THE WORLD BANK 38285 Engaging with Fragile States An IEG Review of World Bank Support to Low-Income Countries Under Stress THE WORLD BANK GROUP 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), the Multilateral Investment Guarantee Agency (MIGA), and the International Centre for the Settlement of Investment Disputes (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 and private sectors. THE INDEPENDENT EVALUATION GROUP ENHANCING DEVELOPMENT EFFECTIVENESS THROUGH EXCELLENCE AND INDEPENDENCE 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 on assessment of IFC’s work toward private sector development, and IEG-MIGA evaluates the contributions of MIGA guarantee 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 the Bank Group’s work, and to provide accountability in the achievement of its objectives. It also improves Bank Group work by identifying and disseminating the lessons learned from experience and by framing recommendations drawn from evaluation findings. WORLD BANK INDEPENDENT EVALUATION GROUP Engaging with Fragile States An IEG Review of World Bank Support to Low-Income Countries Under Stress 2006 The World Bank http://www.worldbank.org/ieg Washington, D.C. ©2006 The International Bank for Reconstruction and Development / The World Bank 1818 H Street NW Washington DC 20433 Telephone: 202-473-1000 Internet: www.worldbank.org E-mail: feedback@worldbank.org All rights reserved 1 2 3 4 5 09 08 07 06 This volume is a product of the staff of the International Bank for Reconstruction and Development / The World Bank. The findings, interpretations, and conclusions expressed in this volume do not necessarily reflect the views of the Executive Directors of The World Bank or the governments they represent. The World Bank does not guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgement on the part of The World Bank concerning the legal status of any territory or the endorsement or acceptance of such boundaries. Rights and Permissions The material in this publication is copyrighted. Copying and/or transmitting portions or all of this work without permission may be a violation of applicable law. The International Bank for Reconstruction and Development / The World Bank encourages dissemination of its work and will normally 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 Clearance Center 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, 1818 H Street NW , Washington, DC 20433, USA; fax: 202-522-2422; e-mail: pubrights@worldbank.org. Photo credit: “The Horror, the Hope.” Child in refugee camp for people fleeing war in Rwanda; Democratic Republic of Congo. ©Mikkel Ostergaard. ISBN-10: 0-8213-6847-8 ISBN-13: 978-0-8213-6847-3 eISBN: 0-8213-6848-6 DOI: 10.1596/978-0-8213-6847-3 Library of Congress Cataloging-in-Publication Data Carvalho, Soniya. Engaging with fragile states : an IEG review of World Bank support to low-income countries under stress / author, Soniya Carvalho. p. cm. — (Independent Evaluation Group study series) Includes bibliographical references. ISBN-13: 978-0-8213-6847-3 ISBN-10: 0-8213-6847-8 ISBN-10: 0-8213-6848-6 (electronic) 1. World Bank—Evaluation. 2. Economic development projects—Evaluation. 3. Economic assistance—Developing countries—Evaluation. I. World Bank. Independent Evaluation Group. II. Title. III. Title: IEG review of World Bank support to low-income countries under stress. HG3881.5.W57C37 2006 332.1'532—dc22 2006034735 World Bank InfoShop Independent Evaluation Group E-mail: pic@worldbank.org Knowledge Programs and Evaluation Capacity Telephone: 202-458-5454 Development (IEGKE) Facsimile: 202-522-1500 E-mail: eline@worldbank.org Telephone: 202-458-4497 Printed on Recycled Paper Facsimile: 202-522-3125 Contents vii Acknowledgments ix Foreword xiii Avant-propos xvii Prefacio xxiii Executive Summary xxxvii Résumé analytique liii Resumen lxix Acronyms and Abbreviations 3 1 Background 6 The LICUS Approach 8 Objectives, Purpose, and Organization of the Review 9 Review Instruments and Methods 15 2 Effectiveness of the Bank’s LICUS Approach 15 Stay Engaged 19 Anchor Strategies in Stronger Sociopolitical Analysis 22 Promote Domestic Demand and Capacity for Positive Change 27 Support Simple and Feasible Entry-Level Reforms 31 Explore Innovative Mechanisms for Social Service Delivery 32 Work Closely with Other Donors 38 Measure and Monitor Results 39 Overall LICUS Approach 42 Conclusion 45 3 Operational Utility of the LICUS Identification, Classification, and Aid-Allocation System 45 LICUS Identification 46 LICUS Classification 47 Aid-Allocation System for LICUS 49 Conclusion iii E N G A G I N G W I T H F R A G I L E S T AT E S 51 4 The Bank’s Internal Support for LICUS Work 51 Analytical Work 52 Staffing and Incentives 54 Operational Policies and Procedures 57 Management Attention and Operational Guidance 61 Conclusion 63 5 Conclusions, Lessons, and Recommendations 63 Conclusions 65 Lessons of Experience for the Bank and Other Donors 71 Recommendations 73 Appendixes 75 Definitions and Data Sources A: 81 LICUS, Fiscal 2003–06 B: 83 Progress on Millennium Development Goals C: 85 Four LICUS Business Models D: 87 Summary of OECD-DAC Principles for International Engagement in E: Fragile States 89 F: Four Bank Themes Based on the OECD-DAC Principles 91 G: Fiscal 2003–05 Country Strategy Documents for Fiscal 2005 LICUS 93 H: Implications of the HIPC Initiative for LICUS 95 I: Breakdown by Country: Lending and Administrative Budget— Total and for Analytical Work 97 J: Political Analysis Relevant for Strategy Development in LICUS 99 K: Capacity Development Projects in LICUS 103 L: Accountability in the Management of Natural Resources in LICUS 105 M: Development Policy Lending 107 N: The Challenges of Engaging Civil Society Organizations in LICUS 109 O: The World Bank’s Economic and Sector Work in LICUS 113 P: Performance of Active Projects 115 Q: Performance of Closed Projects and Lessons in LICUS 125 R: Country Assistance Evaluations and CAS Completion Report Reviews 127 S: Human Resource Proposals for Improving Staffing in LICUS 129 T: Operational Policies and Bank Procedures Relevant to LICUS 131 U: Safeguards in LICUS 133 V: Key Guidance Notes for LICUS 135 W: Overview of the World Bank Institute’s LICUS Activities 137 X: Overview of the Development Economics Vice Presidency’s LICUS Activities 139 Y: Donor Relationships 141 Z: Stakeholder Perceptions Based on Survey Results 155 AA: Statement by the External Advisory Panel 157 BB: Management Response 169 CC: Chairman’s Summary: Committee on Development Effectiveness (CODE) 175 Endnotes 181 Bibliography iv CONTENTS Boxes 4 1.1 The LICUS Challenge: Views from the Field 7 1.2 The 2002 LICUS Approach: Core Principles 22 2.1 Lao People’s Democratic Republic: Good Practice Example of Using Existing Political Knowledge 25 2.2 Afghanistan Priority Restructuring and Reform Process: Developing State Capacity through the Direct Restructuring of the Civil Service 26 2.3 The Complexities of State Building in LICUS: Is the Bank Ready? 28 2.4 Areas for Further Development in the Bank’s Approach to Capacity Development 30 2.5 Three Lessons from Post-Conflict Countries 31 2.6 Timor-Leste: Excessive Optimism, Impatience, and Partial Solutions? 38 2.7 Afghanistan: Lack of a Common Vision among Donors Works against Effective Action 48 3.1 Adjustments to IDA’s Performance-Based Allocation System That Affect LICUS 55 4.1 OP/BP 8.50: A Major Source of Delays in Liberia 56 4.2 Conceptual Problems in Applying OP/BP 8.50 to LICUS Environments 56 4.3 Examples of Procurement Problems in LICUS Figures xxvii ES.1 The Majority of Bank Respondents Said There Has Been No Change When Working on LICUS in Several Areas xlii RA.1 La majorité des personnes interrogées à la Banque ont indiqué que le fait de travailler sur les LICUS n’avait rien changé dans plusieurs domaines lviii R.1 En la mayor parte de las respuestas del personal del Banco se sostiene trabajar en el área de los PIBD no produjo cambio alguno en varias esferas 6 1.1 A Larger Share of LICUS Lending during Fiscal 2003–05 Went to Post-Conflict LICUS, While Administrative Budgets Were More Evenly Distributed across the LICUS Group 10 1.2 LICUS Population Concentrated in Africa 11 1.3 More than Half the Population of Four LICUS (of Eight with Data) Earns Less than $1 a Day 11 1.4 Negative Growth Rate in about Half of LICUS, Lower Growth Rate in Most LICUS Compared with Low-Income Country Average 12 1.5 External Debt More than 175 Percent of GNI for Six LICUS and Higher than Low-Income Country Average for All LICUS in 2004 12 1.6 Human Development Index for LICUS Worse than for Low-Income Countries in 2003 13 1.7 Kaufmann, Kraay, and Mastruzzi (KKZ) Governance Indicators Worse for LICUS than for Non-LICUS Low-Income Countries 19 2.1 Over 40 Percent of Respondents Said That the Bank’s Analytical Work Has Achieved Its Intended Results Only to a Slight Extent or Not at All 21 2.2 Bank Respondents Somewhat More Likely than Other Respondents to Say That the Bank’s Work Has Been Grounded Largely in an Understanding of the Country’s Politics v E N G A G I N G W I T H F R A G I L E S T AT E S 33 2.3 Most Respondents Said That the Use of Nongovernmental or Semi- Autonomous Arrangements Has Made a Small Positive Contribution 36 2.4 The Majority of Bank Respondents Said the Bank Has Pursued Collaboration with Donors to a Large Extent; the Majority of In-Country Respondents Said It Has Done So to a Large or Moderate Extent 37 2.5 The Majority of Respondents Noted a Positive Change in the Bank’s Effectiveness with Donor Collaboration 41 2.6 The Majority of Stakeholder Respondents Said the Bank’s Overall Program Made a Small Positive Contribution to Development of LICUS 41 2.7 The Majority of Stakeholder Respondents Said Development Would Have Been Smaller without Bank Support 41 2.8 The Majority of World Bank Respondents Said the Bank’s Contribution to Development Was Greater than That of Other Donors 41 2.9 The Majority of Stakeholder Respondents Said World Bank Lending and Grant Support to LICUS Has Achieved Its Intended Results to a Moderate or Slight Extent 42 2.10 Comparing the Pre-LICUS and Post-LICUS Initiative Periods, the Majority of Stakeholder Respondents Said There Is Improvement in the Effectiveness of the Bank’s Lending and Nonlending Support 48 3.1 Highly Variable per Capita IDA and Trust Fund Financing across LICUS during Fiscal 2003–05 53 4.1 Change in Various Factors When Working on LICUS 58 4.2 Bank Respondents’ Views on Effectiveness of LICUS Unit 62 4.3 World Bank Institute’s Technical Input 62 4.4 Development Economics Vice Presidency’s Technical Input Tables xxv ES.1 Implementation Experience on the Core Country-Level LICUS Principles xxxix RA.1 Mise en oeuvre des principes fondamentaux définis pour les LICUS lv R.1 Experiencia en materia de aplicación de los principios básicos de la Iniciativa para los PIBD en los países 5 1.1 Twenty-Five LICUS, Fiscal 2005 17 2.1 Lending, Trust Funds, and Administrative Budgets Going to LICUS and Non-LICUS Low-Income Countries during Fiscal 2000–02 and 2003–05 23 2.2 Capacity Development Lending and Trust Funds in LICUS 43 2.3 Implementation Experience with the Core Country-Level LICUS Principles 52 4.1 Overall Balance Between Knowledge and Finance for Administrative Budgets and Lending vi Acknowledgments This review was prepared by a team led by Shonar Lala, and S. Ramachandran commented Soniya Carvalho. Manuel Penalver provided on sections of the review. advice, and Marcello Basani, Alexandru William Kraus, Mary McIntosh, and Matthew Cojocaru, Tara Lonnberg, and Anju Vajja Petri helped design and administer the Stake- contributed to all aspects of the review. Howard holder Survey. William Hurlbut edited the White provided guidance to the work on aid original manuscript and provided document effectiveness and aid allocation. production support. Heather Dittbrenner and Fieldwork and thematic background analysis Caroline McEuen edited and produced the were undertaken by Nils Boesen, Alison Evans, report for publication. The team was assisted by Stephen Jones, Kavita Mathur, Desmond Romayne Pereira. McCarthy, Marina Ottaway, Manuel Penalver, The External Advisory Panel comprised Hans Rothenbuhler, Roger Slade, Inder Sud, and Olayinka Creighton-Randall, John Githongo, Peter Whitford. Pieter Stek, and Gunnar Sørbø. Reviewers included Jens Andvig, Stein Eriksen, The review was conducted in partnership Alan Gelb, Paul Isenman, Hasan Tuluy, and Indra with the Evaluation Department of the Øverland. Alberto Agbonyitor, Ivar Andersen, Norwegian Agency for International Develop- Sarah Cliffe, Meron Desta, Alison Vale Gillies, Blair ment (Norad), which provided both substantive Stephen Glencorse, Barbry Keller, Xavier Legrain, and financial support. Christian Lotz, Trayambkeshwar Sinha, Frederik The review was prepared under the direction van Bolhuis, Melvin Vaz, and Per Egil Wam shared of Alain Barbu, Manager of the Independent knowledge and insights. Deepa Chakrapani, Evaluation Group, Sector, Thematic, and Global Catherine Gwin, Fareed Hassan, Nalini Kumar, Evaluation Division. Director-General, Evaluation: Vinod Thomas Director, Independent Evaluation Group–World Bank: Ajay Chhibber Manager, Sector, Thematic, and Global Evaluation Division: Alain Barbu Task Manager: Soniya Carvalho Note: The World Bank used the term “Low-Income Countries Under Stress (LICUS)” until January 2006, when the Bank adopted the term “fragile states.” Given that this report evaluates the Bank’s fragile state experi- ence—most of which occurred before the change in terminology—it uses the term “LICUS.” The term “fragile states” is used when it appears in titles and when discussing these countries in a broader international context. vii Foreword ome to almost 500 million people, roughly half of whom earn less than H a dollar a day, fragile states, until recently known in the World Bank as Low-Income Countries Under Stress (LICUS), have attracted in- creasing attention. Concern is growing about the ability of these countries to reach development goals as well as about the adverse economic effects they have on neighboring countries and the global spillovers that may follow. With their multiplicity of chronic problems, These 25 countries have a number of similar- LICUS pose some of the toughest development ities: their infant mortality rate is a third higher challenges. Most have poor governance. Many, than that of other low-income countries, life like Sudan, are embroiled in extended internal expectancy is 12 years lower, and their maternal conflicts. Some, like Timor-Leste, are struggling mortality rate is about 20 percent higher. There through tenuous post-conflict transitions. All are also important differences among LICUS. face similar hurdles: weak security, fractured Some, Angola and Cambodia among them, grew societal relations, corruption, breakdown in the at around 4 percent per annum during rule of law, and lack of mechanisms for generat- 1995–2003; others, such as the Solomon Islands, ing legitimate power and authority. As low- the Democratic Republic of Congo, and Guinea- income countries, these countries also have a Bissau, had negative growth rates of similar huge backlog of investment needs and limited magnitude. Some, such as Angola, the Democra- government resources to meet them. tic Republic of Congo, Nigeria, and Papua New Past international engagement with these Guinea, have abundant natural resources, but countries has failed to yield significant improve- others, such as Burundi and Haiti, are resource- ments, and donors and others continue to poor. These differences are recognized in struggle with how best to assist fragile states. specific business models the Bank has LICUS, as the Bank has called fragile states since developed to work with countries in crisis: 2002, are characterized by weak policies, institu- deterioration, prolonged crisis or impasse, post- tions, and governance. The Bank identified 25 conflict or political transition, and gradual such countries in fiscal 2005 based on their improvement. income and Country Policy and Institutional During fiscal 2003–05, lending and adminis- Assessment (CPIA) rating. trative budgets to LICUS stood at $4.1 billion and ix E N G A G I N G W I T H F R A G I L E S T AT E S $161 million—increases of 67 percent and 55 responsibilities of the LICUS Unit and the percent compared with fiscal 2000–02. The Conflict Prevention and Reconstruction Unit. LICUS approach has evolved from general aid Finally, the Bank has yet to address the alloca- effectiveness concerns in 2002 to state-building tion of aid for LICUS in a way that reflects its and peace-building objectives in 2005. IEG’s objectives for these countries and ensures that assessment of experience with the Bank’s LICUS LICUS are not under- or over-aided. approach found some early successes, but signif- With regard to effectiveness, it remains too icant remaining challenges. early to judge the outcomes of the Bank’s efforts. However, some indicators suggest that Early Successes the overall impact may have been limited. The The Bank has improved its operational CPIA rating for LICUS has shown an improving readiness to engage with LICUS. It has increased trend since the launch of the LICUS Initiative, its analytical work and introduced the use of but the Kaufmann, Kraay, and Mastruzzi (KKZ) Interim Strategy Notes to design strategies governance indicator for LICUS shows a deterio- covering a shorter period to accommodate rating trend over the same period. Neither trend volatile LICUS conditions. LICUS managers have is necessarily attributable to the donor actions, also gained greater access to senior Bank but sustained effort seems to be warranted. management and increased guidance on a number of important issues. The Bank has also Lessons for the Bank and Other Donors initiated the LICUS Trust Fund to finance Engagement needs to be quickly followed by a countries in non-accrual; these countries clear and relevant reform agenda. In the Central previously had little access to finance. African Republic, good initial results are now at These moves have helped the Bank contribute risk of being diminished due to inadequate to improved macroeconomic stability and deliver attention to the budget situation. significant amounts of physical infrastructure, Donor efforts need to focus on internalizing especially in post-conflict LICUS. Substantial political analysis in strategy design and progress has also been made on donor coordina- implementation. While the Interim Strategy in tion at the international policy level. Papua New Guinea contained a good discussion of the political system and recognized problems Challenges such as clan loyalties, political patronage, Significant challenges remain, however. The corruption, and lack of capacity, it treated these reforms in some LICUS have lacked selectivity problems as technical in nature and did not and prioritization. The Bank’s effectiveness needs adequately use them to underpin the overall to be improved after the immediate post-conflict approach. phase when structural change is needed. The The analysis does not have to be developed Bank has not yet sufficiently internalized political internally, however. In Lao People’s Democratic understanding in its country strategies. The Republic, the Bank effectively tapped existing strong donor coordination at the international political analysis and avoided the higher costs of policy level has not carried over to the country preparing its own analysis, as well as potential level. Most important, the Bank has made state tension with the government. building a central focus without adequately Appropriate sequencing of reforms and demonstrating how past weaknesses will be sufficient time to implement them are crucial for avoided and better capacity development and achieving results without overwhelming country governance outcomes ensured. capacity. In Afghanistan, donor reforms have not Internally, progress remains unsatisfactory on been selective enough and have led to 120 pieces critical human resource reforms relating to of legislation. In São Tomé and Principe, the Bank staffing numbers, staffing quality, and incentives was far too ambitious and many of the Country to undertake LICUS work. There is significant Assistance Strategy (CAS) objectives were not duplication and confusion about the roles and achieved or were only partially achieved. x FOREWORD Donor coordination cannot succeed without tailoring donor approaches. In addition to a shared vision and purpose. In Afghanistan and recently issued notes, guidance is especially Tajikistan, donors did not subscribe to a single needed for countries in deterioration and clear objective, which made it difficult to achieve prolonged crisis or impasse, and for the transi- policy coherence. tion and development phases that follow the Monitoring and evaluation are at least as immediate reconstruction phase in post-conflict important in LICUS as they are in any other or political transition countries. country. In volatile country environments, where progress is often non-linear, program Recommendations adaptation is essential. Close tracking of • Clarify the scope and content of the Bank’s performance can help determine when adapta- state-building agenda and strengthen the de- tion is necessary and what kind may be sign and delivery of capacity development and appropriate. governance support in LICUS to ensure better Effective country strategy implementation outcomes. requires not only field presence but also • Develop aid-allocation criteria to ensure that adequate communication between field and LICUS are not under- or over-aided. headquarters staff, as well as an adequate • Strengthen internal support for LICUS work number of field staff with the appropriate over the next three years. It is particularly im- authority and skills. In Cambodia, the Bank’s portant to ensure adequate incentives to attract field presence has significantly improved qualified staff—both at headquarters and in understanding of the political situation, but this field offices—to work in LICUS and to ensure knowledge may still be highly concentrated in a an efficient organizational arrangement that few managers and staff, with relatively limited removes duplication and fragmentation be- dissemination to the broader country team. In tween LICUS and the Conflict Prevention and Angola, the initial lack of operational staff in the Reconstruction Units. field office who could work with ministry staff to • Reassess the value added by the LICUS ap- prepare for high-level meetings between proach after three years, when sufficient ex- ministers and the Bank resulted in issues perience on the outcomes of the approach moving too quickly to the top, which created will be available, and base continued Bank sup- unnecessary tensions. port for the LICUS category and approach on Better operational guidance is needed for the findings of that reassessment. Vinod Thomas Director-General Evaluation xi Avant-propos vec près de 500 millions d’habitants dont la moitié gagnent moins de A 1 dollar par jour, les États fragiles (jusqu’ à récemment dénommés « pays à faible revenu en difficulté » par la Banque mondiale) attirent une at- tention grandissante. Il y a tout lieu de douter de l’aptitude de ces pays à at- teindre leurs objectifs de développement et de craindre des effets économiques défavorables sur les pays voisins ainsi que les retombées qui pourraient s’en- suivre au niveau mondial. Accablés de problèmes chroniques, ces pays ont LICUS, terme adopté par la Banque en 2002 pour des impératifs de développement qui présen- désigner les États fragiles, se caractérisent par des tent d’énormes défis. La plupart de ces pays politiques, des institutions et une governance souffrent d’une mauvaise gouvernance. faibles. La Banque a recensé 25 LICUS durant Beaucoup, comme le Soudan, sont en proie à l’exercice 2005, sur la base de leur revenu et de des conflits internes de longue date. Certains, leur performance dans le cadre des évaluations de comme le Timor-Leste, connaissent des la politique et des institutions nationales (CPIA). difficultés de transition post-conflictuelle. Tous Ces 25 pays présentent plusieurs similarités : sont confrontés aux mêmes problèmes : taux de mortalité infantile supérieur de 33 % à manque de sécurité, relations sociales fragmen- celui des autres pays à faible revenu, espérance tées, corruption, détérioration de l’ordre public de vie inférieure de 12 ans, et taux de mortalité et absence de mécanismes de mise en place de maternelle supérieur de quelque 20 %. Il existe pouvoirs légitimes. En tant que pays à faible également des différences importantes entre les revenu, ils ont également d’énormes besoins LICUS. Certains, parmi lesquels l’Angola et le d’investissement en souffrance et des Cambodge, ont enregistré une croissance ressources publiques limitées pour faire face à annuelle de quelque 4 % durant la période 1995- ces besoins. 2003 ; d’autres, tels que les Iles Salomon, la L’action internationale menée jusqu’à présent République démocratique du Congo et la dans ces pays n’a pas produit d’améliorations Guinée-Bissau, ont vu leur croissance ralentir notables et les bailleurs de fonds et tous ceux d’autant. Certains, comme l’Angola, la concernés continuent de s’interroger sur les République démocratique du Congo, le Nigeria meilleurs moyens d’aider les États fragiles. Les et la Papouasie-Nouvelle-Guinée, possèdent xiii E N G A G I N G W I T H F R A G I L E S T AT E S d’abondantes ressources naturelles, tandis que Il reste cependant des défis de taille à relever. d’autres, comme le Burundi et Haïti, sont dotés Par exemple, les réformes mises en place dans de ressources limitées. Les modèles d’interven- certains LICUS n’ont pas été assez sélectives tion spécifiques établis par la Banque pour quant à la définition des priorités. Il importe travailler avec ces pays en crise tiennent compte d’accroître l’efficacité de la Banque dans la de ces divergences : détérioration, crise phase post-conflictuelle, lorsque des change- prolongée ou impasse, situation post- ments structurels s’imposent. La Banque ne conflictuelle ou transition politique, et améliora- tient pas ensore suffisamment compte des tion graduelle. réalités politiques dans ses stratégies-pays. Durant l’exercice 03-05, les prêts et les L’étroite coordination des bailleurs de fonds sur budgets administratifs en faveur des LICUS se le plan international ne s’est pas répercutée au sont établis à 4,1 milliards de dollars et 161 niveau national. Aussi et surtout, la Banque a mis millions de dollars – soit une hausse de 67 % et l’accent sur l’édification de l’État sans démontrer 55 %, respectivement, par rapport à l’exercice de manière adéquate de quelle façon les 00-02. La démarche suivie pour les LICUS a faiblesses passées seront évitées ni comment les évolué entre 2002 et 2005, le centre des résultats seront améliorés sur le plan du préoccupations étant passé de l’efficacité développement des capacités et de la globale de l’aide à l’édification de l’État et à la gouvernance. consolidation de la paix. D’après l’évaluation de Sur le plan interne, les progrès accomplis l’IEG, la démarche suivie par la Banque pour les restent insuffisants en ce qui concerne les LICUS a été initialement couronnée de succès importantes réformes à effectuer dans le mais il reste d’importants obstacles à surmonter. domaine des ressources humaines (effectifs, qualité du personnel et mesures d’encourage- Succès initiaux ment en faveur des travaux effectués au titre des La Banque a amélioré ses capacités LICUS). Les doubles emplois sont fréquents et il opérationnelles d’intervention dans les LICUS. règne une grande confusion quant au rôle et aux Elle a accru ses travaux d’analyse et utilise à responsabilités de l’équipe chargée des LICUS présent des Notes de stratégie intérimaire pour d’une part, et celle chargée de la prévention des élaborer des stratégies portant sur une plus conflits et de la reconstruction d’autre part. courte période afin de tenir compte de l’instabil- Enfin, la Banque doit encore déterminer ité de la situation dans les LICUS. Les respons- comment allouer l’aide aux LICUS en tenant ables des LICUS ont aussi plus facilement accès à compte de ses objectifs pour ces pays et en l’équipe de direction de la Banque et reçoivent évitant que cette aide soit insuffisante ou davantage d’orientations sur les questions excessive. importantes. La Banque a d’autre part créé le S’agissant de l’efficacité, il est encore trop tôt Fonds fiduciaire LICUS pour fournir une aide pour évaluer l’impact des mesures prises par la financière aux pays dont la dette est improduc- Banque. Certains indicateurs donnent tive ; dans le passé, ces pays avaient difficilement cependant à penser que l’impact global risque accès à des moyens de financement. d’être limité. Bien que la performance des Ces initiatives ont permis à la Banque d’aider LICUS se soit améliorée sur le plan de la à améliorer la stabilité macroéconomique et à politique et des institutions nationales depuis le mettre en place une importante infrastructure lancement de l’initiative en faveur de ces pays, matérielle, notamment dans les LICUS sortant l’indicateur de gouvernance KKZ pour les LICUS d’un conflit. D’importants progrès ont reflète une détérioration sur la même période. également été réalisés dans le domaine de la Ni l’une ni l’autre tendance n’est nécessaire- coordination des bailleurs de fonds sur le plan ment attribuable à l’action des bailleurs de de la politique internationale. fonds, mais il y a lieu de poursuivre les efforts. Défis à relever Enseignements à tirer par la Banque et xiv AVA N T- P R O P O S les autres bailleurs de fonds rigoureux des résultats peut aider à déterminer Les engagements pris doivent être rapidement si une adaptation est nécessaire et quelle forme suivis d’un programme de réforme clair et elle doit revêtir. pertinent. En République centrafricaine, les La bonne exécution d’une stratégie-pays résultats initialement favorables risquent aujour- exige à la fois une présence sur le terrain et une d’hui d’être compromis par le manque d’atten- communication adéquate entre les services tion à la situation budgétaire. extérieurs et le siège, ainsi qu’un personnel de Les bailleurs de fonds doivent s’attacher à terrain suffisamment nombreux et doté des tenir compte des réalités politiques dans la pouvoirs et des compétences voulus. Au conception et la mise en œuvre des stratégies. Cambodge, la présence de représentants de la Alors que la stratégie intérimaire en Papouasie- Banque a sensiblement amélioré la compréhen- Nouvelle-Guinée contenait une bonne analyse sion de la situation politique, mais ce savoir reste du système politique et reconnaissait les probablement l’apanage de quelques dirigeants problèmes en présence, tels que les allégeances et leurs collaborateurs, avec une diffusion de clan, le clientélisme politique, la corruption et relativement limitée dans l’ensemble de le manque de capacité, elle les a traités comme l’équipe-pays. En Angola, en raison du manque des problèmes techniques et n’en a pas conven- initial de personnel opérationnel au bureau ablement tenu compte pour étayer la démarche extérieur pouvant travailler avec les services globale. ministériels pour préparer les réunions de haut Cela ne signifie cependant pas que l’analyse niveau entre les ministres et la Banque, les doit être effectuée de manière interne. En questions ont été adressées trop rapidement République démocratique populaire lao, la aux instances supérieures, ce qui a créé des Banque a mis à profit les analyses politiques tensions inutiles. existantes et évité ainsi d’effectuer ell-même une Il faut de meilleures directives analyse plus coûteuse, tout en écartant les opérationnelles pour adapter les démarches des risques de friction avec les autorités locales. bailleurs de fonds. En plus des notes récemment Il est impératif de bien échelonner les publiées, des directives sont tout particulière- réformes et de prévoir suffisamment de temps ment nécessaires pour les pays en situation de pour les mettre en oeuvre si l’on entend détérioration, de crise prolongée ou d’impasse, produire les résultats escomptés sans trop et pour les phases de transition et de développe- lourdement grever les capacités nationales. En ment qui suivent la phase de reconstruction Afghanistan, les réformes introduites par les immédiate dans les pays sortant d’un conflit ou bailleurs de fonds n’ont pas été assez sélectives en transition politique. et ont abouti à la publication de 120 textes de loi. À São Tomé-et-Principe, la Banque s’est montrée Recommandations beaucoup trop ambitieuse et bon nombre des objectifs de la stratégie d’aide-pays n’ont pas été • Préciser la teneur et la portée du programme atteints ou ne l’ont été que partiellement. d’édification de l’État de la Banque et ren- La coordination des bailleurs de fonds ne forcer la conception et la mise en place des peut être efficace sans une vision et un objectif mesures d’aide au développement des capac- communs. En Afghanistan et au Tadjikistan, les ités et à la gouvernance dans les LICUS pour bailleurs de fonds n’avaient pas clairement améliorer les résultats. défini un objectif commun et il a donc été • Formuler des critères d’affectation de l’aide difficile d’assurer la cohérence des politiques. pour faire en sorte que les LICUS reçoivent une Le suivi et l’évaluation sont au moins aussi aide qui n’est ni insuffisante ni excessive. importants dans les LICUS que dans tout autre • Renforcer l’appui interne aux travaux sur les pays. Dans les pays où la situation est instable et LICUS au cours des trois prochaines années. où les progrès sont souvent en dents de scie, il Il est particulièrement important d’offrir des in- est essentiel d’adapter les programmes. Le suivi citations de nature à attirer un personnel qual- xv E N G A G I N G W I T H F R A G I L E S T AT E S ifié – tant au siège que dans les bureaux ex- chargée de la prévention des conflits et de la térieurs – pour travailler sur les LICUS, et de reconstruction. mettre en place une organisation efficace qui • Réexaminer la valeur ajoutée de la démarche élimine les doubles emplois et la fragmentation suivie pour les LICUS