water in four municipalities.Sewage treatment under Tanzania I meant thatthe effluent quality nearly met World HealthOrganization standards in three municipalities.Urban transport was less of an MDP focus inAfrica than in other Regions, but indirectlyUganda II brought considerable improvementsto traffic flows in Kampala by mitigating theimpact of regular flooding prior to the comple-tion of the Nakivubo channel.

Services— Private provisionMDPs made considerable efforts to engageprivate operators in the provision of municipalservice in several countries. Some results wereachieved, but there is still a long way to go. UnderUganda I and III, for instance, the fact thatnearly all the infrastructure investment wasprivately contracted was itself a significant resultfor the municipality of Kampala, which hadtraditionally done its own work. The use of forceaccount also ceased following Madagascar IIand the Zambia MDP. Ghana I and Senegal Istrengthened contractor and consulting in -dustries within their respective countries.

The competitive private sector approach tomunicipal service investment received a boostunder Tanzania I. That project, as well asBurkina Faso I, Ghana III, and Guinea I,provided openings for private solid waste collec-


6 1

tion operators, but a shortage of qualified locallabor was a major constraint in sustaining theseservices. Urban road and street maintenancecontracts under these projects attracted moreprivate sector bids.

Income level of beneficiaries— Poverty reductionAltogether, 60 percent of MDPs have objectivesexplicitly focused on the poor and on povertyreduction. Some MDP results were quite impres-sive. Benin I, Madagascar III, Niger I, andThe Gambia and Togo MDPs generated jobsfor the poor. These jobs came through projectconstruction and required more than 1.3 million person- days of labor- intensive employment,usually through local small and medium- sizedenterprises contracted for road and drainagemaintenance and other works.

Municipal services provided through BurkinaFaso I, Ghana I, Mali I, Mauritania I, Niger I,Nigeria I, and Tanzania I all served low- incomeand squatter settlements through infrastructureand water and sanitation, health facilities, accessroads, public lighting, school fencing, and green

spaces efforts. Under Ghana III, in all projectcities, lower- income communities benefited: theestimated number of beneficiaries for householdlatrines was 60,000, for school latrines 100,000,and for public latrines 30,000. In most cases, roadconstruction and rehabilitation opened accessbetween the poor neighborhoods and theeconomic centers of the cities and improvedscope for informal and small- scale income- generating activities.

Conclusions• Across Regions, MDPs in Sub- Saharan Africa

have kept the greatest focus on improving thelives of the poor.

• Increasing the number of wholesale MDPswould be constrained in a Region with few higher- level agencies that are ready to take onthe intermediation function that such opera-tions require.

• Across countries, MDP performance with M&Evaries considerably, pointing to opportunitiesfor fruitful exchanges of experiences. Theirperformance in the use of cost- benefit analysishas been relatively good, pointing to opportu-nities to apply the techniques in other Regions.

Benin: I— Urban Rehabilitation & Management; II— First Decen-tralized City Management; III— Second Decentralized City Man-agement. Burkina Faso: I— Urban Environment; II— DecentralizedUrban Capacity Building. Burundi: Public Works & Employment Cre-ation. Cameroon: Urban & Water Development Support. Chad:Urban Development. Côte d’Ivoire: Municipal Support. Ethiopia: I— Second Addis Urban Development; II— Capacity Building for De-centralized Service Delivery; III— Public Sector Capacity Building; IV— Urban Water Supply & Sanitation. The Gambia: Poverty Alle-viation & Capacity Building. Ghana: I— Second Urban Development; II— Local Government Development; III— Urban Environment &Sanitation; IV— Fifth Urban Development. Guinea: I— Third UrbanDevelopment (APL); II— Third Urban Development (Phase 2). Kenya:Urban Transport. Madagascar: I— Antananarivo Plain Develop-ment; II— Antananarivo Urban Works Pilot; III— Urban Infrastruc-ture. Malawi: Local Government Development. Mali: I— UrbanDevelopment & Decentralization; II— Second Transport Sector.

Mauritania: I— Urban Infrastructure & Pilot Decentralization; II— Urban Development Program. Mozambique: I— Local GovernmentReform & Engineering; II— Municipal Development; III— MaputoMunicipal Development Program. Niger: I— Urban InfrastructureRehabilitation; II— Local Urban Infrastructure Development. Nige-ria: I— Oyo State Urban Development; II— Lagos Metropolitan De-velopment & Governance. Rwanda: Urban Infrastructure & CityManagement. Senegal: I— Urban Development & DecentralizationProgram; II— Local Authorities Development Program. South Africa:Municipal Financial Management. Swaziland: I— Urban Devel-opment; II— Local Government. Tanzania: I— Urban Sector Reha-bilitation; II— Local Government Support. Togo: Lomé UrbanDevelopment. Uganda: I— First Urban Development; II— NakivuboChannel Rehabilitation; III— Local Government Development Pro-gram; IV— Economic & Financial Management; V— Kampala In-stitutional & Infrastructure Development. Zambia: UrbanRestructuring. Zimbabwe: Urban Sector & Regional Development.

Box B.1: Key to MDPs Referred to in Text

Source: IEG.

6 3

Bank SupportEast Asia and Pacific has the largest urban popula-tion of any Bank Region; 805 million people—42percent of the total population— live in cities, andurbanization continues at a rapid pace, with thepopulation in cities growing by 2.9 percentannually. The World Bank, through 44 MDPsactive during the 1998–2008 decade, committed$5.7 billion to urban development. This portfolioaimed to strengthen the management of 445municipalities across 16 countries. By number ofMDPs, the most active borrowers were China (23projects) and Indonesia (12), followed by a newerMDP borrower, Vietnam (4), and an older one,the Philippines (3). In addition, Mongolia andKorea hosted 1 MDP each. Thus, nearly all of theRegion’s MDPs, 86 percent, were in lower- middle- income countries. Countries in this Region withlarge urban populations (15 million plus) but no Bank- financed MDPs are Thailand and Myanmar.

Portfolio PerformanceOn average, MDPs in the Region are strongperformers, with 80 percent achieving satisfac-tory outcomes. Also, 90 percent had satisfactoryBank performance, and 83 percent had satisfac-tory borrower performance, all well aboveaverages for the worldwide MDP portfolio.

Although no project had an outcome rating ofhighly satisfactory, there are numerous examplesof successful MDPs in several countries in thisRegion that can serve as models for MDPselsewhere. China III, thanks to an outstandingmunicipal team in the megacity of Tianjin(population 10.3 million), succeeded on severalfronts, building and operating a solid wastesanitary disposal facility that became a model for

China, increasing sewage collection andtreatment, improving traffic management, andconsolidating municipal planning capability.Philippines I was particularly successful atupgrading low- income areas by developing newlocal markets and in training 9,129 staff from 74municipalities. These efforts resulted in a newmunicipal management style that was betteradapted to the increasing responsibilities underdecentralization. China IV got good resultsacross the board in Zhejiang Province by improv-ing municipal management as it related to urbanplanning, land development, and environmentin key cities that offered among the best invest-ment climates in China. Indonesia V producedgood results in five municipalities in Kaliman-tan, particularly through the successfulKampung Improvement Program in Pontianak(population 455,173).

APPENDIX C: BANKING ON MUNICIPALITIES: WORLD BANK SUPPORT IN EAST ASIA AND PACIFIC




IBRD commitments (US$ million) 4, 512

IDA commitments (US$ million) 1, 158







Table C.1: Summary of MDP Portfolio, 1998–2008


6 4

Under Indonesia XIX, which focused onmunicipal innovations, the municipality of Bogor(population 769,000) was particularly successfulin developing a lively and informative public Website that was an online version of the earlierpublic information booths. There were signifi-cant environmental gains in improved watersupply and sewage and solid waste disposalthrough China VII and VIII in Shanghai andLiaoning, respectively. Directly focused onretooling municipal management, China IX waseffective in helping four municipalities managethe deep structural reform involving the divesti-

ture of enterprise housing. The Mongolia MDPhelped develop the country’s capability todesign, build, and operate urban servicesthrough the successful improvements theproject brought to the water supply of themunicipality of Ulaanbataar (population844,818).

Some MDPs in the Region performed poorly.Indonesia III achieved little, as the repeatedturnover in municipal leadership in Surabaya(population 2.4 million) undermined commit-ment to agreements. This led to inaction onservice provision to the city that should havecalled for a thorough project reappraisal. ChinaXI did not lead to the hopedfor reduction of airpollution by replacing old industrial plants in themunicipality of Chongqing (population 32million) because of the slow divestiture of suchplants and the cancellation of the project’s creditcomponent. Pollution did decline, but notbecause of the project; instead, that improve-ment occurred because of the slow- down ofindustrial activity.

Indonesia VII intended to improve solid wasteand sewage treatment services in 41 municipali-ties in Sulawesi, but it was only partlyimplemented because of lack of governmentcommitment and concerns over misuse of funds.This left municipalities without the managementprogress intended, especially in operations and

8090

8374 78 75

0

20

40

60

80

100


Perc

ent s

atis

fact

ory

East Asia and Pacific Region All Bank

Figure C.1: MDP Portfolio Performance, Fiscal 1998–2008





Municipal financeIn objectives (%) 60 21





Table C.2: Municipal Management Focus of Region Portfolio

A P P E N D I X C : WO R L D B A N K S U P P O R T I N E A S T A S I A A N D PAC I F I C

6 5

maintenance (O&M), the project’s model ofwhich was too complicated to administer.


More informationSome MDPs in the Region improved the informa-tion available to municipalities. Under China III,for instance, the municipality of Tianjin (popula-tion 10.3 million) was able to develop a real- timeinformation system for the megacity’s intensetraffic. Under Indonesia IX, Bogor’s success inassembling and disseminating information onmunicipal services publicly on the Web is animportant information system achievement.

Monitoring & evaluationM&E systems exhibit the weaknesses found inother Regions and sectors. Thus, under China IVin Zhejiang Province, an operation that excelledin many other respects, M&E did little more thancount and cost the delivery of individual subpro-jects. China XII, in Yunnan Province, did a littlebetter with monitoring the project’s physicalachievements, but it fell short on verifyingprogress on the institutional front. M&E forIndonesia II and VI ventured little beyondcounting the number of subproject contractsawarded and the amount of disbursements. Thismeant that M&E was able to provide preciseinformation about the number of communitytoilets built and their exact unit costs, but not howmuch those facilities were used— which provedto be very little. The Independent EvaluationGroup (IEG) saw communal toilet blocksdesigned for 15 families being used by only 1 or 2.

China X in Hubei had weak M&E, as the indica-tors were defined too broadly to be measurable.The M&E of China IX suffered the classicshortcoming of not providing baseline values for38 indicators that were selected to measureprogress in divesting state- owned enterprise housing.

Even when indicators are good, M&E problemscan arise. This happened when measuring waterquality of the environmentally stressed HuangpoRiver in Shanghai under China VII. Data on

baseline and endline water quality were available,but the samples were drawn from differentlocations on the river. Furthermore, the monitor-ing station built under the project was not fullyoperational. An implementation weakness under -mined the effectiveness of M&E for China XV inSichuan Province, where records of measure-ment of the well- designed performance indica-tors were not systematically kept.

But even when its design is weak, M&E can beimproved during implementation, as when thestrong local team of China III in Tianjin, at itsown initiative, incorporated outcome indicatorsto measure greater municipal managementeffectiveness, which had been overlooked by theinitial M&E design. Finally, one of the mostcomplete M&E systems was introduced throughVietnam I, where four project municipalitiesused indicators that that covered all aspects ofimprovement in water supply service, rangingfrom physical provision to management efficiency.

Urban and spatial planningMDPs have achieved a lot, especially in China,where many local municipalities have embracedcity planning in recent years. Thus, under China I,the Urban Master Plan of Beijing (population 14.9million) incorporated for the first time environ-mental priorities of the municipal environmentalprotection bureau. China III helped Tianjinprepare its Master Plan and consolidate it with theindicative budget for 2005–20, again for the firsttime. Particularly for the city of Ningbo inZhejiang Province, China IV strengthened its long- term land use planning through technicalassistance and firmly embedded the conservationof historic and cultural monuments into its citycenter planning, now recognized as one ofChina’s best.

On the urban transport side, two operations,China II and VI, enabled Shanghai (population14.6 million) to improve its transport planningby providing expert input. Other countries alsosaw some of their city planning improve throughMDP assistance. Thus, under Indonesia X, ninemunicipalities prepared local environmental


6 6

plans and strategies for the first time, one ofwhich was the critical Drainage Master Plan forJakarta (population 8.5 million). ThroughPhilippines I, some 70 municipalities learnedhow to incorporate specific investments of theoperation in subprojects into their local city plans.

More innovative approaches to planning did notalways succeed. Under Indonesia II support forIntegrated Urban Infrastructure DevelopmentPlanning, a modernized and multisectoralapproach to planning, made only modestinroads in smaller municipalities, which found ittoo complex and were more comfortable withthe traditional sectoral approach they knew well.

Investment planning and strategiesMDPs in the Region generally did not requireclient municipalities to strengthen the manage-ment of their investment planning and strategies.Larger municipalities in particular often had theirown investment plans in place before the MDP.


Better financial managementIn China, a number of municipalities improvedtheir financial management and accountingprocedures with the help of MDPs. China III,for instance, helped the mega- municipality ofTianjin integrate different financial networksacross its very large organization, where comput-erization of all accounts within a local areanetwork has now become standard. China Vhelped improve cost recovery for water supply,allowing four municipalities in southern JiangsuProvince to cover operating, if not investment,costs. The municipal audit bureau of Shanghai(population 14.6 million) was quickly able toadopt international accounting standards, asrequired by China VII.

Mobilizing own revenuesRevenue enhancement through MDPs in theRegion focused particularly on increasing directcost recovery from the project investmentsthemselves, rather than seeking broader

improvements in general revenues. In practice,cost recovery has been as challenging in thisRegion as in others. China VII was unable toraise tariffs enough to enable five municipalsanitation companies in Liaoning Province tocover their operating costs. China XII did notenable the five municipal sanitation utilities inYunnan Province to achieve full cost recovery,but there has been some progress in tariff adjust-ment. China X reported similar constraints inlimiting cost recovery, but for solid wastemanagement in Hubei Province.

Although the details are scarce, Indonesia II didreport enhanced revenue collection among the45 municipalities assisted by the project in EastJava and Bali. Under Indonesia IV, however,inflation eroded effective cost recovery ofmunicipal water utilities in Semarang (popula-tion 1.3 million) and Surakarta (population555,308). Mongolia’s MDP achieved a lot onthe municipal finance front, but not the fullfinancial autonomy for the municipal sanitationutility of Ulaanbataar (population 844,818)promised by the project’s ambitious objectives.Nevertheless, computerized billing worked welland considerably enhanced tariff collections.Philippines I achieved significant results acrossthe 74 client municipalities, especially throughproperty tax cadastres that more than doubledassessed values; actual tax collections increasedby 64 percent over the 1994–2001 project period.Vietnam’s MDP enabled municipal watersupply utilities in Hanoi (population 1.4 million)and Haiphong (population 602,695) to covertheir O&M costs and even build up some reserves.

Municipal creditworthiness and debt managementThis aspect of municipal management wasexplored on a small scale. Philippines I’sMunicipal Development Fund established a long- term credit window that loaned $34 million toeligible municipalities. Although the lending wassmall scale relative to municipal needs, the creditmechanism did introduce 74 municipalitiesacross the country to debt service management.


6 7

Private finance participationIEG found little evidence of significant effort byMDPs to enhance private finance for municipalservices in the Region.


Investment prioritiesSome 60 percent of MDPs provided ERRestimates for project investments at appraisaland completion. They were widely used forMDPs completed in China. China XIII yielded a39 percent ERR based on users’ willingness topay for sanitation services in 38 municipalities inShandong Province. China II and VII reportedERRs of 28 percent, from the benefits of time andoperating cost savings from improved trafficflows in Shanghai. China III led to ERRs of 23percent Tianjin, based on benefits accruingprincipally from urban land development forhousing and industrial uses.

But a more robust economic analysis, distin-guishing new businesses from those that hadsimply transferred to the China IV project areain Shaoxing (population 421,283) in ZhejiangProvince, would have evaluated the project’sland development more precisely. SatisfactoryERRs in the 14–18 percent range were reportedelsewhere through Mongolia’s and Vietnam’sMDPs, as well as Indonesia II. In some cases,unpersuasive reasons were given for projectteams not estimating even simplified internalrates of return. Thus, excessive cost and timeneeded were cited as reasons for not estimatingan ERR for China VII, despite the high cost ofthe project investment incurred to improve thequality of the water supply to Shanghai.

ProcurementMDPs in this Region reported few significantresults, as far as changes in procurement practiceat the municipal level are concerned. Oneexception was China III, through which themunicipality of Tianjin conducted successfulinternational competitive bidding to establishthe Shuangkou solid waste disposal site, China’sfirst fully sanitary landfill, complete with an

onsite leachate treatment plant that became amodel operation for the country.

Operations and maintenanceThe results of some MDPs call for municipalitiesto pay more attention to assuring financing forongoing operations of existing infrastructure andmunicipal services. For instance, a municipalwater supply system provided under Indone-sia II for the Kintamani district of Denpasar(population 405,923) in Bali fell into disuse, asthe local authorities could not afford to pay tooperate the necessary pumps for more than afraction of the time needed. In Sulawesi,Indonesia VII’s 41 client municipalities wereunable to adopt the project’s “performance- oriented maintenance management systems,”which they found too complicated. In WesternJava Indonesia X saw that continuing uncertain-ties about the funding mechanisms for municipalwaste management corporations put theproject’s urban environmental achievements atrisk. China XV, completed in 2007, reported thatthe four beneficiary municipalities needed toraise more revenues to ensure O&M funding.

Services— Most affected sectorsMDPs in this Region provided support to munici-palities to improve services connected withwater supply, basic sanitation, and other environ-mental improvements. In addition, they helpedimprove urban transport, through new urbanroads, street paving, and drainage and trafficmanagement measures. The upgrading of low- income areas through the introduction of basicinfrastructure continues through MDPs in EastAsia, but on a smaller scale than before, and elsewhere.

To improve the municipal management of watersupply, MDPs made some notable achievementsin China in particular. A significant environmen-tal and public health gain for more than 8million inhabitants of Shanghai was the result ofChina VII’s provision of a safer water supply.This was done by implementing a major intakeupriver in less- polluted reaches of the environ-mentally stressed Huangpo River, as well as


6 8

implementing mitigation measures in solidwaste collection and disposal and restrictinguse of agricultural fertilizers to prevent runofffrom further polluting the river. Major munici-pal water treatment plants under China IVimproved service quality to the people living inkey cities of Zhejiang Province, Hangzhou(population 1.9 million), Ningbo (population719,867), and Wenzhou (population 865,672).

MDPs also improved basic sanitation. Theinnovative, low- cost, small- scale “modular”approach to sewage treatment was adopted bythe municipality of Malang (population 747,000)under Indonesia VI. However, its success waslimited, as low- income residents continued todischarge sewage without charge into stormdrains, rather than paying the (modest) feeimposed by the new system.

MDPs made more progress helping municipali-ties improve their solid waste management,especially in the final disposal of waste. China IIIled to the building and operation of the country’sfirst sanitary landfill at Shuangkou near Tianjin— now considered a successful model nationwide.This experience built on earlier successful effortsto improve solid waste disposal in Beijing underChina I. Under China X, the municipality ofXianfang (population 462,956) in Hubei Provincesucceeded in disposing of 100 percent of itscollected solid waste in a sanitary landfill built bythe project. Indonesia V introduced controlledlandfills to five municipalities in Kalimantan thatalso closed down their earlier unsanitary dumps,which had polluted the surface water of nearby settlements.

MDPs made significant improvements to urbantransport. The municipality of Shanghai was ableto complete its high- capacity inner ring roadunder China II, an operation like others in thecountry that paid little attention to publictransport. Mostly through traffic managementimprovements, with construction limited towidening existing streets, China III introducedbetter traffic surveillance and monitoring to thecity of Tianjin. Traffic management was a priorityunder Vietnam II, too, especially through the

successful introduction of computer- controlledtraffic lights in Hanoi (population 1.4 million),which led to average trip time savings of 30percent, well above the 10 percent targeted.

Compared with other Regions, MDP coverage ofslum upgrading was thin, although Philippines Isupported investments in this area across 74municipalities. This focus was also found underIndonesia V in Kalimantan. The most successfulcomponent of the latter MDP was the KampungImprovement Program in five municipalities, aprogram component that the Bank hassupported for more than two decades through-out the country. In contrast, an activity supportedin East Asia but not found often elsewhere wasthe support under China I that enabled twice thecoverage by Beijing’s interconnected districtheating network.

Services— Private provisionVery few MDPs focused attention on expanding therole of the private sector in providing municipalservices. Consequently, IEG found few examples ofsignificant results in this area within the Region.Efforts were made through some MDPs tostimulate private commercial operations in servicedelivery by closing old municipal service depart-ments and replacing them with agencies, such asthe Beijing Drainage Company under China I andthe Shanghai Public Transport Company underChina II, but these new enterprises remainedfirmly harnessed to the state sector. The first stepstoward a private- public partnership in watersupply at the municipal level were taken inShandong Province under China XIII.

Income level of beneficiaries— Poverty reductionFew MDPs in the Region— mostly those in Indone-sia and the Philippines— focused clearly on theurban poor. Indonesia VIII, for instance,supported 18,000 infrastructure microprojectsthat had been identified by participatorycommunity development plans in low- incomeurban areas. Indonesia V, particularly through itskampung improvement program of upgrading low- income areas with basic services, is estimatedto have benefited nearly half the population of the


6 9

five client municipalities in Kalimantan. Philip-pines I started out with a strong focus on benefit-ing the poor, but this became less clear in the faceof incentives for municipalities to embark on revenue- generating subprojects that wouldbenefit higher- income groups.

Conclusions• In countries with unitary municipal adminis-

trations for very large cities (even megacities),such as China, the retail approach to strength-ening municipal management can be an ap-propriate model.

• MDPs have enabled many municipalities tostrengthen their management of service pro-vision, especially for improving the urban en-vironment. The sectoral focus varies acrosscountries in the Region, pointing to possibili-ties of fruitful exchanges of successful experi-ences among them.

• Results in strengthening municipal financeshave been less evident across this Region, call-ing for more MDP efforts to enhance revenuemobilization for municipalities to fund theO&M necessary to sustain the service provisionachievements obtained thus far.

China: I— Beijing Environment; II— Shanghai Metropolitan Trans-port; III— Tianjin Urban Development Project; IV— Zhejiang Mul-ticities Development; V— Southern Jiangsu EnvironmentalProtection; VI— Shanghai Environment; VII— Second ShanghaiMetropolitan Transport; VIII— Liaoning Environment; IX— Enterprise Housing and Social Security Reform; X— Yunnan En-vironment; XI— Hubei Urban Environment; XII— ChongqingIndustrial Pollution Control and Reform; XIII— Shandong Envi-ronment; XIV— Liaoning Urban Transport; XV— Sichuan UrbanEnvironment; XVI— Chongqing Urban Environment; XVII— UrumqiUrban Transport; XVIII—Shijiazhuang Urban Transport; XIX— Tianjin Second Urban Development; XX— Wuhan Urban Trans-port; XXI— Chongqing Small Cities Infrastructure Improvement; XXII— Liuzhou Environment Management; XXIII— Second Shang-

hai Urban (APL). Indonesia: I— Sulawesi– Irian Jaya Urban De-velopment; II— East Java/Bali Urban Development; III— Semarang Surakarta Urban Development; IV— Surabaya UrbanDevelopment; V— Kalimantan Urban Development; VI— SecondEast Java Urban Development; VII— Second Sulawesi UrbanDevelopment; VIII— Urban Poverty; IX— Municipal Innovations; X— Western Java Environmental Management; XI— SecondUrban Poverty; XII— Urban Sector Development and Reform.Korea: Pusan Urban Transport. Mongolia: Urban Services Im-provement. Philippines: I— Third Municipal Development; II— Local Government Unit Finance and Development; III— Supportfor Strategic Local Development and Investment. Vietnam: I— Water Supply; II— Urban Transport Improvement; III— UrbanUpgrading; IV— Coastal Cities Environmental Sanitation.

Box C.1: Key to MDPs Referred to in Text

Source: IEG.

7 1

APPENDIX D: BANKING ON MUNICIPALITIES: WORLD BANK SUPPORT IN EUROPE AND CENTRAL ASIA

Bank SupportMore than 280 million people live in cities in theEurope and Central Asia Region—64 percent ofthe total population. Through 28 MDPs activeduring the 1998–2008 decade, the World Bankmade commitments of $1.7 billion. This portfo-lio aimed to strengthen the management of 292municipalities in 16 countries. By number ofMDPs, the most active borrowers were theRussian Federation (4 projects), Bosnia andHerzegovina (4), Georgia (3), Turkey (3),Kazakhstan (2), and the Kyrgyz Republic (2). Theremaining 10—Armenia, Croatia, Kosovo, Latvia,Lithuania, Poland, Tajikistan, Turkmenistan,Ukraine, and Uzbekistan— hosted 1 MDP each.Thus, half of the Region’s MDPs were in upper- middle- income countries. Countries in theRegion with large urban populations but no Bank- financed MDPs are Romania and Belarus.

Portfolio PerformanceSixty- three percent of completed MDPs achievedsatisfactory outcomes, and the percentage ofsatisfactory Bank and borrower performance wasa little higher. These figures are somewhat belowthe Bank- wide averages.

Among successful cases in the Region’s portfo-lio, Russia IV stands out for its outcome ratingof highly satisfactory. The project considerablystrengthened the financial management of thenewly created municipality of Kazan (population1.2 million) by helping local officials organize andunify municipal accounts, debts, and otherobligations. The municipality turned a deficitinto a small surplus. Outstanding payables, amajor problem at the outset, were substantiallyreduced. Two- thirds of the project funding wasused for urgent repairs to abandoned and

derelict schools and health centers, bringingthem back into full use. Kazan considerablyimproved its asset management, divesting someunnecessary inventory. Real estate assets remain-ing on the municipal books are now leased at 90percent of their market values, up from 50percent prior to the project.

Other successful examples included Bosnia andHerzegovina I, which helped develop a munici-pal credit market as intended. In the process, itstrengthened financial management both by themunicipalities and by five commercial banks thatentered this market for the first time, making 28loans for $13.3 million.

Turkey I helped improve the efficiency of wateruse in municipalities by substantially improvingworker productivity per connection as well as the bill- collection ratio. The Kyrgyz Republic project




IBRD commitments (US$ million) 1, 496

IDA commitments (US$ million) 207







Table D.1: Summary of MDP Portfolio, 1998–2008


7 2

was successful in separating road planning,budgeting, and contract administration from roadconstruction, as intended, and it helped themunicipalities of Jalalabad, Bishkek, and Osh setup their own passenger transport authorities toplan, contract, and monitor the private provisionof local services. Georgia III helped nine munici-palities, housing three- quarters of the country’surban population, to become creditworthy andparticularly increased the effectiveness of theirdelivery of street paving and water supply services.