au bout de trois ans, entre l’équipe chargée des LICUS et celle lorsqu’on possédera suffisamment de don- Vinod Thomas Directeur général de l’évaluation xvi Prefacio n los Estados frágiles, hasta recientemente conocidos en el Banco Mundial E como países de ingreso bajo en dificultades, viven casi 500 millones de personas, aproximadamente la mitad de las cuales ganan menos de un dólar por día; por ese motivo, la situación de esos países despierta cada vez más atención. Es causa de creciente preocupación la posibilidad de que esos países no logren alcanzar sus objetivos de desarrollo, así como los efectos económi- cos desfavorables que provoca su situación en países vecinos y la consiguiente posibilidad de que sus problemas se propaguen al resto del mundo. Dada la multiplicidad de problemas crónicos batallando por hallar la manera más eficaz de que padecen, esos países plantean algunos de ayudarlos. Los PIBD, como el Banco denomina los desafíos más arduos en materia de desde 2002 a los Estados frágiles, se caracterizan desarrollo. En la mayoría de ellos la gestión de por lo insatisfactorio de sus políticas, institu- los asuntos públicos es insatisfactoria. Muchos ciones y gestión pública. En el ejercicio de 2005, de esos países —por ejemplo, Sudán— están el Banco identificó a 25 de esos países, sumidos en vastos y complicados conflictos basándose en sus ingresos y en el puntaje que internos. Algunos, como Timor-Leste, se les corresponde en la evaluación de las políticas debaten en medio de delicados procesos de e instituciones nacionales (CPIA). transición posteriores a conflictos. Todos tienen Esos 25 países presentan ciertas semejanzas: ante sí obstáculos similares: inadecuada seguri- en comparación con otros países de ingreso dad, fractura de las relaciones sociales, corrup- bajo, su tasa de mortalidad infantil es un tercio ción, desintegración del Estado de derecho y más alta; la esperanza de vida es 12 años menor, falta de mecanismos de generación de poder y y la tasa de mortalidad materna es alrededor de autoridad legítimos. Además, por tratarse de un 20% más alta. A esto se agregan importantes países de ingreso bajo, han acumulado enormes diferencias entre distintos PIBD. Algunos, como necesidades de inversión y no disponen de Angola y Camboya, registraron un crecimiento suficientes recursos públicos para atenderlas. económico de alrededor del 4% por año en el La labor internacional llevada a cabo con esos período 1995–2003; en otros, como Islas países no ha generado mejoras significativas, por Salomón, República Democrática del Congo y lo cual los donantes y otras instituciones siguen Guinea-Bissau, las tasas de crecimiento xvii E N G A G I N G W I T H F R A G I L E S T AT E S económico fueron de magnitud similar, pero a PIBD en situaciones posteriores a conflictos. negativas. Algunos, como Angola, Nigeria, Papua También se han logrado avances sustanciales en Nueva Guinea y República Democrática del materia de coordinación de los donantes a nivel Congo, poseen abundantes recursos naturales; de la política internacional. otros, como Burundi y Haití, son pobres en recursos. Esas diferencias se reconocen en Dificultades modelos económicos específicos —de deteri- No obstante, subsisten considerables dificul- oro, crisis o estancamiento prolongados, tades. Por ejemplo, las reformas introducidas en situaciones posteriores a conflictos o de transi- algunos PIBD no han sido selectivas ni se han ción política, y mejora gradual— que el Banco priorizado. El Banco debe actuar más eficaz- ha elaborado para trabajar con países en crisis. mente al concluir la fase que sigue inmediata- En los ejercicios de 2003–05, el monto del mente a un conflicto, en que se requieren presupuesto para otorgamiento de préstamos y reformas estructurales. Las estrategias para los del presupuesto administrativo destinados a los países que elabora el Banco no permiten afirmar PIBD fue de US$4.100 millones y US$161 que se hayan captado, en suficiente medida, las millones, respectivamente; esas sumas superan realidades políticas de los países. No se ha en 67% y 55% las de los ejercicios de 2000–02. El llevado al nivel de países la firme coordinación enfoque de la Iniciativa para los PIBD ha de los donantes lograda en el nivel de las políti- evolucionado: las preocupaciones generales cas internacionales. Lo más importante es que el sobre la eficacia de la ayuda en 2002 dejaron Banco ha centrado la atención en la construc- paso, en 2005, a objetivos de fortalecimiento del ción del Estado, sin poner adecuadamente de Estado y de consecución de la paz. La evaluación manifiesto la manera de evitar las fallas del de la experiencia en la aplicación del enfoque pasado y garantizar mejores resultados en sobre los PIBD del Banco realizada por el Grupo cuanto a creación de capacidad y gestión de los de Evaluación Independiente (IEG) revela asuntos públicos. algunos éxitos iniciales, pero subsisten consid- En la esfera interna, el avance sigue siendo erables dificultades. insatisfactorio en relación con las reformas esenciales en materia de recursos humanos Éxitos iniciales referentes a número de funcionarios, calidad de El Banco se ha puesto en mejores condiciones sus aptitudes e incentivos para la realización de la operativas para ocuparse de los PIBD. Ha labor de la Iniciativa para los PIBD. Existe consid- intensificado su labor de análisis y ha erable duplicación de esfuerzos y confusión con comenzado a usar las Notas de la Estrategia respecto a las funciones y responsabilidades de la Provisional para diseñar estrategias para un Unidad de PIBD y la Unidad de prevención de período más breve, a fin de dar cabida a las conflictos y de reconstrucción. Finalmente, el condiciones inestables de los PIBD. Además los Banco aún no ha abordado el tema de una gerentes que se ocupan de esos países han asignación de ayuda para los PIBD que refleje los obtenido un acceso más amplio a la adminis- objetivos de la institución en esos países y tración superior del Banco, y reciben mayor confiera certeza de que los PIBD no reciben orientación en relación con varios temas menos ni más asistencia de la que necesitan. importantes. Por otra parte, el Banco ha Con respecto a la eficacia, aún sería comenzado a utilizar el Fondo Fiduciario PIBD prematuro abrir juicio sobre los resultados de la para financiar a países excluidos del régimen de labor del Banco, pero algunos indicadores llevan contabilidad en valores devengados, que tenían a pensar que el impacto puede haber sido limitado acceso al financiamiento. limitado. El puntaje de la CPIA referente a los Esas medidas han ayudado al Banco a lograr PIBD pone de manifiesto un mejoramiento de la mejoras en cuanto a estabilidad tendencia desde la puesta en marcha de la Inicia- macroeconómica y a proporcionar un volumen tiva para los PIBD, pero el indicador de gobern- significativo de infraestructura física, en especial abilidad de Kaufmann, Kraay, Mastruzzi (KKZ) xviii P R E FA C I O muestra, en el mismo período, una tendencia al políticas. deterioro de la situación. Ninguna de las dos El seguimiento y la evaluación no revisten tendencias es necesariamente atribuible a las menos importancia en los PIBD que en los actividades de los donantes, pero parece justifi- restantes países. En entornos nacionales inesta- carse la realización de un esfuerzo sostenido. bles, en que el progreso suele no ser lineal, es esencial adaptar los programas a esa realidad. Enseñanzas para el Banco y para otros Un estrecho seguimiento del desempeño puede donantes contribuir a establecer el momento en que Iniciada la participación, es preciso establecer, deben realizarse adaptaciones, y de qué tipo. sin dilación, objetivos de reforma claros y Una eficaz ejecución de la estrategia para el pertinentes. En la República Centroafricana país requiere no sólo una presencia in situ, sino existe el peligro de que una inadecuada atención también adecuada comunicación entre el de la situación presupuestaria comprometa personal que opera sobre el terreno y el de la ahora los satisfactorios resultados iniciales. sede, así como un adecuado número de Es preciso que la labor de los donantes se funcionarios sobre el terreno, dotados de centre en la inserción del análisis político en el facultades y aptitudes apropiadas. En Camboya, diseño y la ejecución de la estrategia. Aunque en la presencia in situ del Banco ha permitido Papua Nueva Guinea la estrategia provisional conocer bastante mejor la situación política del daba cabida a un adecuado debate del sistema país, pero ese saber tal vez esté muy concen- político y en ella se reconocían problemas tales trado en unos pocos gerentes y funcionarios; como lealtades de clanes, clientelismo político, poco se ha difundido al resto del grupo a cargo corrupción y falta de capacidad, esos problemas del país. En la oficina de Angola, la falta inicial de se trataban como dificultades técnicas; y no se personal operativo en condiciones de colaborar los utilizó del modo adecuado para basar en con el personal de los ministerios en los prepar- ellos el enfoque general. ativos para reuniones de alto nivel entre No obstante, puede prescindirse de un ministros y representantes del Banco, aceleró análisis interno. En la República Democrática excesivamente el avance del trámite hacia la Popular Lao, el Banco aprovechó eficazmente cúspide administrativa, lo que creó tensiones los análisis políticos existentes y evitó los innecesarias. mayores costos que implicaba la preparación de Se requiere una mejor orientación operativa sus propios análisis, así como las posibles para adaptar a las necesidades locales los tensiones con el gobierno. enfoques de los donantes. Además de los Establecer una adecuada secuencia de estudios recientemente publicados, existe una reformas y disponer de tiempo suficiente para especial necesidad de orientación para países en ejecutarlas son factores esenciales para lograr proceso de deterioro y sujetos a crisis o resultados sin hacer recaer una carga estancamiento prolongados, y para las fases de abrumadora sobre la capacidad del país. En transición y desarrollo que siguen a la fase de Afganistán las reformas de los donantes no han reconstrucción inmediata en los países en sido suficientemente selectivas y han dado lugar situaciones posteriores a conflictos o en transi- a 120 leyes. En Santo Tomé y Príncipe, el Banco ción política. adoptó objetivos excesivamente ambiciosos, lo que impidió alcanzar —por lo menos Recomendaciones totalmente— muchos de los objetivos de la Estrategia de asistencia al país. • Establecer claramente el alcance y contenido No puede lograrse la coordinación de los de los objetivos de fortalecimiento del Estado donantes sin una visión y una finalidad compar- que persigue el Banco y reforzar el diseño y el tidas. En Afganistán los donantes no coinci- suministro de respaldo en materia de desarrollo dieron en un único objetivo claro, lo que de capacidad y gestión de los asuntos públicos impidió la consecución de coherencia en las en los PIBD, para lograr mejores resultados. xix E N G A G I N G W I T H F R A G I L E S T AT E S • Elaborar criterios de asignación de la ayuda, elimine la duplicación y fragmentación de es- para que los PIBD no reciban menos ni más fuerzos entre la Unidad de PIBD y la Unidad de ayuda de la necesaria. prevención de conflictos y de reconstrucción. • Reforzar el apoyo interno para la labor refer- • Al cabo de tres años, volver a evaluar el valor ente a los PIBD en los próximos tres años. Re- agregado del enfoque de la Iniciativa para los viste especial importancia contar con PIBD, cuando se disponga de experiencia sufi- adecuados incentivos que atraigan personal ciente sobre los resultados de ese enfoque, y calificado —en la sede y en las oficinas en los basar en las conclusiones de esa nueva evaluación países— que se ocupe de los PIBD, y estable- el continuo apoyo del Banco para la categoría de cer una estructura institucional eficiente que los PIBD y para el enfoque que a ellos se refiere. Vinod Thomas Director General, Evaluación xx Main Evaluation Messages • Low-Income Countries Under Stress (LICUS) present some of the toughest development chal- lenges, and the donor community continues to grapple with the question of how best to as- sist them. The World Bank has been an active participant in international policy discussions on LICUS and has improved its operational readiness to support them since introducing the LICUS Initiative in 2002. • Before the LICUS Initiative, outcomes of the Bank’s assistance programs in LICUS were mostly in the unsatisfactory range. The initiative has increased Bank attention to LICUS, but it is too early to assess outcomes. Implementation experience has been mixed, and outcomes of the few country strategies that have been assessed by the Independent Evaluation Group (IEG) mostly indicate underachievement of objectives. • By adopting state building as a central objective, the Bank has made an area of traditional weakness (capacity development and governance) a part of its main focus in LICUS. Focusing the LICUS Initiative on the complex state-building agenda requires that the Bank clarify its areas of comparative advantage and the scope and content of the agenda. The Bank also needs to identify innovative approaches to improve the weak capacity development and governance record, and performance indicators to measure state-building outcomes. • Little progress has been made on critical human resource reforms relating to staffing num- bers, staffing quality, and incentives to undertake LICUS work in the three years since the LICUS approach was implemented. • Although the Bank has recently emphasized the need to increase its field presence in LICUS, that emphasis alone will be insufficient for the effective implementation of country strategies. Increased field presence needs to be complemented by stronger communica- tion between the Bank’s field and headquarters staff. An adequate number of field staff with the appropriate authority and skills is also required. • Donor reform agendas in LICUS could be more selective. In complex LICUS environments, where virtually every sector requires reform, appropriate sequencing of reforms and suf- ficient time to implement them are crucial for achieving results without overwhelming limited LICUS capacity. Executive Summary ome to almost 500 million people, roughly half of whom earn less than H a dollar a day, fragile states, until recently known in the World Bank as Low-Income Countries Under Stress (LICUS), have attracted in- creasing attention. Concern is growing about the ability of these countries to reach the Millennium Development Goals (MDGs) as well as about the adverse economic effects they have on neighboring countries and the global spillovers that may follow. With their multiplicity of chronic problems, Development focuses on stagnant low-income LICUS pose some of the toughest development countries (defined by gross national product per challenges. Most have poor governance and are capita and growth rates), and the Failed States embroiled in extended internal conflicts or are Index of Foreign Policy focuses on state failure, struggling through tenuous post-conflict transi- identifying countries based on such factors as the tions. They face similar hurdles of widespread level of economic decline, security, factionalized lack of security, fractured relations among societal elites, displaced persons, human rights breaches, groups, significant corruption, breakdown in the and external intervention. The U.S. Agency for rule of law, absence of mechanisms for generat- International Development aims to address issues ing legitimate power and authority, a huge surrounding vulnerability and crisis, many pertain- backlog of investment needs, and limited govern- ing to the political environment. The U.K. Depart- ment resources for development. Past interna- ment for International Development (DFID) and tional engagement with these countries has the Organisation for Economic Co-operation and generally failed to yield significant improvements. Development–Development Assistance Commit- The donor community is grappling with the tee’s (OECD-DAC’s) definitions of fragile states question of how best to assist countries faced with are similar to those used by the World Bank. such challenging problems. With their differing As defined by the World Bank, all LICUS are motivations and objectives, donors and characterized by weak policies, institutions, and researchers have chosen to address different governance. The Bank has used two criteria to aspects of these problems, which has led them to define core and severe LICUS (henceforth LICUS focus on slightly varying groups of countries. For refers to core and severe LICUS, not marginal instance, recent research by the Center for Global LICUS, which are identified by the Bank only for xxiii E N G A G I N G W I T H F R A G I L E S T AT E S monitoring purposes): per capita income within would have received minuscule amounts of the threshold of International Development administrative budgets prior to the initiative Association (IDA) eligibility and performance of because of their non-accrual status). 3.0 or less (2.5 or less for severe and 2.6–3.0 for The Bank’s LICUS approach has evolved since core) on both the overall Country Policy and its initial articulation in 2002, which was Institutional Assessment (CPIA) rating and the grounded in country-level core principles (see CPIA rating for Public Sector Management and table ES.1). The original rationale for the initiative Institutions. was that of improving aid effectiveness by using Some low-income countries without CPIA data other instruments, such as analytical work and are also included. In fiscal 2005, the Bank identi- knowledge transfers where necessary, supple- fied 25 countries as LICUS. Six fiscal 2005 LICUS mented by financial transfers to promote change. did not have a CPIA rating: Afghanistan, Liberia, In 2005, the objectives and scope of the Myanmar, Somalia, Timor-Leste, and the territory LICUS Initiative shifted from general aid of Kosovo. This review bases its evaluation on the effectiveness to state-building and peace- Bank’s assistance to the 25 countries classified as building objectives. The LICUS Initiative also in- LICUS in fiscal 2005. troduced four business models (deterioration, Lending and administrative budgets to LICUS prolonged political crisis or impasse, post- have increased since the LICUS Initiative began. conflict or political transition, and gradual Lending to LICUS increased from about $2.5 improvement) that provided for varying billion during fiscal 2000–02 (before the LICUS treatment of different types of LICUS. Learning Initiative) to about $4.1 billion during fiscal by doing and the focus on organizational issues 2003–05 (since the launch of the LICUS Initia- in the 2002 approach were retained and further tive). On a per capita basis, lending to LICUS reinforced in the 2005 approach. ranged from $0 to $25.4 during fiscal 2003–05. This review set out to answer three Administrative budgets for LICUS increased from questions: about $104 million during fiscal 2000–02 to about $161 million during fiscal 2003–05. On a per • How effective has the Bank’s LICUS approach capita basis, administrative budgets for LICUS been? ranged from $0.002 to $4.5 during fiscal 2003–05. • How operationally useful are the Bank’s crite- A large share of lending to LICUS during fiscal ria for identifying and classifying LICUS, and how 2003–05 went to post-conflict LICUS (post- useful is the aid-allocation system for them? conflict countries are identified based on Post- • How appropriate and adequate has the Bank’s Conflict Progress Indicators, for purposes of internal support for LICUS work been? determining exceptional IDA grants), while administrative budgets have been more evenly Main Findings and Conclusions distributed across the LICUS group (7 post- conflict LICUS out of the 25 received 64 percent Effectiveness of the Bank’s LICUS approach of total LICUS lending, and 34 percent of the to- Implementation experience across the core tal LICUS administrative budget). country-level LICUS principles has been mixed While the large proportion of lending to post- (see table ES.1). Problems encountered in im- conflict LICUS might have occurred even plementation sometimes arose from overambi- without the LICUS Initiative (given that IDA’s tious Bank objectives (thus requiring the scaling exceptional post-conflict allocations predate the down of objectives) and sometimes from initiative), the initiative likely contributed to the inadequate Bank effort or inappropriate input, as more even distribution of administrative suggested by IEG’s fieldwork and its CAS Comple- budgets across the group (given an increase of tion Report Reviews (thus requiring scaling up of 400 percent or more in administrative budgets effort). between fiscal 2000–02 and 2003–05 for three The majority of stakeholders interviewed in LICUS—Liberia, Somalia, and Sudan—that IEG’s Stakeholder Survey said that the Bank’s xxiv EXECUTIVE SUMMARY Table ES.1: Implementation Experience on the Core Country-Level LICUS Principles LICUS principle Implementation experience rating Stay engaged Substantial Anchor strategies in stronger sociopolitical analysis Medium • Political understanding • Medium-substantial • Internalizing political understanding in strategy design and implementation • Medium-low Promote domestic demand and capacity for positive change Low Support simple and feasible entry-level reforms Medium-low • Macroeconomic reforms • Substantial • Delivery of physical infrastructure • Substantial • Transition from the immediate post-conflict reconstruction phase to the development phase • Low • Selectivity and prioritization • Low Explore innovative mechanisms for social service delivery Medium Donor collaboration Medium • At international policy level • Substantial • At country level • Medium-low Measure and monitor resultsa Low Sources: Fieldwork and thematic background analysis done for this review by IEG, 2005. a. Not specifically mentioned as a separate core principle by the Bank, but included by IEG because it is pivotal to the Bank’s learning-by-doing LICUS agenda. overall program in LICUS has made a small are likely to permit a more tailored response to positive contribution to development—a view LICUS. The percentage of closed LICUS that refers to Bank support generally, and not to projects rated satisfactory on outcome by IEG the LICUS approach per se. increased from 50 percent in fiscal 2002, before There have been some notable early the LICUS Initiative, to 58 percent in 2003, 65 successes with regard to the LICUS principles. percent in 2004, and 82 percent in 2005. The The Bank’s LICUS Initiative has allowed for in- corresponding numbers for projects in non- creased Bank engagement in countries where LICUS low-income countries ranged from 70 to such engagement would likely have been lower. 79 percent. The Bank has recently engaged with a number But several significant challenges remain. The of LICUS from the early days of peace or political Bank’s initial engagement with a number of transition. The Bank has also contributed to LICUS has not been adequately followed up by a macroeconomic stability and to the delivery of focused and well-sequenced reform agenda. significant amounts of physical infrastructure, Furthermore, the Bank has yet to internalize especially in post-conflict LICUS. Substantial sufficient political understanding in country progress has been made in donor coordination strategy design and implementation. The Bank at the international policy level, as exemplified also needs to strengthen the quality of its by the recent agreement of a wide spectrum of country-level coordination with other donors, donors, including the Bank, to the 12 OECD- especially in implementation follow-through DAC principles of international engagement. that goes beyond policy agreements. The Bank has often played a leading role as In addition, the Bank has made one of its co-chair of international donor events and co- areas of traditional weaknesses (capacity author of joint policy papers. The Bank’s development and governance) a central part of recently introduced business models, which its focus by adopting the more complex state- differentiate among different types of LICUS, building objective. This new emphasis requires xxv E N G A G I N G W I T H F R A G I L E S T AT E S the Bank to identify its comparative advantage exclusively on the CPIA to identify LICUS. The more effectively; improve performance, includ- CPIA, however, fails to capture some key aspects ing through the development of innovative of state fragility (such as accommodation of approaches; and identify partners who can political dissent) and conflict (such as political complement its work to ensure achievement of instability and security or susceptibility to the intended outcomes. Finally, the choice of conflict), and may need to be supplemented. A the term state building may itself be inap- stronger approach to the identification of LICUS propriate, given its political and ideological would require an analytical framework that connotations. more explicitly focuses on the objectives of the The Bank needs to develop its operational LICUS Initiative. approaches in LICUS, especially for the deterio- The policy selectivity of the system the Bank ration and prolonged crisis or impasse business uses to allocate IDA resources (called perform- models. Further refinement of the business ance-based allocation, or PBA) has increased models by more explicitly factoring in differ- over the years, and less IDA funding has been ences in capacity to perform core state functions available for countries with weaker policies, (for example, resource generation, resource institutions, and governance. This has raised the allocation, basic social service and infrastructure question of whether LICUS are receiving provision, and political accommodation of appropriate amounts of IDA resources. Adjust- dissent and security) is also needed to enable ments to the PBA have resulted in increased IDA the Bank to achieve a better fit between its financing, including to some post-conflict LICUS operational approaches and the varying institu- and LICUS undergoing political transitions. Yet it tional environments of LICUS. remains far from clear whether the current The Bank’s work on post-conflict countries levels of IDA funding ensure that LICUS are not predates the LICUS approach, and the under- or over-aided. corresponding business model for post-conflict The aid-allocation issue has once again LICUS is articulated more clearly than the other come to the fore with some research question- business models. However, it has shortcomings ing the empirical evidence for the positive link and needs to be further developed to guide the between policies and aid effectiveness (which transition and development phases that follow underlies the PBA). Other research argues that the immediate post-conflict reconstruction aid can be effective in promoting sustainable phase. Furthermore, while the Bank has given policy turnarounds in failing states by building increasing attention to conflict prevention, and strengthening the preconditions for there is limited knowledge about the effective- reform or by enhancing the chances that the ness of its efforts in this area. reform will be sustained once it is set in place. The Bank’s role and comparative advantage The latter research finds that potential returns in conflict prevention have yet to be clearly from aid to LICUS can be extraordinarily high, established, especially because conflict preven- even though the risks of failure are substantial. tion requires the Bank to give greater attention For its part, the Bank has yet to address the to the root causes of conflict and address ethnic, aid-allocation issue for LICUS in a way that sociological, and political factors. The Bank reflects its objectives for these countries and needs to define better what its peace-building ensures that LICUS are not under- or over-aided. objective does and does not include and how it will be achieved. Appropriateness and adequacy of internal Bank support for LICUS work Operational utility of the Bank’s LICUS The Bank’s internal support for LICUS work has identification, classification, and progressed in several areas: aid-allocation mechanisms Despite the move to state- and peace-building • Expanding analytical work by de-linking ad- objectives, the Bank continues to rely almost ministrative budgets for economic and sector xxvi EXECUTIVE SUMMARY work and technical assistance from lending LICUS and a LICUS country demonstrate volumes inadequate recognition of LICUS work within • Using Interim Strategy Notes that allow for the the Bank and point to an incentive system in design of strategies that cover a shorter period need of reform. to accommodate the volatile LICUS conditions The uneven attention of individual country • Providing LICUS managers access to the Bank’s directors, especially if they are also covering a senior management larger, more “successful,” or higher-profile • Introducing the LICUS Trust Fund to finance country, was mentioned by staff as an issue, countries in non-accrual (for which the Bank indicating the need to ensure consistent previously lacked an instrument). attention to LICUS work throughout the management hierarchy. In IEG’s Stakeholder Based on country experience, the LICUS Unit Survey, the majority of Bank respondents said has distilled guidance on a number of important that there has been no change when working on issues and has fed this guidance into both LICUS with respect to several human resource operational advice to country teams and broader matters (see figure ES.1). external policy debates. There is significant duplication and confusion However, three years after the Bank surrounding the roles and responsibilities of the recognized the need for an internal culture shift LICUS and the Conflict Prevention and to implement the LICUS approach effectively, Reconstruction (CPR) Units. Staff is concerned the Bank’s internal support for LICUS work has with the practical questions of which unit to turn progressed little. It remains unsatisfactory on to for specific types of advice and what kinds of critical human resource reforms relating to support to expect from each unit. In IEG’s staffing numbers, staffing quality, and incentives Stakeholder Survey, about two-thirds of Bank to undertake LICUS work. Bank staff comments respondents saw some problem with the about the importance of working on both a non- current organizational arrangement: 37 percent Figure ES.1: The Majority of Bank Respondents Said There Has Been No Change When Working on LICUS in Several Areas (listed below) 100 75 Percent 50 25 0 Career prospects Financial compensation Realism in expectations Level of support from Efforts made by the by management management Bank to ensure your personal security and safety Improved Same Worsened Source: Appendix Z (Stakeholder Survey results). Note: Number of valid responses ranges from 213 to 238. The question in the survey did not differentiate between staff who had worked in a LICUS and those who had worked in a non- LICUS in their previous assignment. xxvii E N G A G I N G W I T H F R A G I L E S T AT E S said that there is some duplication between the when it dampens internal pressure for reform, support of the Bank’s LICUS Unit and that of the thus potentially hindering the emergence of CPR Unit, 15 percent that there is a lot of conditions needed to bring about serious and duplication, and 12 percent that there is a sustainable political reform. conflict or contradiction. In the deterioration and prolonged crisis or impasse business models, where there is often Lessons of Experience for the Bank little consensus between donors and govern- and Other Donors ment on development strategy, engagement Several lessons emerge from this review’s assess- needs to include policy dialogue aimed at ment of the Bank’s experience in implementing creating an opening for reform, while simulta- the core principles of the LICUS approach. Many neously working on a reform agenda should a of the issues covered under these lessons were window of opportunity appear. In the post- noted as areas in need of improvement in the conflict or political transition and gradual im- 2002 LICUS Task Force report (World Bank provement business models, engagement will 2002)—such as the need to anchor strategies in need to have more of a technical content and a stronger sociopolitical analysis or to support stronger focus on implementing the reform highly focused reform agendas—and have also agenda, given the greater reform consensus been emphasized in the Bank’s 2005 LICUS between donors and government. reports. The lessons derive from the Bank’s own The Bank’s guidance for prolonged conflict implementation experience, but may also be or political impasse countries states that useful in guiding other donor assistance in “relatively non-controversial development LICUS. issues may provide an entry point for construc- tive dialogue between the parties to a conflict.” LICUS engagement For deteriorating governance countries, the Bank’s guidance states that the Bank should Staying engaged is only a means to an end and needs provide “input on specific economic issues to be quickly followed by a clear and relevant reform which are important for mediation efforts and agenda in LICUS. In the absence of a clear and may serve as a way to restart dialogue” (World relevant reform agenda, early successes of Bank 2005e). engagement may be short lived and contribute little to the achievement of country strategy Country ownership and absorptive capacity objectives. The examples of the Central African constraints apply as much to knowledge products as Republic and Haiti show that various obstacles to financial products. The involvement of country may make the follow-up to a successful initial counterparts in the Bank’s analytical work LICUS engagement difficult. Because political remains limited to administrative aspects, with successes were insufficiently backed up on the much less country-client participation in select- economic side, the government of the Central ing topics and undertaking analysis, thereby African Republic is now faced with a potentially reducing national buy-in. Yet the involvement of disastrous budget crisis. In Haiti, the donor country counterparts is essential to ensuring community seems to have given inadequate client ownership and improving the impact of attention to ensuring a minimum level of security. analytical work. In both cases, good initial results of the LICUS In Tajikistan, the lack of government in- Initiative are now at risk of being diminished. volvement in the selection and preparation of In certain instances, strategic disengagement— the Bank’s analytical work limited the govern- with the exception of in-house analytical work— ment’s interest in the results, which hindered may be needed, at least for periods of time. This is effective implementation. In Angola, some a particularly appropriate strategy when involve- Bank-led analytical work (for instance, the ment with the Bank is seen as inappropriately recent Country Economic Memorandum) was giving legitimacy to the LICUS government or perceived by senior government officials as an xxviii EXECUTIVE SUMMARY imposition of Bank views on their internal tension with the government, by allowing the affairs, which led to limited ownership and Bank to avoid getting bogged down in some of capacity development. Without country the sensitivities surrounding the analysis. For the ownership, the chance of analytical work Bank, the acquisition of existing knowledge, as influencing government policy is small. well as its dissemination, proved more important LICUS governments’ absorptive capacity and effective in this case than knowledge creation. constraints in using analytical work may also limit possible knowledge transfer. The Angolan govern- The main focus of donor efforts needs to be on helping ment, for instance, endorsed the Bank Interim staff internalize political analysis in strategy design Strategy Note but expressed concern about the and implementation. Although the Bank has amount of analytical and advisory activities conducted or had access to good political analysis foreseen. This has raised doubt about whether the in some LICUS, such analysis has not been analytical products would be fully utilized by the adequately reflected in its strategy. For example, government. The absorptive capacity of the the Interim Strategy in Papua New Guinea con- government is severely limited, and analytical and tains a good discussion of the political system and advisory activities undertaken mostly by the Bank recognizes problems such as clan loyalties, politi- risk straining relations with the government, cal patronage, corruption, and lack of capacity. regardless of their technical quality. In Cambodia, Yet the strategy treats these problems as technical plans for analytical and advisory services in the matters and does not adequately use them to 2005 Country Assistance Strategy (CAS)—totaling underpin the overall approach. 