At the same time, three MDPs turned in a weakperformance. The Poland MDP had little impact

on the development of a commercial credit marketfor municipal investment. Its performance wasundermined by a 1999 Finance Law that preventedmunicipal borrowing, and the project disbursedvery little. The Kazakhstan project failed topromote efficient municipal management of socialassets divested by state- owned enterprises, mainlybecause the project gave insufficient attention tothe financial challenges the municipalities faced.The Lithuania MDP did not strengthen munici-pal management, as intended, because the Associ-ation of Local Authorities of Lithuania, slated as theexecuting agency for the project, lacked thenecessary capability and resources to perform thisfunction effectively.


More informationFew MDPs in the Region aimed specifically tostrengthen information systems for municipalmanagement and planning. But the need forsuch information was acute in Georgia in themid-1990s; three successive MDPs in thatcountry introduced computer equipment andmade municipal information more transparentunder the law. However, much progress remainsto be made, and Georgia can learn from othergood experiences, such as Chile’s InformationSystem on Municipalities. Russia I successfullycreated a territorial information and analyticsystem for land, real estate, and infrastructureand a developer’s manual, both of which have

63 69 6974 78 75

0

20

40

60

80

100


Perc

ent s

atis

fact

ory

Europe and Central Asia Region All Bank

Figure D.1: MDP Portfolio Performance, Fiscal 1998–2008










Table D.2: Municipal Management Focus of Region Portfolio

A P P E N D I X D : WO R L D B A N K S U P P O R T I N E U R O P E A N D C E N T R A L A S I A

7 3

had a positive impact on the housing market andhave become valuable references for potentialforeign investors and other parties.

Monitoring and evaluationM&E design in MDPs, as for many other projects,was often weak because it focused on the deliveryof component outputs rather than achievingproject outcomes. Even some output indicatorswere not always clear or measurable, lackingbaseline and endline (target) data. Thisprevented M&E implementation and use as afeedback mechanism to inform and improveproject performance. M&E in the Latvia andKosovo projects did not distinguish betweenoutput and outcome indicators. These projectsespecially lacked those indicators that couldmeasure actual institutional improvementsagainst those planned and relied too heavily onusers’ opinions, expressed through beneficiaryassessments. Performance indicators could notbe measured for lack of data— for example, onmunicipal action plans in the Kyrgyz RepublicMDP or on municipal finances under the firstGeorgia MDP. During Georgia II and III, M&Efocused more on management information onmunicipalities, and this information slowlybecame available in the country. The lack ofbaseline data undermined the effectiveness ofM&E in the Turkey MDP, which did not explicitlycite preproject levels of pollution in the Sea ofMarmara, for instance.

The Kazakh MDP’s M&E could not capitalize ontechnical assistance relating to financial manage-ment methods and was unable to producemonitoring data on a continuous basis tomeasure its operating performance. Russia IV inKazan, in contrast, incorporated a strong M&Esystem, whose design included easy- to- measureindicators such as municipal debt and level oftargeted cash- transfer subsidies that were alsopart of the conditions of tranche release of thestructural adjustment design of this loan.

Urban and spatial planningMost MDPs in the Region did not include signifi-cant urban planning activities. An exception wasRussia I in St. Petersburg, which drafted several

laws to aid the planning process in improving theavailability of serviced land. The laws had yet tobe ratified at the time of project completion.

Investment planning and strategiesBy training 550 staff in 30 municipalities, Kosovo’sMDP helped prepare five- year rolling financialplans for the first time and incorporatedcommunity inputs through participatory pro -cesses. The Kyrgyz Republic MDP helpedJalalabad and Bishkek municipalities prepare plansfor financing and contracting urban road building,which was separate from the construction itself.


Better financial managementSeveral MDPs achieved positive results in thisarea. Among the most notable was Russia IV inKazan, where project technical assistance helpedlocal staff unify the municipal accounts for thefirst time and make them more transparent.Among other things, computerized accountsallowed Kazan’s employees to receive theirsalaries on time.

Other experiences show that equipment, technicalassistance, or training alone is not enough toensure better financial management. Initially,municipalities made limited use of computingequipment provided through Georgia I.Uzbekistan’s MDP did not lead to improvedfinancial management by the municipal solid wasteutility, despite the technical assistance provided.The actual modernization of municipal manage-ment did not occur in the Turkmenistan’s MDPwithout the widely expected devolution of respon-sibilities to the local level. Under the Kazakhstanproject, municipal water utilities did not implementmodern financial management techniques, forwhich training was provided, because of disconti-nuities in leadership. Similar weaknesses in theLatvia MDP were partly overcome by twinning theDaugavpils city water utility with the water works ofthe city of Tampere in Finland.

Mobilizing own revenuesThe Region’s MDP portfolio obtained some goodresults in this area. Technical assistance through


7 4

Bank supervision of Russia IV helped Kazan findalternative sources of revenue, which wereurgently needed, as large federal transfers weresoon to lapse. Municipalities participating inGeorgia II increased own- source revenues by11 percent during 1998–2002, compared with adecrease of 16 percent for other municipalities,thanks in part to the incentives to become creditworthy.

The Kyrgyz Republic MDP helped establishurban road funds and raised taxes threefold onprivate vehicles. The Uzbekistan MDP did notsucceed in establishing a stable self- financingmechanism for solid waste management throughplanned tariff adjustments, although a last- minutereprieve prevented the operation’s collapse.Specific cost recovery from MDP investmentsthemselves obtained good results in the TurkeyMDP, where the Bursa municipal water andsanitation utility successfully maintained tariffs atlevels sufficient to meet its financial obligations.

Municipal creditworthiness and debt managementUnder Georgia I the Municipal DevelopmentFund of Georgia became Georgia’s main fundingsource for municipalities and an instrument forstrengthening municipal management. Todaythe fund has become the government’s principalagency for financing major developmentprograms beyond just municipal development.The latest Georgia III project saw 9 municipali-ties, home to 73 percent of the urban popula-tion, become creditworthy, also giving some ofthem access to additional concessional funding.Under the previous Georgia II, some munici-palities had overborrowed and defaulted on theirloan repayments.

Technical assistance to 32 municipalities and,most significantly, to 5 commercial banks underBosnia and Herzegovina I helped municipali-ties become more creditworthy by increasingrevenue collections and helping banks under -stand their debt portfolio better. The govern-ment supported a similar approach under theKyrgyz Republic MDP, partly in the hope ofreducing the financial burden of subsidies. The

potential foreign exchange risk inherent inexternal funding of municipal credit does notappear to have constrained municipal creditwor-thiness in the Region. In Poland, with memoriesof that country’s recent high inflation, commer-cial banks tried unsuccessfully to transfer this riskto equally reluctant municipalities, stallingproject implementation. But in most othercountries, central governments and their agentshave been willing to assume this risk.

Private finance participationTo date, MDP efforts and results in getting privatefinance into municipal services have been limited.Poland’s project did not succeed in channelingprivate bank funds into municipalities because ofthe overall project failure. Private funding ofTashkent’s solid waste management underUzbekistan’s MDP could not proceed while theoperation continued to operate at a loss. Privatefunding of Kazan’s water utility under Russia IVwas held back because the public operatorreceived support from other donors.


Investment prioritiesAs in other Regions and sectors, few MDPs gavemuch attention to estimating ERRs, either atappraisal or at completion. To be eligible forproject financing, Bosnia and Herzegovina Idid require participating municipalities todemonstrate that subprojects achieved at least12 percent ERR, but information on the actualrates achieved was not systematically monitored.After a poor start in neglecting ERR estimates inits earlier projects, Georgia III ensured that allsubprojects met minimum rates of return. TheKyrgyz Republic MDP also reported, thanks tohuge savings in operating costs, high ERRs for itsurban road investments in Bishkek, Jalalabad,and Osh.

Although the project itself was barelyimplemented, the Poland MDP did leadcommercial banks to require that municipalities’proposals for subprojects meet minimum ERRrequirements. Unconvincing reasons for the lackof ERR estimates in this Region are similar to


7 5

those given in other Regions, such as Bankguidelines not requiring them for emergencyprojects and lines of credit.

ProcurementSeveral MDPs gave municipalities a first opportu-nity to become involved in the competitiveprocurement of works and services, with somepositive results. Under Russia IV, Kazan munici-pality voluntarily adopted local competitive bidding— not required with a structural adjust-ment loan— which resulted in lower-price con -tracts. Sole- source purchasing was reduced from55 percent of total to 25 percent, which wasbetter than the target of 35 percent.

The Uzbekistan project introduced interna-tional competitive bidding to Tashkent munici-pality that resulted in significant cost savings inthe acquisition of a new fleet of 270 solid wastecollection vehicles. Georgia III enabled munici-palities to play a greater role in procurementthan had been possible under the earlieroperations, although local management therestill needs to be strengthened more. Morecentralized political arrangements in some of theRegion’s countries leave procurement as agovernment responsibility.

Operations and maintenanceThere was little evidence of MDP attention toongoing O&M. In the Kyrgyz Republic MDP,however, each participating municipality had toadopt a prioritized five- year road maintenanceprogram for its urban roads and streets. TheUzbekistan MDP established a repair andmaintenance depot, generously equipped withspare parts at the outset, to keep the fleet of newsolid waste collection vehicles on the road. Underthe Latvia MDP, the lack of such a facility put thevehicles of one major Riga bus company at risk.

Services— Most affected sectorsAmong all activities undertaken, the projectsperformed best in improving services and relatedinfrastructure, especially for urban street pavingand drainage, neighborhood upgrading, and basicsanitation and other environmental improve-ments. Georgia II successfully completed 89

subprojects in 11 municipalities, with two- thirdsof project investment in Tbilisi and three- fourthsof investments in road rehabilitation and watersupply. The greatest improvements were in urbanroad paving and clearing blocked drainage thatcaused periodic flooding.

Under Georgia I, during the country’s post- independence phase, infrastructure and serviceswere preserved and improved for power,heating, and water. The Tbilisi Metro, which isused by 90 percent of the city’s population, was“rescued” through the emergency funding ofsignaling systems and spare parts. Dysfunctionalsewerage systems in the municipalities of Batumiand Poti were restored to working order, thoughservice levels were still short of desired goals.Restoring heating to hospitals and schoolbuildings in five municipalities allowed contin-ued operation throughout the winter.

The Kazakhstan project allowed the water utilityto supply good quality drinking water to 37,000people in two municipalities. The project’scleanup of sewage spills had an immediate healthimpact: between 1999 and 2002, the number ofdysentery cases fell from 83 to 8, and typhoidcases fell from 83 to 0. According to the benefici-ary assessment of the Kosovo MDP, 90 percentof respondents felt that the 115 (mostly schooland water supply) subprojects implemented in30 municipalities did respond to their needs; 84percent were satisfied with results that they felthelped reduce water- borne diseases.

The Kyrgyz Republic project exceeded itstarget by substantially improving 105 kilometersof roads, making the municipalities of Jalalabad,Bishkek, and Osh more accessible. Latvia’sMDP helped improve drinking water quality, anduntreated water was no longer being dischargedinto the Daugava River. The Lithuania MDPcontributed only modestly to improvement inconditions of municipal service infrastructure,but it did help reduce street lighting energy costsin Vilnius. The Turkmenistan project helpedincrease bus and trolley services in Ashgabat,fully meeting targets, as well as contributing toenhanced reliability and frequency of services.


7 6

Wastewater collection systems and networkswith 80 percent treatment were put in place inBursa city under Turkey I, which led toincreased service coverage between 1993 and2000—from 93 percent to 97 percent of thepopulation for water supply and from 73 percentto 82 percent for sewerage. Over the sameperiod, unaccounted- for water fell from 65percent to 45 percent. Uzbekistan obtainedpositive environmental results by helping restoreTashkent’s solid waste management system. Forthe most part, the environmental impacts ofroads, water, and solid waste subprojects ofGeorgia I–III were positive, especially throughimproving air and water quality in the poorerneighborhoods of the beneficiary municipalities.

Services— Private provisionPrivate provision of services was relatively lowamong this Region’s MDPs. The Kyrgyz RepublicMDP helped three municipalities— Jalalabad,Bishkek, and Osh— establish passenger transportauthorities and plan, contract, and monitor theprivate provision of services. Construction is nowawarded to private contractors, and majorequipment has been sold to the private sector.

Under Turkey I, the Bursa Metropolitan Munici-pality contracted waste collection and landfilloperations to private contractors. It alsopromoted private participation for meter

reading, billing, and invoicing. The Turk -menistan MDP helped increase private partici-pation of the suburban and intercity transport to70 percent, but greater effort could have beenmade at deregulation, to allow greater competi-tion. Under the Uzbekistan project, an interna-tional tender in 2003 for the private operation ofMakhsustrans’s Chilanzar and Shaihantaurdistricts of Tashkent failed to yield any bidsbecause of doubts about the profitability of theoperations without subsidies.

Income level of beneficiaries— Poverty reductionOnly four MDPs have explicit poverty- reductionobjectives. The Kosovo MDP broadly met itsRegional poverty goals. The project directed themajority of its social services toward the disabled,women, and youth and generated 26,188 days oftemporary employment. But there is lessevidence on outreach to other vulnerable groupssuch as widows, victims of conflict, inhabitants ofremote villages, and so forth.

Under Russia IV in Kazan, the newly createdMunicipal Department of Social Protection identi-fied eligible poor recipients through the munici-pality’s new computerized database of 23,900assisted families to replace earlier untargetedsubsidies with direct cash payments, which alsoproduced an overall savings to the municipality.

Armenia: Third Social Investment Fund Project. Bosnia andHerzegovina: I—Local Development Project; II—CommunityDevelopment Project; III—Solid Waste Management Project;IV—Urban Infrastructure & Services Project. Croatia: CoastalCities Pollution Control Project. Georgia: I—Municipal Infra-structureRehabilitation Project; II—Municipal Development Project; III—Second Municipal Development & Decentralization Project.Kazakhstan: I—Social Protection Project; II—Atyrau Pilot Water.Kosovo: Second Community Development Fund. Kyrgyz Repub-lic: I—Urban Transport Project; II—Small Towns Infrastructure

& Capacity Building Project. Latvia: Municipal Services Devel-opment Project. Lithuania: Municipal Development Project.Poland: Municipal Finance Project. Russia: I—St. PetersburgCenter City Rehabilitation Project; II—Northern RestructuringProject; III—St. Petersburg Economic Development Project;IV—Kazan Municipal Development Project. Tajikistan: Munic-ipal Infrastructure Project. Turkey: I—Bursa Water & SanitationProject; II—Municipal Services Project; III—Istanbul MunicipalInfrastructure Project. Turkmenistan: Urban Transport Project.Ukraine: Urban Infrastructure Project. Uzbekistan: TashkentSolid Waste Management Project.

Box D.1: Key to MDPs Referred to in Text

Source: IEG.


7 7

Several ”one- stop shops” were created to providethese services to the beneficiaries.

Conclusions• By continuing to do more wholesale MDPs

that emulate the successful cases in this Region,positive MDP impact can be broadened to ben-efit more municipalities.

• New MDPs could make better and more in-tensive use of simple ERR estimates to deter-mine investment priorities and measureefficiency of results.

• M&E of new MDPs needs to be stronger thanin the past, especially in measuring the achieve-ment of objectives through quantified baselinesand targets.

7 9

Bank SupportLatin America and the Caribbean is the mosturbanized of the Bank’s Regions. Currently, some435 million people, 78 percent of the totalpopulation, live in cities. The urban populationnow grows at only half the annual rate of the 1970s.

During the past decade (1998–2008), the WorldBank had a portfolio of 36 active MDPs andcommitted $2.6 billion to them. The Region’sMDPs aimed to strengthen the management of1,098 municipalities in 13 countries. The mostactive borrowers were Colombia (7 projects),Brazil (5), Honduras (4), Venezuela (3), Peru (3),and Argentina (3). Mexico, Ecuador, Chile,Bolivia, Nicaragua, Haiti, and Belize had 1 or 2projects each. Half the MDPs were in higher- middle- income countries; the other half— exceptHaiti (low- income)—were in lower- middle- income countries. The Region’s MDP portfoliocovered all countries in the Region with largeurban populations.

Portfolio PerformanceThe Region’s MDP portfolio has a strongperformance record, with 86 percent ofcompleted operations rated satisfactory. ThisRegion reports the best MDP performanceamong the six Bank Regions.

An outstanding performer, rated highly satisfac-tory, was Colombia II. It successfully strength-ened the capacity of institutions in charge ofplanning, managing, and maintaining urbantransport infrastructure in Bogotá (population7.1 million). This MDP lowered sector adminis-tration costs from 17 percent in 1996 to 10percent in 1998, and road maintenance costs

were lowered by 77 percent, despite a sevenfoldincrease in the network between 1996 and 1999.

Venezuela I introduced basic infrastructure on alarge scale to low- income barrios in 45 municipali-ties across the country, benefiting 66,000 poorfamilies; this was 43 percent above target. Theproject exposed these municipalities to lendingoperations for the first time, supporting theirfinancial management and revenue growththrough detailed technical assistance. Colom bia Iand its follow- on Colombia IV together helpedcreate a local credit market among 179 municipali-ties around the country, although municipaldemand for credit was weaker than expected.Municipalities with conservative financial adminis-trations were reluctant to take on debt, and othercreditworthy borrowers had alternative sources ofcredit. Between them, Brazil I and III, in thestates of Minas Gerais and Bahia, respectively,brought technical assistance for improving

APPENDIX E: BANKING ON MUNICIPALITIES: WORLD BANK SUPPORT IN LATIN AMERICA AND THE CARIBBEAN

Completed MDPs (number) 21



IBRD commitments (US$m) 2, 485

IDA commitments (US$m) 166

Commitments per completed MDP (US$ million) 86





Municipalities served (number) 1, 098Source: World Bank data.Note: IBRD = International Bank for Reconstruction and Development; IDA = International Develop-ment Association; MDP = municipal development project.

Table E.1: Summary of MDP Portfolio, 1998–2008


8 0

financial management to 179 municipalities, eachusing a wholesale arrangement through their state intermediaries.

Among the most important physical resultsobtained were improvements in the urbanenvironment that were attained through basicsanitation investments financed by the projects.Valuable results in the form of a nationwidemunicipal information system came from Chile II.A wholesale operation on a larger scale involving77 municipalities, the Bolivia MDP helpedstrengthen administrative and financial controls. Italso focused physical investment on basic sanita-

tion in the poorer municipalities, especially in theBeni region. Ecuador I successfully helped makethe fiscal transfers to municipalities more transpar-ent, as intended, while helping improve theadministrative efficiency of 99 municipalitiesthroughout the country.

In contrast, three MDPs had unsatisfactoryoutcomes. The Haiti operation (retail) failed toexpand the water supply to the capital Port auPrince or make it more efficient. Water meteringtargets were not met, and illegal consumptioncontinued unabated. Most of the loan for Mexico Iwas cancelled, as six municipalities on the U.S.border region were unprepared to meet theproject’s environmental requirements, in compli-ance with the norms of the North American FreeTrade Agreement. Deteriorating national economicconditions undermined the ability of Mexico II tofinance the modern municipal solid waste landfills,although there was some progress in improvingmunicipal planning of solid waste management,thanks to project technical assistance that wentahead. The Bank was slow to restructure theproject, which resulted in the cancellation of 70percent of the loan.


More informationThere were few instances in the Region of MDPsgenerating or using information for planning.Chile was a notable exception. Chile I and II

8676 7674 78 75

0

20

40

60

80

100


Perc

ent s

atis

fact

ory

Latin American and the Caribbean Region All Bank

Figure E.1: MDP Portfolio Performance, Fiscal 1998–2008










Table E.2: Municipal Management Focus of Region Portfolio

A P P E N D I X E : WO R L D B A N K S U P P O R T I N L AT I N A M E R I C A A N D T H E C A R I B B E A N

8 1

launched and consolidated the National Systemfor Municipal Information (SINIM), which hasreported about the situation and performancethrough more than 250 indicators for all 345municipalities since 2000. Available over theInternet, SINIM covers local finances, administra-tion, health and education services, spatialplanning, poverty, other social indicators, as wellas geographic characteristics of all Chile’smunicipalities. But Colombia III failed to createa national environmental information systembased on municipalities, because the Ministry ofthe Environment did not play the coordinationrole expected of it. Brazil I and III, in therespective states of Minas Gerais and Bahia,implemented similar databases at the state leveland helped individual municipalities build theirown information systems.

Most progress in compiling information wasmade by extending and consolidating localproperty tax registers, or cadastres. Colombia IIimproved these to such an extent that itexceeded its target of updating 4.5 million titleregisters by 57 percent. At the municipal level,performance in using the additional informationfor strong taxation flows varied.

Monitoring and evaluationAs in other Regions— and other sectors, too— MDPs in this Region obtained at best modestresults in designing, implementing, and usingM&E. The usual culprits were found: focusmainly on outputs rather than outcomes, lack ofbaseline data to compare against actual achieve-ments, and inadequate collection of data onactual project performance.

Under Argentina I, a well- conceived logframewith performance indicators was established atmidterm review for infrastructure works, but asimilar effort for institutional development wasless successful. Under Brazil II, the M&Eframework to verify the achievement of projectobjectives was weak. Under Brazil III, threeyears after Board approval, the Bank andborrower agreed to adopt a set of indicators tomonitor outputs and outcomes. However, actualoutcome data appeared infrequently and

seemed inconsistent, providing only anecdotalevidence of increased tax- collection rates,improved health conditions, and improvedaccess to water and sanitation services.

Under Chile I, a lack of M&E on outcomesmeant that claims of strengthened municipalmanagement as a result of project technicalassistance on information technology could notbe substantiated. Even Chile II, which otherwiseexcelled in providing information for planning,did poorly on M&E. Its design included 18performance indicators, but these were mostlyabout the delivery of outputs, such as thenumber of municipalities served and the numberof technical assistance contracts made. The twoindicators that came closest to monitoringachievement of project objectives were thosethat considered municipal own- source revenuesand municipal operational surpluses.

Honduras also lacked explicit and quantifiableindicators able to demonstrate progress towardsustained coastal tourism in the project region.For Mexico II, the logframe developed duringsupervision was specified only in broad terms.The Implementation and Completion andResults Report cites several examples of projectoutputs used to justify conclusions on outcomes.Venezuela II lacked appropriate performanceindicators altogether, but Venezuela III paidmore attention to the design of the M&E system.However, it was barely used, as baseline datawere not collected because of lack of inadequate resources.

Urban and spatial planningAlthough nine MDPs had objectives focused onstrengthening municipal planning, there is littleevidence of what was actually achieved. Co -lombia III led to the preparation of 17 municipalenvironmental plans and the incorporation ofenvironmental aspects into land- use plans.

Investment planning and strategiesIEG found limited evidence of achievements inthis area. One reported instance was Argentina I,which helped many municipalities plan cost- effective investment programs.


8 2


Better financial managementThe majority of MDPs in this Region aimed toimprove municipalities’ financial management,starting with better accounting and financialreporting systems. On balance, the results fromthese efforts have been positive, sometimes withvaluable demonstration effects on municipalitiesthat were not part of the projects. Bolivia Ihelped 77 municipalities strengthen their financialcontrol systems.

Brazil I in Minas Gerais provided technicalassistance for financial management to about 50municipalities, a good number but well short ofthe ambitious plans to cover all urban municipali-ties in the state. Under Chile I, municipalpractices and technical capacity have beenimproved, and equipment (computers, communi-cations, drivers’ license testing) has been updated.Similar improvements were reported underMexico I. In addition, Venezuela I provided on- the- job learning opportunities for 45 municipali-ties to manage credit operations for the first time.

Mobilizing own revenuesThrough technical assistance, Bolivia I municipali-ties improved their resource mobilization. Brazil IIreported that 26 municipalities assisted by theproject increased their own revenues more thanother municipalities, but IEG found the statisticalsignificance of this evidence questionable.

More significant is that own revenues for projectmunicipalities grew faster than higher- level transfersover the 2001–03 period. Under Brazil III, a surveyindicated that the majority of participating munici-palities increased efficiency in financial manage-ment and tax administration and showed asustained increase in the collection of propertytaxes and services between 1996 and 2000. UnderEcuador I, of a random sample of 99 municipali-ties, 53 percent had doubled revenues in real termsover the project period.

Municipal creditworthiness and debt managementSeveral MDPs successfully introduced municipal-

ities to credit operations, providing them withassistance for managing such operations. Inparticular, Colombia I and IV effectively est -ablished a local credit market with the officialLocal Development Fund, called FINDETER; thiscurrently has a credit rating of AAA, refinancingcommercial bank loans to municipalities to fundtheir investments in infrastructure and services.One municipality, Pereira (population 0.4million), was able to issue bonds that wereoversubscribed, and another (Barranquilla,population 1.4 million) was able to pay off its short- term debt thanks to project advice onportfolio management.

Brazil I, II, and III enhanced the credit man -agement capabilities of poor municipalitiesespecially, as did Ecuador I. But such efforts toconsolidate local credit in the Region have beenthwarted in recent years by national efforts tocontrol fiscal deficits at the local level. In Chile,borrowing by municipalities is forbidden altogether.

Private finance participationProgress in this direction was not widelyachieved through MDPs. Colombia IV, however,helped municipalities increase water, gas, andsolid waste tariffs, for instance, making someservices profitable for private investors. Thissituation continues to this day, although servicesare less affordable to the poor. Averagehousehold expenditure on basic sanitation roseby 204 percent between 1997 and 2003. Attemptsto stimulate private funding and operation ofmunicipalities made little headway under BrazilI and III, mainly because of a lack of interest bythe municipalities themselves. Venezuela IIfound a similar reluctance toward privatization ofurban transport.


Investment prioritiesAbout half the completed MDPs reported ERRestimates at appraisal and completion. High ERRs(34–42 percent) at completion were estimatedfor basic sanitation and slum upgrading invest-ments under Venezuela III and Brazil I and III.

A P P E N D I X E : WO R L D B A N K S U P P O R T I N L AT I N A M E R I C A A N D T H E C A R I B B E A N

8 3

Strong ERRs (29–34 percent) were also reportedfor municipal urban transport investments underColombia II and Venezuela II. MDPs that didnot report internal rates of return at completionincluded Mexico I and II, Ecuador I and II,Argentina I, and Honduras II. Operations suchas these, which included investment in municipalinfrastructure and services, would have beenamenable to simple estimates of ERRs that wouldhave informed the evaluation about performance efficiency.

ProcurementWith long histories of decentralized municipalresponsibilities, many municipalities have signifi-cant experience with managing procurement.Nevertheless, Brazil II did introduce 49 poorermunicipalities in the state of Ceará to handlingcompetitive procurement for works. Similarly,Venezuela I helped 45 municipalities learn tomanage procurement themselves as decentral-ization gathered pace in that country.

Operations and maintenanceAlthough municipalities are typically responsiblefor O&M of infrastructure and services withintheir jurisdictions, few MDPs paid attention tothis aspect of municipal management. Oneexception was Venezuela I’s provision oftechnical assistance and training to participatingmunicipalities for carrying out urban roadmaintenance activities, affecting 360 kilometersof pavement. Another was Venezuela III, whichtried to build up local O&M capabilities but cameup against municipalities’ unwillingness to curtailinvestments by allocating more resources toO&M, especially at times of financial crisis in thecapital Caracas (population 1.8 million).