30 tasks to be completed over fiscal 2005–07— appear overly ambitious considering the country’s Focused reform agenda limited institutional capacity. In complex LICUS environments, where virtually every Political understanding and its use in country sector requires reform, appropriate sequencing of strategy reforms and sufficient time to implement them are crucial for achieving results without overwhelming Commissioning and consuming—not necessarily limited LICUS capacity. While donors must strive for producing—good political analysis is critical for collective donor selectivity, this is far from being donors in LICUS. The objective of a country team achieved, as the examples of Afghanistan’s donor- should be to commission or consume (not endorsed reform agenda and Haiti’s Interim necessarily produce) analysis that is directly Cooperation Framework (ICF), presented below, relevant to, and usable in, the development of a indicate. However, even if collective donor strategy. In LICUS, especially in environments selectivity is not immediately achieved, the Bank where speed is critical, donors need to ensure needs to ensure focus and selectivity in its own that existing political analysis is mined before assistance program, based on its core commissioning new analysis. competences. Such Bank selectivity has been In Lao People’s Democratic Republic, the Bank increasing in recent years but remains a challenge. effectively tapped existing political analysis and In Afghanistan, the reforms covered by invited a political scientist who had published donors are wide ranging, show lack of sufficient extensively about the country to make a presenta- priority, and have led to 120 pieces of pending tion to the country team on politics and reform in legislation. These reforms, dealing with virtually the country. This allowed for the preparation of an every economic and social aspect of the country, independent summary of relevant political need to be carefully prioritized and sequenced, analysis (tailored to the needs of the donor but donors have yet to do so. In Haiti, the ICF is community in general and the Bank in particular) meant to guide international assistance and and its dissemination to a relevant group of Bank cooperation with Haiti through September staff and other donors. It avoided the higher costs 2006, and covers practically all basic state func- of preparing a “Bank” analysis, as well as potential tions, ranging from security, to national xxix E N G A G I N G W I T H F R A G I L E S T AT E S dialogue, to economic governance, to economic expertise. If foreign experts are brought in to recovery, to basic services. Individually, each of provide technical assistance, it must be ensured these areas seems important, but together they that this will not compromise the long-term add up to a formidable agenda. development of local capacity. With respect to the Bank’s own assistance program, São Tomé and Principe is an example Donor coordination where the Bank was far too ambitious in relation to the resources allocated to the country, with the Donor coordination cannot succeed without a result that many of the CAS objectives were not common vision and purpose among donors—when achieved or were only partially achieved. donor objectives cannot be fully harmonized, it is Beyond selectivity in CASs, it is critical to important that they at least be complementary. The ensure that actual reform agendas on the Bank’s approach has not fully recognized the ground are focused and well prioritized. The differing motivations of donors for engaging lack of selectivity and prioritization in the reform with LICUS. Although the broad concept of agendas raises questions of effectiveness, fragility is widely understood and accepted, the especially given the limited capacity in LICUS. countries identified by donors as fragile vary. While it is difficult to be selective in a country The motivations for supporting fragile states where there is an urgent need to fix many things, range from security, to aid effectiveness, to the appropriate sequencing of reforms is crucial equitable development, to poverty reduction, to to ensuring that limited LICUS capacity is not state building, to peace building and conflict overtaxed, while also avoiding partial solutions. prevention. Well-sequenced reforms spanning a sufficient In both Afghanistan and Tajikistan, IEG’s number of years, along with donor commitment fieldwork found that major donors did not to see them through, will be essential. subscribe to a single clear objective. Without a In Timor-Leste, donors may have pulled out common overall objective, policy coherence is too quickly, without sufficiently dealing with the unlikely. The Bank’s donor coordination efforts country’s pressing capacity needs. In Haiti, and modalities are insufficiently informed by the development assistance has greatly fluctuated objectives of the different players in a country. over the years. The country has gone through That said, donor coordination is a form of collec- several “feast or famine” cycles in its relations tive action, requiring that other donors similarly with the donor community. This might have improve their outreach to the Bank and subordi- been avoided had various donors better timed nate bilateral agendas to agreed multilateral and sequenced their aid. objectives. Capacity development in post-conflict LICUS Coordination needs to begin within each donor agency. Coordination is not only important Capacity development and governance programs among multilateral and bilateral donor agencies. need to start early, even in post-conflict LICUS. It is also a vital issue within each donor agency. Immediately following the cessation of conflict, Projects in different sectors of the same country the international donor community tends to often work in parallel and fail to tap synergies. focus its assistance on physical reconstruction. This was the case in the Bank’s Community Because capacity to use aid effectively in post- Empowerment and Agricultural Projects in conflict LICUS is low and governance is often Timor-Leste. poor, the focus from the beginning also needs to A side-effect of the Bank’s decentralization to be on the development of capacity and improve- country offices has been the concentration of ment of governance, not merely the reconstruc- country knowledge among local staff and tion of physical infrastructure. This may require inadequate dissemination of this knowledge to the creation or strengthening of public institu- the country team, especially to those based in tions, civil service reform, and use of local Washington. Addressing the problems of coordi- xxx EXECUTIVE SUMMARY nation across the various departments of donor yardsticks and ensure that the bar of success is agencies (such as among Bank departments set at an appropriate height. dealing with public sector management, conflict prevention and reconstruction, LICUS, capacity Improving internal organizational support for development, and research) is particularly LICUS work important in LICUS, where problems are complex and widespread and often require Field presence alone is insufficient for effective multisectoral solutions. country strategy implementation—it needs to be complemented by adequate communication between Results measurement and monitoring field and headquarters donor agency staff, as well as by an adequate number of field staff with the Monitoring and evaluation are at least as important appropriate authority and skills. Understanding of in LICUS as they are in any other country. Monitor- country circumstances is often best achieved ing and evaluation are crucial in LICUS for a through substantial field presence, although number of reasons: that alone is not enough. Internalizing analysis of the country conditions throughout all donor • First, the Bank, like other donors, is still learn- agency departments involved and applying its ing what approaches work in LICUS contexts. lessons to all interventions is equally important. Closely monitoring experiences in order to In Cambodia, for example, the Bank’s field draw lessons is critical, and learning and shar- presence has significantly improved understand- ing needs to become a more prominent feature ing of the political situation, but discussions with of LICUS work. country team members and other stakeholders • Second, given that progress is often slow in suggest that this knowledge may still be highly these countries, it is important to reassess con- concentrated among a few managers and staff tinually whether the program is on course to (mostly in the country office and Bangkok hub), achieve the desired outcomes. with relatively limited dissemination to the • Third, a constantly changing and volatile LICUS broader country team. environment where progress is often nonlin- The issue appears to have shifted from a ear means that program adaptation is partial understanding of the political realities of essential—closely tracking performance will Cambodia to a question of where within the help determine when and what kind of adap- Bank’s country team this knowledge is located tation is necessary. and how it is used to guide decision making in strategy and program implementation. The Effective learning by doing to improve the concentration of in-depth country knowledge Bank’s future effectiveness in LICUS can only among just a few staff members implies that only happen with strong monitoring and evaluation. some Bank activities and interventions benefit. The Bank has stated that state and peace In general, greater knowledge transfer is needed building should be the goals used to measure the between donor country offices and their LICUS Initiative’s success. But the Bank has yet headquarters-based country and sector staff. to identify performance indicators for this Despite the cost, field offices need to be purpose or yardsticks against which perform- adequately staffed if they are to engage ance may be measured. Change is often more effectively with clients. In Angola, the initially process oriented, especially in the deterioration small group of field staff faced a multiplicity of and prolonged crisis or impasse business tasks, from strategic dialogue with government models, and outputs and outcomes that may be and donors to logistics such as moving the office expected in the other business models may not to new premises. The situation was made more be appropriate yardsticks of success. Objectives difficult by the lack of operational-level staff in should be appropriate to particular LICUS the field office who could, in consultation with contexts, which would, in turn, determine Ministry staff, prepare the ground before high- xxxi E N G A G I N G W I T H F R A G I L E S T AT E S level meetings between the ministers and the In the gradual improvement business model, Bank. Moving issues to the top too quickly— the primary staff skill needed is the ability to because the lower levels were not staffed—led provide customized technical assistance and to unnecessary tensions. Donor decisions work hand-in-hand with a client that is already regarding the number of staff in each LICUS reforming. should reflect the extent and nature of intended engagement, considering respective donors’ Sharing experiences —both positive and negative— objectives in those countries. is essential for learning, but doing so effectively Apart from the absolute numbers, field office requires a receptive institutional environment and staff also need sufficient authority in relation to management support. Sharing experiences of what headquarters to ensure that not every decision is working in different LICUS situations, and has to be approved by headquarters. An effective what is not, can foster learning. Learning is field presence requires that the right kind of staff especially important in LICUS work because the be involved in the country. In semistructured donor community is continuing to grapple with interviews done for this review, several donors the question of how best to assist these emphasized that coordination is unusually challenging countries. Although the Bank has susceptible to the strengths and the foibles of shared some lessons through its LICUS Learning the individuals involved. More appropriate Group Seminar Series, much more attention is training for staff being posted to difficult field needed to intensify the systematic stock-taking assignments and improved incentives within the and dissemination of emerging LICUS Bank that encourage staff to collaborate with experiences—both those of the Bank and of other donors might ameliorate these idiosyn- other donors, and both positive and negative. cratic risks. Creating a more receptive institutional In the deterioration business model, where environment and ensuring management there might be a breakdown of dialogue with support for the sharing of negative experiences the government, donor agency staff will need will be critical. So far, the Bank seems mainly strong diplomatic and persuasive skills to willing to share positive examples, as in its ensure that the door remains open for a recent LICUS reports. dialogue with the government, while simultane- ously mobilizing nongovernmental groups, Effective communication is essential both for including civil society. ensuring country acceptance of donor approaches In the prolonged crisis or impasse business for LICUS and for tempering unrealistic country model, where problems are chronic or there is expectations about what can be achieved, political stalemate, the necessary staff skills will especially immediately following the cessation of include immense patience as well as creativity, with conflict. Better communication of donor constant innovation to break persistent logjams. objectives and approaches in LICUS will be In the post-conflict or political transition needed to ensure country buy-in. It can also business model, the necessary staff skills will prevent disillusionment by tempering unrealis- include specific technical knowledge of how to tic expectations among stakeholders about what develop sound economic systems, institutions, can be achieved in a specific period of time. and key infrastructure. Staff should also possess In the Bank’s deterioration and prolonged the ability to act quickly and decisively in these crisis or impasse business models, where the environments, before the optimism following economic and social situation is for the most peace dissipates, and to help guard against the part worsening or stagnant, the communication countries falling back into conflict. Since these strategy would need to disseminate actively the situations often attract massive international aid, benefits of reform to both the government and donor staff need strong coordination and civil society. In the Bank’s post-conflict or politi- sequencing skills to organize both development cal transition business model, in order to partners and their activities. prevent the disillusionment that follows unreal- xxxii EXECUTIVE SUMMARY istic expectations, the communication strategy without harming long-term government capacity should target the entire population and be development; ways to foster political reconcilia- explicit about what donors will do, when, how, tion while also contributing to effective and and what results should be expected. The legitimate governance; ways to internalize politi- communication strategy in the gradual improve- cal understanding in country strategy design and ment business model will need to be more implementation; and ways to address linkages informational, presenting relevant cross-country among politics, security, and development and cross-sectoral experiences. effectively. Immediately following the cessation of The balance of the Bank’s recent guidance on conflict, international donors, including the LICUS is tilted more toward what instruments Bank, have often committed large amounts of should be used rather than outlining actual aid coupled with overly ambitious agendas. This operational approaches for what needs to be has frequently created high expectations among done differently—and how—in varying groups the population and led to disillusionment when of LICUS. LICUS country teams would also expectations have remained unfulfilled and benefit from more narrative-based guidance, of there are few tangible improvements day to day. the kind presented in chapter 2 of this review, Avoiding overambitious agendas and utilizing and from short, problem-oriented notes, rather better communication are critical, and the Bank than the more formal guidance notes that are needs to invest in such strategies. often too condensed and devoid of sufficient country context. Better operational guidance is needed for tailoring donor approaches to the special conditions of LICUS. Recommendations The LICUS Initiative has raised awareness of the need to act differently in LICUS, but the Bank • Clarify the scope and content of the Bank’s state- and other donors have yet to identify precisely building agenda and strengthen the design and how to do so. The extent to which donor delivery of capacity development and governance approaches to LICUS need to, and can, support in LICUS. efficiently address the causes—not just symp- Given its weak record on capacity develop- toms—of countries becoming or remaining ment and governance, as well as its current LICUS also needs greater attention. Solutions focus on the more ambitious and complex state- that view causes as givens may miss all- building objective in LICUS, the Bank needs to important contextual factors. Donor operational clarify its areas of comparative advantage in guidance must ensure that areas outside the relation to other donors and adopt innovative comparative advantage of particular donors be approaches that ensure better capacity and left to others, while their own work both governance outcomes. Innovative approaches adequately factors in and complements the need to be developed for achieving a better fit work done by others. between the Bank’s interventions and the The Bank’s deterioration and prolonged crisis capacity of a LICUS to perform core state or impasse business models, and the transition functions; ensuring implementation of focused and development phases that follow the immedi- and well-sequenced interventions in LICUS ate reconstruction phase in the post-conflict or environments, where virtually every aspect of political transition business model, pose some of capacity and governance may need significant the biggest challenges faced by the donor improvement; and effectively monitoring community. These are also areas where there has capacity and governance outcomes. been relatively little innovative thinking. There is a pressing need for operational guidance in • Develop aid-allocation criteria for LICUS that en- several areas, including ways to prioritize and sure they are not under- or over-aided. sequence reforms while avoiding partial The Bank needs to conduct a technical review solutions; ways to deliver services quickly of the cumulative effect of the various adjust- xxxiii E N G A G I N G W I T H F R A G I L E S T AT E S ments to the performance-based allocation skills, such as in public sector management; are system on aid volumes to LICUS. Aid-allocation capable of seeking and using political criteria that reflect the Bank’s objectives in knowledge; and are willing and able to work in LICUS and ensure that these countries are not interdisciplinary teams. Current plans to address under- or over-aided need to be developed. these issues in the forthcoming Strengthening Whether and to what extent the criteria should the Organizational Response to Fragile States be based on factors other than policy perform- paper are welcome, even if late. ance (such as levels of other donor assistance, More systematic thinking is needed about assessment of potential risks and rewards, and staffing decisions for LICUS within the context regional and global spillovers) needs to be of the Bank’s overall staffing, recognizing that examined, keeping in mind that aid is limited assigning more and better-qualified staff to work and trade-offs will have to be made. on LICUS would likely mean trade-offs for other Bank country teams. Trade-offs to benefit LICUS • Strengthen internal Bank support for LICUS work may or may not be justified, depending on the over the next three years. Bank’s objectives for LICUS as well as other Bank Two aspects of internal Bank support need clients’ needs for assistance. attention. First, staffing numbers, skills, and Second, the organizational structure for incentives for working on LICUS need to be priori- LICUS and conflict work needs to be stream- tized. Ensuring adequate incentives to attract lined. The Bank needs to ensure an efficient qualified staff—both at headquarters and in field organizational arrangement that removes offices—to work on LICUS will require giving clear duplication and fragmentation of support signals of what is deemed to be success in LICUS, between the LICUS and the CPR Units. what outcomes staff will be held accountable for, how much risk it is reasonable to take, how failure • Reassess the value added of the LICUS approach will be judged, and how overall performance after three years. evaluation ratings and staff career development The value of the LICUS category and will take these into account. approach, including the operational usefulness As in Olympic diving, where the scoring of the business models, needs to be inde- system factors in both the technical perfection pendently evaluated after three years, when and the difficulty of the dive, staff performance sufficient experience with the outcomes of the in LICUS should be judged by assigning appropri- approach will be available. At that time it should ate weight to the extent of challenges presented be possible to address the more fundamental by varying LICUS environments. Signaling the question of whether and to what extent Bank importance of LICUS work throughout the assistance can effectively support sustainable management hierarchy will also be required. state building. Continued Bank support for the Apart from incentives, the Bank needs to LICUS category and approach should be based ensure that staff working on LICUS has relevant on the findings of that reassessment. xxxiv Principaux messages de l’évaluation • Le développement est un défi extrêmement difficile à relever dans les États fragiles et la communauté des bailleurs de fonds continue de s’interroger sur la meilleure façon de leur venir en aide. La Banque mondiale participe activement aux débats internationaux sur les LICUS et a amélioré ses capacités opérationnelles d’assistance depuis le lancement de l’ini- tiative en faveur des LICUS en 2002. • Avant la mise en place de cette initiative, les programmes d’aide de la Banque aux LICUS avaient donné des résultats généralement peu satisfaisants. L’initiative a attiré l’attention de la Banque sur les LICUS mais il est encore trop tôt pour évaluer les résultats. L’expéri- ence sur le terrain est mitigée et les conclusions de l’évaluation effectuée par le Groupe indépendant d’évaluation (IEG) indiquent pour la plupart que les objectifs n’ont pas été en- tièrement atteints. • En adoptant l’édification de l’État comme objectif central, la Banque a placé le renforce- ment des capacités et la gouvernance, deux domaines historiquement faibles, au cœur de son action dans les LICUS. Pour centrer l’initiative en faveur des LICUS sur le programme complexe d’édification de l’État, la Banque doit préciser les domaines dans lesquels elle a un avantage relatif, ainsi que la portée et la teneur de ce programme. Elle doit également trouver de nouveaux moyens d’améliorer les résultats en matière de renforcement des ca- pacités et de gouvernance et définir des indicateurs de performance pour mesurer les ré- sultats obtenus sur le plan de l’édification de l’État. • Trois ans après le lancement de l’initiative en faveur des LICUS, des progrès limités ont été accomplis sur le front des importantes réformes à effectuer dans le domaine des ressources humaines, qu’il s’agisse des effectifs, de la qualité du personnel ou des mesures d’en- couragement offerts au personnel qui mène des travaux au titre des LICUS. • Bien que la Banque ait récemment souligné la nécessité d’accroître sa présence dans les LICUS, cela ne suffira pas à garantir la bonne mise en œuvre des stratégies-pays. L’ac- croissement de sa présence sur le terrain doit être allié à une meilleure communication entre les bureaux extérieurs de la Banque et le siège. Il importe également de déployer un per- sonnel de terrain suffisamment nombreux et doté des pouvoirs et des compétences voulus. • Les programmes de réformes des bailleurs de fonds dans les LICUS pourraient être plus sélec- tifs. Dans l’environnement complexe des LICUS, où des réformes s’imposent dans pra- tiquement tous les secteurs, il est impératif de bien échelonner les réformes et de prévoir suffisamment de temps pour les mettre en oeuvre si l’on entend produire les résultats es- comptés sans trop lourdement grever les capacités limitées des LICUS. Résumé analytique vec près de 500 millions d’habitants dont la moitié gagnent moins de 1 A dollar par jour, les pays à faible revenu en difficulté (jusqu’ à récemment dénommés LICUS) attirent une attention grandissante. Il y a tout lieu de douter de l’aptitude de ces pays à atteindre les objectifs de développement pour le Millénaire (ODM) et de craindre des effets économiques défavorables sur les pays voisins ainsi que les retombées mondiales qui pourraient s’ensuivre. Accablés de problèmes chroniques, les LICUS l’accent sur différents groupes de pays. Par ont des impératifs de développement qui exemple, les études récemment menées par le présentent d’énormes défis. La plupart de ces Center for Global Development portent sur les pays souffrent d’une mauvaise gouvernance et pays à faible revenu en stagnation (sur la base du sont en proie à des conflits internes de longue produit national brut par habitant et des taux de date ou connaissent des difficultés de transition croissance), tandis que l’indice des pays en post-conflictuelle. Tous sont confrontés aux situation de faillite de Foreign Policy met mêmes problèmes : manque de sécurité, l’accent sur l’échec de l’État en identifiant les relations sociales fragmentées, grave corrup- pays sur la base de critères tels que le niveau de tion, détérioration de l’ordre public, absence de déclin économique, la sécurité, les élites mécanismes de mise en place de pouvoirs divisées en factions, les personnes déplacées, légitimes, énormes besoins d’investissement en les violations des droits de l’homme et les souffrance et ressources publiques limitées pour interventions extérieures. L’Agence des États- le développement. Les interventions interna- Unis pour le développement international tionales dans ces pays n’ont jusqu’à présent pas (USAID) se concentre sur les problèmes de produit d’améliorations notables. vulnérabilité et les situations de crise, générale- Les bailleurs de fonds s’interrogent sur les ment dans l’arène politique. L’Agence britan- meilleurs moyens d’aider les pays confrontés à nique pour le développement international des problèmes aussi épineux. Du fait qu’ils ont (DFID) et l’Organisation de coopération et de des motivations et des objectifs différents, les développement économiques/Comité d’aide au bailleurs de fonds et les chercheurs ont choisi développement (OCDE/CAD) utilisent la même de s’attaquer à différents aspects de ces définition que la Banque mondiale pour identi- problèmes, ce qui les a conduits à mettre fier les États fragiles. xxxvii E N G A G I N G W I T H F R A G I L E S T AT E S Tels que définis par la Banque mondiale, tous exceptionnel), tandis que les budgets adminis- les LICUS se caractérisent par des politiques, des tratifs ont été plus également répartis entre les institutions et une gouvernance faibles. La Banque LICUS (7 LICUS sortant d’un conflit sur les 25 a utilisé deux critères pour définir les pays à faible LICUS ont reçu 64 % du montant total des prêts revenu les moins performants et les plus en aux LICUS et 34 % du budget administratif total difficulté (le terme LICUS ne désigne donc pas les affectué à ces pays). pays à faible revenu qui éprouvent tout juste Même si les LICUS sortant d’un conflit auraient quelques difficultés, lesquels sont identifiés par la peut-être reçu une grande partie des prêts en Banque uniquement aux fins de suivi) : revenu par l’absence de l’initiative en faveur des LICUS (étant habitant inférieur au seuil d’éligibilité défini par donné que l’IDA accordait déjà des fonds à titre l’Association internationale du développement exceptionnel aux pays sortant d’un conflit avant (IDA) et note égale ou inférieure à 3 (égale ou le lancement de cette initiative), l’initiative a inférieure à 2,5 pour les pays à faible revenu les probablement contribué à une répartition plus plus en difficulté) dans le cadre des Évaluations de égale des budgets administratifs entre les pays du la politique et des institutions nationales (CPIA) et groupe (sachant que les budgets administratifs des Évaluations CPIA de la gestion et des institu- ont augmenté de 400 % entre les exercices 00–02 tions du secteur public. et 03–05 pour trois LICUS, le Liberia, la Somalie et Certains pays à faible revenu pour lesquels il le Soudan, qui auraient reçu une minuscule n’existe pas de données CPIA sont également portion des budgets administratifs avant l’initia- inclus. Pour l’exercice 05, la Banque a identifié 25 tive car les prêts accordés à ces pays sont classés pays faisant partie des LICUS. Six LICUS recensés improductifs). durant l’exercice 05 n’avaient pas reçu de La démarche adoptée par la Banque pour les notation CPIA (Afghanistan, Liberia, Myanmar, LICUS a évolué depuis sa formulation initiale en Somalie, Timor-Leste et territoire du Kosovo). 