Services— Most affected sectorsAs in other Regions, the most popular servicesprovided through MDPs included urban trans -port, slum upgrading, basic sanitation, solidwaste management, and other urban environ-mental improvements.

Municipalities were able to improve urbantransport through MDPs across the Region.Colombia II, for instance, brought the very

successful Transmilenio bus- operated publictransport to Bogotá (population 7.2 million),leading other municipalities, including Barran-quilla (population 1.4 million), Pereira (popula-tion 0.4 million), and Cali (population 2.4 million),to plan similar bus projects. There was alsointerest from other countries. Through financingand training of 250 municipal staff, Venezuela IIenabled municipalities to make simple trafficmanagement improvements such as road signals,intersection improvements, and rationalization ofbus routes to reduce traffic congestion. In Belize,street and traffic improvements— including trafficsignal systems, improved street drainage,widened sidewalks for pedestrians, and bicycle lanes— had a positive impact on road safety.

According to two surveys made at the completionof Brazil III, respiratory and intestinal diseaseswere reduced in municipalities in Bahia state in low- income areas where street paving hadreduced dust particles and basic sanitation hadprevented the pollution of the water supply bysewage. In contrast, sewage treatment remains amajor challenge in the Region. Pereira is stillwithout sewage treatment for its 440,000 inhabi-tants, who live in an ecologically sensitive zone,despite the successful participation of the munici-pality in several projects, including Colombia I.Similar shortcomings in final sewage treatmentwere evident under Brazil I, which neverthelessbrought other basic sanitation improvements to150 municipalities in the state of Minas Gerais.The introduction of the final disposal andtreatment facilities for solid waste provedchallenging under Mexico II, where deteriorat-ing macroeconomic conditions meant that onlythree of the seven facilities intended were built,and only partially.

Other urban environmental improvementsintroduced by MDPs included, in Colombia II,controls over discarding used tires and batteries,as well as the reduction of noise pollution byurban traffic through the deployment and use ofnew monitoring equipment. Other MDPs helpedreinforce municipal management of the urbanenvironment. Thus, Colombia III helped 17municipalities prepare environmental plans, and


8 4

the Ecuador MDP helped 23 municipalitiesestablish specific Environmental ManagementUnits within their municipal administrations.

Income levels of beneficiaries— Poverty reductionAbout one- third of MDPs in the Region hadobjectives explicitly focused on assisting theurban poor. Even for Argentina I, an MDP thatwas not specifically poverty focused, about one- fifth of the beneficiaries were poor. Bolivia I,which emphasized beneficiary participation inthe choice of investments, made most invest-ments in municipalities where poor people lived,such as in the Beni region of the Amazon.

Under Brazil I, municipalities in Minas Geraisstate invested in lower- standard basic sanitationand upgrading only of interest to lower- incomegroups. Brazil III went one stage further inurban poverty mapping of the changes broughtabout by municipal investment in street paving,provision of drainage, and water supply andsanitation; however, the mapping was discontin-ued because of lack of resources. The povertyimpact of Colombia I’s work with 179 munici-palities can be inferred from national data,which show that service coverage for those inthe lowest quintile of income distributionimproved significantly between 1993 and 2003,from 80 percent to 91 percent for electricity, 77

percent to 83 percent for basic sanitation, and18 percent to 33 percent for fixed- linetelephones. Colombia II survey data showedthat most users of the Transmilenio urbantransport system in Bogotá are within the twolowest quintiles.

Conclusions• Doing more wholesale MDPs and scaling them

up is likely to yield positive results in a Regionwhere 100 percent of wholesale MDPs ob-tained satisfactory outcomes.

• More can be done to disseminate the goodMDP practices in the Region. Globally, munic-ipalities in other Regions could benefit from thisexperience in municipal information systems,municipal creditworthiness and financial man-agement, urban transport, and poverty reduc-tion. Within the Region itself, the Bank is poisedto share MDP experiences among borrowercountries. Finally, within individual countries,national and state authorities have opportuni-ties to share and exchange experiences amongmunicipalities from different parts.

• Successful experience from other Regions canbe put to good use in those areas where short-comings have been noted in the Region, suchas in M&E, private financing of municipal serv-ices, O&M, and key environmental services,such as sewage and solid waste disposal and treatment.

Argentina: I— Second Municipal Development; II— Basic Mu-nicipal Services; III— Subnational Government Public SectorModernization. Belize: Belize City Infrastructure. Bolivia: I— Municipal Development; II— Urban Infrastructure. Brazil: I— Minas Municipal Development; II— Ceará Urban Development &Water Resource; III— Bahia Municipal Infrastructure Development& Management; IV— Bahia Poor Urban Areas Integrated Devel-opment; V— Recife Municipal APL. Chile: I— Municipal Devel-opment; II— Second Municipal Development. Colombia: I— Municipal Development; II— Bogota Urban Transport; III— Urban Environment; IV— Urban Infrastructure Services Develop-

ment; V— Bogota Urban Services Project; VI— Integrated MassTransit Systems; VII— Disaster Vulnerability Reduction Project.Ecuador: I— First Municipal Development; II— EnvironmentalManagement. Haiti: Port- au- Prince Water Supply. Honduras: I— Natural Disaster Mitigation; II— Sustainable Coastal Tourism Proj-ect; III— Barrio Ciudad; IV— Water & Sanitation Program. Mexico: I— Northern Border Environment; II— Solid Waste. Nicaragua: Nat-ural Disaster Vulnerability Reduction. Peru: I— Lima Transport; II— Vilcanota Valley Rehabilitation & Management; III— Second RealProperty Rights. Venezuela: I— Low- Income Barrios Improve-ment; II— Urban Transport; III— Caracas Slum Upgrading.

Box E.1: Key to MDPs Referred to in Text

Source: IEG.

8 5

Bank SupportAbout 180 million people live in cities in theMiddle East and North Africa Region, about 57percent of the total population. During thedecade 1998–2008 the World Bank had a portfo-lio of 18 MDPs spanning 8 countries in theRegion. The Bank commitments of $845 millionfocused on strengthening the municipal manage-ment of 379 municipalities in Tunisia (4 projects),West Bank and Gaza (4), the Republic of Yemen(3), Jordan (2), Lebanon (2), the Arab Republicof Egypt (1), the Islamic Republic of Iran (1), andMorocco (1). More than 70 percent of theprojects are in lower- middle- income countries,17 percent in low- income countries, and 11percent in upper- middle- income countries. TheRegion’s portfolio covered all countries in theRegion with large urban populations (15 millionplus) except for Algeria.

Portfolio PerformanceSome 63 percent of completed MDPs in theRegion achieved satisfactory outcomes. Bankperformance was also satisfactory 63 percent ofthe time. These ratings are below Bank- wideaverages. In contrast, 88 percent of MDPs havesatisfactory borrower performance, well abovethe Bank average. The disconnect reflects goodefforts by the borrower in West Bank and Gaza,where exogenous factors of conflict preventedcommensurate project outcomes.

The strongest performing MDPs in the Region,each awarded satisfactory ratings for theiroutcomes and Bank and borrower performance,were in Tunisia and the West Bank and Gaza.Tunisia I, a wholesale operation assisting 257municipalities throughout the country, pro duced

excellent results that continue more than eightyears after completion. Not only did MDP-partici-pating municipalities increase their own revenuesmore than other municipalities, but the partici-pants also produced a current surplus that wastwice the target. The project helped the remotemunicipality of Kasserine (population 82,000)upgrade the Ezzouhour district of town and keptit in good condition through careful mainte-nance, sometimes involving local residents.

West Bank and Gaza III succeeded in meetingmore modest objectives that focused on repair-ing municipal infrastructure damaged during theintifada rather than providing completely newservices. Despite the difficult circumstances of itsimplementation, the project succeeded inmaking timely and effective repairs, thanks inpart to strong and enthusiastic local leadership.

APPENDIX F: BANKING ON MUNICIPALITIES: WORLD BANK SUPPORT IN THE MIDDLE EAST AND NORTH AFRICA




IBRD commitments (US$ million) 652

IDA commitments (US$ million) 94







Table F.1: Summary of MDP Portfolio, 1998–2008


8 6

63 63

8874 78 75

0

20

40

60

80

100


Perc

ent s

atis

fact

ory

Middle East and North Africa Region All Bank

Figure F.1: MDP Portfolio Performance, Fiscal 1998–2008


Weaker performance was turned in by theMorocco MDP, which failed to improve thehousing stock of the ancient city of Fez (popula-tion 964,891) as intended. Public- to- privateleverage of investments for the rehabilitation ofthe Medina is likely to remain at a 1:1 ratio, wellbelow the projected target of 1:13. However, theFez municipality was consolidated through theamalgamation of six local governments aroundthe time of project restructuring in 2003, andmunicipal management improved on the techni-cal but not the financial side. West Bank andGaza I also performed poorly. The start- up wasat the time of the 2000 intifada and the Israeli

military response to it. Because of events beyondthe control of the project, the MDP was unableto assume any effective role in the nationalsystem of central and local government that theproject hoped to constitute.


More informationMDPs in the Region rarely set out to make moreinformation available for municipal manage-ment. The best results were obtained underTunisia I, although they could have been mademore widely available to the municipalitiesthemselves to help them improve their manage-ment. Instead, detailed information on munici-pal financial performance remained in the handsof the national Municipal Funding and Support Agency. A newer operation, Jordan I aims toimprove information on municipal finances atthe national level, as well as information for assetmanagement at the municipal level, too.

Monitoring and evaluationThere is little information on the extent towhich M&E frameworks were designed andused in projects in this Region. Whereverperformance indicators were available, theymostly related to outputs (in the form ofdelivery of project components) rather thanoutcomes (in the form of achievement ofproject objectives). Even in such cases, baselinedata were rarely available.









Table F.2: Municipal Management Focus of Region Portfolio

A P P E N D I X F : WO R L D B A N K S U P P O R T I N T H E M I D D L E E A S T A N D N O R T H A F R I C A

8 7

The Republic of Yemen I, for instance, did nothave baseline data on before- project conditionsto track the impact of new pipelines andhousehold connections on improving watersupply. In Tunisia II’s M&E, the chosenperformance indicators measured projectoutputs, such as the provision of project techni-cal assistance, rather than moving toward theproject objective of increasing the efficiency ofpublic sector management at the municipal level,for which no baseline condition or targets werespecified in the project design. IEG estimatedthat this project alone accounted for one quarterof all municipal investments in the countryduring the 1997–2003 period of its implementa-tion. Despite this high profile, M&E was unableto show what impact the project had, only thelevels of municipal services in the country as a whole.

West Bank and Gaza IV operated under thevery difficult circumstances of the intifada. Inthe rush to plan and deliver emergency servicesat the outset, baseline indicators were notadequately set up. Overall, there was a persistentinadequacy of information about governmentprocesses, including budget and transfer datathat should have improved under the project.

Urban and spatial planningLittle was achieved in strengthening municipalplanning capabilities in the Region. Under WestBank and Gaza I, three municipalities pre -pared three- year development plans for the first time.

Investment planning and strategiesThis too was not a common feature of MDPs inthe Region. Tunisia II required 76 municipali-ties to prepare investment plans to be eligible forproject funding of municipal infrastructure, butit is not clear how many actually did prepare them.


Better financial managementUnder Tunisia I and II during 1993–2003,financial management by many of the 257

municipalities assisted by the projects improved,which led to better financial results. Strongerfinancial management was initially the outcomeof rapid loan disbursements to finance prioritylocal investments, which then progressed to theadoption of computerized accounting in 32municipalities for the first time and to three- yearbudgeting and of outsourcing municipal services.

The municipality of Ariana (population 237,395)became one of the country’s top 10 tax- collection districts— it ranks 23rd in popu lation— aftermaking its own tax administration more efficient,following intense training its officials had at thenew municipal training center specially createdby the project. Altogether, 10,000 local andcentral government staff received project trainingthat covered more than 50 percent of all munici-pal staff in Tunisia at the time.

Under Republic of Yemen I, municipalmanagement of the local water supply became areality as the water authority of Sana’a (popula-tion 1.9 million) became a fully autonomouscorporation able to cover operating costs for thefirst time in this sector; this also happened in 12other municipalities. Municipal financial manage-ment began to improve under West Bank andGaza II, as local governments began to institutesolid waste collection fees, for instance, but thedeteriorating security situation after 2000 stalledfurther progress.

Mobilizing own revenuesTunisia I produced excellent results thatcontinue to this day. Not only did participatingmunicipalities increase their own revenues morethan other municipalities, but the participantsproduced current surpluses that were twice thetarget. In the municipality of Monastir (popula-tion 64,222), for instance, municipal ownrevenues as a share of the total rose from 30percent to 38 percent between 1991 and 1998.Resources for Tunisian municipalities enabledthem to finance more investments than initiallyexpected. Under Tunisia II, several municipali-ties were able to use these additional revenues topay off short- term debts and build up net


8 8

savings. Morocco’s MDP helped improve costrecovery and contributed to mobilizing internaland external resources in a nondeficit, noninfla-tionary way to help finance minor investmentsby the municipality of Fez. This was a positiveresult for a project that otherwise failed toachieve its objectives.

Municipal creditworthiness and debt managementTunisia I introduced 257 municipalities to themanagement of credit. As the agency responsiblefor implementing the project and collectingmunicipal debt service, the Municipal Fundingand Support Agency saw its own creditworthi-ness enhanced when it obtained a credit ratingof AA+ and successfully issued its own bonds inthe local market, to the value of $23.5 million.

Private finance participationOnly a few of the Region’s MDPs assigned aspecific role for private sector funding to helpstrengthen municipal management. Apart fromthe bond issue under Tunisia I, there is littleevidence of a concerted effort by MDPs in theRegion to stimulate private funding of municipalservices at all. Even under that project, onlyminor private financing occurred at the munici-pal level, such as for detailed service design workfor historic sites in Monastir. Significant privateparticipation in municipal water authority inSana’a has yet to occur as the Republic ofYemen I had hoped. Under West Bank andGaza II, private financiers on whom the projectdesign had initially relied to fund some munici-pal investments shied away as the conflictworsened in 2000.


Investment prioritiesOnly two MDPs used estimates of ERRs to assessthe priority of the project investments at appraisaland to measure the efficiency of project achieve-ments at completion. Following careful analysis atcompletion, the West Bank and Gaza IIIyielded a very high ERR of 55 percent, exceedingeven the appraisal estimate of 47 percent.

Project improvements to road and waterinfrastructure in 10 municipalities that accountedfor 76 percent of all project costs generated verystrong benefit streams, when compared with thedire without- project counterfactual. At comple-tion, Republic of Yemen I yielded a 28 percentERR (up from 25 percent at appraisal) thatdemonstrated the significant benefits obtainedwhen municipal water supply shifts from high- cost tanker delivery to low- cost networkprovision. As well as demonstrating the positiveresults of the projects themselves, these examplesdemonstrate the feasibility of estimating ERRseven in the most challenging circumstances.

ProcurementThere is little evidence of municipalities takingcharge of procurement in a Region where thishas largely remained a responsibility of centralgovernment authorities. Under Republic ofYemen I, delays were caused by the division ofprocurement responsibilities between theautonomous water authority for the municipalityof Sana’a and the Ministry of Energy and Water.The ministry finally oversaw the internationalcompetitive bidding for the works, which led tocost savings at the outset, but these were off-set by unfavorable foreign exchange rate movements.

Operations and maintenanceThere were mixed results in this Region. UnderTunisia I, for example, the remote municipalityof Kasserine (population 82,000) upgraded theEzzouhour district of town and kept it in goodcondition through careful maintenance, some -times involving local residents. But performanceat the municipal level can vary under the sameproject. Thus, officials of the municipality ofAriana (population 237,395) saw the advantagesof neglecting routine maintenance in upgradedareas. They felt that leaving drains blocked andpavement broken gave them a better chance ofreceiving central government aid to finance acomplete replacement. The main shortcomingof the otherwise successful Republic of YemenI was its inability to provide for adequate ongoingO&M of the facilities built under the project.

A P P E N D I X F : WO R L D B A N K S U P P O R T I N T H E M I D D L E E A S T A N D N O R T H A F R I C A

8 9

Services— Most affected sectorsMunicipal infrastructure and services constitutedthe most numerous objectives in MDPs of theRegion. The sectors for which MDPs soughtstrengthened municipal management in theRegion included low- income neighborhoodupgrading, urban street paving and drainage,water supply and basic sanitation, as well as otherenvironmental improvements such as solidwaste management.

MDPs achieved mixed results in upgrading andurban street and road improvements. WestBank and Gaza I, for instance, attended toplanned rehabilitation work as well assubsequent damage caused by conflict. Thiswould be done through 54 damage repairsubprojects in 9 municipalities, sometimesexceeding targets, as in the case of 184 kilome-ters of roads built against a target of 100 kilome-ters. However, later border closures preventedthe use of physical assets, which would deterio-rate through lack of upkeep. West Bank andGaza III successfully completed two roadprojects (13.2 kilometers) and rehabilitated 67kilometers of roads. These improvementsreduced travel costs and times by almost 50percent in the project area.

West Bank and Gaza IV reached 61 municipali-ties through 2,200 subprojects in water and sanita-tion, roads, electricity, and solid waste, but the fullextent to which this augmented services is notfully known. Under Tunisia I, the number ofsubprojects financed and their outlay exceededexpectations by 250 percent and 50 percent,respectively, but their impact on service levels wasnot fully documented. Improved streets gavepeople better access to their homes andbusinesses, as well as providing drainage andproper public lighting in central and residentialareas of the client cities. A participatory approachadopted by many municipalities encouragedcommunities to contribute to the costs of someimprovements, as IEG saw in Kasserine; there, 50community leaders met with local officials toidentify the priority investments for their ownneighborhoods. This dialogue appears to have

developed greater understanding of the need forcost recovery. Twice as much in direct taxation wascollected in Kasserine after the project as before.

Republic of Yemen I helped improve the livingconditions in the Akama neighborhood of Sana’aby reducing raw sewage flooding in residentialareas, thanks to 7,500 additional householdsbeing connected to the sewerage system, seventimes the original target. Water supply alsoincreased, though not as much as targeted,because only 5,000 households of the targeted18,500 were connected. However, despite replac-ing 21,500 water meters and rehabilitating 30kilometers of pipelines, the project did notsucceed in reducing unaccounted- for water.

West Bank and Gaza II constructed or rehabil-itated 64 kilometers of water lines (as well as 77kilometers of roads, which was several times theoriginal targets), but it is not clear if this was dueto any dilution in design criteria. West Bank andGaza III expanded the water network in severalsmall settlements that reported 90 percent oftheir population receiving a 24-hour piped watersupply. The extent of this achievement cannot befully evaluated, however, for lack of baseline dataabout the level of before- project service, or evendata on the population served.

Services— Private provisionMDPs in the Region did not put much emphasison increasing the private provision of municipalservices, and there was little progress where suchattempts were made. Under Republic ofYemen I, the intended private management ofthe Sana’a water agency had yet to beimplemented and the government remaineduncommitted. Under West Bank and Gaza II,renewed conflict in 2000 precluded any increasein private sector participation.

The otherwise very successful Tunisia I madelittle progress in improving the incentiveframework for building partnerships with theprivate sector and municipalities. UnderMorocco’s MDP, both the government’scommitments for the rehabilitation process and


9 0

projections for leveraging private sector invest-ments fell far short of expectations during theproject period, and it appears too early to assesswhether projections made at project closing willbe realized to any significant extent.

Income level of beneficiaries— Poverty reductionThere was no explicit focus on the income levelsof beneficiaries or on poverty reduction in mostMDPs in the Region. Even in the few cases wherethis was directly or indirectly attempted, theresults fell short of targets. West Bank andGaza IV managed to create 270,000 person- daysof employment for unskilled workers, but thiswas short of the target of 400,000, after some ofthe resources allocated to employment genera-tion were transferred to service provision instead.

Under Morocco’s MDP, only 20 percent of thebeneficiaries were classified as poor. UnderTunisia I, there was no clear focus on poverty

reduction. Several stakeholders, especially in themunicipalities themselves, appeared to beunfamiliar with the Bank’s mission relating topoverty reduction and saw no contradiction inproject investment being made in higher- income areas.

Conclusions• Development programs in the Region can

make more use of municipalities as partners inservice provision, even where central govern-ments prefer to retain overall responsibility themselves.

• Robust evidence from the Region shows thatMDPs can improve the performance of mu-nicipal finance, and the potential for strength-ening this dimension of municipal managementappears to be under- exploited.

• Frequent claims that M&E and ERR exercisesare too complex and costly to implement involatile country conditions are not borne outby experience in this Region, where a few ex-periences have been quite successful.

Arab Republic of Egypt: Alexandria Development. Islamic Re-public of Iran: Urban Upgrading & Housing Reform. Jordan: I— Regional & Local Development; II— Cultural Heritage, Tourism &Urban Development. Lebanon: I— First Municipal Infrastructure; II— Cultural Heritage & Urban Development. Morocco: Fes Med-ina Rehabilitation. Tunisia: I— Municipal Sector Investment; II— Second Municipal Development; III— Transport Sector

Investment; IV— Third Municipal Development. West Bank andGaza: I— Municipal Development; II— Bethlehem 2000; III— Second Municipal Infrastructure Development; IV— EmergencyMunicipal Services Rehabilitation. Republic of Yemen: I— Sana’aWater Supply & Sanitation; II— Taiz Municipal Development &Flood Protection; III— Port Cities Development Program.

Box F.1: Key to MDPs Referred to in Text

Source: IEG.

9 1

Bank SupportAlthough South Asia is one of the world’s lessurbanized regions, more than 431 millionpeople, 29 percent of the total population, live inthe Region’s cities. Through just 12 MDPs activeduring the past decade (1998–2008), the WorldBank committed $1.3 billion. This small portfolioaimed to strengthen the management of 146municipalities in 6 countries. By number ofMDPs, the most active borrowers were India (4projects); Pakistan (3); and Bangladesh (2);Afghanistan, Bhutan, and Sri Lanka hosted 1each. Thus, all but two of the Region’s MDPswere in low- income countries. With an urbanpopulation of 4.5 million—16 percent of the total— Nepal was the only large country in theRegion not to host an MDP.

Portfolio PerformanceOnly three of the seven completed MDPs in thisRegion (43 percent) achieved a satisfactoryoutcome, making this the weakest of the Bank’sRegional MDP portfolios. Only four of them hadsatisfactory ratings for both Bank and borrower performance.

The only fully satisfactory completed operationin the portfolio was India II, which set up amunicipal development fund that financedinvestments by municipalities in the southernstate of Tamil Nadu. This operation built on morethan 20 years of continuous Bank assistance tothe urban development of Tamil Nadu and itscapital Chennai. By introducing computerizedaccounting and modern financial managementmethods, India II helped 45 municipalities inthe state prepare “corporative developmentplans” to help determine their priority invest-ments. This support also helped Tamil Nadu’s

second- largest city Madurai (population 909,908)successfully issue municipal bonds for the firsttime, to finance an inner ring road.

Several MDPs performed poorly. Implementa-tion of Bangladesh I was hostage to land- acquisition problems and a lack of coordinationbetween borrower agencies, resulting in resettle-ment not complying with Bank guidelines. Inaddition, municipal financial managementremained weak. Sri Lanka’s MDP suffered frompoor design that did not take into account thepublic opposition to the project’s plans for solidwaste disposal. The design was also based on anincomplete understanding of the baseline waterquality of the polluted Beira Lake in the capitalColombo (population 2.3 million) that theproject aimed to improve. The municipalmanagement in Colombo barely changed as aresult of the project.

APPENDIX G: BANKING ON MUNICIPALITIES: WORLD BANK SUPPORT IN SOUTH ASIA




IBRD Commitments (US$ million) 671

IDA Commitments (US$ million) 673

Commitments per completed MDP (US$ million) 88






Table G.1: Summary of MDP Portfolio, 1998–2008


9 2

4357 57

74 78 75

0

20

40

60

80

100


Perc

ent s

atis

fact

ory

South Asia Region All Bank

Figure G.1: MDP Portfolio Performance, Fiscal 1998–2008


Pakistan II did not succeed in the NorthwestFrontier largely because efficiency criteria forselecting subprojects and financing them wereoutweighed by political factors that determinedthe choices made. Bhutan’s MDP did notsucceed primarily because the design overesti-mated the management capabilities of local municipalities.


More informationIndia II provided technical assistance to 50municipalities, called urban local bodies in Tamil

Nadu. As a result, 46 local municipalities haveprepared city corporate plans with a mapping ofurban infrastructure using important baselinedata. However, these plans have yet to becomekey drivers of local municipal development, eventhough they have facilitated municipal access toloans by the Tamil Nadu Urban DevelopmentFund, itself established by an earlier MDP project.The need for systematic data collection is nowbetter appreciated; in fact, the State MunicipalAdministration and Water Supply Departmentplans to assist local municipalities prepare ahuman development index for such basicservices as water supply, sanitation, health,poverty alleviation, and access to basic needs.

Under the Sri Lanka MDP, the Colombo masterplan benefited from the project’s provision of ageographic information system for the UrbanDevelopment Authority, but little expertise waspassed on to the Colombo municipality that isresponsible for cadastral and land- use applica-tions for the geographic information system.

Monitoring and evaluationM&E was weak in MDPs. Its weakness came fromtoo much focus on the delivery of project outputsand too little on project impacts gained throughachieving the MDP objectives. This M&Eshortcoming was even evident in the otherwise well- performing India II, which gave littleattention to measuring the achievement ofmunicipal service improvements, let alone the









Table G.2: Municipal Management Focus of Region’s Portfolio

A P P E N D I X G : WO R L D B A N K S U P P O R T I N S O UT H A S I A

9 3

impact on beneficiaries. For instance, a completedbus stand was treated as fully achieved, eventhough it had not started functioning because itdid not have the necessary official permits. Eventargets that had measurable goals did not havebaseline data to compare against.

However, some evaluation studies carried out atthe end of India II to inform the follow- upproject did provide useful information on urbanenvironmental indicators. Some of the largermunicipalities collect regular and reliableinformation on service status and achievements,such as water supply per capita per day, thoughthis is still not typical of most municipalities inthe state. In Pakistan II, too, the M&E systemwas overly focused on inputs and outputs, andeven data that were collected were not used toimprove implementation.

Under the Sri Lanka MDP, the lack of baselinedata on the original condition of the Beira Lakewater and the absence of systematic monitoringof changes to it made it impossible to assessproperly the results obtained through theproject.

Urban and spatial planningApart from the update to the Colombo master planunder the Sri Lanka MDP, there was not muchevidence of MDP impact on this in the Region.

Investment planning and strategiesIndia II in Tamil Nadu provided consultanttechnical assistance to and facilitated exchangesamong 45 municipalities to help each preparecorporate plans that set out their investmentpriorities for the following 10 years.