2002, suivant six principes fondamentaux définis Ces données sont basées sur l’évaluation de pour ces pays (voir le tableau ES.1). Au départ, l’aide fournie par la Banque aux 25 pays classés l’initiative visait à fournir une aide plus efficace dans le groupe des LICUS durant l’exercice 05. en utilisant d’autres instruments, tels que des Les prêts et les budgets administratifs affectés travaux d’analyse et des transferts de connais- aux LICUS ont augmenté depuis le lancement de sances, assortis de transferts financiers pour l’initiative en faveur de ces pays. Les prêts aux promouvoir le changement. LICUS ont augmenté de quelque 2,5 milliards de En 2005, les objectifs et la portée de l’initia- dollars durant l’exercice 00–02 (avant le tive en faveur des LICUS ont été modifiés en lancement de l’initiative) à environ 4,1 milliards privilégiant l’efficacité générale de l’aide au de dollars durant l’exercice 03–05 (depuis le détriment de l’édification de l’État. Dans le cadre lancement de l’initiative). En valeur par habitant, de l’initiative, quatre modèles d’intervention les prêts aux LICUS ont varié entre 0 et 25,4 dollars (détérioration, crise prolongée ou impasse, durant l’exercice 03–05. Les budgets administrat- situation post-conflictuelle ou transition ifs affectés aux LICUS ont augmenté de quelque politique, et amélioration graduelle) ont 104 millions de dollars durant l’exercice 00–02 à également été adoptés pour traiter séparément environ 161 millions de dollars durant l’exercice les différents types de LICUS. La démarche 03–05. En valeur par habitant, les budgets adoptée en 2005 est restée axée sur les thèmes administratifs affectés aux LICUS se sont établis retenus en 2002, à savoir l’apprentissage par entre 0,002 et 4,5 durant l’exercice 03–05. l’action et les questions organisationnelles, tout Une grande partie des prêts alloués aux en renforçant l’action dans ces domaines. LICUS durant l’exercice 03–05 est allée aux La présente étude a pour objet de répondre à LICUS sortant d’un conflit (les pays sortant d’un trois questions : conflit sont identifiés sur la base des indicateurs de progrès post-conflit, ou PCPI, afin de • Dans quelle mesure la démarche suivie par la déterminer les dons accordés par l’IDA à titre Banque pour les LICUS est-elle efficace ? xxxviii R É S U M É A N A LY T I Q U E • Dans quelle mesure les critères retenus par la efforts). Banque pour identifier et classer les LICUS La majorité des parties prenantes interrogées sont-ils utiles sur le plan opérationnel, et dans dans le cadre de l’enquête menée par l’IEG a quelle mesure le système de répartition de indiqué que le programme général de la Banque l’aide est-il utile à cet égard ? dans les LICUS avait contribué dans une faible • L’appui interne fourni par la Banque aux travaux mesure au développement—opinion qui sur les LICUS est-il approprié? concerne l’appui global de la Banque et non la démarche proprement dite adoptée pour les Principaux résultats et conclusions LICUS. Les principes définis pour les LICUS ont Efficacité de la démarche suivie par la Banque initialement produit des résultats positifs. pour les LICUS L’initiative de la Banque en faveur des LICUS a L’expérience montre que les principes permis à la Banque d’accroître ses opérations fondamentaux définis pour les LICUS ont été dans les pays où elles auraient probablement été plus ou moins bien appliqués (voir le tableau moins importantes. La Banque a récemment RA.1). Les problèmes rencontrés au niveau de lancé des opérations dans plusieurs LICUS où la l’exécution sont attribuables dans certains cas paix vient d’être rétablie ou qui amorcent une aux objectifs trop ambitieux de la Banque (qui phase de transition politique. La Banque a doit donc réviser ses objectifs à la baisse), et également contribué à la stabilité macroé- dans d’autres cas au niveau inadéquat des efforts conomique et à la mise en place d’une ou des apports de la Banque, comme l’indiquent importante infrastructure matérielle, les enquêtes réalisées par l’IEG et son évalua- notamment dans les LICUS sortant d’un conflit. tion des rapports d’achèvement des stratégies D’importants progrès ont été réalisés en matière d’aide-pays (ce qui nécessite d’intensifier les de coordination des bailleurs de fonds sur le Tableau RA.1 : Mise en oeuvre des principes fondamentaux définis pour les LICUS Principes définis pour les LICUS Mise en oeuvre Rester mobilisés Substantielle Ancrer les stratégies dans une analyse sociopolitique plus robuste Moyenne • Comprendre la situation politique • Moyenne à substantielle • Intégrer la compréhension politique à la conception et à la mise en œuvre des stratégies • Moyenne à faible Promouvoir la demande et les capacités internes de changement positif Faible Promouvoir une première série de réformes simples et réalisables Moyenne à faible • Réformes macroéconomiques • Substantielle • Mise en place d’une infrastructure matérielle • Substantielle • Transition de la phase de reconstruction post-conflit à la phase de développement • Faible • Sélectivité et définition des priorités • Faible Étudier de nouveaux mécanismes de fourniture des services sociaux Moyenne Collaboration des bailleurs de fonds Moyenne • Au niveau des politiques internationales • Substantielle • Au niveau national • Moyenne à faible Mesure et suivi des résultatsa Faible Sources : Fnquêtes et analyses thématiques de base effectuées pour cette étude, IEG, 2005. a. Pas considéré par la Banque comme un principe fondamental en soi, mais inclus par l’IEG car cette activité est cruciale pour le programme d’apprentissage par l’action de la Banque dans les LICUS. xxxix E N G A G I N G W I T H F R A G I L E S T AT E S plan de la politique internationale, comme en en situation de détérioration, de crise témoigne l’adhésion récente d’un large éventail prolongée ou d’impasse. Il est également de bailleurs de fonds, dont la Banque, aux douze nécessaire de perfectionner les modèles en principes d’intervention internationale définis tenant plus explicitement compte des écarts de par l’OCDE/CAD. capacité d’exécution des fonctions centrales de La Banque a souvent joué un rôle de chef de l’État (mobilisation des ressources, répartition file en tant que coprésident des réunions interna- des ressources, fourniture des services sociaux tionales des bailleurs de fonds et co-auteur de et des infrastructures de base, et dispositions documents d’orientation communs. Les modèles politiques pour faire face à l’opposition et aux d’intervention récemment adoptés par la Banque problèmes de sécurité) pour permettre à la pour tenir compte des différents types de LICUS Banque de mieux adapter ses méthodes devraient permettre de prendre des mesures opérationnelles à l’environnement institution- mieux adaptées à la situation de chaque pays. Le nel des LICUS. pourcentage de projets achevés dans les LICUS Les travaux de la Banque sur les pays sortant qui ont donné des résultats jugés satisfaisants par d’un conflit sont plus anciens que l’initiative en l’IEG est passé de 50 % durant l’exercice 02, avant faveur des LICUS et le modèle d’intervention le lancement de l’initiative en faveur des LICUS, à correspondant pour les LICUS sortant d’un 58 % en 2004 et à 82 % en 2005. Les chiffres conflit est plus clairement défini que les autres correspondants pour les projets réalisés dans les modèles d’intervention. Il présente cependant pays à faible revenu non classés dans le groupe des lacunes et doit être encore mis au point pour des LICUS varient entre 70 et 79 %. guider les phases de transition et de développe- Mais il reste de sérieux défis à relever. Les ment qui suivent la phase de reconstruction opérations initiales de la Banque dans certains post-conflit. D’autre part, bien que la Banque LICUS n’ont pas toujours été suivies par un accorde une attention grandissante à la préven- programme de réformes ciblées et bien échelon- tion des conflits, on dispose de données limitées nées. D’autre part, il reste à mieux intégrer les sur l’efficacité de ses efforts dans ce domaine. réalités politiques dans la conception et la mise Il reste à définir clairement le rôle et l’avan- en œuvre des stratégies-pays de la Banque. Celle- tage relatif de la Banque en matière de préven- ci doit également améliorer la qualité de la tion des conflits, d’autant plus que les mesures à coordination de ses opérations dans les pays prendre dans ce domaine nécessitent que la avec celles des autres bailleurs de fonds, Banque accorde une plus grande attention aux notamment en assurant un suivi de l’exécution causes véritables des conflits et examine les qui aille au-delà des accords de principe. facteurs ethniques, sociologiques et politiques. Par ailleurs, la Banque a placé le renforce- La Banque doit mieux définir en quoi consiste ment des capacités et la gouvernance, deux son objectif de consolidation de la paix et domaines historiquement faibles, au cœur de comment atteindre cet objectif. son action en adoptant l’objectif plus complexe d’édification de l’État. Ce recentrage signifie que Utilité opérationnelle des mécanismes retenus la Banque doit identifier plus efficacement son par la Banque pour l’identification et le avantage relatif, améliorer sa performance, classement des LICUS, ainsi que pour notamment en définissant de nouvelles l’affectation de l’aide à ces pays approches, et identifier des partenaires à même Malgré le recentrage sur des objectifs d’édifica- de compléter ses travaux pour produire les tion de l’État et de consolidation de la paix, la résultats visés. Enfin, le choix du terme édifica- Banque continue d’utiliser presque exclusive- tion de l’État n’est peut-être pas heureux, car il ment les CPIA pour identifier les LICUS. Ces a des connotations politiques et idéologiques. évaluations ne tiennent cependant pas compte La Banque doit élaborer ses méthodes de certains aspects fondamentaux de la fragilité opérationnelles pour les LICUS, notamment des États (tels que les moyens de faire faire à pour les modèles d’intervention dans les pays l’opposition politique) et des conflits (tels que xl R É S U M É A N A LY T I Q U E l’instabilité politique et la sécurité ou le risque plusieurs fronts : de conflit) et il serait bon de les compléter par • Etoffement des travaux d’analyse en dissociant d’autres dispositifs. Un meilleur moyen d’identi- les budgets administratifs affectés aux études fier les LICUS serait d’utiliser un cadre économiques et sectorielles et à l’assistance analytique qui est plus explicitement axé sur les technique d’une part, des volumes de fi- objectifs de l’initiative en faveur des LICUS. nancement d’autre part Le système utilisé par la Banque pour allouer • Utilisation de Notes de stratégie intérimaire les ressources de l’IDA (appelé système d’affec- qui permettent de concevoir des stratégies tation des fonds en fonction de la performance) portant sur une plus courte période pour tenir est devenu plus sélectif au fil des années et les compte de l’instabilité de la situation dans les pays dont les politiques, les institutions et la LICUS gouvernance sont plus faibles reçoivent moins • Accès des responsables des LICUS à l’équipe de ressources de l’IDA. Cela a soulevé la de direction de la Banque question de savoir si l’IDA alloue des montants • Création du Fonds fiduciaire LICUS pour appropriés aux LICUS. Les modifications fournir des financements aux pays dont la apportées au système d’affectation des fonds en dette est improductive (la Banque n’avait fonction de la performance se sont traduites par jusqu’alors pas d’instrument à cet effet). un accroissement des ressources allouées par l’IDA, notamment à certains LICUS sortant d’un À la lumière des données d’expérience des conflit ou en transition politique. Mais il est pays, l’équipe chargée des LICUS a établi des difficile de dire si les niveaux actuels de finance- directives sur un certain nombre de questions ment IDA garantissent que les LICUS ne importantes pour enrichir les conseils reçoivent pas une aide insuffisante ou excessive. opérationnels fournis aux équipes-pays et les La question de la répartition de l’aide a débats plus généraux sur la politique extérieure. ressurgi, certaines études mettant en cause les Cependant, trois ans après avoir reconnu la preuves empiriques de la corrélation positive nécessité de modifier sa culture interne afin de entre les politiques et l’efficacité de l’aide (sur bien mettre en œuvre la démarche adoptée laquelle repose le système d’affectation des fonds pour les LICUS, la Banque a peu progressé sur le en fonction de la performance). D’autres études plan de l’appui interne fourni aux travaux sur les montrent que l’aide peut contribuer à promou- LICUS. Il reste beaucoup à faire en matière de voir des changements d’orientation durables dans réformes des ressources humaines, qu’il s’agisse les États en situation de faillite en créant et des effectifs, de la qualité du personnel et des renforçant les conditions nécessaires aux incitations offertes pour encourager à réformes ou en accroissant les chances que les poursuivre les travaux sur les LICUS. Les réformes mises en place s’inscrivent dans la commentaires du personnel de la Banque sur durée. Ces études concluent que l’aide fournie l’importance de travailler à la fois sur les LICUS aux LICUS peut produire des résultats exception- et sur les autres pays montrent que les travaux nels, bien que les risques d’échec soient accomplis au titre des LICUS ne sont pas substantiels. suffisamment pris en compte au sein de la Pour sa part, la Banque doit aborder la Banque et soulignent la nécessité de réformer le question de l’affectation de l’aide aux LICUS système d’incitation. d’une manière qui tienne compte de ses objectifs Les membres du personnel interrogés ont pour ces pays et garantisse que les LICUS ne indiqué que les directeurs des opérations pour reçoivent pas une aide insuffisante ou excessive. les pays n’accordent pas tous la même attention aux LICUS, tout particulièrement s’ils sont Validité et efficacité de l’appui interne de la également chargés d’un pays plus vaste, plus Banque aux travaux menés au titre des LICUS « performant » ou plus médiatisé, ce qui souligne L’appui interne fourni par la Banque aux travaux la nécessité d’accorder une attention soutenue menés au titre des LICUS s’est amélioré sur aux travaux sur les LICUS à tous les niveaux de la xli E N G A G I N G W I T H F R A G I L E S T AT E S direction. Dans l’enquête menée par l’IEG qu’il y a un conflit ou une contradiction. auprès des parties prenantes, la majorité des personnes interrogées à la Banque ont indiqué Leçons tirées de l’expérience de la que le fait de travailler sur les LICUS n’avait rien Banque et des autres bailleurs de fonds changé dans plusieurs aspects des ressources Il y a plusieurs enseignements à tirer de cette humaines (voir la figure RA.1). évaluation de l’application par la Banque des Il y a de nombreux doubles emplois et il principes fondamentaux régissant la démarche règne une grande confusion sur le rôle et les suivie pour les LICUS. Le rapport 2002 du responsabilités de l’équipe chargée des LICUS groupe de travail sur les LICUS (Banque d’une part, et celle chargée de la prévention des mondiale 2002) indique qu’il y a des améliora- conflits et de la reconstruction d’autre part. Les tions à apporter dans bon nombre des domaines personnes interrogées ont indiqué qu’elles ne couverts dans le cadre de ce bilan et les rapports savent pas à quelle équipe s’adresser pour 2005 de la Banque sur les LICUS soulignent obtenir des conseils spécifiques ni quel type de également cette nécessité. Par exemple, il soutien elles peuvent attendre de chaque importe d’ancrer les stratégies dans une analyse équipe. Environ les deux tiers des membres du sociopolitique plus rigoureuse et de promou- personnel de la Banque ayant participé à voir des programmes de réformes très ciblés. l’enquête de l’IEG auprès des parties prenantes Les enseignements tirés reposent sur l’expéri- ont indiqué que l’organisation actuelle n’était ence de la Banque mais ils peuvent également pas satisfaisante : 37 % ont dit qu’il y avait des fournir des orientations aux autres bailleurs de doubles emplois entre l’appui fourni par fonds sur l’aide à fournir aux LICUS. l’équipe de la Banque chargée des LICUS et celui fourni par l’équipe chargée de la prévention des Intervention dans les LICUS conflits et la reconstruction ; 15 % estiment qu’il y a de nombreux doubles emplois et 12 % disent Rester mobilisés n’est pas une fin en soi et il faut Figure RA.1 : La majorité des personnes interrogées à la Banque ont indiqué que le fait de travailler sur les LICUS n’avait rien changé dans plusieurs domaines (voir ci-dessous) 100 75 Pourcentage 50 25 0 Perspectives Rémunération Attentes réalistes Niveau de soutien Efforts de la Banque de carrière de la direction de la directon pour assurer la sécurité de personnel Mieux Pareil Pire Source : Annexe Z (Résultats de l’enquête auprès des parties prenantes). Note : Le nombre de réponses valides varie entre 213 et 238. La question a été posée sans distinction à des personnes qui avaient travaillé sur un LICUS ou sur des pays ne faisant pas partie de ce groupe. xlii R É S U M É A N A LY T I Q U E rapidement mettre en place un programme de Les directives de la Banque pour les pays en réforme clair et approprié dans les LICUS. En conflit prolongé ou dans l’impasse politique l’absence d’un programme de réforme clair et stipulent que « les questions de développement approprié, les premiers succès remportés par relativement peu controversées peuvent fournir les opérations dans les pays risquent d’être de le point de départ d’un dialogue constructif courte durée et de peu contribuer à la réalisa- entre les parties en conflit ». Pour les pays où la tion des objetifs des stratégies-pays. Les gouvernance se dégrade, les directives exemples de la République centrafricaine et de indiquent que la Banque devrait « donner son Haïti montrent qu’il peut être difficile de avis sur les questions économiques particulière- poursuivre sur la lancée des premiers succès ment importantes pour les efforts de médiation remportés dans les LICUS. Les progrès et susceptibles d’aider à relancer le dialogue » politiques ne s’étant pas répercutés dans l’arène (Banque mondiale 2005e). économique, le gouvernement de la République centrafricaine est aujourd’hui confronté à une Les obstacles à la prise en main des opérations par crise budgétaire qui pourrait être lourde de le pays et à l’amélioration des capacités d’absorp- conséquences. À Haïti, il semble que les tion concernent tant les produits de diffusion du bailleurs de fonds n’aient pas accordé l’attention savoir que les produits financiers. La participation voulue au maintien d’un niveau minimum de des homologues nationaux aux travaux sécurité. Dans les deux cas, les résultats initiale- d’analyse de la Banque reste limitée aux ment favorables de l’initiative en faveur des questions administratives. Les pays clients LICUS risquent de ne pas s’inscrire dans la participent beaucoup moins à la sélection des durée. sujets d’étude et à la réalisation des analyses, ce Dans certains cas, le retrait stratégique – à qui réduit l’adhésion nationale. La participation l’exception des travaux d’analyse internes – des homologues nationaux est cependant pourrait être nécessaire, du moins temporaire- indispensable pour assurer la prise en charge ment. C’est une stratégie particulièrement des opérations par le pays client et accroître appropriée lorsque l’intervention de la Banque l’impact des travaux d’analyse. donne l’impression de légitimer à tort le Au Tadjikistan, la faible participation des gouvernement d’un LICUS ou lorsque que cela autorités locales à la sélection et à la réalisation atténue les pressions exercées en faveur des des travaux d’analyse de la Banque a limité leur réformes, ce qui pourrait entraver la mise en intérêt pour les résultats, ce qui a empêché le place des conditions nécessaires à une réforme bon déroulement des travaux. En Angola, politique réelle et durable. certains travaux d’analyse menés par la Banque Dans les scénarios de détérioration ou de (tels que le récent Mémorandum économique crise prolongée, où les bailleurs de fonds et les sur le pays) ont été considérés par les hauts autorités locales ne sont souvent pas d’accord responsables politiques comme une ingérence sur la stratégie de développement, l’interven- de la Banque dans leurs affaires internes, ce qui tion doit inclure un dialogue sur l’action à mener a limité l’adhésion nationale et le renforcement pour ouvrir la voie aux réformes, tout en des capacités. Sans adhésion nationale, il y a peu préparant un programme de réforme au cas où de chances que les travaux d’analyse influencent les conditions deviendraient favorables. Dans les la politique du gouvernement. scénarios de situation post-conflictuelle ou de Le manque de capacité d’absorption des transition politique et d’amélioration graduelle, autorités des LICUS pour utiliser les travaux l’intervention devra être de nature plus d’analyse pourrait également limiter les possibil- technique et mettre davantage l’accent sur la ités de transfert de connaissances. Par exemple, mise en œuvre du programme de réforme, le gouvernement angolais a approuvé la Note de compte tenu du plus haut degré de consensus stratégie intérimaire de la Banque mais exprimé entre les bailleurs de fonds et le gouvernement son inquiétude face au volume des travaux sur les réformes à entreprendre. d’analyse et de conseil envisagé, ce qui a soulevé xliii E N G A G I N G W I T H F R A G I L E S T AT E S la question de savoir si le gouvernement utilis- Les bailleurs de fonds doivent s’attacher à aider les erait pleinement les services d’analyse et de services concernés à intégrer les analyses conseil. La capacité d’absorption du gouverne- politiques à la conception et à la mise en œuvre de la ment est très faible, et les travaux d’analyse et de stratégie. Bien que la Banque ait effectué ou ait conseil effectués essentiellement par la Banque eu accès à de bonnes analyses politiques dans risquent de créer des tensions avec le gouverne- certains LICUS, elle n’a pas suffisamment tenu ment, quelle que soit leur qualité technique. Au compte de ces analyses dans sa stratégie. Par Cambodge, les services d’analyse et de conseil exemple, la stratégie intérimaire en Papouasie- prévus dans la stratégie d’aide au pays (CAS) Nouvelle-Guinée contient une bonne analyse du définie en 2005, qui représentent 30 activités à système politique et reconnaît les problèmes en mener à bien durant l’exercice 05-07, semblent présence, tels que les allégeances de clan, le trop ambitieux compte tenu des faibles clientélisme politique, la corruption et le capacités institutionnelles du pays. manque de capacité, mais elle les traite comme des problèmes techniques et n’en tient pas Analyse politique et son utilisation dans la convenablement compte pour étayer la stratégie-pays démarche globale. Il est essentiel que les bailleurs de fonds demandent Un programme de réforme ciblé et utilisent de bonnes analyses politiques dans les LICUS, sans nécessairement les réaliser eux-mêmes. Dans l’environnement complexe des LICUS, où des L’objectif d’une équipe-pays devrait être de réformes s’imposent dans pratiquement tous les demander ou utiliser (sans forcément les secteurs, il est impératif de bien échelonner les réaliser) des analyses qui se rapportent ou réformes et de prévoir suffisamment de temps pour servent directement à l’élaboration d’une les mettre en oeuvre si l’on entend produire les stratégie. Dans les LICUS, notamment lorsqu’il résultats escomptés sans trop lourdement grever les est essentiel d’agir vite, les bailleurs de fonds capacités limitées des LICUS. Alors que les bailleurs doivent veiller à ce que les analyses politiques de fonds doivent collectivement faire preuve de existantes soient exploitées avant de faire sélectivité, c’est loin d’être le cas, comme le réaliser de nouvelles analyses. montrent les examples du programme de En République démocratique populaire lao, la réforme parrainé par les bailleurs de fonds en Banque a mis à profit les analyses politiques Afghanistan et aussi le cadre de coopération existantes et invité un politologue qui avait intérimaire à Haïti décrits ci-après. Cependant, publié de nombreux articles sur le pays à présen- même si tous les bailleurs de fonds ne parvien- ter un exposé à l’équipe-pays sur la situation nent pas immédiatement à appliquer le principe politique et la réforme dans le pays. Cela a de sélectivité, la Banque doit veiller à ce que son permis d’établir un résumé indépendant des propre programme d’aide soit ciblé et sélectif, en analyses politiques pertinentes (adapté aux faisant appel à ses compétences de base. La besoins des bailleurs de fonds, en particulier de Banque a fait preuve d’une plus grande sélectiv- la Banque) qui a été distribué aux membres du ité au cours des dernières années, mais cela reste personnel de la Banque concernés et à d’autres une gageure. bailleurs de fonds. Cela a également évité En Afghanistan, les réformes parrainées par d’effectuer une analyse « maison » plus coûteuse, les bailleurs de fonds sont de nature très diverse, ainsi que des tensions potentielles avec le manquent de sélectivité et ont conduit à l’élabo- gouvernement, en permettant à la Banque de ne ration de 120 projets de loi. Il est impératif que pas se noyer dans les détails de l’analyse. Pour la les bailleurs de fonds définissent les priorités et Banque, l’acquisition et la diffusion des connais- le calendrier de ces réformes, qui portent sur sances existantes se sont avérées plus pratiquement tous les secteurs économiques et importantes et efficaces que la création de savoir. sociaux du pays. À Haïti, le cadre de coopération intérimaire est censé fournir des orientations xliv R É S U M É A N A LY T I Q U E pour l’aide internationale et la coopération avec conflit. Au lendemain d’un conflit, la le pays jusqu’en septembre 2006. Il couvre communauté internationale des bailleurs de pratiquement toutes les fonctions de base de fonds a tendance à concentrer son assistance sur l’État, que ce soit la sécurité, le dialogue la reconstruction de l’infrastructure. Vu les national, la gouvenance économique, le faibles capacités d’utilisation de l’aide et la redressement économique ou les services mauvaise gestion des affaires publiques dans la essentiels. Individuellement, chacun de ces plupart des LICUS sortant d’un conflit, la priorité domaines semble important, mais ensemble ils initiale devrait être de renforcer les capacités et représentent un programme monumental. d’améliorer la gouvernance, pas seulement de S’agissant du propre programme d’aide de la reconstruire l’infrastructure matérielle. Cela Banque, l’évaluation par l’IEG du rapport pourrait passer par le renforcement des institu- d’achèvement de la CAS à São Tomé-et-Principe tions publiques ou la création de nouvelles pour les exercices 01-05 montre que la Banque institutions, la réforme de la fonction publique avait des objectifs beaucoup trop ambitieux par et le recours aux compétences locales. Si on fait rapport aux ressources allouées au pays. Au appel à des experts étrangers pour l’assistance total, nombre des objectifs de la CAS n’ont pas technique, il faut veiller à ce que cela ne compro- été réalisés ou ne l’ont été que partiellement. mette pas le développement à long terme des Outre la sélectivité dans les CAS, il faut bien capacités locales. veiller à ce que les réformes mises en œuvre dans les pays soient ciblées et hiérarchisées. Le Coordination des bailleurs de fonds manque de sélectivité et l’absence d’un ordre de priorité dans les programmes de réforme La coordination des bailleurs de fonds ne peut être soulèvent des questions d’efficacité, compte efficace sans une vision et un objectif communs. tenu des capacités limités des LICUS. Bien qu’il Lorsque les objectifs des bailleurs de fonds ne soit difficile d’être sélectif dans un pays où il faut peuvent pas être pleinement harmonisés, il est trouver une solution rapide à de nombreux important qu’ils soient au moins complémentaires. problèmes, il est essentiel de bien échelonner La démarche adoptée par la Banque n’a pas les réformes pour éviter de trop lourdement entièrement tenu compte des différentes mo- grever les capacités limitées des LICUS, tout en tivations des bailleurs de fonds dans leurs évitant des solutions partielles. L’étalement des opérations dans les LICUS. Bien que la notion réformes sur un nombre suffisant d’années est globale de fragilité soit généralement comprise essentiel, de même que la volonté des bailleurs et acceptée, les pays jugés fragiles diffèrent de fonds de veiller à leur mise en œuvre. selon les bailleurs de fonds. Les raisons de Au Timor-Leste, les bailleurs de fonds se sont soutenir les États fragiles vont de la sécurité à la peut-être retirés trop tôt, sans accorder l’atten- prévention des conflits, en passant par l’efficac- tion voulue aux pressants besoins de capacités ité de l’aide, le développement équitable, la du pays. À Haïti, l’aide au développement a été réduction de la pauvreté, l’édification de l’État très variable au fil du temps. Le pays a connu des et la consolidation de la paix. hauts et des bas dans ses relations avec la En Afghanistan et au Tadjikistan, les enquêtes communauté des bailleurs de fonds. Cela aurait réalisées par l’IEG montrent que les principaux pu être évité si les bailleurs de fonds avaient bailleurs de fonds n’ont pas le même objectif. En mieux coordonné et échelonné leur aide. l’absence d’un objectif global commun, les politiques adoptées ne peuvent être cohérentes. Renforcement des capacités dans les LICUS L’ampleur et la nature des moyens déployés par sortant d’un conflit la Banque pour coordonner l’action des bailleurs de fonds ne tiennent pas suffisamment Les programmes de renforcement des capacités et compte des objectifs des différents acteurs dans d’amélioration de la gouvernance doivent être lancés chaque pays. Ceci étant, la coordination des rapidement, même dans les LICUS sortant d’un bailleurs de fonds est une forme d’action collec- xlv E N G A G I N G W I T H F R A G I L E S T AT E S tive qui demande que les autres bailleurs de • Sachant que les progrès sont souvent lents fonds se montrent également solidaires de la dans ces pays, il est important de réévaluer Banque et subordonnent leurs programmes régulièrement le programme pour déterminer bilatéraux aux objectifs multilatéraux convenus. s’il a des chances de produire les résultats es- comptés. Coordination bien ordonnée commence par soi- • L’instabilité et l’évolution constante de la situ- même. La coordination n’est pas seulement ation dans les LICUS, où les progrès sont sou- importante entre les donateurs multilatéraux et vent en dents de scie, font qu’il est essentiel bilatéraux. C’est également un impératif au sein d’adapter les programmes : le suivi étroit des de chaque bailleur de fonds. Les projets réalisés résultats aidera à déterminer quand des adap- dans différents secteurs d’un même pays sont tations sont nécessaires et sous quelle forme. souvent exécutés parallèlement sans exploiter les synergies, comme dans le cas des projets L’apprentissage par l’action indispensable d’autonomisation des populations locales et de pour améliorer l’efficacité des opérations de la promotion de l’agriculture parrainés par la Banque dans les LICUS demande un rigoureux Banque au Timor-Leste. système de suivi et d’évaluation. L’une des conséquences indirectes de la La Banque a indiqué que l’édification de l’État décentralisation des opérations de la Banque et la consolidation de la paix devraient être les dans les bureaux extérieurs a été la concentra- critères utilisés pour évaluer le succès de l’initia- tion des données sur les pays entre les mains tive en faveur des LICUS. Mais elle n’a toujours du personnel local et le manque de communi- pas défini d’indicateurs de performance à cet cation de ces données aux membres de effet ni de critères d’évaluation des résultats. l’équipe-pays, notamment ceux basés à Lorsque le changement est plus souvent axé sur Washington. Il est essentiel de régler les les processus, notamment dans les scénarios problèmes de coordination entre les différents d’intervention en cas de détérioration et de crise services des bailleurs de fonds (tels que les prolongée ou d’impasse, les résultats et les services de la Banque chargés de la gestion du impacts visés dans les autres scénarios ne sont secteur public, de la prévention des conflits et pas forcément des critères de succès appropriés. de la reconstruction, des LICUS, du renforce- Les objectifs devraient être adaptés au contexte ment des capacités et des études) dans les particulier des LICUS, qui devrait lui-même LICUS, où les problèmes sont nombreux et déterminer les critères et garantir que les complexes et nécessitent souvent des interven- objectifs visés ne sont pas trop ambitieux. tions multisectorielles. Amélioration de l’appui fourni aux travaux sur Mesure et suivi des résultats les LICUS au niveau de l’organisation interne Le suivi et l’évaluation sont au moins aussi La présence sur le terrain ne suffit pas à garantir la importants dans les LICUS que dans les autres pays. bonne mise en œuvre des stratégies-pays. Elle doit Le suivi et l’évaluation sont d’une importance être assortie d’une communication adéquate entre les cruciale pour plusieurs raisons : bureaux extérieurs et le siège des bailleurs de fonds. Il importe également de déployer un personnel de • La Banque, comme les autres bailleurs de terrain suffisamment nombreux et doté des pouvoirs fonds, ne sait pas encore quelles démarches et des compétences voulus. La meilleure façon sont les plus efficaces dans les LICUS. Il est es- d’apprécier la situation dans un pays est souvent sentiel de suivre de près les résultats obtenus d’accroître la présence sur le terrain, mais cela pour tirer les leçons de l’expérience, et il faut ne suffit pas. Il est également important de faire une plus grande place à l’apprentissage et procéder à une analyse interne dans tous les au partage des données dans les travaux ef- services compétents des bailleurs de fonds et fectués au titre des LICUS. d’appliquer les leçons tirées de cette analyse à xlvi R É S U M É A N A LY T I Q U E toutes les interventions. Au Cambodge, par autorité suffisante par rapport au siège pour exemple, la présence de la Banque sur le terrain faire en sorte que celui-ci ne doive pas approu- a permis de beaucoup mieux apprécier la ver toutes les décisions. Pour avoir une vértable situation politique, mais les entretiens menés présence sur le terrain, il faut déployer un avec les membres de l’équipe-pays et les autres personnel compétent. Dans les entretiens semi- parties prenantes montrent que ces données directifs organisés pour la présente étude, restent concentrées entre les mains de quelques plusieurs bailleurs de fonds ont souligné que la responsables et leurs collaborateurs (principale- coordination dépend tout particulièrement des ment au sein de la représentation locale de la forces et des faiblesses des individus concernés. Banque et au centre de Bangkok), sans être En offrant une formation plus appropriée au systématiquement communiquées aux autres personnel déployé dans des situations difficiles membres de l’équipe-pays. et en mettant en place des mesures pour Il s’agit moins aujourd’hui d’un problème l’encourager à collaborer avec les autres d’appréciation des réalités politiques au bailleurs de fonds, la Banque pourrait atténuer Cambodge que de savoir qui détient ces ces risques particuliers. informations