Better financial managementSignificant efforts to improve financial manage-ment were mainly confined to India and Bhutan,with positive results in India but less so in Bhutan.India I and II helped strengthen local municipalcapacity in finance and accounting, includingcomputerization. This occurred through training(35 freestanding courses) for finance and account-

ing officials, as well as for elected representatives.This training helped most local municipalitiesadopt the accrual accounting system and comput-erize the collection of municipal taxes and fees.Now the collection performance of several localmunicipalities can be monitored in real time,making information quickly accessible fordecision makers at the municipal and state level.Compliance has become easier to monitor, andusers find it easier to pay their taxes.

Under the Bhutan MDP, financial reportingduring implementation was weak and was madeworse by lack of technical support within thecountry for the computerized financial manage-ment system that had been customized for theproject. On a related matter, much needs to bedone to build financial systems for cost recovery.

Mobilizing own revenuesUnder India I, Tiruchirapalli municipality(population 775,484) reported an increase of 60percent in revenue between fiscal 2004 and 2005.The municipality used the additional revenue toundertake new infrastructure investments of itsown, such as the water supply in the Srirangamarea. The Madurai municipality increasedproperty tax collections by 20 percent duringfiscal 2004–05, compared to 6 percent for theprevious year. In terms of direct cost recoveryfrom MDP investment, there is little evidence ofsignificant results in the Region.

Under India I, little was done to simplifyprocedures for revising bus fares or tostrengthen the transport corporation in Chennai(population 4.3 million), which continued to bea loss- making entity, unable to invest in orexpand services. Under Bangladesh I, the costrecovery component through Agrani Bank loanswas cancelled, and the property tax collectionsystem remained unchanged.

Municipal creditworthiness and debt managementIndia II encouraged municipalities in Tamil Naduto become creditworthy to have better access toloans awarded by the Tamil Nadu Urban Develop-ment Fund (TNUDF). Under this arrangement,


9 4

local municipalities were given easy access todiscuss and access the TNUDF’s knowledge baseof innovative funding, for which there was a highdemand. By March 2005, 39 percent of TNUDF’sportfolio related to investments in bridges androads, 38 percent to sewerage and sanitation, and17 percent to water supply.

The state’s second largest city, Madurai (popula-tion 909,908), went one step further. With techni-cal assistance provided through the project, themunicipality issued bonds to raise funds to pay forthe construction of an inner ring road that todayyields $1.4 million per annum in toll charges.

Private finance participationMDPs in this Region did little to achieve privatefinancing of projects. India I and II came closestby encouraging the TNUDF to create near- market conditions for municipal investment thatwould begin to interest private financiers. ButIndia I failed to promote the intended privateparticipation of shelter and land development inthe slums of Chennai because of the lack ofinterest on the part of private developers.


Investment prioritiesEstimates for ERRs were rarely made for MDPs inthe Region. One exception was Bhutan’s MDP,which yielded a 25.8 percent ERR at completion,according to government estimates.

ProcurementAs in other Regions, MDPs involved local munici-palities more in preparing and sometimes fullymanaging procurement. Although someshortcomings were still reported, the procure-ment experience of Pakistan I highlighted theeffectiveness of spot checks, especially in themunicipality of Lahore (population 6.3 million), onthe good faith of bids. Such checks helped preventinsider trading and the formation of local cartels.

Under the Sri Lanka MDP, both the municipal-ity of Colombo and the national authoritiesperfected their skills in prequalifying bidders, sothat tenders always included high- quality techni-

cal solutions. The Bhutan MDP introducedcompetitive procurement for municipal worksfor the first time in that country, even thoughmunicipal capacity in this area remains weak.

Operations and maintenanceMDPs in the Region rarely addressed municipalresponsibilities for O&M, and this neglect remainsan ongoing concern. Thus, the benefits fromphysical works under Bangladesh I, for instance,are unlikely to be sustained because of continuedneglect of maintenance. Despite progress inimproving sewerage under Pakistan I, there toomunicipalities’ O&M is adequate. As result, theuncollected waste accumulating in sewers anddrains undermines the benefits of the upgradingthat was done.

Services— Most affected sectorsMDPs in India, Bhutan, Bangladesh, Pakistan, andSri Lanka helped improve urban services andrelated infrastructure. In Tamil Nadu, India I andII contributed positively to services, infrastruc-ture, and security in slums and made someimprovements to urban roads and transportservices. The projects improved living conditionsin 489 slums (against a target of 590), housing76,000 people—or 5 percent of the slum popula-tion—in the 10 largest agglomerations in the state.This was done by providing paved pathways,drains, streetlights, public fountains and baths,and tenure security. Beneficiaries reported healthimprovements and greater social acceptance.There were some shortfalls in transport services:only 4 of 10 depots and 7 of 10 terminals werecompleted. The widening of the inner ring road inChennai was only partially completed because ofdifficulties in acquiring necessary land. For thetransport corporation in Chennai, 1,595 buschassis were procured, but they could not all beused because of financial constraints.

Under Pakistan I, about 300,000 low- incomepeople in Lahore, Gujranwala, Sialkot, andMultan benefited from slum upgrading. InLahore, 21 major roads were improved and newstreet lighting and traffic signals were installed.But the construction of stabilization ponds forsewage treatment in Lahore was deferred for a

A P P E N D I X G : WO R L D B A N K S U P P O R T I N S O UT H A S I A

9 5

later project. Pakistan II’s upgrading report-edly reached 90 communities with 550,000beneficiaries and engaged in road constructionthat saved travel time and improved environ-mental conditions, but evidence for theseassertions was not always clear. They must be indoubt, given the reports of lack of coordinationamong stakeholders and of failed projectinfrastructure that had to be rebuilt prematurely.

Following Bangladesh I, 90 percent of slumdwellers in three Dhaka slums (Islambagh,Raulpur, and Shaheednagar) reported improvedliving conditions through construction of 482latrines. But the project faced implementationshortfalls, again because of difficulties in acquir-ing land. The Bhutan MDP helped improve thequality of life in 10 towns through enhancementsto water supply and other urban infrastructure,although the results were short of targets. Theproject experience enabled municipalities thereto participate in the environmental screening ofsubprojects. Under the Bhutan MDP, there wasimproved interaction between central and localgovernments on the environmental screening ofurban investments. A beneficiary survey revealedthat 73–83 percent of respondents in 3 townsconsidered the water supply to have improved,but this result has to be set against surveys intowns that were not covered by the project thatalso reported similar improvements.

The Sri Lanka MDP did not improve solid wastemanagement in Colombo, despite the citybuilding a large- scale compost operation, whichthe Bank had initially suggested was not the besttechnical solution to the problem. The projecthad greater success in reducing wastewaterpollution in the Beira Lake catchment area. Anindustrial waste system was completed under theproject, and several lakeside dwellings werehooked up to the sewerage system.

Income levels of beneficiaries— Poverty reductionThere is evidence that the modest results of theRegion’s MDP portfolio did nevertheless bringsome benefits to the poor. Pakistan II aimed toreach low- income groups in the Northwest

Frontier Province, but the effectiveness ofpoverty targeting was unclear because of politi-cal interference in beneficiary identification.Improvements in living conditions cannot beattributed to the project.

Under India I, a cross- subsidy from the sale of asmall number of lots for middle- and higher- income households helped finance a number ofserviced plots for the poorest households, whichincluded common open spaces that made bestuse of the land available. Under India II, benefitsreached the poor through slum upgrading inparticular. Also an integrated sanitation programprovided public complexes with toilets andwashing areas in underserved areas such asslums. Beneficiaries— who are typically poor— reported a substantial improvement in theirquality of life. As a result, open defecation wasreported to have decreased by 80 percent.

Conclusions• The positive experiences of India I and II, in-

volving almost 20 years of continuous Bank as-sistance to urban development in the state ofTamil Nadu, suggest that adapting a whole-sale, step- by- step approach to a particular con-text over a sustained period can yield positive results.

• India II has contributed to improving urbaninfrastructure services in Tamil Nadu, directlythrough projects funded by TNUDF and indi-rectly through capacity building in municipal-ities that have made additional infrastructureinvestments using their own funds.

Afghanistan: Kabul Urban Reconstruction. Bangladesh: I— Urban De-velopment; II— Municipal Services. Bhutan: Urban Development. India: I— Tamil Nadu Urban Development; II— Tamil Nadu Second Urban De-velopment; III— Karnataka Municipal Reform; IV— Third Tamil NaduUrban Development. Pakistan: I— Punjab Urban Development; II— Northwest Frontier Province Community Infrastructure; III— PunjabMunicipal Services Improvement. Sri Lanka: Colombo Environment Improvement.

Box G.1: Key to MDPs Referred to in Text

Source: IEG.


9 6

• Even in projects that otherwise performweakly, municipal management can bestrengthened by increasing the responsibili-ties of local government for procurement of

works and goods, as experience in Bhutanand Sri Lanka showed. To minimize risks, spotchecks can be necessary, such as those madein Pakistan I.

9 7

PrinciplesAs a meta- evaluation, this assessment wasdesigned to assemble and review existing IEGevaluation findings about Bank support for MDPsfrom Project Performance Assessment Reports(PPARs) and IEG Reviews of ImplementationCompletion Reports (ICR Reviews). As such, it isanalogous to a literature review, where the litera-ture in this case consists of previous IEG assess-ments, particularly in PPARs.

The 1998–2008 period of the review, chosen forits immediate relevance to ongoing work in thisarea, encompasses a portfolio of all MDPscompleted since 1998, as well as those stillongoing. MDPs that exited between 1998 and2008 generally have an ICR, a self- evaluationprepared by the Bank’s Operations Region, andan ICR Review, an independent assessment doneby IEG based on the ICR. About one- third of theclosed MDPs were approved within the samedecade; the approvals of the remaining two- thirds in some cases dated as far back as 1988.This study considered all MDPs that werecompleted since 1998 and those that are still ongoing.

Municipalities and CitiesThe study used the online World Gazetteerdatabase in Germany, which contains details ofmore than 167,000 named municipalities.Records include census populations, geographiccoordinates of location, and the type of localauthority, in English and in the local language. Inextracting municipality- level population figuresfrom this database, IEG found that 31,000 largermunicipalities, each having 12,500 or moreinhabitants, were home to 3.25 billion people,approximately half the world’s population, and

very close to the 50 percent now reported to livein urban areas.

The correspondence is not exact, however, forthree reasons. First, a larger urban municipalitywith an extensive jurisdiction might containsome rural inhabitants on its periphery,especially if the jurisdiction is large. Second,because concepts of urban population vary fromcountry to country, a local definition of “urban”may not always be comparable with the criterionused in this study. Third, it will not be the case ofa single metropolitan area composed of multiplemunicipal jurisdictions— in such cases, therewould be one city, yet many municipalities. IEG’smethodology may overstate the number of cities,because it counts peripheral rural populations inlarger municipalities as urban. To ensure thatestimates of the number of cities are reliable, IEGtriangulated the results with estimates of theurban population from the World DevelopmentIndicators. The calibration at the country levelconfirmed IEG estimates for this study to bewithin ±10 percent of the World DevelopmentIndicators estimates.

The MDP PortfolioThe study portfolio of MDP operations wasidentified through an internal Bank database. Asa first cut, IEG identified 231 operations classi-fied by one of the Bank’s four related activitycodes: #71 Municipal Management, #72 Munici-pal Finance, #73 Municipal Services, and #74Subnational Government Administration. ThenIEG conducted a keyword search for projectswithout these codes, but with the words munici-pal (and variants), city/cities, local government,and local authorities to identify projects thatworked closely with municipalities and cities, but

APPENDIX H: NOTES ON METHODOLOGY OF EVALUATION


9 8

not classified as such by one of the Bank’s activity codes.

Next the study team eliminated 68 of theseoperations after finding that they did not havemunicipal management objectives or com -ponents despite the activity coding. That left 163MDP projects.

IEG sent the preliminary listings (by Region) ofthis portfolio to Bank urban staff in eachOperational Region, requesting that they helpidentify any Type I and Type II errors in the listsby pointing out operations that had beenincluded that were not MDPs, and otheroperations that were MDPs but were not in thelists. Thanks to excellent responses, IEG was ableto exclude some projects that did not fully meetthe criterion of a direct focus on improvingmunicipal management. IEG could also includeadditional projects overlooked in its first search,ones that lacked a municipal activity code butthat were focused on strengthening municipalmanagement. As a result of this dialogue with theRegions, 14 projects were dropped from theportfolio and 41 projects were added, resultingin a portfolio of 190 MDPs.

The final study portfolio of 190 MDPs included114 closed MDPs and 76 ongoing MDPs. IEGproject reviews are only available for closed MDPs,of course. Entry MDPs have not yet beenevaluated by IEG, nor will they be through thisstudy. But they are considered in the presentstudy, where they stand as evidence of the lessonsof evaluated closed MDPs being carried forward.

About 92 percent of the MDPs in the portfolio aremapped to the Sustainable Development Network. Sixty- six percent are mapped to the Urban SectorBoard, with 12 percent to the Water Sector Board, 9percent to the Transport Sector Board, and 5percent to the Environment Sector Board.

IEG EvaluationsDuring fiscal 1998–2008, IEG completed 17 PPARs— all in different countries, covering 24MDPs, about one- fifth of completed MDPs. TheMDPs chosen for the PPARs were not randomly

selected. When choosing them, IEG appliedvarious considerations: providing input for IEGthematic studies and Country Assistance Evalua-tions as well as ensuring that all six Bank operat-ing Regions were covered.

Criteria for selection of the 24 MDPs chosen forIEG field review through PPARs were varied. Asfar as this study itself is concerned, the mostrelevant criterion was to use the PPAR as an input.This applied to The Gambia MDP, Tanzania I,Indonesia II, VI, and IX, Russia IV, and IndiaI and II. Others were chosen to feed into IEGCountry Assistance Evaluations, includingGeorgia I, II, and III and Colombia I and IV.Some were selected from countries where IEGevaluations of urban projects had been thin,namely China III, IV, and VII, Sri Lanka, andUzbekistan. The remaining projects were partof IEG’s regular program of PPAR assessments.