au sein de l’équipe-pays de la Dans un scénario de détérioration de la Banque et comment elles sont utilisées pour situation, où il pourrait y avoir une rupture du éclairer les décisions concernant la mise en dialogue avec le gouvernement, le personnel œuvre de la stratégie et du programme. Du fait des bailleurs de fonds devra déployer ses talents que seuls quelques membres du personnel de négociation et de persuation afin de possèdent des données détaillées sur le pays, poursuivre le dialogue tout en mobilisant des toutes les activités et interventions de la Banque groupes non gouvernementaux, notamment la ne peuvent pas en bénéficier. D’une manière société civile. générale, il faut accroître le transfert de connais- Dans un scénario de crise prolongée ou sances entre les bureaux extérieurs des bailleurs d’impasse, où les problèmes sont chroniques et de fonds et leurs équipes-pays et sectorielles où la situation politique est bloquée, le person- basées au siège. nel devra faire preuve d’une patience et d’une Bien que cela coûte cher, les bureaux créativité exceptionnelles et chercher des extérieurs doivent être dotés du personnel voulu solutions inédites pour sortir de l’impasse. pour intervenir efficacement dans les pays Dans un scénario de transition post- clients. En Angola, la petite équipe locale initiale- conflictuelle ou politique, le personnel devra ment déployée a dû s’atteler à une multitude de disposer de connaissances techniques partic- tâches allant du dialogue stratégique avec le ulières pour savoir comment mettre en place gouvernement et les bailleurs de fonds aux des systèmes économiques valables, des institu- dispositions logistiques à prendre pour tions et des infrastructures de base. Il devra s’installer dans ses nouveaux locaux. La situation également être capable d’agir rapidement et a été compliquée par le manque de personnel résolument dans ces situations, avant que l’opti- opérationnel présent sur place pour préparer les misme soulevé par le retour à la paix se dissipe réunions de haut niveau entre les ministres et la et pour aider les pays à éviter une reprise des Banque. Faute de personnel aux échelons hostilités. Étant donné qu’une aide interna- inférieurs, les questions ont été adressées trop tionale massive est souvent mobilisée dans ces rapidement aux instances supérieures, ce qui a situations, le personnel des bailleurs de fonds créé des tensions inutiles. Les bailleurs de fonds doit posséder de solides aptitudes de coordina- devraient déterminer les effectifs à déployer dans tion et de programmation pour organiser les chaque LICUS en tenant compte de la portée et partenaires du développement et leurs activités. de la nature des opérations prévues ainsi que de Dans un scénario d’amélioration graduelle de leurs objectifs respectifs dans ces pays. la situation, les principales compétences requises Outre la question des effectifs, les bureaux sont l’aptitude à fournir une assistance technique extérieurs doivent également jouir d’une adaptée aux besoins et à travailler de pair avec un xlvii E N G A G I N G W I T H F R A G I L E S T AT E S pays qui a déjà entrepris des réformes. s’agit d’éviter les lendemains qui déchantent en L’apprentissage passe par la mise en commun des tenant la population informée et en exposant données d’expérience – positives et négatives, mais clairement les tenants et les aboutissements de cela demande un environnement institutionnel l’action des bailleurs de fonds, ainsi que les réceptif et l’appui de la direction. La mise en résultats visés. Dans un scénario d’amélioration commun des données d’expérience sur les graduelle de la situation, la stratégie de bonnes et les mauvaises solutions aux communication doit être davantage axée sur problèmes des LICUS peut promouvoir l’information en présentant les données de l’apprentissage, ce qui est particulièrement l’expérience dans différents pays et secteurs. important pour les travaux effectués au titre des Dans la période immédiate qui suit la LICUS car les bailleurs de fonds continuent de cessation des hostilités, la communauté interna- s’interroger sur la meilleure façon d’aider ces tionale des bailleurs de fonds, y compris la pays en difficulté. Bien que la Banque ait partagé Banque, a souvent engagé d’importants volumes certains enseignements dégagés dans le cadre d’aide assortis d’objectifs trop ambitieux. Cela de la série de séminaires de son groupe de crée généralement des attentes excessives dans réflexion sur les LICUS, il faut s’attacher la population et conduit au désenchantement davantage à faire régulièrement le point et à lorsque ces attentes ne sont pas satisfaites et qu’il diffuser les nouvelles leçons tirées de l’expéri- y eu peu d’améliorations tangibles dans la vie ence de ces pays, tant par la Banque que par les quotidienne. Il est essentiel d’éviter de autres bailleurs de fonds. poursuivre des objectifs trop ambitieux et de Il est essentiel de créer un environnement mieux utiliser la communication, et la Banque institutionnel plus réceptif et d’obtenir l’appui doit investir dans ces stratégies. de la direction pour partager les données d’expérience négatives. Jusqu’à présent, la Des directives opérationnelles sont nécessaires pour Banque semble avoir surtout fait part des adapter les démarches suivies par les bailleurs de exemples positifs, comme on peut le constater fonds à la situation particulière des LICUS. L’initia- dans ses derniers rapports sur les LICUS. tive en faveur des LICUS a attiré l’attention sur la nécessité d’agir différemment dans les LICUS, La communication est indispensable pour faire en mais la Banque et les autres bailleurs de fonds sorte, d’une part, que les LICUS acceptent les n’ont pas défini clairement la marche à suivre. Ils démarches suivies par les bailleurs de fonds dans doivent également s’employer plus activement à ces pays, et d’autre part, qu’ils n’attendent pas trop déterminer dans quelle mesure les démarches de ces efforts, surtout dans la période qui suit adoptées dans les LICUS doivent, et peuvent, immédiatement la cessation des hostilités. Il s’attaquer aux raisons pour lesquelles les pays importe de mieux informer les LICUS des sont ou deviennent des LICUS, et pas seulement objectifs et des démarches des bailleurs de aux symptômes. En adoptant des solutions qui fonds pour obtenir leur adhésion. Cela pourrait considèrent les causes du mal comme un fait également éviter les déceptions en modérant les acquis, on risque d’ignorer des facteurs attentes des parties prenantes sur ce qui peut contextuels fondamentaux. Les directives être accompli sur une période donnée. opérationnelles des bailleurs de fonds doivent Dans les scénarios de détérioration de la insister sur l’importance d’axer les efforts situation et de crise prolongée ou d’impasse, où uniquement sur les domaines où ils ont un la situation économique et sociale stagne ou se avantage relatif, de sorte que les travaux effectués dégrade sur pratiquement tous les fronts, la par chaque bailleur de fonds prennent en stratégie de communication doit viser à compte et complètent ceux effectués par les promouvoir activement les avantages de la autres. réforme auprès des autorités et de la société Les plus grands défis à relever par la civile. Dans le modèle de transition post- communauté des bailleurs de fonds se posent conflictuelle ou politique établi par la Banque, il dans les situations de détérioration et de crise xlviii R É S U M É A N A LY T I Q U E prolongée ou d’impasse, de même que dans les sur le front du renforcement des capacités et de phases de transition et de développement qui la gouvernance. Il y a trois impératifs à suivent la phase de reconstruction dans le respecter : les interventions de la Banque scénario de situation post-conflit ou de transition doivent mieux tenir compte de la capacité des politique. Ce sont également des domaines dans LICUS d’assurer les fonctions centrales de l’État, lesquels il y a relativement peu d’idées nouvelles. elles doivent être bien échelonnées et adaptées Il est urgent de formuler des directives à l’environnement des LICUS, où de grandes opérationnelles dans plusieurs domaines : améliorations sont nécessaires dans pratique- comment organiser et échelonner les réformes ment tous les aspects des capacités et de la selon les priorités tout en évitant les solutions gouvernance, et il faut mettre en place un partielles ; comment mettre rapidement en place système efficace de suivi dans ces domaines. des services sans compromettre le développe- ment à long terme des capacités de l’État ; • Définir des critères d’affectation de l’aide en faveur comment promouvoir la réconciliation politique des LICUS pour faire en sorte que cette aide ne soit tout en aidant à rétablir une gouvernance efficace ni insuffisante ni excessive. et létigime ; comment intégrer les réalités La Banque doit effectuer un examen politiques dans la conception et la mise en œuvre technique de l’effet cumulatif des différentes des stratégies-pays ; et comment tenir compte modifications du système d’affectation des fonds des liens entre politique, sécurité et développe- en fonction de la performance sur les volumes ment. d’aide accordés aux LICUS. Il importe d’établir Les récentes directives de la Banque sur la des critères d’affectation des fonds qui tiennent marche à suivre dans les LICUS portent compte des objectifs de la Banque dans ces pays davantage sur les instruments à utiliser que sur et de veiller à ce que l’aide fournie ne soit ni les différentes manières de procéder selon les insuffisante ni excessive. Il convient également catégories de LICUS. Il serait également bon que de déterminer dans quelle mesure ces critères les équipes chargées des LICUS puissent se devraient être basés sur des facteurs autres que référer à des textes explicatifs, comme ceux l’efficacité des mesures prises par les pays (tels présentés au chapitre 2 de la présente étude, que les volumes d’aide des autres bailleurs de ainsi qu’à des notes succinctes adaptées aux fonds, l’évaluation des risques et des résultats problèmes, au lieu des notes d’orientation potentiels, et les retombées sur le plan régional officielles qui sont souvent trop condensées et et mondial), en tenant compte du fait que l’aide sans rapport avec le contexte du pays. est limitée et qu’il y aura des choix à faire. Recommandations • Renforcer l’appui interne de la Banque aux travaux accomplis au titre des LICUS au cours des trois • Préciser la portée et la teneur du programme d’éd- prochaines années. ification de l’État formulé par la Banque et ren- L’appui interne de la Banque doit être forcer la conception et la mise en place des renforcé dans deux domaines. Tout d’abord, il mécanismes d’appui au renforcement des capac- faut définir les priorités en ce qui concerne les ités et d’amélioration de la gouvernance dans les effectifs, les compétences et les mesures à mettre LICUS. en place pour encourager les travaux au titre des Compte tenu de l’insuffisance des mesures LICUS. Pour fournir des incitations à même prises dans le domaine du renforcement des d’attirer des individus qualifiés pour travailler sur capacités et de la gouvernance, et de la priorité les LICUS, au siège comme dans les bureaux accordée à l’objectif plus ambitieux d’édification extérieurs, il faut que les membres du personnel de l’État dans les LICUS, la Banque doit préciser sachent clairement quels sont les critères de ses avantages relatifs par rapport aux autres succès dans les LICUS, de quels résultats ils bailleurs de fonds et adopter des démarches seront tenus comptables, quels risques ils novatrices pour obtenir de meilleurs résultats peuvent raisonnablement prendre, comment les xlix E N G A G I N G W I T H F R A G I L E S T AT E S échecs seront évalués et comment ces résultats contexte des effectifs globaux de la Banque, en influenceront leur notation générale et leurs tenant compte du fait que l’affectation d’un plus perspectives de carrière. grand nombre d’individus plus qualifiés aux Tout comme dans les épreuves de plongeon travaux à mener au titre des LICUS imposera aux Jeux olympiques, où la note finale tient probablement des arbitrages avec les autres compte de la qualité technique et de la difficulté équipes-pays de la Banque. La redistribution des du plongeon, la performance des membres du effectifs en faveur des LICUS peut être justifiée personnel dans les LICUS devrait être évaluée en ou non, selon les objectifs de la Banque dans ces utilisant un coefficient de pondération pays et les besoins d’aide des autres pays clients. correspondant au degré de difficulté de l’envi- Ensuite, il convient de simplifier la structure ronnement de chaque LICUS. Il convient organisationnelle des travaux effectués au titre également de signaler l’importance des travaux des LICUS et de la prévention des conflits. La effectués au titre des LICUS à tous les niveaux de Banque doit mettre en place une organisation la direction. efficace qui élimine les doubles emplois et la En plus des mesures d’encouragement, la fragmentation de l’appui entre l’équipe chargée Banque doit faire en sorte que les membres du des LICUS et celle chargée de la prévention des personnel travaillant sur les LICUS possèdent les conflits et de la reconstruction. compétences voulues, par exemple en gestion du secteur public, qu’ils soient capables • Réexaminer la valeur ajoutée de la démarche suivie d’obtenir les connaissances politiques pour les LICUS au bout de trois ans. nécessaires et de les utiliser, et qu’ils soient L’utilité d’une classification spéciale pour les désireux et capables de travailler dans des LICUS et de la démarche adoptée pour ces pays, équipes pluridisciplinaires. À noter une mesure notamment l’intérêt opérationnel des modèles positive, quoique tardive, à cet égard : il est prévu d’intervention, devra faire l’objet d’une évalua- d’aborder ces questions dans le Rapport 2006 sur tion indépendante au bout de trois ans, le renforcement des interventions dans les États lorsqu’on disposera de suffisamment de fragiles. données sur les résultats obtenus. Il sera alors Il importe de réfléchir plus systématique- possible de se poser la question plus fondamen- ment au personnel à affecter aux LICUS dans le tale de savoir si l’aide fournie par la Banque l Principales mensajes de la evaluación • Los países de ingreso bajo en dificultades (PIBD) experimentan algunas de las dificultades más sev- eras en materia de desarrollo, y la comunidad de los donantes sigue batallando por hallar la man- era más eficaz de ayudarlos. El Banco Mundial ha participado activamente en los debates internacionales de políticas referentes a los PIBD y desde que introdujo la Iniciativa para los PIBD, en 2002, se ha puesto en mejores condiciones operativas para darles respaldo. • Antes de la Iniciativa para los PIBD, la mayor parte de los resultados de los programas del Banco en PIBD se situaban en la gama de lo insatisfactorio. La Iniciativa ha hecho que el Banco preste más atención a los PIBD, pero sería prematuro abrir juicio sobre los resultados. La experiencia en cuanto a ejecución de la Iniciativa ha sido despareja, y la mayoría de los resultados de las pocas estrategias para los países evaluadas por el Grupo de Evaluación Independiente (IEG) indican que los objetivos no se han alcanzado en la medida necesaria. • Al adoptar como objetivo central el fortalecimiento del Estado, el Banco ha hecho de un ámbito de debilidad tradicional (creación de capacidad y gestión de los asuntos públicos) un componente de su enfoque principal en los PIBD. Para que la Iniciativa para los PIBD se centre en el complicado programa de fortalecimiento del Estado es necesario que el Banco establezca claramente sus ám- bitos de ventajas comparativas y el alcance y el contenido del programa. Asimismo, debe identi- ficar enfoques innovadores para lograr mejoras en el historial poco satisfactorio en materia de desarrollo de la capacidad y gestión de los asuntos públicos, así como indicadores de desempeño que permitan medir los resultados en cuanto a fortalecimiento del Estado. • En los tres años transcurridos desde que se implementó el enfoque de la Iniciativa, poco se ha avan- zado en la ejecución de reformas esenciales en la esfera de los recursos humanos relacionadas con número de funcionarios, calidad de sus aptitudes e incentivos para la realización de la labor referente a los PIBD. • Aunque el Banco ha hecho hincapié recientemente en un necesario incremento de su presencia in situ en los PIBD, ello no basta para una eficaz ejecución de estrategias para los países. Se re- quiere, como complemento, una más sólida comunicación entre el personal del Banco in situ y en la sede, así como un número adecuado de funcionarios in situ poseedores de las facultades y ap- titudes adecuadas. • Podrían ser más selectivos los objetivos de reforma de los donantes en los PIBD. En los complica- dos entornos de los PIBD, en que se requieren reformas prácticamente en todos los sectores, es esencial establecer una apropiada secuencia de reformas y disponer de tiempo suficiente para ll- evarlas a cabo, a fin de lograr resultados sin hacer recaer una carga abrumadora sobre la escasa capacidad de los PIBD. Resumen os países de ingreso bajo en dificultades (hasta recientementa conoci- L dos en el Banco Mundial como PIBD), poblados por casi 500 millones de personas, aproximadamente la mitad de los cuales ganan menos de un dólar por día, suscitan cada vez más atención. Es causa de creciente pre- ocupación la posibilidad de que esos países no logren alcanzar sus objetivos de desarrollo, así como los efectos económicos desfavorables que provoca su situación en países vecinos y la consiguiente posibilidad de que sus proble- mas se propaguen al resto del mundo. Dada la multiplicidad de problemas crónicos que ayudarlos. Dadas sus diferencias en cuanto a padecen, los PIBD plantean algunos de los propósitos y objetivos, los donantes y los desafíos más arduos en materia de desarrollo. En investigadores han optado por abordar la mayor parte de ellos, la gestión de los asuntos diferentes aspectos de esos problemas, lo que públicos es insatisfactoria, y esos países están los ha llevado a centrar la atención en grupos inmersos en amplios conflictos internos o se de países levemente distintos. Por ejemplo, en están debatiendo en el proceso de culminación estudios recientes del Centro para el de precarias transiciones en situaciones posteri- Desarrollo Mundial se centra la atención en los ores a conflictos. Todos tienen ante sí obstáculos países de ingreso bajo en situación de similares: generalizada falta de seguridad, estancamiento (definida por el producto fractura de las relaciones sociales, significativos nacional bruto per cápita y las tasas de niveles de corrupción, desintegración del Estado crecimiento económico), y, en el Índice de de derecho y falta de mecanismos de generación Estados fallidos de Foreign Policy, la atención de poder y autoridad legítimos, una enorme está centrada en las fallas del Estado y se identi- acumulación de necesidades de inversión insatis- fica a los países según factores tales como nivel fechas, y limitados recursos públicos para el de deterioro económico, seguridad, sectores desarrollo. En general, la labor internacional sociales predominantes organizados como llevada a cabo en el pasado en relación con esos facciones, personas desplazadas, violación de países no ha generado mejoras significativas. derechos humanos e intervención externa. La La comunidad de los donantes sigue Agencia de los Estados Unidos para el batallando por hallar la manera más eficaz de Desarrollo Internacional procura hacer frente a liii E N G A G I N G W I T H F R A G I L E S T AT E S problemas relativos a vulnerabilidad y crisis, presupuestos administrativos para los PIBD se muchos de los cuales pertenecen al entorno incrementó de aproximadamente US$104 político. Las definiciones de Estados frágiles millones en los ejercicios de 2000–02 a alrede- adoptadas por el Departamento del Reino dor de US$161 millones en los de 2003–05. En Unido de Desarrollo Internacional (DFID) y cifras per cápita, dichos presupuestos oscilaron por el Comité de Asistencia para el Desarrollo entre US$0,002 a US$4,5 en los ejercicios de de la Organización de Cooperación y 2003–05. Desarrollo Económicos (CAD de la OCDE) son Gran parte del financiamiento otorgado a los similares a las utilizadas por el Banco Mundial. PIBD en 2003-05 se destinó a PIBD en Tal como los define el Banco Mundial, todos situaciones posteriores a conflictos (los países los PIBD se caracterizan por lo inadecuado de de esa categoría se identifican en función de sus políticas, instituciones y gestión de sus indicadores de progreso en situaciones posteri- asuntos públicos. El Banco ha utilizado dos ores a conflictos, a los efectos determinar las criterios para definir PIBD básicos y PIBD en donaciones excepcionales de la AIF), en tanto situación grave (de aquí en adelante, por “PIBD” que los presupuestos administrativos se han se entiende los de esas dos categorías, y no los distribuido en forma más pareja entre los distin- marginales, que el Banco identifica tan sólo con tos grupos de PIBD (siete PIBD en situaciones fines de control): ingreso per cápita compren- posteriores a conflictos de los 25 considerados dido en el rango de admisibilidad de la recibieron un 64% del total del financiamiento Asociación Internacional de Fomento (AIF), y un para los PIBD y un 34% del total del presupuesto desempeño de 3,0 puntos o menos (2,5 o administrativo para este tipo de países). menos en el caso de los PIBD en situación grave Si bien la elevada proporción de y entre 2,6 y 3,0 para la categoría básica) según la financiamiento para PIBD en situaciones posteri- calificación global de la Evaluación de las políti- ores a conflictos pudo haberse alcanzado aunque cas e instituciones nacionales (CPIA, por su sigla no se hubiera creado la Iniciativa para los PIBD, en inglés) y la calificación de la CPIA referente a dado que las asignaciones excepcionales de la gestión e instituciones del sector público. AIF para países en situaciones posteriores a También se incluyen algunos países de bajo conflictos anteceden a la Iniciativa, es probable ingreso sin datos de la CPIA. En el ejercicio de que ésta haya contribuido a una distribución más 2005, el Banco identificó 25 países como PIBD. pareja de los recursos del presupuesto adminis- Seis PIBD identificados en el ejercicio de 2005 trativo dentro del grupo (dado el incremento de —Afganistán, Liberia, Myanmar, Somalia, no menos del 400% registrado por los Timor-Leste y el territorio de Kosovo— presupuestos administrativos entre los ejercicios carecían de puntajes de la CPIA. El presente de 2000-02 y de 2003-05 en relación con tres examen basa su evaluación en la asistencia PIBD —Liberia, Somalia y Sudán— que habrían otorgada por el Banco a los 25 países clasifica- recibido recursos administrativos ínfimos antes dos como PIBD en el ejercicio de 2005. de la Iniciativa, dada su situación de excluidos El monto de los presupuestos para del régimen de contabilidad en valores devenga- otorgamiento de préstamos y administrativo dos). destinados a los PIBD se ha incrementado desde El enfoque de la Iniciativa para los PIBD del la puesta en marcha de la Iniciativa. El Banco ha variado en relación con su estructura financiamiento otorgado a los PIBD aumentó de original, de 2002, que se basaba en seis princip- alrededor de US$2.500 millones en los ejercicios ios fundamentales (véase el Cuadro R.1). El de 2000–02 (antes de la Iniciativa para los PIBD) fundamento original de la Iniciativa consistía en a alrededor de US$4.100 millones en los ejerci- dar más eficacia a la ayuda mediante la cios de 2003–05 (a partir de la puesta en marcha utilización de otros instrumentos, como la labor de la Iniciativa). En cifras per cápita, dicho de análisis y las transferencias de conocimien- financiamiento osciló entre US$0 y US$25,4 en tos, complementados por transferencias los ejercicios de 2003–05. El monto de los financieras para promover reformas. liv RESUMEN En 2005 los objetivos y el alcance de la Inicia- vista operativo los criterios utilizados por el tiva para los PIBD pasaron de una eficacia Banco para identificar y clasificar a los PIBD, y general de la ayuda a objetivos de fortalec- en qué medida es útil el sistema de asignación imiento del Estado y consecución de la paz. En de la asistencia que se les brinda? la Iniciativa se introdujeron asimismo cuatro • ¿En qué medida ha sido apropiado y adecuado modelos (deterioro, crisis políticas o el respaldo interno otorgado por el Banco a la estancamiento prolongados, situaciones labor referente a los PIBD? posteriores a conflictos o transición política, y mejora gradual) que dieron lugar a la aplicación Principales constataciones y de un tratamiento diferente a distintos tipos de conclusiones PIBD. En 2005, se mantuvieron y se reforzaron los conceptos de aprendizaje práctico y atención Eficacia del enfoque del Banco para los PIBD a las cuestiones de organización contenidos en La experiencia obtenida en la aplicación de los el enfoque de 2002. principios básicos de la Iniciativa para los PIBD a A través del presente examen se procura dar nivel de países ha sido heterogénea (véase el respuesta a tres preguntas: Cuadro R.1). Los problemas de aplicación experimentados obedecen, en algunos casos, a • ¿En qué medida ha sido eficaz el enfoque del la adopción de objetivos excesivamente Banco para los PIBD? ambiciosos por parte del Banco (se requieren • ¿En qué medida son útiles desde el punto de objetivos más modestos) y, en otros casos, a que Cuadro R.1: Experiencia en materia de aplicación de los principios básicos de la Iniciativa para los PIBD en los países Calificación de la experiencia Principio de la Iniciativa para los PIBD en materia de aplicación Perseverancia en la participación Considerable Estrategias ancladas en un mejor análisis sociopolítico Mediana • Comprensión de la situación política • Mediana-considerable • Asimilación de esa comprensión de la situación política en el diseño y la • Mediana-baja ejecución de la estrategia Promoción de la demanda interna y la capacidad de realizar reformas positivas Baja Reformas sencillas y factibles en el nivel de ingreso Mediana-baja • Reformas macroeconómicas • Considerable • Creación de infraestructura física • Considerable • Transición de la fase de reconstrucción inmediata en situaciones posteriores a conflictos • Baja a la fase de desarrollo • Selectividad y priorización • Baja Examen de mecanismos innovadores para la prestación de servicios sociales Mediana Colaboración de los donantes Mediana • A nivel de políticas internacionales • Considerable • A nivel de países • Mediana-baja Medición y seguimiento de los resultadosa Baja Fuentes: Labor sobre el terreno y análisis temático de antecedentes realizados para este examen, IEG, 2005. a. El Banco no lo menciona específicamente como principio básico autónomo, pero el IEG lo incluyó porque es un puntal de los objetivos de aprendizaje práctico de la Iniciativa para los PIBD. lv E N G A G I N G W I T H F R A G I L E S T AT E S los esfuerzos del Banco o los aportes no han tades. La participación inicial del Banco en sido apropiados, como lo indican la labor sobre relación con algunos PIBD no siempre ha sido el terreno del IEG y las revisiones de los seguida por un programa de reforma debida- informes de terminación de las estrategias de mente focalizado y dotado de una adecuada asistencia a los países (deben intensificarse los secuencia de etapas. Además el Banco aún no ha esfuerzos). asimilado suficientes conocimientos políticos en La mayoría de los entrevistados en la el diseño y la ejecución de estrategias de países. encuesta de interesados realizada por el IEG También es necesario que la institución dijeron que la contribución positiva al desarrollo coordine mejor sus actividades en los países con que implica el programa global del Banco para otros donantes, especialmente en cuanto al los PIBD ha sido escasa, opinión que se refiere al seguimiento de todas las etapas de la ejecución, respaldo otorgado por el Banco en general, no que van más allá de los acuerdos de políticas. al enfoque mismo de la Iniciativa para los PIBD. Además el Banco ha hecho de un ámbito de En relación con los principios de la Iniciativa debilidad tradicional (creación de capacidad y para los PIBD, se han logrado algunos notables gestión de los asuntos públicos) un aspecto éxitos iniciales. La Iniciativa ha hecho posible cardinal de su enfoque, al adoptar el objetivo de una mayor participación del Banco en países en fortalecimiento del Estado, que es más que probablemente habría sido menor. Recien- complejo. Para ello debe identificar con mayor temente, el Banco atendió la situación de varios precisión su ventaja comparativa; mejorar su PIBD desde los primeros días de paz o de transi- desempeño, incluso a través de la creación de ción política, y ha contribuido a la estabilidad enfoques innovadores, e identificar asociados macroeconómica y al suministro de un volumen que puedan complementar su labor para hacer significativo de infraestructura física, especial- efectivo el logro de los resultados deseados. mente en PIBD en situaciones posteriores a Finalmente, el propio término fortalecimiento conflictos. Se han logrado avances sustanciales del Estado puede ser inapropiado, dadas sus en materia de coordinación de los donantes a connotaciones políticas e ideológicas. nivel de políticas internacionales, como lo pone Es necesario que el Banco elabore sus de manifiesto la reciente aceptación, por un enfoques operacionales en el marco de la Inicia- amplio espectro de donantes, incluido el Banco, tiva para los PIBD, especialmente en relación de los 12 principios de participación interna- con los modelos de deterioro y crisis o cional del CAD del OCDE. estancamiento prolongados. Es necesario, El Banco ha cumplido repetidamente un asimismo, perfeccionar aún más los modelos, papel orientador como copresidente de teniendo en cuenta en forma más explícita las reuniones internacionales de donantes y diferencias de capacidad para el cumplimiento coautor de documentos de políticas conjuntos. de funciones estatales fundamentales (por Recientemente introdujo modelos en que se ejemplo, generación de recursos, asignación de establecen diferencias entre distintos tipos de recursos, suministro de infraestructura y PIBD, lo que probablemente permita atender servicios sociales básicos, y admisión de la mejor las necesidades específicas de los PIBD. El discrepancia y la seguridad en la esfera política), porcentaje de proyectos de PIBD cerrados y para que el Banco pueda hacer coincidir mejor clasificados por el IEG como satisfactorios desde sus enfoques operacionales con los diversos el punto de vista de los resultados aumentó entornos institucionales de los PIBD. desde el 50% en el ejercicio de 2002, antes de La labor del Banco en los países en que se estableciera la Iniciativa para los PIBD, situaciones posteriores a conflictos antecede al hasta el 58% en 2003, 65% en 2004 y 82% en enfoque de la Iniciativa para los PIBD, por lo cual 2005. Las cifras correspondientes de proyectos el modelo que corresponde a los PIBD en ejecutados en países de ingreso bajo que no situaciones posteriores a conflictos tiene una eran PIBD oscilaron entre 70% y 79%. estructura más diáfana que la de los restantes No obstante, subsisten considerables dificul- modelos. No obstante, presenta imperfecciones lvi RESUMEN y es necesario desarrollarlo mejor para orientar algunos PIBD en situaciones posteriores a las fases de transición y desarrollo que siguen a conflictos y a PIBD que realizan transiciones la fase de reconstrucción inmediata posterior a políticas. No obstante, dista mucho de ser claro conflictos. Por otra parte, aunque el Banco ha si los actuales niveles de financiamiento de la venido prestando cada vez mayor atención a la AIF permiten afirmar que los PIBD no reciben prevención de conflictos, poco se sabe sobre la menos ni más ayuda de la necesaria. eficacia de los esfuerzos realizados en ese La cuestión de la asignación de la ayuda ha ámbito. vuelto a primer plano, y en algunos estudios se Aún no se han establecido claramente el ponen en tela de juicio las pruebas empíricas papel y la ventaja comparativa del Banco en referentes a la existencia de un vínculo positivo materia de prevención de conflictos, en especial entre políticas y eficacia de la ayuda (en que se porque, para cumplir esa labor, la institución basa el sistema de asignación basado en el debe prestar mayor atención a las causas desempeño). En otros estudios se sostiene que profundas de los conflictos y tener en cuenta la ayuda puede promover un cambio sostenible factores étnicos, sociológicos y políticos. El de las políticas en los Estados en descomposi- Banco debe definir mejor el contenido del ción, al crear y reforzar las condiciones previas objetivo de consecución de la paz y lo que es de la reforma o al aumentar la probabilidad de ajeno a éste, así como la manera de alcanzarlo. que ésta, una vez establecida, se mantenga. En esos últimos estudios se concluye que conceder Utilidad operativa de los mecanismos de asistencia a los PIBD puede ser extremadamente identificación, clasificación y asignación de beneficioso, aunque los riesgos de fracaso sean asistencia para los PIBD que aplica el Banco sustanciales. A pesar del énfasis en los objetivos de fortalec- Por su parte, el Banco aún no ha abordado la imiento del Estado y consecución de la paz, el cuestión de la asignación de la ayuda para los Banco sigue basándose casi exclusivamente en PIBD a través de un método que refleje los la CPIA para identificar a los PIBD. Esas evalua- objetivos correspondientes a esos países e ciones, sin embargo, no permiten captar impida que esa asistencia sea insuficiente o algunos aspectos esenciales de fragilidad del excesiva. Estado (como la admisión de las discrepancias políticas) y conflictos (por ejemplo, inestabili- Pertinencia y suficiencia del respaldo interno dad política y seguridad, o susceptibilidad a del Banco a la labor referente a los PIBD conflictos), lo que puede requerir mecanismos El respaldo interno otorgado por el Banco a la complementarios. Para reforzar el método de labor referente a los PIBD ha avanzado en varios identificación de los PIBD se requeriría un ámbitos: marco analítico centrado en forma más explícita en los objetivos de la Iniciativa para los PIBD. • Ampliación de la labor de análisis, gracias a A lo largo de los años, el sistema utilizado por que se desvinculan, de los volúmenes de fi- el Banco para asignar recursos de la AIF nanciamiento, los presupuestos administra- (denominado asignación basada en el tivos destinados a financiar los estudios desempeño) se ha ido haciendo más selectivo económicos y sectoriales y la asistencia téc- con respecto a las políticas, y se ha reducido el nica financiamiento de la AIF disponible para países • Utilización de las notas provisionales sobre la con políticas e instituciones más débiles y una estrategia, que permiten diseñar estrategias gestión pública menos satisfactoria. Esto ha que abarcan un período más breve para dar llevado a preguntarse si los PIBD están recibi- cabida a las condiciones inestables de los PIBD endo de la AIF un volumen apropiado de • Acceso a la administración superior del Banco asistencia. Los ajustes a la asignación basada en para los administradores de la Iniciativa el desempeño han dado lugar a un aumento del • Creación de un Fondo Fiduciario de los PIBD financiamiento de la AIF, incluido el otorgado a para otorgar financiamiento a países excluidos lvii E N G A G I N G W I T H F R A G I L E S T AT E S del régimen de contabilidad en valores deven- dad de reformar el sistema de incentivos. gados (en relación con los cuales el Banco El personal señaló como un problema el carecía anteriormente de instrumentos). desigual nivel de atención que prestan al tema De la experiencia en los países, la Unidad de los diferentes directores a cargo de países, en PIBD ha extraído orientaciones sobre algunos especial cuando se ocupan de un país mayor, temas importantes y las ha hecho gravitar en el más “exitoso” o más destacado. Puso también de asesoramiento operativo a los grupos a cargo de manifiesto la necesidad de que, en toda la los países y en más amplios debates externos jerarquía de la administración se preste atención sobre políticas. sistemática a la labor referente a los PIBD. En la No obstante, tres años después que el Banco mayor parte de las respuestas del personal del reconociera la necesidad de una modificación Banco a la encuesta de interesados realizada por cultural interna tendiente a una eficaz aplicación el IEG, se señala que trabajar en el área de los del enfoque de PIBD, poco se ha logrado en PIBD no ha producido cambio alguno en cuanto al otorgamiento de respaldo interno del relación con varias cuestiones de recursos Banco para la labor de la Iniciativa. Sigue siendo humanos (véase el gráfico R.1). insatisfactorio el avance de reformas esenciales Existe considerable duplicación de esfuer- sobre recursos humanos relativas a número de zos y confusión en torno a las funciones y funcionarios, calidad de sus aptitudes e responsabilidades de la Unidad de PIBD y la incentivos para la realización de la labor relativa Unidad de prevención de conflictos y de a los PIBD. Los comentarios del personal del reconstrucción. Preocupan al personal las Banco sobre la importancia que reviste trabajar cuestiones prácticas de saber a qué unidad hay tanto en países pertenecientes a la Iniciativa que acudir para obtener determinados tipos de como en otros países revelan que dentro del asesoramiento y qué tipos de apoyo cabe Banco no se reconoce en la medida apropiada la esperar de cada unidad. En la encuesta de labor referente a los PIBD, e indican la necesi- interesados del IEG, alrededor de los dos Gráfico R.1: En la mayor parte de las respuestas del personal del Banco se sostiene trabajar en el área de los PIBD no produjo cambio alguno en varias esferas (que a continuación se mencionan) 100 75 Porcentaje 50 25 0 Perspectivas Compensación Realismo en las Nivel de Esfuerzos del Banco profesionales financiera expectativas de respaldo de la para garantizar la la administración administración seguridad e integridad física de las personas Mejoró Fue igual Empeoró Fuente: Apéndice Z (Resultados de la encuesta de interesados). Nota: El número de respuestas válidas está comprendido entre 213 y 238. En la pregunta formulada en la encuesta no se distinguió entre los funcionarios que, en su cometido anterior, habían trabajado con un PIBD y los que lo habían hecho con un país que no pertenecía a dicha categoría. lviii RESUMEN tercios de los funcionarios del Banco que presupuestaria potencialmente desastrosa. En respondieron mencionó algún problema de la Haití, la comunidad de los donantes parece no actual estructura institucional: un 37% señaló haber prestado adecuada atención al logro de cierta duplicación de esfuerzos entre el un nivel mínimo de seguridad. En ambos respaldo de la Unidad de PIBD y el de la Unidad casos, los resultados iniciales satisfactorios de de prevención de conflictos y de reconstruc- la Iniciativa para los PIBD corren el riesgo de ción del Banco; el 15% mencionó un alto grado deteriorarse. de duplicación y el 12%, la existencia de conflic- En ciertos casos puede requerirse, al menos tos o contradicciones. por ciertos períodos, una salida estratégica, salvo en lo referente a la labor interna de Enseñanzas de la experiencia para análisis. Se trata de una estrategia especialmente Banco y otros donantes apropiada cuando se considera que una labor en De este examen de la experiencia recogida por común con el Banco confiere, inadecuada- el Banco en la aplicación de los principios mente, legitimidad al gobierno de un PIBD o básicos del enfoque de la Iniciativa para los PIBD reduce la presión interna en procura de se extraen varias enseñanzas. Muchos de los reformas y, por lo tanto, dificulta el surgimiento problemas a los que se refieren esas enseñanzas de las condiciones necesarias para suscitar se percibieron como ámbitos que requerían reformas políticas serias y sostenibles. mejoras en el Informe del grupo de estudio de En los modelos de deterioro y crisis o los PIBD de 2002 (Banco Mundial 2002) —por estancamiento prolongados, en que suele haber ejemplo, la necesidad de anclar las estrategias limitado consenso entre los donantes y el en mejores análisis sociopolíticos o respaldar gobierno con respecto a la estrategia de objetivos de reforma muy focalizados— y desarrollo, es necesario que la participación también se hizo hincapié en ellos en los comprenda un diálogo de políticas encaminado informes del Banco de 2005 sobre los PIBD. Las a dar cabida a la reforma a la vez que se trabaja enseñanzas emanan de la experiencia del Banco en un programa de reforma, por si surgiera una en materia de ejecución, pero también pueden ventana de oportunidad. En los modelos de ser útiles para orientar otra asistencia de situación posterior a conflictos o transición donantes en países de la Iniciativa. política y mejora gradual, la participación deberá tener un carácter más pronunciadamente Presencia en los PIBD técnico y centrarse en mayor medida en la ejecución del programa de reforma, dado el Perseverar en la participación no es más que un mayor consenso entre los donantes y el medio encaminado a un fin, por lo cual, en la Inicia- gobierno al respecto. tiva para los PIBD, esa presencia debe ser seguida La directriz del Banco para relacionarse con sin tardanza por un programa de reforma claro y países con conflictos o impases políticos prolon- pertinente. Si no se cuenta con un programa de gados establece que “temas de desarrollo relati- reforma claro y pertinente, los éxitos iniciales vamente no polémicos pueden ser un punto de suscitados por la presencia de los donantes partida para un diálogo constructivo entre las pueden ser fugaces y no contribuir más que en partes en conflicto”. Para aquellos países donde escasa medida al logro de los objetivos de la hay un deterioro en su gobernabilidad, la estrategia para el país. Los ejemplos de la directriz establece que el Banco debe aportar República Centroafricana y Haití muestran que “ideas sobre cuestiones económicas específicas varios obstáculos pueden dificultar la contin- que sean importantes para la mediación y que uación de una participación inicial exitosa en puedan ayudar a reiniciar el diálogo” (Banco un PIBD. Como los éxitos políticos no Mundial 2005e). contaron con suficiente respaldo económico, el Gobierno de la República Centroafricana se Los problemas de identificación y capacidad de ve ahora confrontado con una crisis absorción de los países afectan tanto a los productos lix E N G A G I N G W I T H F R A G I L E S T AT E S del conocimiento como a los productos financieros. parecen excesivamente ambiciosos, dada la La participación de contrapartes nacionales en escasa capacidad institucional del país. la labor de análisis del Banco sigue estando limitada a los aspectos administrativos; mucho Conocimiento de la situación política y su menor es la participación de los países clientes utilización en la estrategia para el país en la selección de temas y la realización de análisis, lo que va en detrimento del interés de Encargar y aprovechar —no necesariamente los países en el programa. No obstante, la partic- producir— análisis políticos acertados es esencial ipación de contrapartes nacionales es esencial para los donantes en los PIBD. El objetivo de un para lograr la identificación de los clientes y grupo a cargo de un país debería consistir en mejorar el impacto de la labor de análisis. encargar o utilizar (no necesariamente En Tayikistán, la falta de participación del producir) análisis directamente pertinentes para gobierno en la selección y preparación de la la preparación de una estrategia, que puedan labor de análisis del Banco redujo el interés del utilizarse en el curso de esa labor. En los PIBD, gobierno en los resultados, lo que fue en especialmente en entornos en que sea esencial detrimento de una eficaz ejecución. En Angola, actuar con celeridad, los donantes, antes de altas autoridades públicas percibieron ciertos encomendar nuevos análisis políticos, deben aspectos de la labor de análisis orientada por el comprobar que se aprovechan los existentes. Banco (por ejemplo, el reciente memorando En la República Democrática Popular Lao, el económico sobre el país) como una imposición Banco aprovechó eficazmente análisis políticos de los puntos de vista del Banco en los asuntos existentes e invitó a un especialista en Ciencia internos de su país, lo que fue en detrimento de Política que había realizado amplias publica- la identificación del país y del desarrollo de su ciones sobre Laos a efectuar una exposición capacidad. Sin esa identificación, es poco sobre política y reforma ante el equipo a cargo probable que la labor de análisis influya sobre la del país. Esto permitió preparar un resumen política pública. independiente de análisis políticos pertinentes Las dificultades de capacidad de absorción de (adaptados a las necesidades de la comunidad los gobiernos de los PIBD a los efectos de utilizar de los donantes en general y del Banco en la labor de análisis pueden también reducir una particular) y difundirlo entre un grupo posible transferencia de conocimientos. El pertinente de funcionarios del Banco y otros gobierno angoleño, por ejemplo, se manifestó donantes. Se evitó así el elevado costo que de acuerdo con la nota provisional sobre la suponía preparar un análisis “del Banco”, así estrategia del Banco, pero expresó preocu- como una posible tensión con el gobierno, dado pación con respecto al volumen de actividades que permitió evitar que el Banco se viera de análisis y asesoramiento previstos, lo que trabado por algunos de los problemas de generó dudas acerca de la plena utilización de susceptibilidad que rodeaban el análisis. Para el los productos analíticos por parte del gobierno. Banco, en este caso, la adquisición de La capacidad de absorción del gobierno está conocimientos existentes, así como su difusión, sujeta a severos límites, y las actividades de resultaron más importantes y eficaces que la análisis y asesoramiento realizadas principal- creación de conocimientos. mente por el Banco crean el riesgo de generar tensiones en las relaciones con el gobierno, sea Los esfuerzos de los donantes deben centrarse en cual fuere su calidad técnica. En Camboya, los ayudar al personal a asimilar análisis políticos al planes de prestación de servicios de análisis y diseñar y aplicar estrategias. Aunque en algunos asesoramiento en el marco de la Estrategia de PIBD el Banco ha realizado o ha tenido acceso a asistencia al país (EAP) de 2005 —con un total acertados análisis políticos, éstos no se han de 30 tareas que han de completarse a lo largo reflejado adecuadamente en la estrategia de la de los ejercicios económicos de 2005-07— institución. Por ejemplo, la estrategia provisional lx RESUMEN para Papua Nueva Guinea contiene un adecuado pero en conjunto constituyen un programa de análisis del sistema político y en ella se recono- enormes proporciones. cen problemas tales como lealtades de clanes, Con respecto al programa de asistencia del clientelismo político, corrupción y falta de propio Banco, en la revisión del informe de capacidad. No obstante, la estrategia trata esos terminación de la estrategia de asistencia al país problemas como cuestiones técnicas; no las realizado por el IEG con respecto a los ejercicios utiliza adecuadamente como pilares del enfoque 2001–05, referente a Santo Tomé y Príncipe, se global. concluyó que el Banco era excesivamente ambicioso en relación con los recursos asigna- Programa de reforma centrado en objetivos dos al país. Muchos de los objetivos de la EAP siguieron sin alcanzarse, o sólo se lograron en En entornos complejos de PIBD, en que prácticamente parte. todos los sectores requieren reformas, es esencial Aparte de la cuestión de la selectividad en las establecer una adecuada secuencia de reformas y EAP, es esencial centrar y priorizar adecuada- disponer de tiempo suficiente para aplicarlas, a fin de mente los programas actuales de reforma sobre lograr resultados sin hacer recaer una carga el terreno. Sin estos pasos, surgen dudas acerca abrumadora sobre la escasa capacidad de los PIBD. de la eficacia, en especial porque la capacidad de Aunque los donantes deben esforzarse en lograr los PIBD es limitada. Aunque es difícil actuar una selectividad colectiva, esto dista mucho de selectivamente en un país que padece la urgente haberse alcanzado, como lo indican los ejemplos necesidad de arreglar muchas cosas, es esencial del programa de reforma de Afganistán, respal- diseñar una adecuada secuencia de reformas, dado por donantes, y el Marco de Cooperación para no gravar excesivamente la reducida capaci- Provisional de Haití, que más abajo aparecen. No dad de los PIBD y evitar soluciones parciales. obstante, aunque no se logre de inmediato dicho Será fundamental establecer reformas en una objetivo, es preciso que el Banco, basándose en secuencia adecuada, que abarquen un número sus competencias esenciales, aplique un enfoque suficiente de años, y lograr que los donantes se apropiado y el principio de selectividad en su comprometan a seguir su evolución. propio programa de asistencia. En los últimos En Timor-Leste, quizá los donantes se años, esa selectividad del Banco ha ido en retiraron antes de tiempo, sin ocuparse en aumento, pero la dificultad persiste. medida suficiente de las apremiantes necesi- En Afganistán, las reformas realizadas por dades de capacidad que experimenta el país. En donantes ocupan una amplia gama; sus priori- Haití, la asistencia para el desarrollo ha fluctu- dades no se han desarrollado en forma ado pronunciadamente a lo largo de los años. El suficiente y han dado lugar a 120 proyectos de país ha atravesado varios ciclos “de abundancia leyes. Es necesario establecer cuidadosamente o hambruna” en sus relaciones con la el orden de prelación y la secuencia de esas comunidad de los donantes, lo que podría reformas, que abarcan prácticamente todos los haberse evitado si diversos donantes hubieran aspectos económicos y sociales del país, pero establecido un adecuado cronograma y secuen- los donantes aún no lo han hecho. En Haití, el cia para su ayuda. Marco de Cooperación Provisional está destinado a orientar la asistencia y cooperación Desarrollo de la capacidad en PIBD en internacionales con ese país hasta septiembre situaciones posteriores a conflictos de 2006. El programa comprende prácticamente todas las funciones estatales básicas, tales como La ejecución de los programas de desarrollo de la seguridad, diálogo nacional, buen gobierno en capacidad y buena gestión de los asuntos públicos la esfera económica, recuperación económica y debe iniciarse tempranamente, aun en PIBD en servicios básicos. Cada uno de esos ámbitos, situaciones posteriores a conflictos. Inmediata- tomados separadamente, parece importante, mente después de la cesación de un conflicto, la lxi E N G A G I N G W I T H F R A G I L E S T AT E S comunidad internacional de los donantes tiende donantes, análogamente, tengan un contacto a centrar su asistencia en obras físicas de más estrecho con el Banco y subordinen sus reconstrucción. Como los PIBD en situaciones programas bilaterales a objetivos multilaterales posteriores a conflictos tienen escasa capacidad adoptados de común acuerdo. de utilizar eficazmente la ayuda, y como la La coordinación debe iniciarse dentro de cada gestión pública suele ser insatisfactoria, es entidad donante. No basta con coordinar la labor necesario centrar la atención, desde el inicio, en de las entidades multilaterales y bilaterales la creación de capacidad y una mejor gestión de donantes. Se trata también de una cuestión los asuntos públicos, y no tan sólo en la esencial dentro de cada uno de los organismos reconstrucción de la infraestructura física. Para donantes. Los proyectos realizados en diferentes ello puede ser necesario crear o reforzar institu- sectores de un mismo país suelen operar en ciones públicas, realizar la reforma del servicio forma paralela, sin aprovechar sinergias. Así civil y recurrir a expertos técnicos locales. Si se sucedió en Timor-Leste con los proyectos agríco- utilizan los servicios de expertos extranjeros las y de potenciación comunitaria del Banco. para obtener asistencia técnica, ello no debe Un efecto secundario de los esfuerzos de comprometer el desarrollo a largo plazo de la descentralización del Banco hacia sus oficinas en capacidad local. los países ha sido la concentración de los conocimientos sobre los países en los funcionar- Coordinación de los donantes ios locales y una inadecuada difusión de ese saber entre los grupos a cargo de los países, en No puede lograrse la coordinación de los donantes especial los que tienen su centro de operaciones si éstos no adoptan una visión común y persiguen en la ciudad de Washington. Hacer frente a los una misma finalidad; cuando los objetivos de los problemas de coordinación entre los diversos donantes no puedan armonizarse plenamente, es departamentos de las entidades donantes (por importante que, por lo menos, sean comple- ejemplo entre los departamentos del Banco que mentarios. En su enfoque, el Banco no ha recon- se ocupan de la gestión del sector público, la ocido plenamente las diferentes razones que prevención de conflictos y la reconstrucción, los llevan a los donantes a ocuparse de los PIBD. PIBD, la creación de capacidad y la investigación) Aunque el amplio concepto de fragilidad se reviste especial importancia en los PIBD, cuyos conoce y acepta en forma generalizada, varían problemas, complejos y generalizados, suelen los criterios de los donantes para calificar como requerir soluciones multisectoriales. frágil la situación un país. Se otorga respaldo a los Estados frágiles por diversos motivos: seguri- Medición y seguimiento de resultados dad, eficacia de la ayuda, desarrollo equitativo, reducción de la pobreza, fortalecimiento del El seguimiento y la evaluación no son menos Estado, consecución de la paz, prevención de importantes en el caso de los PIBD que en el de conflictos, etc. cualquier otro país. Varias razones hacen esencial Tanto en Afganistán como en Tayikistán, el la labor de seguimiento y evaluación en los IEG concluyó, a través de su labor sobre el PIBD: terreno, que los principales donantes no perseguían un objetivo único y claro. A falta de • Primero, el Banco, al igual que otros donantes, un objetivo global común, es improbable lograr aún se encuentra en el proceso de compren- coherencia de políticas. En los esfuerzos y der qué enfoques son apropiados en contex- modalidades de coordinación de los donantes tos de PIBD. Es esencial realizar un seguimiento del Banco, no se tienen en cuenta en medida de las experiencias para extraer enseñanzas, y suficiente los objetivos de los distintos protago- el aprendizaje y el intercambio de informa- nistas que actúan en un país. Pero la coordi- ción deben convertirse en un componente nación de los donantes es una modalidad de más destacado de la labor referente a los PIBD. acción colectiva que requiere que otros • Segundo, como en esos países el avance suele lxii RESUMEN ser lento, es importante evaluar continuamente de las entidades donantes participantes asimilen la cuestión de si el programa sigue estando en- los resultados de los análisis y apliquen las caminado hacia el logro de los resultados enseñanzas emanadas de ellos a todas las deseados. intervenciones. En Camboya, por ejemplo, la • Tercero, un entorno de PIBD constantemente presencia del Banco sobre el terreno ha cambiante e inestable, en que el avance suele permitido mejorar significativamente el no ser lineal, hace esencial la adaptación del conocimiento de la situación política. Pero los programa, por lo cual un estrecho seguimiento debates con los miembros del grupo a cargo del del desempeño contribuirá a establecer cuándo país y otros participantes indican que ese se requiere adaptación y qué tipo de adaptación conocimiento quizá sigue estando sumamente es necesaria. concentrado en unos pocos administradores y funcionarios (en su mayor parte en la oficina en Sólo a través de una firme labor de el país y en el centro de actividades de seguimiento y evaluación puede lograrse un Bangkok), y que es relativamente limitada la aprendizaje práctico eficaz para aumentar la difusión en resto del grupo a cargo del país. eficacia de la futura labor del Banco en los PIBD. El problema parece haber variado: de un El Banco ha declarado que el fortalecimiento conocimiento parcial de las realidades políticas del Estado y la consecución de la paz deben ser de Camboya a la cuestión de quién posee ese los objetivos utilizados para medir el éxito de la conocimiento dentro del grupo del Banco a Iniciativa para los PIBD. No obstante, el Banco cargo del país y la manera en que se lo utiliza aún no ha identificado indicadores de para orientar la adopción de decisiones en desempeño a esos efectos ni patrones de materia de ejecución de programas y estrategias. medición del desempeño. En un contexto en El hecho de que el conocimiento en profundi- que el cambio está orientado más bien por dad de un país se concentre en tan sólo unos procesos, especialmente en los modelos de pocos funcionarios implica que sólo beneficia a deterioro y crisis o estancamiento prolongados, algunas actividades e intervenciones del Banco. los resultados que pueden esperarse en otros En general se requiere una mayor transferencia modelos quizá no sean patrones de éxito apropi- de conocimientos entre las oficinas de los ados. Los objetivos deberían adaptarse al donantes en los países y su personal de países y contexto especial de cada PIBD, lo que, a su vez, sectores que opera en la sede. determinaría patrones de medición y garanti- Pese a su costo, las oficinas fuera de la sede zaría que los valores para medir el éxito se deben estar adecuadamente dotadas de personal establecieran en los niveles apropiados. para que puedan trabar eficaces relaciones con los clientes. En Angola, el inicialmente reducido Mejor respaldo institucional interno para la grupo de funcionarios en el terreno se vio labor referente a los PIBD confrontado con múltiples tareas: desde el diálogo estratégico con gobiernos y donantes La presencia sobre el terreno no basta para una hasta actividades de logística, como el traslado eficaz ejecución de la estrategia para el país; es de la oficina a nuevas instalaciones. La situación preciso complementarla a través de una adecuada se hizo aún más difícil por la falta de personal de comunicación entre el personal en el terreno y el de nivel operativo en la oficina en el país que, en las oficinas centrales de las entidades donantes, y consulta con funcionarios del ministerio, pudiera de un adecuado número de funcionarios en el preparar el terreno antes de que los ministros y terreno, poseedores de las atribuciones y aptitudes el Banco celebraran reuniones de alto nivel. Una apropiadas. La mejor manera de comprender las tramitación demasiado acelerada hacia los circunstancias de los países suele consistir en niveles jerárquicos superiores —provocada por una significativa presencia sobre el terreno, el hecho de que los niveles inferiores carecían de aunque esto de por sí no basta. No menos personal— provocó innecesarias tensiones. Las importante resulta que todos los departamentos decisiones de los donantes con respecto al lxiii E N G A G I N G W I T H F R A G I L E S T AT E S número de funcionarios en cada PIBD deberían gran escala, el personal de los donantes debe reflejar la escala y las características de la partici- poseer sólidas aptitudes de coordinación y pación deseada, a la luz de los objetivos persegui- establecimiento de secuencias de actividades dos por los respectivos donantes en esos países. para organizar a los asociados para el desarrollo Es necesario que el personal en el terreno, y a sus actividades. además de ser suficiente, posea adecuadas En el modelo de mejora gradual, la principal potestades, en relación con la sede, para que no de las aptitudes que requiere el personal todas las decisiones deban ser aprobadas por la consiste en la capacidad de proporcionar sede. Una eficaz presencia sobre el terreno asistencia técnica adaptada a las necesidades del requiere la actuación en el país del tipo de caso y de trabajar en estrecha relación con un funcionarios apropiado. En entrevistas cliente que ya ha emprendido reformas. semiestructuradas realizadas para el presente estudio, varios donantes hicieron hincapié en El intercambio de experiencias —positivas y que las virtudes y defectos de las personas cuya negativas— es esencial para el aprendizaje, pero actividad haya de coordinarse influyen poderosa- requiere respaldo de la administración y un entorno mente sobre la coordinación. Podrían mitigarse institucional favorable. Se puede promover el esos riesgos de idiosincrasia a través de una aprendizaje a través del intercambio de experi- capacitación más apropiada para los funcionarios encias sobre éxitos y fracasos en diferentes a quienes se encomienden difíciles cometidos de situaciones de PIBD. El aprendizaje reviste campo, y de mejores incentivos dentro del Banco especial importancia en la labor relativa a los para que el personal colabore con otros PIBD porque la comunidad de los donantes donantes. sigue batallando por hallar la manera más eficaz En el modelo de deterioro, en que el diálogo de ayudar a esos países, cuya situación con el gobierno puede interrumpirse, el representa un desafío. Aunque el Banco ha personal de las entidades donantes deberá difundido ciertas enseñanzas a través de la serie poseer firmes aptitudes diplomáticas y de de seminarios del grupo de estudio sobre PIBD, persuasión que garanticen que se mantenga es mucho mayor la atención necesaria para abierta la puerta para un diálogo con el evaluar y difundir sistemáticamente experien- gobierno, a la vez que se movilizan grupos no cias emergentes relativas a los PIBD, recogidas gubernamentales, incluida la sociedad civil. por el Banco y otros donantes. En el modelo de crisis o estancamiento Será esencial crear un entorno institucional prolongados, en que los problemas son crónicos más receptivo y lograr el respaldo de la adminis- o se produce una parálisis política, el personal, tración para el intercambio de experiencias entre otras aptitudes, deberá poseer extraordi- negativas. Hasta ahora el Banco parece estar naria paciencia, así como creatividad, con principalmente dispuesto a intercambiar constantes innovaciones para despejar atascos ejemplos positivos, como ocurre en sus persistentes. recientes informes sobre la Iniciativa para los En el modelo de situaciones posteriores a PIBD. conflictos o de transición política, el personal, entre otras cosas, deberá contar con Una comunicación eficaz es esencial para lograr la conocimientos técnicos específicos para aceptación, por parte de los países, de los enfoques desarrollar una infraestructura esencial, institu- de los donantes referentes a PIBD y para atemperar ciones y sistemas económicos sólidos. Además las expectativas poco realistas de los países sobre deberá estar en condiciones de actuar rápida y lo que puede lograrse, en especial inmediatamente enérgicamente en esos contextos, antes de que después que cesa el conflicto. Será preciso dar a se disipe el optimismo que sigue a la paz, y de conocer mejor los objetivos y enfoques de los contribuir a evitar que los países vuelvan a donantes en los PIBD, para que éstos se identi- sumirse en conflictos. Como esas situaciones fiquen con el programa. Con ello se podrá suelen atraer una asistencia internacional en además prevenir desilusiones, moderando las lxiv RESUMEN expectativas poco realistas de los interesados eficiente las causas —y no simplemente los acerca de lo que puede lograrse en determinado síntomas— que hacen que los países se período. conviertan en PIBD o mantengan la condición En los modelos del Banco sobre deterioro y de tales. Una solución que considere las causas crisis o estancamiento prolongados, en que la como datos inmutables puede no tener en situación económica y social en su mayor parte cuenta factores contextuales extremadamente está empeorando o se ha estancado, sería importantes. La orientación operativa de los necesario que a través de la estrategia de donantes debe garantizar que los ámbitos comunicación se dieran a conocer activamente ajenos a la ventaja comparativa de determina- al gobierno y a la sociedad civil los beneficios de dos donantes se dejen en manos de otros, en la reforma. En el modelo del Banco sobre tanto que en su propia labor se tenga en cuenta situaciones posteriores a conflictos o transición y se complemente adecuadamente la labor política, para prevenir la disolución que sigue a realizada por otros. las expectativas faltas de realismo, la estrategia En los modelos de deterioro y crisis o de comunicación debería estar orientada hacia estancamiento prolongados preparados por el la totalidad de la población y establecer expresa- Banco, y en las fases de transición y desarrollo mente lo que han de hacer los donantes, cuándo que siguen a la fase inmediatamente posterior a y cómo deberían hacerlo, y qué resultado cabría la reconstrucción en el modelo de situaciones prever. La estrategia de comunicación posteriores a conflictos o transición política, se correspondiente al modelo de mejoramiento plantean algunos de los mayores desafíos que gradual deberá tener un carácter más informa- tiene ante sí la comunidad de los donantes. Se tivo, y en ella deberían presentarse experiencias trata asimismo de ámbitos en que el pen- comparadas de países o de sectores. samiento innovador ha sido relativamente En muchos casos, inmediatamente después limitado. Existe una apremiante necesidad de de la cesación del conflicto los donantes interna- orientación operativa en varias esferas, incluida cionales, incluido el Banco, han comprometido la de hallar mecanismos para priorizar y estable- un gran volumen de asistencia aunado a cer la secuencia de las reformas y a la vez evitar objetivos excesivamente ambiciosos. Ello ha soluciones parciales; mecanismos que permitan creado a menudo grandes expectativas en la una rápida prestación de servicios que no vaya población y ha conducido a desilusiones cuando en detrimento de la creación de capacidad a esas expectativas han quedado insatisfechas y largo plazo del sector público; mecanismos los resultados tangibles cotidianos han sido tendientes a promover la reconciliación política limitados. Evitar los programas demasiado y al mismo tiempo contribuir a una labor pública ambiciosos y utilizar mejores procedimientos de eficaz y legítima; mecanismos que permitan comunicación son factores esenciales, por lo aprovechar los conocimientos políticos en el cual es necesario que el Banco invierta en diseño y la ejecución de las estrategias, y estrategias de ese género. mecanismos que permitan abordar eficazmente los vínculos entre política, seguridad y Se requiere una mejor orientación operativa para desarrollo. adaptar los enfoques de los donantes a las La orientación reciente del Banco con condiciones especiales de los PIBD. La Iniciativa respecto a los PIBD se concentra más en la para los PIBD ha creado conciencia sobre la determinación de los instrumentos que deben necesidad de tratar en forma diferente a esos utilizarse que en la formulación de enfoques países, pero el Banco y otros donantes aún no operacionales con respecto a lo que es han determinado con precisión la manera de necesario hacer en forma diferente —y la hacerlo. También es preciso establecer con manera de hacerlo— en distintos grupos de mayor certeza la medida en que, en los PIBD. Además, los grupos a cargo de los países enfoques de los donantes frente a los PIBD, de la Iniciativa se beneficiarían de una mayor deben y pueden tenerse en cuenta en forma cantidad de orientaciones expositivas, como la lxv E N G A G I N G W I T H F R A G I L E S T AT E S que se presenta en el capítulo 2 del presente asistencia que reflejen los objetivos del Banco examen, así como notas breves, orientadas por en esos países y hagan que la asistencia que se problemas, en lugar de las notas de orientación les conceda a éstos no sea insuficiente ni más formales, que suelen ser excesivamente excesiva. Es preciso establecer si los criterios concentradas, y en las cuales no se describe en deben basarse en