Prior to this study and as per normal practice,IEG carried out 114 desk ICR reviews, covering100 percent of the completed MDPs. From theICR Reviews, information on the objectives,components, and lessons of each operation werecompiled into the study database.

Municipal Management ThemesTo identify whether an MDP supported one ofthe study’s three municipal managementthemes, IEG conducted keyword searches of theobjectives formulated for each operation. Whenthe appropriate keyword was found, the MDPwas classified as being focused on the particulartheme in question. For each theme, the follow-ing key words (in parentheses) were used: (i)city planning (plan*, strateg*, program*,*tech*, *inst*, *train*, *capa*, *manag*); (ii)municipal finances (finance*, fund*, budget*,fin*/manag*); and (iii) service provision (service,infras*, water, env*). Because the three are notmutually exclusive categories, it was possible foran individual MDP to focus on more than onetheme at the same time.

IEG adopted a similar procedure to identify therelevance of the design of an MDP, throughsimilar keyword searches of the description of a

A P P E N D I X H : N OT E S O N M E T H O D O L O G Y O F E VA L U AT I O N

9 9

project’s components. As with the focus onobjectives, an MDP’s design could cover morethan just one of the study’s themes.

Levels of Assessment of MDP PortfolioThe most intensive assessment in this study,presented in the main report, was based onearlier evaluation findings of the 24 MDPsreviewed by PPARs. The study also looked morebroadly, reporting the findings in the Regional

annexes to this report, at evaluation findings of all114 completed MDPs for which there are ICRReviews. Finally, the study also considered,without evaluation, the 76 ongoing MDPs, inorder to review how the current portfolio contin-ues to address the issues raised by this evaluation.Among other things, the different sets of MDPsexplain the discontinuous nomenclature ofindividual MDPs in the main report and the extracountries and MDPs referred to in the annexes.

Using new municipal services in Pereira, Colombia. Photo courtesy of Roy Gilbert.

1 0 1

Management Comments1. IEG confirms that its assessment is based on how

often MDPs refer explicitly to the poor or poverty re-

duction in their objectives. MDPs with poverty reduc-

tion components or actions that lack a supporting

objective would not be counted. Explicit poverty re-

duction objectives would increase the count of MDPs

covering poverty. They would also make the aims and

purpose of their poverty-related actions clear.

Chapter 11. Launched in 2006 and reporting in 2008, the Com-

mission had 22 leading practitioners, mostly from the

developing world. They were charged with drafting the

policy implications for sustained economic growth. The

Commission was sponsored by bilateral and multilateral

donors (including the World Bank).

2. In some countries, municipalities might go by

other designations not used in this report. Commune,

county, opstina, and wilayat are just a few examples.

Whatever term is used, the generic municipality is typ-

ically headed by an elected or designated council and

mayor, who appoint technical staff and officials to carry

out day-to-day municipal management. Also, as used in

this report, the term city refers to a built-up spatial

concentration of wealth, population, and economic ac-

tivity, as in the 2009 WDR Reshaping Economic Geog-

raphy (World Bank 2008).

3. Currently, the largest municipality in the world is

Chongqing, China, with an estimated population of

31.6 million people.

4. Figures are taken from www.world-gazetteer.com

in Germany.

5. The study does not cover two exceptional arrange-

ments. The first is where several contiguous munici-

palities manage a single megacity, often constituting a

metropolitan area; this example did not arise in the

Bank portfolio reviewed here. The second arrange-

ment is where a single municipality manages several very

small settlements, but this usually occurs in lightly pop-

ulated rural areas, which are not covered by the pres-

ent study.

6. This portfolio was identified in three stages. First,

a keyword search identified all operations within the

1998–2008 period whose formal Bank coded activity in-

cluded the word “municipal” or “subnational.” This gave

a preliminary total of 231 projects. Second, 68 of these

operations were eliminated when closer review revealed

that, despite their activity coding, they did not have mu-

nicipal management objectives or components. This

lowered the count to 163 projects. For the third stage,

IEG sent the list of 163 to all Bank urban sector staff, invit-

ing them to comment and make corrections. Feedback

from these staff led to the removal of 14 and the addi-

tion of 41 projects. These adjustments resulted in a

portfolio of 190 MDPs (details in appendix H).

7. To directly assess the effectiveness of municipal

management support, 17 of these 24 PPAR MDPs were

purposefully selected from all Regions to serve as build-

ing blocks for the present study. The remaining 7 were

chosen for other reasons, including ensuring IEG cov-

erage of under-evaluated countries and more detailed

evaluations of operations whose rating by IEG differed

from the Region’s.

Chapter 21. Project documentation and government reports

may inform the number of municipalities served, but

without identifying each one by name, especially when

a large number of municipalities is involved. Evaluation

would have been easier if all Implementation Comple-

tion Reports for MDPs routinely reported the name, pop-

ulation, and project investment in each municipality

supported. In using these sources to estimate the total

number of municipalities and cities served for this study,

IEG exercised care to avoid double counting munici-

palities that may have been assisted by more than one

MDP operation in a particular country. Finding out just

ENDNOTES

how many cities were assisted, something that has not

been clearly done previously by the Bank, was one of

the questions driving this evaluation.

2. IEG uses 20,000 municipalities as the denomina-

tor here. Countries hosting Bank-financed MDPs ac-

count for the vast majority of all developing country

municipalities—some 18,000, or 90 percent of the total.

3. Examples of apex agencies in wholesale MDPs re-

viewed by PPARs include Georgia II and III—The Mu-

nicipal Fund of Georgia, www.mdf.org.ge/; Brazil

II—Ceará state government, Secretariat of Cities,

http://www.cidades.ce .gov.br/; Chile II—Subsecretariat

of Regional Development, http://www.subdere.gov.cl;

Colom bia I and IV—Local Development Fund (FINDETER),

http://www.findeter.gov.co/.

4. An extreme case of this is the municipality of

Chongquing, whose 82,000 square kilometer jurisdic-

tion embraces more than 30 million people in 7 large

and 25 small and medium-sized cities, as well as nu-

merous tiny rural settlements. This extensive area is

more than 10 times the 8,051 square kilometers of the

municipality of Sao Paulo, Brazil, which is home to 10.4

million people.

5. For this, the study transformed the six category out-

come ratings into a six-point numerical scale, where

highly satisfactory = 6, satisfactory = 5, and so on. The

difference of the mean scores proved to be highly sig-

nificant (t statistic = 2.3012, significant at 99 percent).

On this scale, the mean of the wholesale MDP rating was

4.46 (satisfactory), and the mean of the retail MDP out-

come was 3.97 (moderately satisfactory).

6. This difference of means was found to be statis-

tically significant (t statistic = 2.5821, significant at 99

percent).

7. This compares unfavorably with an earlier IEG find-

ing for the urban portfolio as a whole, which was 53 per-

cent of completed projects with poverty-focused

objectives; 69 percent of ongoing projects had this

focus (IEG 2004, pp. 11–12).

Chapter 31. This report provides ratings of 11 specific MDP

achievements across the municipal management di-

mensions of planning, finance, and service provision. For

the municipal information system results in this in-

stance and for all others, the assessments are based on

the observed efficacy of the actual results obtained by

the specific management improvement. A rating of

“substantial” means that the expected result was fully

achieved; a rating of “modest” means that the expected

result was only partly achieved.

2. Since IEG began rating M&E performance in July

2007, only 18 percent of all completed MDPs achieved

a rating of substantial for their M&E.

3. This is now known more generically as a Devel-

opment Policy Loan.

4. For more detailed evaluation findings on the Cities

Alliance, see IEG (2008).

Chapter 41. Taking own revenues and transfers together, total

municipal revenues can account for up to 6 percent of

gross domestic product (Shah 2006, p. 35).

2. An earlier IEG evaluation of MDPs in Brazil (IEG

1999) found that these operations obtained robust fi-

nancial results; the financial performance of munici-

palities receiving MDP support was significantly stronger

than that of unassisted municipalities.

3. A primary objective of the World Bank Group’s new

subnational finance facility is to strengthen a (munici-

pal) borrower’s ability to deliver key infrastructure serv-

ices and to improve their efficiency and accountability

as services providers: http://www.ifc.org/ifcext/ subna-

tionalfinance.nsf/ Content/Home.


1 0 2

1 0 3

BIBLIOGRAPHY

Annez, Patricia, Gwenaelle Huet, and George Peterson.

2008. Lessons for the Urban Century: Decentral-

ized Infrastructure Funds in the World Bank. Di-

rections for Development Series— Infrastructure.

Washington, DC: World Bank.

Boadway, Robin, and Anwar Shah (eds.). 2007. Inter-

governmental Fiscal Transfers. Washington DC:

World Bank.

Spence, Michael. 2008. “The Growth Report: Strategies

for Sustained Growth and Inclusive Development.”

Commission on Growth and Development Final Re-

port, Washington, DC.

Davis, Mike. 2006. Planet of Slums. London, Verso.

Duranton, Gilles. 2008. “Cities: Engines of Growth and

Prosperity for Developing Countries?” Working Paper

No. 12 of Commission on Growth and Develop-

ment, World Bank, Washington DC.

IEG (Independent Evaluation Group). 2008. “The Cities

Alliance.” Global Program Review Vol. 1 (4).

———. 2004. Improving the Lives of the Poor through

Investment in Cities: An Update on the Performance

of the World Bank’s Urban Portfolio. Washington, DC:

World Bank.

———. 1999. Developing Towns and Cities: Lessons

from Brazil and the Philippines. Washington, DC:

World Bank.

Leautier, Frannie. 2006. Cities in a Globalizing World:

Governance, Performance and Sustainability. Wash-

ington DC, World Bank Institute.

Peterson, George E., and Patricia Clarke Annez. 2007. Fi-

nancing Cities: Fiscal Responsibility and Urban In-

frastructure in Brazil, China, India, Poland and

South Africa. Los Angeles: Sage Publications.

Ravallion, Martin, Shaohua Chen, and Prem Sangraula.

2007. “New Evidence on the Urbanization of Global

Poverty.” Policy Research Paper 4199, World Bank,

Washington, DC.

Shah, Anwar (ed.). 2006. Local Governance in Devel-

oping Countries. Washington, DC: World Bank.

UNFPA (United Nations Population Fund). 2007. State

of the World Population 2007: Unleashing the Po-

tential of Urban Growth. New York: UNFPA.

World Bank. 2008. World Development Report 2009: Re-

shaping Economic Geography. Washington, DC,

and New York: World Bank and Oxford University

Press.

———. 2006. “China: Governance, Investment Climate,

and Harmonious Society: Competitiveness En-

hancements for 120 Cities in China.” Poverty Re-

duction and Economic Management, Financial and

Private Sector Development Unit, East Asia and Pa-

cific Region Report No. 37759-CN, World Bank, Wash-

ington DC.

———. 2004. World Development Report 2004: Mak-

ing Services Work for Poor People. Washington, DC,

and New York: World Bank and Oxford University

Press.

———. 2003. World Development Report 2003: Sus-

tainable Development in a Dynamic World—

Transforming Institutions, Growth, and Quality of

Life. Washington, DC, and New York: World Bank and

Oxford University Press.

———. 2002. World Development Report 2002: Build-

ing Institutions for Markets. Washington, DC, and

New York: World Bank and Oxford University Press.

———. 2001. World Development Report 2001: At-

tacking Poverty. Washington, DC, and New York:

World Bank and Oxford University Press.

———. 2000a. Cities in Transition: World Bank Urban

and Local Government Strategy. Washington, DC:

World Bank.

———. 2000b. World Development Report 1999/2000:

Entering the 21st Century. Washington, DC, and

New York: World Bank and Oxford University Press.

———. 1989. “Report of the Task Force on Financial Sec-

tor Operations.” Report No. R89-163, World Bank,

Washington, DC.

Yussuf, Shahid. 2007. “About Urban Mega Regions:

Knowns and Unknowns.” Policy Research Paper

4252, World Bank, Washington, DC.

WORKING FOR A WORLD FREE OF POVERTY

The World Bank Group consists of five institutions—the International Bank for Reconstruction and Development(IBRD), the International Finance Corporation (IFC), the International Development Association (IDA), theMultilateral Investment Guarantee Agency (MIGA), and the International Centre for the Settlement of InvestmentDisputes (ICSID). Its mission is to fight poverty for lasting results and to help people help themselves and their envi-ronment by providing resources, sharing knowledge, building capacity, and forging partnerships in the public andprivate sectors.

THE WORLD BANK GROUP

IMPROVING DEVELOPMENT RESULTS THROUGH EXCELLENCE IN EVALUATION

The Independent Evaluation Group (IEG) is an independent, three-part unit within the World Bank Group. IEG-World Bank is charged with evaluating the activities of the IBRD (The World Bank) and IDA, IEG-IFC focuses onassessment of IFC’s work toward private sector development, and IEG-MIGA evaluates the contributions of MIGAguarantee projects and services. IEG reports directly to the Bank’s Board of Directors through the Director-General,Evaluation.

The goals of evaluation are to learn from experience, to provide an objective basis for assessing the results of theBank Group’s work, and to provide accountability in the achievement of its objectives. It also improves Bank Groupwork by identifying and disseminating the lessons learned from experience and by framing recommendations drawnfrom evaluation findings.

THE INDEPENDENT EVALUATION GROUP

SKU 18043

ISBN 978-0-8213-8043-7

Ë|xHSKIMBy380437zv*:+:!:+:!

improving municipal management for cities to succeed - the … · 2016-07-10 · the welfare impact...

Documents