factores distintos de los forma suficiente el contexto del país. resultados de políticas (por ejemplo, los niveles de la asistencia de otros donantes, la evaluación Recomendaciones de potenciales riesgos y premios, y los efectos secundarios regionales y mundiales), y en qué • Establecer claramente el alcance y contenido de medida corresponde hacerlo, teniendo en los objetivos de fortalecimiento del Estado que cuenta que la ayuda es limitada, por lo cual persigue el Banco y reforzar el diseño y el sumin- habrá que adoptar soluciones de compromiso. istro de respaldo en materia de desarrollo de la ca- pacidad y gestión de los asuntos públicos en los • Reforzar el apoyo interno, en el Banco, para la labor PIBD. referente a los PIBD en los próximos tres años. Dado su historial insatisfactorio en materia Es necesario prestar atención a dos aspectos de desarrollo de la capacidad y gestión de los del respaldo interno en el Banco. En primer asuntos públicos, así como el hecho de que lugar debe priorizarse el objetivo de establecer actualmente centra la atención en el objetivo, adecuadas cifras de funcionarios dotados de más ambicioso y complejo, de fortalecer el aptitudes e incentivos apropiados para que se Estado en su labor referente a los PIBD, es ocupen de los PIBD. Para establecer adecuados necesario que el Banco delimite claramente sus incentivos que atraigan a personal calificado — ámbitos de ventajas comparativas en relación en la sede y en las oficinas fuera de la sede— que con los demás donantes y adopte enfoques se ocupe de los PIBD habrá que impartir claras innovadores que garanticen mejores resultados señales sobre lo que se considera éxito en los en materia de capacidad y gestión de los asuntos PIBD, de qué resultados se hará responsable al públicos. Es preciso elaborar enfoques personal, en qué medida es razonable asumir innovadores para establecer una corresponden- riesgos, cómo se juzgarán los fracasos y cómo se cia más estrecha entre las intervenciones del tendrán en cuenta los puntajes de las evalua- Banco y la capacidad de un PIBD de cumplir ciones globales de desempeño y el desarrollo funciones estatales básicas; hacer efectiva la profesional de los funcionarios. aplicación de intervenciones centradas con Al igual que en las zambullidas olímpicas, en precisión y en adecuada secuencia en contextos que en el puntaje depende de la perfección de PIBD —en que prácticamente todos los técnica y de la dificultad de la zambullida, el aspectos relativos a la capacidad y al buen desempeño del personal en los PIBD debería gobierno pueden requerir considerables evaluarse asignando apropiada ponderación a la mejoras—, y realizar un efectivo seguimiento de magnitud de las dificultades que presentan los los resultados en materia de capacidad y buen distintos entornos de los PIBD. También será gobierno. preciso poner de manifiesto la importancia de la labor referente a los PIBD en toda la línea • Elaborar criterios de asignación de la ayuda para jerárquica de la administración. que los PIBD no reciban menos ni más ayuda de la Además de esos incentivos es necesario que necesaria. el Banco se asegure de que el personal que se Es necesario que el Banco lleve a cabo un ocupa de los PIBD posea aptitudes adecuadas examen técnico del efecto acumulativo que para su tarea, por ejemplo en materia de tienen, sobre los volúmenes de ayuda a los administración del sector público; sea capaz de PIBD, los diversos ajustes introducidos al buscar y utilizar conocimientos políticos, y sistema de asignación basada en el desempeño. quiera y pueda realizar una labor basada en Deben elaborarse criterios de asignación de equipos interdisciplinarios. Merecen plácemes, lxvi RESUMEN aunque sean tardíos, los planes actuales encami- Segundo, es necesario simplificar la estruc- nados a atender esos temas en el estudio de tura orgánica de la labor relativa a los PIBD y a 2006 sobre fortalecimiento de la respuesta los conflictos. Es preciso que el Banco establezca orgánica a los Estados frágiles. una eficiente estructura orgánica que elimine la Se requiere un pensamiento más sistemático duplicación y fragmentación del apoyo entre la con respecto a las decisiones sobre dotación de Unidad de PIBD y la Unidad de prevención de personal para la labor referente a los PIBD en el conflictos y de reconstrucción. contexto de los planes globales de dotación de personal del Banco, dado que, para asignar • Al cabo de tres años, volver a evaluar el valor agre- personal más numeroso y más capacitado a la gado del enfoque de la Iniciativa para los PIBD. labor que se realice en beneficio de esos países, Transcurridos tres años, cuando se disponga probablemente habrá que llegar a soluciones de de experiencia suficiente con respecto a los compromiso con otros grupos del Banco a cargo resultados del enfoque de los PIBD, será de países, lo que se justificará o no según los necesario someter a una evaluación independi- objetivos que adopte el Banco en relación con ente el valor de la categoría PIBD y el enfoque esos países y las necesidades de asistencia que correspondiente, incluida la utilidad, para las experimenten otros clientes de la institución. operaciones, de los modelos. A esa altura podrá lxvii ACRONYMS AND ABBREVIATIONS AAA Analytical and advisory activities ADB Asian Development Bank ARPP Annual Report on Portfolio Performance BP Bank procedure CAS Country Assistance Strategy CASCR Country Assistance Strategy Completion Report CDD Community-driven development CDF Comprehensive Development Framework CEM Country Economic Memorandum CEP Community Empowerment Project CFAA Country Financial Accountability Assessment CPAR Country Procurement Assessment Report CPIA Country Policy and Institutional Assessment CPR Conflict Prevention and Reconstruction Unit CRN Country Reengagement Note CSO Civil society organization DEC Development Economics Vice Presidency DFID Department for International Development (United Kingdom) DO Development objective DPL Development Policy Lending DPR Development Policy Review EA Environmental assessment ESW Economic and sector work EU European Union FRM Financial Resource Mobilization Department FSG Fragile States Group GDP Gross domestic product GEMAP Governance and Economic Management Assistance Program GNI Gross national income HDI Human Development Index HIPC Heavily indebted poor country IBRD International Bank for Reconstruction and Development ICF Interim Cooperation Framework (Haiti) ICR Implementation Completion Report IDA International Development Association IDF Institutional Development Funds IEG Independent Evaluation Group (formerly Operations Evaluation Department, OED) IFA Integrative Fiduciary Assessment IGR Institutional and Governance Review ISN Interim Strategy Note lxix E N G A G I N G W I T H F R A G I L E S T AT E S ISR Implementation Status and Results Report (formerly Project Status Report, PSR) JAM Joint Assessment Mission KKZ Kaufmann, Kraay, and Mastruzzi indicators LIC Low-income country LICUS Low-Income Countries Under Stress MDGs Millennium Development Goals NGO Nongovernmental organization Norad Norwegian Agency for International Development ODA Official development assistance OECD-DAC Organisation for Economic Co-operation and Development’s Development Assistance Committee OP Operational policy OPCS Operations Policy and Country Services Department PA Poverty Assessment PBA Performance-based allocation PCF Post-Conflict Fund PCPI Post-Conflict Progress Indicator PEFA Public Expenditure and Financial Accountability PER Public Expenditure Review PIU Project Implementation Unit PREM Poverty Reduction and Economic Management Network PRSC Poverty Reduction Support Credit PRSP Poverty Reduction Strategy Paper PSR Project Status Report (now ISR) QAG Quality Assurance Group QEA Quality at entry SVP Rural medical centers (Uzbekistan) SWAp Sectorwide approach TFET Trust Fund for East Timor TRM Transitional Results Matrix TSS Transitional Support Strategy UN United Nations UNDG United Nations Development Group UNDP United Nations Development Programme UNTAET United Nations Transitional Administration in East Timor USAID U.S. Agency for International Development VDC Village development committee WAEMU West African Economic and Monetary Union WBI World Bank Institute OED changed its official name to the Independent Evaluation Group (IEG) in December 2005. The new designation “IEG” will be inserted in all IEG’s publications, review forms, databases, and Web sites. lxx Chapter 1: Evaluation Highlights • Plagued by a multitude of chronic problems, LICUS pose some of the toughest development challenges. • Donors and researchers are grappling with how best to respond to LICUS and have chosen to focus on different aspects of the problem. • Bank lending and administrative budgets to LICUS have increased since the start of the LICUS Initiative and have amounted to about $4.1 billion and $161 million, respectively, during fiscal 2003–05. • Post-conflict LICUS absorbed a large share of LICUS lending dur- ing fiscal 2003–05; administrative budgets were more evenly dis- tributed across the LICUS group. • General aid effectiveness concerns in LICUS have been replaced by state-building and peace-building objectives that remain inad- equately defined. • The Bank has yet to identify appropriate performance indicators for its state- and peace-building objectives. 1 Background he ongoing debates on aid effectiveness as well as international events, T especially the attacks of September 11, 2001,1 have attracted increas- ing attention to the problems facing Low-Income Countries Under Stress (LICUS). Concern is growing about the ability of these aspects of these problems, which has led them countries to reach the Millennium Development to focus on slightly different groups of Goals (MDGs), as well as about the adverse countries. economic effects they have on neighboring For instance, recent research by the Center countries and the global spillovers that may for Global Development focuses on stagnant follow.2 low-income countries (defined by gross national With their multiplicity of chronic problems, product per capita and growth rates), and LICUS pose some of the toughest development Foreign Policy’s Failed States Index focuses on challenges (box 1.1). Most have poor state failure. It identifies countries based on such governance; are embroiled in extended internal factors as the level of economic decline, security, conflicts or are struggling through tenuous factionalized elites, displaced persons, human post-conflict transitions; and face similar rights breaches, and external intervention. The hurdles of widespread lack of security, fractured U.S. Agency for International Development relations among societal groups, significant (USAID) aims to address issues surrounding corruption, breakdown in the rule of law, vulnerability and crisis. Many of these issues absence of mechanisms for generating legiti- pertain to the political environment. The U.K. mate power and authority, a huge backlog of Department for International Development investment needs, and limited government (DFID) and the Organisation for Economic Co- resources for development. Past international operation and Development’s (OECD) defini- engagement with these countries has generally tions of fragile states are similar to that used by failed to yield significant improvements.3 the Bank.4 The donor community is grappling with the As defined by the Bank, all LICUS are charac- question of how best to assist countries faced terized by weak policies, institutions, and with such challenging problems. With their governance. The World Bank has used two differing motivations and objectives, donors and criteria to define core and severe LICUS: per researchers have chosen to address varying capita income within the threshold of Interna- 3 E N G A G I N G W I T H F R A G I L E S T AT E S tional Development Association (IDA) eligibility, LICUS category over time. The Bank has recently and performance of 3.0 or less on both the replaced the term LICUS with fragile states, overall Country Policy and Institutional Assess- while retaining the same criteria to identify ment (CPIA) rating and on the CPIA rating for these countries.8 Public Sector Management and Institutions.5 LICUS share a number of similarities. They Some low-income countries without CPIA data have dismal economic and social indicators are also included.6 Depending on the income (figures 1.3–1.7). Besides being home to almost level and CPIA rating, a LICUS country is classi- 500 million people, roughly half of whom earn fied in one of three subgroups: severe, core, or less than a dollar a day, they have an infant marginal.7 Marginal LICUS score on the edge of mortality rate a third higher than that of other what is considered LICUS, and hence are identi- low-income countries, a life expectancy that is 12 fied by the Bank only for monitoring purposes years lower, and a maternal mortality rate that is (henceforth, LICUS refers to core and severe about 20 percent higher.9 If the trend continues, LICUS, not marginal LICUS). most LICUS will be unable to meet the MDGs In fiscal 2005, the Bank characterized 25 (appendix C). A vast majority of LICUS are countries as LICUS (see table 1.1 and figure 1.2). conflict-affected. This review focuses on these 25 countries. Despite their similarities, there are also Appendix B presents the important differences among LICUS. While Dismal social indicators list of core, severe, and some LICUS, such as Angola and Cambodia, and poor prospects for marginal LICUS for fiscal grew at around 4 percent a year during 2003–06, and illustrates 1995–2003, others, such as the Solomon Islands, achieving the MDGs are which countries have the Democratic Republic of Congo, and Guinea- common to LICUS. moved in and out of the Bissau, experienced negative growth rates of Box 1.1: The LICUS Challenge: Views from the Field Haiti The rural economy has suffered from prolonged drought and also “Preval’s task is colossal,” said a Haitian-born professor and author. because donors have had little success in supporting projects in “Everything has to be built . . . There are no institutions in Haiti. The rural areas because of concerns about security. Donors are challenge is really monumental.” Preval will be forced to confront under growing criticism for not having delivered on their much- the problems of a nation with almost no functioning judicial system, publicized aid pledges, and for having channeled a large part of corrupt and inept law enforcement, deep poverty, and abominable what they did deliver into the high fees and salaries of consult- public sanitation. Then there are the violent gangs that rule urban ants and nongovernmental organizations. slums, the kidnapping rings and a flourishing drug and money-laun- dering trade. There are also tens of thousands of children who do Kosovo not attend school, hundreds of miles of unpaved or poorly maintained “Out of all our non-luck came luck,” said the owner of a highly highways, and a national budget kept afloat primarily by the largess popular Thai restaurant in Pristina, referring to the vast amount of international aid groups and foreign countries. of international aid that has been poured into the province since its liberation following the 1999 war. “It is as though we have Afghanistan been given a second chance to rebuild our own home.” Despite many accomplishments, the general perception among Constructing the peace has, however, proved to be far from the Afghans more than three years into the reconstruction easy. Although life has demonstrably improved under the UN’s program is that there has been only minimal improvement in their guardianship, Kosovo’s transformation into a modern, multi- lives. Many in Kabul complain about the persistent unreliability of ethnic society—the international community’s much-vaunted the power supply, poor condition of the roads, and a lack of jobs. aim—continues to remain elusive. Sources: For Haiti, The Washington Post, “Challenges Loom for Preval in Haiti,” February 21, 2006. For Afghanistan, work undertaken for this review, IEG, 2005. For Kosovo, The Guardian, “A Second Chance to Rebuild Our Home,” October 31, 2003. 4 BACKGROUND Table 1.1: Twenty-five LICUS, Fiscal 2005 Severe LICUS Core LICUS a,b Afghanistan Burundia,b Angolaa,b Cambodiaa a,c Central African Republic Comorosa Haitia Democratic Republic of Congoa,b a,c Liberia Guinea-Bissaua,b Myanmara,c Kosovo (territory) a Solomon Islands Lao People’s Democratic Republic Somaliaa,c Nigeriaa Sudana,c Papua New Guinea Zimbabwec Republic of Congoa,b São Tomé and Príncipe Tajikistana Timor-Lestea,b Togoc Uzbekistan Source: OPCS, World Bank. Note: The countries classified as LICUS change slightly from year to year (appendix B). a. Conflict-affected countries in fiscal 2005. b. Post-conflict countries in fiscal 2005. c. Countries in non-accrual in fiscal 2005. Loans to, or guaranteed by, a sovereign are placed in non-accrual status when the oldest payment arrears are six months overdue—that is, when the second consecutive payment is missed on the loans with the oldest arrears. similar magnitudes. Vastly higher levels of ing fiscal 2003–05 went But LICUS are otherwise external debt as a percentage of gross national to 7 post-conflict LICUS income prevail in Liberia and São Tomé and (28 percent of the total a heterogeneous group. Principe than in Uzbekistan and Haiti. number of LICUS)10 A number of LICUS have abundant natural (figure 1.1 and table 2.1). resources, including Angola, the Democratic During fiscal 2000–02, the Sector Boards that Republic of Congo, Nigeria, and Papua New received the most Bank LICUS lending were Guinea, but not Burundi or Haiti. Furthermore, Economic Policy (25 percent); Health, Nutrition, the LICUS group includes countries such as São and Population (12 percent); and Private Sector Tomé and Principe, the Solomon Islands, Development (11 percent). During fiscal 2003–05, Tajikistan, and Uzbekistan, where the Human the Sector Boards with the greatest LICUS lending Development Index (HDI) is above the low- were Transport (22 percent), Rural Development income country (LIC) average, as well as countries (13 percent), and Economic Policy (10 percent). such as Burundi, the Central African Republic, and Comparing the two time periods, 2000–02 and Guinea-Bissau, where the HDI is considerably 2003–05, the Sector Boards with increases in below the LIC average (figures 1.4–1.6). LICUS lending were Transport and Rural Develop- During fiscal 2003–05 (the period after the ment; and those with a launch of the LICUS Initiative), the Bank decline in lending were Lending to LICUS has provided about $4.1 billion in lending to the 25 Private Sector Develop- increased, with a large LICUS, compared with about $2.5 billion during ment and Economic fiscal 2000–02 (the period before the initiative). Policy. share of lending going to Sixty-four percent of the total LICUS lending dur- With respect to ad- post-conflict LICUS. 5 E N G A G I N G W I T H F R A G I L E S T AT E S Figure 1.1: A Larger Share of LICUS Lending during Fiscal 2003–05 Went to Post-Conflict LICUS, While Administrative Budgets Were More Evenly Distributed across the LICUS Group 4,000 160 3,000 120 US$ million US$ billion 2,000 80 1,000 40 0 0 Lending Administrative budget (US$4.1 billion) (US$161 million) 18 Non-post-conflict LICUS 7 Post-conflict LICUS Source: World Bank database. Note: For definitions of LICUS, post-conflict LICUS, lending, and administrative budgets, see “Definitions and Data Sources” in appendix A. ministrative budgets, the Bank allocated about its LICUS approach in 2002. The approach has $161 million during fiscal 2003–05, compared since evolved and was rearticulated in 2005. Key with about $104 million during fiscal 2000–02. elements of both these stages in the develop- Sixty-six percent of the total administrative ment of the initiative are presented below, and budget to LICUS during 2003–05 went to 18 non- the main differences between them highlighted. post-conflict LICUS (72 percent of the total number of LICUS), and was thus more evenly The 2002 LICUS approach distributed across the LICUS group than lending The Bank coined the term LICUS and Administrative budgets (figure 1.1 and table established the LICUS Task Force in November 2.1).11 2001. The initiative thus both reflected and to LICUS have also While the large contributed to broader concerns in the donor increased. proportion of lending to community about aid effectiveness in difficult post-conflict LICUS might countries. The Task Force Report, published in have occurred even without the LICUS Initiative 2002 (henceforth called the 2002 LICUS Task (given that IDA’s exceptional post-conflict alloca- Force Report, World Bank 2002), aimed to tions predate it), the initiative likely contributed to describe how the Bank could best help chroni- the more even distribution of administrative cally weak-performing countries get onto a path budgets across the LICUS group (given an leading to sustained growth, development, and increase of 400 percent or more in administrative poverty reduction. budgets between fiscal 2000–02 and 2003–05 for The rationale provided for the LICUS Initia- three LICUS—Liberia, Somalia, and Sudan— tive, as stated in the 2002 LICUS Task Force which would have received minuscule amounts of Report, was that: administrative budgets The share of before the initiative Aid does not work well in these [LICUS] administrative budgets is because of their non- environments because governments lack accrual status). the capacity or inclination to use finance more evenly distributed effectively for poverty reduction. Yet across the LICUS group The LICUS Approach neglect of such countries perpetuates than lending is. The Bank first articulated poverty in some of the world’s poorest 6 BACKGROUND countries and may contribute to the Force Report, the Bank General aid effectiveness collapse of the state, with adverse elaborated its LICUS concerns have been regional and even global consequences. approach (reaffirming The challenge of aid effectiveness in LICUS some aspects, changing replaced by state-building is thus to use other instruments, supple- the emphasis of others, and peace-building mented by financial transfers where and adding some new objectives that have not necessary, to promote change (World elements) in the 2005 Bank 2002, p. 1). Fragile States Good yet been well defined. Practices in Country The “other instruments” referred to by the Assistance Strategies Report (henceforth called Task Force included analytical work and the 2005 Fragile States Report, World Bank knowledge transfer (which were to receive much 2005e) and the 2005 Low-Income Countries more emphasis than financial transfers, although Under Stress Update (henceforth called the precisely how much more was not defined). 2005 LICUS Update, World Bank 2005h). Since The LICUS approach was also to include the 2002 LICUS Task Force, the objectives and greater management attention and support of scope of the LICUS Initiative have shifted from LICUS work within the Bank. The approach general aid effectiveness concerns to state- outlined in the LICUS Task Force Report was building and peace-building objectives (World subsequently summarized as core country-level Bank 2005e). and Bank-level principles (box 1.2). Implemen- State building and peace building have not tation of the approach began following discus- been well defined, however, and remain some- sion of the Task Force Report by the Bank’s what abstract, especially from an operational Board of Executive Directors in June 2002, thus point of view. This leaves several questions insuffi- launching the LICUS Initiative. In October 2002, ciently answered. For example, what precise the Bank established the LICUS Unit in the balance between state and non-state capacity Operations Policy and Country Services (OPCS) does state building imply? To what extent are the Vice Presidency to coordinate LICUS implemen- common political and ideological connotations tation. The LICUS Initiative was meant to be a of the terms state learning-by-doing initiative. building and peace LICUS is intended to be a building intended? What The 2005 LICUS approach is the exact role of the learning-by-doing Taking stock of experience since the LICUS Task Bank in the security initiative. Box 1.2: The 2002 LICUS Approach: Core Principles Country level Bank level • Stay engaged. • Give much more attention to analytical work and transferring • Anchor strategies in stronger sociopolitical analysis. knowledge, and much less to transferring financial resources. • Promote domestic demand and capacity for positive change. • Ensure high-quality staff in LICUS. • Support simple and feasible entry-level reforms. • Further clarify, disseminate, and revise operational policies • Explore innovative mechanisms for social service delivery. and procedures for LICUS work to enable a faster and more • Work closely with other donors. effective response. • Support a more balanced approach to LICUS country pro- grams, underpinned by enhanced institutional support and management attention. Source: World Bank 2004b. 7 E N G A G I N G W I T H F R A G I L E S T AT E S sector? Furthermore, LICUS-specific approaches Learning by doing and the focus on organiza- to achieve state-building and peace-building tional issues in the 2002 approach were retained objectives have not been adequately articu- and further reinforced in the 2005 approach. lated—how will these approaches differ from The 2005 approach is based on a two-way past approaches and ensure a higher chance of knowledge flow: global knowledge is to inform success than in the past? staff guidance and country operations, and According to the Bank, state building and country experiences are to be distilled into staff peace building should be the goals by which to guidance and global knowledge. measure the LICUS Initiative’s success. However, the Bank has yet to identify performance indica- Differences between the 2002 and 2005 LICUS tors that would permit this to be done, beyond approaches stating that the “logical corollary of a central focus There are four main differences between the on peace-building and state-building in the Bank’s Bank’s LICUS approach as articulated in the assistance strategy for fragile states is that short- 2002 LICUS Task Force Report and recent term results measurement should also emphasize elaborations of the approach in the 2005 Fragile these dimensions while continuing to focus on States Report and the 2005 LICUS Update: growth, poverty reduction, and the Millennium Development Goals within the long-term vision • The 2005 approach emphasizes state building for recovery” (World Bank 2005h, p. 7). and puts greater focus on building state than Recognizing the diversity among LICUS, the 2005 non-state capacity compared with the 2002 Fragile States Report distinguishes among four approach. types of LICUS: those experiencing deterioration; • Compared with the 2002 approach, which those facing prolonged focused on capacity building,13 the 2005 The Bank’s recent political crisis or impasse; approach adopts the more expansive state- business models aim to those that are post-conflict building objective. or in political transition; • Peace building is one of the key objectives of address the diversity of and those experienc- the 2005 approach, and greater prominence LICUS. ing gradual improvement. is given to conflict prevention. The 2002 The Bank has proposed a approach did not mention peace building or different business model for intervening in each of conflict prevention among its core principles. the four types (appendix D). The Bank does not • The 2005 LICUS approach distinguishes intend to maintain a list of LICUS that fall under among LICUS and recommends a separate each business model, but instead to use the business model for each of the four groups of business models as an aid to planning scenarios LICUS; the 2002 approach was presented in when country teams are designing assistance terms of core principles applicable across all strategies. The expectation is that a country could LICUS. fall under more than one business model and move in or out of given models over time. Objectives, Purpose, and Organization The 12 Principles for International Engagement of the Review in Fragile States, agreed to by the OECD’s The review responds to the interest of the World Development Assistance Committee (OECD- Bank’s Board of Executive Directors in ensuring DAC) at the January 2005 London Forum (appen- the effectiveness of Bank support to LICUS. This dix E), also inform the Bank’s LICUS approach.12 review aims to answer three questions: The Bank has clustered the 12 principles into 4 main themes to structure its own work: building • How effective has the Bank’s LICUS approach state capacity and accountability; peace, security, been? and development linkages; donor coordination • How operationally useful are the Bank’s for results; and institutional flexibility and respon- criteria for identifying and classifying LICUS siveness (appendix F). and the aid-allocation system for them? 8 BACKGROUND • How appropriate and adequate has the evaluation of the Bank’s The recent approach Bank’s internal support for LICUS work been? support to Regional emphasizes state capacity programs will shed light Given the relative newness of the LICUS Initia- on the performance of over non-state capacity. tive, this review assesses implementation experi- multicountry projects and ence rather than outcomes. It uses the Bank’s partnership programs. stated LICUS approach as the benchmark—how Finally, this review does not, at this early well or badly the Bank followed its core country- stage, question the need for the LICUS Initia- level LICUS principles (chapter 2). The review tive itself, rather deferring that judgment to the also assesses the Bank’s criteria for identifying follow-up review recommended in three years, and classifying LICUS and for determining when sufficient evidence on outcomes will be lending allocations for them (chapter 3), as well available. The focus of this review is on how the as the appropriateness and adequacy of the Bank’s stated LICUS approach has been Bank’s internal support for LICUS work (chapter implemented, what has been learned about 4). The final chapter presents the conclusions effectiveness, and how the Bank can do better and recommendations (chapter 5). in the future (which may or may not be good It is beyond the scope of this review to assess enough to merit the existence and continua- the effectiveness of the Bank’s fiduciary controls tion of the LICUS Initiative). in LICUS or the extent of fraud and corruption In three years’ time, based on the outcomes associated with Bank projects in LICUS. This is a achieved, it will be opportune to ask—and topic that needs careful review, especially in light answer—the question of whether the Bank of the Quality Assurance Group’s finding that should have a LICUS category and approach at “fraud and corruption problems affect some all. At that time, it should be possible to address projects as demonstrated by detailed implemen- the more fundamental tation reviews in several risky countries” (World question of whether and Peace building and Bank 2006b, p. iv). to what extent Bank conflict prevention are The Independent Evaluation Group’s (IEG’s) assistance can effectively forthcoming evaluation of the effectiveness of the support sustainable more prominent in the Bank’s fiduciary work examines Country Financial state building. In aca- 2005 approach. Accountability Assessments (CFAAs) and Country demic debates about Procurement Assessment Reports (CPARs) with a state reconstruction, two main views prevail. view to assessing how these instruments One view questions “the dominant idea that influenced Bank assistance and strengthened failing states should always be rebuilt [consistent public financial management reform in client with the liberal democratic model] as most state countries, including some LICUS. reconstruction efforts have failed and bred new Because the Bank’s LICUS business models problems. . . . Until very recently, failing states were introduced in December 2005, it was were dismantled, not rebuilt,” (Carnegie found to be premature to assess their Endowment for International Peace 2004). implementation experience. Efforts to examine The second view recognizes the difficulties retrospectively the extent to which the Bank and imperfections in rebuilding states following followed the guidance contained in the business this model, but stresses that the rebuilding of models in different groups of LICUS yielded little states is necessary to improve the social and insight, given the still broad and general nature economic viability of failing states and to of the business model guidance. prevent conflicts from spilling over (Carnegie During fiscal 2000–05, the Bank approved 26 Endowment for International Peace 2004). Regional (multicountry) programs, amounting to about $2.9 billion, that included one or more of the Review Instruments and Methods 25 LICUS.14 This review does not, however, address Several instruments were used to conduct this Regional programs in LICUS. A forthcoming IEG review: 9 E N G A G I N G W I T H F R A G I L E S T AT E S • Literature review review should be treated with caution because • Portfolio assessment the response rates, especially for in-country • Thematic reviews stakeholders, are very low. • Fieldwork in 10 LICUS—Afghanistan, Angola, Cambodia, the Central African Republic, Haiti, The 25 countries classified by the Bank as Lao People’s Democratic Republic, Sudan, severe and core LICUS in fiscal 2005 constitute Tajikistan, Timor-Leste, and Zimbabwe (chosen the population for this review. The review fo- to ensure Regional representation, represen- cuses on the effectiveness of the Bank’s LICUS tation of post-conflict and other LICUS, and in- approach, but where possible, comments are clusion of countries of interest to the made on the effectiveness of the Bank’s overall IEG-Norad partnership) program in LICUS, noting that the two are not • Semistructured interviews of Bank staff synonymous. Specific aspects of the Bank’s • A survey of 455 persons, including in-country engagement in LICUS are compared with those stakeholders, Bank staff, and other donor staff in various other groups of LICUS and non- (henceforth referred to as the Stakeholder LICUS: post-conflict LICUS and non-LICUS low- Survey), with response rates of 16 percent (24 income countries (non-LICUS LICs). Appendix A respondents), 31 percent (382 respondents), contains the definitions of these groups of and 35 percent (49 respondents), respectively LICUS and non-LICUS, other concepts used in (appendix Z). The survey data presented in this the review, and the respective data sources. Twenty-Five Fiscal 2005 LICUS at a Glance Figure 1.2: LICUS Population Concentrated in Africa Europe and Central Asia Latin America and the Caribbean 34 (8%) 8 (2%) East Asia and Pacific South Asia 26 (6%) 78 (18%) Sub-Saharan Africa 285 (66%) Source: World Bank 2005j. Note: Number indicates population in millions (percentages in parentheses). Total population in the 25 LICUS = 432 million. 10 BACKGROUND Figure 1.3: More than Half the Population of Four LICUS (of Eight with Data) Earns Less than $1 a Day Nigeria ‘96 25 Uzbekistan ‘00 Cambodia ‘97 Population (million) 11 Zimbabwe ‘95 Burundi ‘98 6 Tajikistan ‘03 Lao People’s Democratic Republic ‘97 Central 3 African Republic ‘93 0 25 50 75 Percent of population earning below $1 a day Source: World Bank 2005j. Note: Data not available for 17 of the 25 LICUS. Year for which data were available is indicated along with the name of the country. Figure 1.4: Negative Growth Rate in about Half of LICUS, Lower Growth Rate in Most LICUS Compared with Low-Income Country Average (1995–2004) 15 GDP per capita growth rate 10 5 0 –5 nds sau go we nea lic i go oros te ti cipe o eria an n an ola ia ria . und Rep Tog ista Hai bod s pub Con Con kist Sud Libe bab r-Le Ang Isla -Bis Gui Nig Prin Bur Com k . Cam Dem n Re e Taji Zim o . of c of Uzb nea mon ew Tim and Rep ca ubli ua N Gui ple’s Solo Afri é Tom Rep . Dem Pap Peo tral São Lao Cen LIC average Source: World Development Indicators 2006. Note: Data not available for Afghanistan, Kosovo, Myanmar, and Somalia. For Haiti, growth rate represented is during 1998–2004. GDP = gross domestic product; LIC = low-income country. 11 E N G A G I N G W I T H F R A G I L E S T AT E S Figure 1.5: External Debt More than 175 Percent of GNI for Six LICUS and Higher than Low-Income Country Average for All LICUS in 2004 800 Debt (percent of GNI) 600 400 200 0 ria cipe ola n tan ti sau i ngo go we c ia nds nea eria an o p. oros und ubli ista Tog Hai . Re bod Con Sud Libe Ang bab ekis -Bis Isla f Co Gui Nig Prin Bur Com Rep k Cam Dem Taji Zim c of Uzb nea p. o mon ew and can ubli ua N ple’s Gui . Re Solo Afri é Tom Rep Dem Peo Pap tral São Lao Cen LIC average Source: World Development Indicators 2006. Note: Data not available for Afghanistan, Kosovo, Myanmar, Somalia, and Timor-Leste. GNI = gross national income; LIC = low-income country. Figure 1.6: Human Development Index for LICUS Worse than for Low-Income Countries in 2003 Angola Zimbabwe Burundi Uzbekistan Cambodia 0.593 Togo Central African Republic Timor-Leste Comoros Tajikistan Republic of Congo 0.300 Sudana Dem. Rep. of the Congo Solomon Islands Guinea-Bissau São Tomé and Principe Haiti Papua New Guinea Lao PDR Nigeria Myanmar Source: UNDP 2005. Note: Dark circle indicates HDI (Human Development Index) of LICs (low-income countries). LICs include all low-income countries (including LICUS) as defined by UNDP. HDI is a composite index produced by the United Nations and measures average achievement in three basic dimensions of human development—a long and healthy life (measured by life expectancy at birth), knowledge (measured by adult literacy rate and gross enrollment ratio), and a decent standard of living (measured by GDP per capita [purchasing power parity U.S.]). Countries with a value greater than 0.593 (outside the circle) are, on average, doing better than LICs; and countries with values smaller than 0.593 (inside the circle) are, on average, doing worse than LICs. a. Based on an estimate for northern Sudan. 12 BACKGROUND Figure 1.7: Kaufmann, Kraay, and Mastruzzi (KKZ) Governance Indicators Worse for LICUS than for Non-LICUS Low-Income Countries 1996 1998 2000 2002 2004 –0.25 –0.50 –0.75 –1.00 –1.25 LICUS Non-LICUS Source: http://www.worldbank.org/wbi/governance/data.html. Note: Graph presents the aggregate of: (i) control of corruption, (ii) governance effectiveness, (iii) political stability, (iv) rule of law, (v) regulatory quality, and (vi) voice and accountability. Unweighted average excludes Kosovo. The KKZ scale ranges from –2.5 to + 2.5. The KKZ indicators are a statistical compilation of responses on the quality of governance, given by a large number of enterprise, citizen, and expert survey respondents in industrial and developing countries, as reported by a number of survey institutes, think tanks, nongovernmental organizations, and international organiza- tions (including the World Bank and its CPIA). The KKZ results should be interpreted with caution, because even the most recent aggregate indicators, for 2004, have substantial margins of error. The margins of error are not unique to perception data—measurement error is pervasive among all measures of governance and institutional quality. An advantage of KKZ meas- ures of governance is that they are able to be explicit about the accompanying margins of error, whereas these are most often left implicit with objective measures of governance. Ag- gregation of separate sources of data, the six indicators and over countries, on average, reduces the margin of error compared with an individual data source. At an individual country and indicator level, very few countries would show significant change over 2000–04. 13 Chapter 2: Evaluation Highlights • The Bank’s implementation experience in LICUS has been mixed. • Its operational readiness to engage in LICUS has improved. • The Bank has contributed to macroeconomic stability and the de- livery of significant amounts of infrastructure, especially in post- conflict countries. • The Bank’s effectiveness needs to be improved after the immedi- ate post-conflict phase, when structural change is needed. • The Bank has not yet sufficiently internalized political under- standing in its country strategies. • Its reform agenda in LICUS has lacked selectivity and prioritization. • The strong donor coordination at the international policy level needs to be replicated at the country level. • State building has been made a central objective, despite the weak record in capacity development and governance. 2 Effectiveness of the Bank’s LICUS Approach his chapter assesses the effectiveness of the Bank’s LICUS approach based T on implementation experience since adoption of the initiative in June 2002. The main reason for the focus on implementation experience rather than on outcomes is that the LICUS Initiative is relatively new, and most available outcome data pertains to the period before 2002. The implementa- tion narratives presented in this chapter are intended to determine early on what is working and what is not and to provide lessons for the future. The effectiveness of the approach is first commended that much more attention be given assessed against the Bank’s stated approach for in LICUS to analytical work and transferring each of the country-level LICUS principles and knowledge than to transferring financial their subsequent elaborations. An additional resources, and to grants rather than to loans. aspect of the approach is also discussed— The 2005 Fragile States Report and the 2005 measuring and monitoring results—that is not LICUS Update continued to emphasize the specifically mentioned as a separate principle by importance of staying engaged. the Bank, but that is pivotal to the Bank’s learning-by-doing LICUS agenda. Aggregate data Implementation experience on the overall LICUS approach are then The Bank has made substantial progress on this presented in the last section of this chapter. principle. Since 2002, the Bank has improved its operational readiness to support LICUS and has Stay Engaged been more likely to engage, be it in the Central African Republic, Haiti, Somalia, Sudan, or Stated approach Zimbabwe. The Bank has engaged with LICUS in The 2002 LICUS Task Force Report (World Bank a number of ways over fiscal 2003–05: prepara- 2002) noted that Bank disengagement from tion of country strategies with shorter time LICUS could put these countries at risk of state frames; provision of lending; provision of trust failure and discourage other donor support. funds, including through the LICUS Trust Fund The report recognized that the nature of (appendix A); and provision of administrative engagement with LICUS would be somewhat budgets. The LICUS Trust Funds are specific to different from that with a typical LIC. It re- LICUS. 15 E N G A G I N G W I T H F R A G I L E S T AT E S Interim Strategy Notes The Bank’s work on 2003–05. A full evaluation of the extent and nature strategy design in LICUS of benefits resulting from the projects financed by have allowed continued continued during fiscal the LICUS Trust Fund, the sustainability of their strategy design in LICUS. 2003–05 through the benefits, and consistency with activities financed preparation of Interim by other trust funds, including the Post-Conflict Strategy Notes (ISNs).1 The use of ISNs for Fund, is still pending. LICUS has allowed the design of strategies that Based on stakeholder perceptions, the early cover a shorter period (up to 24 months), which experience with the Trust Fund seems to have is more appropriate to the generally volatile been generally positive, although there have conditions prevailing in LICUS. During fiscal been significant delays in disbursing funds. In 2003–05, the Bank prepared an increasing Haiti, the Trust Fund facilitated Bank reengage- number of ISNs or Country Assistance Strategies ment with the country and has been instrumen- (CASs) to stay engaged in LICUS—7 in fiscal tal in supporting the implementation of the 2003, 4 in 2004, and 13 in 2005 (appendix G). Interim Cooperation Framework, which was The Bank’s lending to LICUS increased by 67 established to guide international assistance and percent between fiscal 2000–02 and 2003–05, cooperation with the government. In the compared with a 10 percent increase in lending Central African Republic, the Trust Fund helped to non-LICUS LICs. Lending to LICUS increased finance activities in the Transitional Results from $2.0 per capita in fiscal 2000–02 to $3.2 in Matrix (TRM) and helped intensify the reengage- fiscal 2003–05, compared with an increase for ment process by multiplying the contacts and non-LICUS LICs (excluding India) from $5.2 in exchanges among all parties. In Liberia, the fiscal 2000–02 to $6.0 in fiscal 2003–052 LICUS Trust Fund facilitated enactment of public (appendix I). Post- procurement legislation and implementation of Lending to LICUS has conflict LICUS absorbed government-wide procurement procedures, a large share of lending and contributed to donor coordination through increased compared with during fiscal 2003–05 support for implementation of the multidonor lending to non-LICUS (with 7 post-conflict results framework. LICs. LICUS, comprising 28 The proportion of grants relative to loans percent of the total going to LICUS has increased. Consistent with number of LICUS, receiving 64 percent of total the recommendation of the 2002 LICUS Task LICUS lending) (table 2.1). Lending to post- Force Report that a greater proportion of grants conflict LICUS averaged $8.2 per capita yearly relative to loans be used in LICUS, the grant compared with $1.5 to non-post-conflict LICUS component of total project lending to LICUS (appendix I). has increased from less than 20 percent in fiscal At the same time, there was a 70 percent 2003 to about 50 percent in fiscal 2005, while increase in trust fund financing to LICUS the grant component of total project lending to between fiscal 2000–02 and 2003–05, compared non-LICUS LICs stayed around 15 percent. If with a decline of 87 percent in trust fund financ- grants from trust funds are included, LICUS ing to non-LICUS LICs. As with lending, post- would show an even higher proportion of grant conflict LICUS absorbed a large share of trust financing. Moreover, during the discussions to fund resources going to LICUS during fiscal replenish IDA funding in 2004 (IDA 14), it was 2003–05 (table 2.1). agreed that debt sustainability will be the basis The LICUS Trust Fund for the allocation of grants to IDA-only countries Post-conflict LICUS have was introduced by the under which all LICUS are expected to qualify absorbed a large share of Bank in 2004, mainly to for 100 percent grant financing.4 Five LICUS are LICUS lending. finance engagement in currently between the decision and completion LICUS that were in non- points for debt relief through the Highly accrual.3 The Trust Fund accounted for 1.3 Indebted Poor Country (HIPC) Initiative, and percent of trust funds going to LICUS during fiscal eight additional countries may qualify in the 16 EFFECTIVENESS OF THE BANK’S LICUS APPROACH Table 2.1: Lending, Trust Funds, and Administrative Budgets Going to LICUS and Non-LICUS Low-Income Countries during Fiscal 2000–02 and 2003-–05 Percentage Fiscal change, 2003–05 Fiscal Fiscal fiscal 2000–02 to share of 2000–02 2003–05 fiscal 2003–05 total LICUS (%) Lending (million US$) Post-conflict LICUS — 2,664 — 64 Non-post-conflict LICUS — 1,473 — 36 Total LICUS 2,480 4,137 67 Non-LICUS LICs 18,557 20,400 10 Trust fund (million US$) Post-conflict LICUS — 1,816 — 92 Non-post-conflict LICUS — 159 — 8 Total LICUS 1,159 1,974 70 Non-LICUS LICs 11,090 1,485 –87 Administrative budget (million US$) Post-conflict LICUS — 54 — 34 Non-post-conflict LICUS — 107 — 66 Total LICUS 104 161 55 Non-LICUS LICs 380 450 18 Sources: Appendix I; Trust Fund Database. Note: For definitions of LICUS, post-conflict LICUS, non-LICUS LICs, lending, trust funds, and administrative budgets, see “Definitions and Data Sources” in appendix A. The share of lending to the post-conflict LICUS is further exaggerated on a per capita basis because only a quarter of the total LICUS population resides in the seven post-conflict LICUS. For every dollar lent per capita in non-post-conflict LICUS, $5.4 is lent in post-conflict LICUS. future, but no LICUS has yet reached the Determining whether Proportion of grants to completion point and received HIPC grants the increased overall LICUS relative to loans (appendix H; IEG 2006b). LICUS administrative Administrative budgets going to LICUS also budgets are adequate has increased. increased by 55 percent between fiscal 2000–02 for LICUS work requires and 2003–05, compared with an 18 percent assessing the efficiency of resource use and is increase in non-LICUS LICs.5 Unlike lending and beyond the scope of this review. Instead, staff trust funds that have gone mostly to post- views on the adequacy of the administrative conflict countries, administrative budgets budgets are presented here. during fiscal 2003–05 have been distributed The Liberia team noted the challenges for more evenly across the LICUS group (with 18 staff of working on a reengagement strategy and non-post-conflict LICUS, comprising 72 percent the added time cost of of the total number of LICUS, receiving 66 working in a volatile Administrative budgets percent of total LICUS administrative budgets; institutional and politi- table 2.1). In Cambodia, a higher administrative cal environment. It have increased and have budget for the country office (97 percent found that requests for been more evenly increase between fiscal 2000 and 2004) has further staff support distributed across the helped to deepen the Bank’s understanding of could not be met by the the political underpinnings of the Cambodian existing Bank budget, LICUS group than has state and improve donor coordination. while the strategy of lending. 17 E N G A G I N G W I T H F R A G I L E S T AT E S minimum engagement itself made it difficult to and capacity-building work in LICUS, even when justify further increases in the Bank budget. lending is low. In LICUS with no lending program, staff While the administrative budget for econ- pointed out that they must claim additional omic and sector work in LICUS has more than Bank budget as a special dispensation, which doubled since the start of the LICUS Initiative, can also add to the state of uncertainty—in the one-fourth or more of LICUS do not have any Africa Region, a part of the Bank budget is held economic and sector work (ESW) being back during the year, and LICUS country teams conducted in Sector Boards such as Education; need to reapply for budget based on milestones Environment; Health, Nutrition and Population; achieved. With respect to supervision budgets, Social Development; Social Protection; Trans- staff noted that country norms tended to apply, port; Urban Development; and Water Supply irrespective of whether the country was LICUS and Sanitation. This lack of ESW in important or not, despite the much higher supervision Sector Boards raises some questions about intensity of projects in LICUS. the effectiveness of future Bank assistance The appropriate levels of administrative (appendix O). budget for LICUS need Although there is some variability, the quality But Bank staff in some to be determined in the of economic and sector work seems to be broader context of the satisfactory overall in LICUS, and the analytical LICUS say that allocation of administra- work has enabled the Bank to maintain administrative budgets tive budgets across all operational readiness in a number of LICUS. are still inadequate. Bank departments. The Quality Assurance Group (QAG) assessments above-average increase show that the quality of economic and sector in LICUS administrative budgets for fiscal work in LICUS is improving over time. There 2003–05 meant that non-LICUS LICs received a also do not appear to be systematic differences below-average increase. Non-LICs, in contrast, between the quality of economic and sector received their share of the increase in the work based on CPIA status in fiscal 2001 and administrative budget. The Bank needs to 2002. No comparable figures are available for determine whether this is an appropriate distri- later years. QAG assessments of analytical and bution of the Bank’s administrative budget. advisory activities (AAA) rated “internal quality” While six LICUS show a decline in the as satisfactory for Angola and marginally satisfac- administrative budget for analytical work in fiscal tory for Uzbekistan. 2003–05 compared with fiscal 2000–02, the Bank Improvements in the process aspects of the has nearly doubled the administrative budget for Bank’s ESW would help enhance country-level analytical work in aggregate in LICUS, from effectiveness. The involvement of country about $25 million in fiscal 2000–02 to about $50 counterparts in the Bank’s analytical work million in fiscal 2003–05. The administrative remains limited to administrative aspects, with budget for analytical work in non-LICUS LICs has much less country-client participation in select- also increased, from about $69 million in fiscal ing topics and undertaking analysis, which has 2000–02 to about $110 million in fiscal 2003–05, reduced national buy-in. a 59 percent increase over the two time periods This was the case in Tajikistan, where lack of (appendix I). The increase may partly be government involvement in the selection and explained by the delinking of administrative preparation of analytical work limited its interest budgets for economic and sector work and in the results, which hindered implementation. technical assistance (the two main components In Angola, some Bank-led analytical work (for of analytical work) from instance, relating to the recent Country Budgets for analytical lending volumes by Economic Memorandum) was perceived by the Bank, in recognition senior government officials as an imposition of work have increased for of the importance of Bank views on internal affairs, which led to all but six LICUS. maintaining analytical limited ownership and capacity development. In 18 EFFECTIVENESS OF THE BANK’S LICUS APPROACH the Stakeholder Survey, over 40 percent of in- used by the govern- ESW quality seems to be country, Bank, and other donor respondents said ment. The absorptive that analytical work has achieved intended results capacity of the govern- satisfactory overall in only to a slight extent or not at all (figure 2.1). ment is severely limited, LICUS. Analytical work done in collaboration with and analytical and other donors can also improve the Bank’s policy advisory activities done mostly by the Bank risk influence. For example, Sudan’s Joint Assess- straining relations with the government, no ment Mission—which involved multiple donors matter what their technical quality. In Cambodia, and culminated in the preparation of a needs plans for analytical and advisory services in the assessment report—helped the Bank to gain the 2005 CAS—a total of 30 tasks to be completed confidence of the government and to increase over fiscal 2005–07—appear excessive. its own policy influence. This highlights the importance of designing programs of analytical Anchor Strategies in Stronger work as part of a coordinated process with other Sociopolitical Analysis donors, rather than by the Bank alone. Princi- ples of donor coordination should apply as Stated approach much to analytical work and policy capacity The 2002 LICUS Task Force Report (World Bank development as to other areas of donor 2002) emphasized the importance of sociopoliti- programs. cal analysis to help identify feasible reforms and Absorptive capacity constraints apply at least the best ways to promote them. The rationale was as much to knowledge products as to financial that understanding local dynamics, perceptions, products, so the amount of knowledge transfer and circumstances allows clearer understanding that can be usefully undertaken will differ across of the effects of proposed reforms on various LICUS. In Angola, the government endorsed the societal groups, their Interim Strategy Note but expressed concern likely response to them, Collaboration with other about the amount of foreseen analytical and and thus the likely advisory activities. This has raised doubt about success of the reform donors can increase the whether the analytical products would be fully agenda. Bank’s policy influence. Figure 2.1: Over 40 Percent of Respondents Said That the Bank’s Analytical Work Has Achieved Its Intended Results Only to a Slight Extent or Not at All 60 40 Percent 20 0 Large extent Moderate extent Slight extent Not at all World Bank [N = 262] Donors [N = 38] In-country [N = 21] Source: Appendix Z (Stakeholder Survey results). Note: N indicates the number of valid responses. 19 E N G A G I N G W I T H F R A G I L E S T AT E S Absorptive capacity The Fragile States strategy in the government’s Interim Coopera- Report (World Bank tion Framework (ICF), the piece analyzes what constraints apply as 2005e) and the 2005 media Haitians are using as the source of their much to knowledge LICUS Update (World information and offers recommendations on products as to financial Bank 2005h) have con- how to improve the information policy and tinued to emphasize the practices of the transitional government.6 products produced. importance of political The risk sections in the Bank’s strategy factors, including the documents for the Central African Republic need to complement invisible reforms with highlight the country’s political instability, lack more visible ones that show tangible changes, of reform commitment and champions, weak thereby enhancing popular support. capacity to implement reforms, and inadequate external support as major dangers that could Implementation experience thwart the success of the Bank’s operations. The understanding of Bank staff of the political They also seem to suggest an appreciation by environment in which they work has been Bank staff of the country’s political situation. mixed. Even where such understanding exists, it Sensitivity to day-to-day politics in Timor- has not necessarily influenced strategy design Leste, specifically unrest among military and implementation. Political analysis is all the veterans, led the Bank to start a small program more important in LICUS, where decision- to deal with their grievances before the issue making processes are not institutionalized and could become a factor of destabilization in a still- may be influenced by personal and political fragile country. interests. The Bank has increasingly encouraged more open discussion and treatment of political Examples of inadequate attention to political issues. issues in its activities, including through In other countries, the Bank could have guidance from the LICUS Unit. However, the improved its strategy by better reflecting the Bank has yet to internalize political understand- political situation. For example, in Zimbabwe, the ing sufficiently in LICUS country strategy design limited political analysis is apparent in the focus and implementation. of the Bank’s 2005 ISN on what Zimbabwe has to do on economic reforms, reestablishing social Examples of good political understanding. IEG’s service delivery, and rebuilding infrastructure, fieldwork for Sudan found that Bank staff have rather than on the more difficult question of how demonstrated good understanding of the such a process can be encouraged and initiated country’s political environment and been tactful to balance the political and technical aspects of a in a complex environ- future reform process. Bank staff understanding ment. Overall, the Bank In Haiti, a number of donors see the lack of of the political has managed to avoid more thorough political analysis as the reason being seen as siding with for what they consider a key shortcoming in the environment in which any one of the parties to execution of donor strategy—that is, the danger- they work has been the conflict, as other ous neglect by the donor community of the mixed. international actors have need to provide adequate resources for dealing been perceived to be with the current security problems. doing, although serious challenges remain for the While the Bank’s political understanding of Bank with the ongoing situation in Darfur. Cambodia has improved recently, it has been a Similarly, a useful internal piece of political voyage of slow and gradual discovery. Progress analysis of Haiti’s communication demonstrated has been punctuated by successive over- that Bank staff are aware of the importance of estimations of the role of the formal government the political environment and are drawing some institutions in relation to that of informal institu- practical conclusions from such insights. tions based on patronage and political and Focusing on the inadequate information military power. The Bank did not appear to have 20 EFFECTIVENESS OF THE BANK’S LICUS APPROACH internalized such understanding sufficiently (for 2.1), but this has not Sound political example, in its treatment of demobilization, always been the case. The including the subsequent IDA-financed project), Bank should commission understanding and its nor of the role of institutionalized corruption in such analysis only in cases appropriate distillation supporting these less formal institutions (for where good political into Bank strategy example, in its approach to forestry). analysis does not already remains a function of The forestry project in Cambodia foundered, exist. at least in part, because of insufficient attention Critical to the Bank’s specific personalities by the Bank to the problem of corruption. Many effectiveness is its ability rather than something non-Bank stakeholders cited this project as an to adequately reflect that can be expected example of the Bank’s neglect of the political sound political analysis reality of the country, arguing that the outcome in its strategy. This has more commonly. of the project was entirely predictable. been an area of Respondents to the Stakeholder Survey noted weakness in the Bank. For example, the Interim an improvement over time in the grounding of Strategy in Papua New Guinea has a good discus- the Bank’s work in an understanding of a sion of the political system. It recognizes the country’s politics (appendix Z). However, only problems of clan loyalties, political patronage, about a third of Bank respondents, and a quarter corruption, lack of capacity, and other factors, of both other donor and in-country respondents, but the Strategy then goes on to disregard some said the Bank’s work is largely grounded in an of this vital knowledge and treat these issues understanding of the country’s politics. About a as technical problems. Commissioning and quarter of in-country respondents, Bank respon- The political analysis dents, and other donor respondents said it is so and reasons for past consuming—not only to a slight extent or not at all (figure 2.2). failures should have necessarily producing— Commissioning and consuming—not nec- underpinned the Bank’s good political analysis is essarily producing—good political analysis is crit- strategy in the country. ical for the Bank in LICUS. In Lao PDR, the Bank While in Lao PDR the critical for the Bank in effectively tapped existing political analysis (box Bank effectively tapped LICUS. Figure 2.2: Bank Respondents Somewhat More Likely than Other Respondents to Say That the Bank’s Work Has Been Grounded Largely in an Understanding of the Country’s Politics 75 50 Percent 25 0 Large extent Moderate extent Slight extent Not at all World Bank [N = 305] Donors [N = 42] In-country [N = 23] Source: Appendix Z (Stakeholder Survey results). Note: N indicates the number of valid responses. 21 E N G A G I N G W I T H F R A G I L E S T AT E S Box 2.1: Lao People’s Democratic Republic: Good ing the political dynamics that shaped the Practice Example of Using Existing Political conflict and that will determine the course of the Knowledge peace process” (Rogier 2005). This will inevitably take the Bank into addressing ethnic, sociologi- cal, and political factors. Although the Bank has The Bank invited a political scientist who had published extensively on given more attention to preventive aspects the Lao People’s Democratic Republic to make a presentation to the recently,7 there is limited knowledge about the country team on politics and reform in the country. This allowed for effectiveness of these efforts, and the Bank’s role preparation of an independent summary of relevant political analysis and comparative advantage in these areas have (tailored to the needs of the donor community in general and to the Bank yet to be clearly established. in particular) and dissemination of this information to relevant Bank staff and other donors. It also avoided the higher costs of preparing a “Bank” Promote Domestic Demand and Capacity analysis, as well as potential tension with the government. In other for Positive Change words, acquisition of existing knowledge as well as its dissemination proved more important and effective than knowledge creation. Stated approach Source: Fieldwork undertaken for this review in Lao People’s Democratic Republic, IEG, 2005. The 2002 LICUS Task Force Report argued that donors should work strategically to build existing political analysis, as noted above; even capacity for reform both inside and outside in that case the internalization of the government. Shifting the emphasis, the 2005 knowledge gained in Bank strategy remained Fragile States Report stresses the centrality of insufficient. building capacity inside the state. With respect The main focus of the to non-state actors, it recommends balancing The Bank has not yet Bank’s efforts to improve state capacity-building efforts with support for sufficiently internalized attention to political civil society and the private sector. factors needs to be on political understanding helping staff internalize Implementation experience in strategy design and political analysis in strat- There has been a marked jump in the Bank’s implementation. egy design and imple- lending and trust funds devoted to capacity mentation. Appendix J development in LICUS (see definition of outlines the many types and layers of political capacity development in the note to table 2.2). analysis that are useful in strategy development. The level has increased from less than $90 The operational aspects of the Bank’s LICUS million in fiscal 2002 to over $366 million in approach do not fully reflect the conflict-preven- 2005 (table 2.2, appendix K). Capacity develop- tion goal, which was given more prominence in ment, however, has not historically been an the 2005 LICUS approach than in the 2002 area of strength for the Bank. Continuing iteration. The 2005 Fragile States Report includes problems with the design and implementation conflict prevention as part of the peace-building of its capacity development work in LICUS objective, and more specifically under its deterio- suggest that its effectiveness is likely to remain ration business model, limited. The conflict-prevention where it emphasizes the According to the 2005 Comprehensive goal will require the Bank importance of contribut- Development Framework (CDF) Progress to address root causes, ing to multidonor and Report (World Bank 2005c), capacity-building but its role in this area is community-level conflict- support in most LICUS tends to consist of prevention efforts. isolated interventions in specific areas, and it is yet to be clearly Addressing conflict not always responsive to country priorities. Of established. prevention will require 39 countries where this was found to be the the Bank to give much case, 18 (46 percent) were LICUS. greater attention to the root causes of conflict— IEG’s 2005 review of the Bank’s capacity- conflict prevention “necessarily entails address- building interventions in Sub-Saharan Africa 22 EFFECTIVENESS OF THE BANK’S LICUS APPROACH Table 2.2: Capacity Development Lending and Trust Funds in LICUS (US$ million) Fiscal year 2000 2001 2002 2003 2004 2005 2000–05 Total 122 22 87 55 16 366 669 Percentage of total resources 23 5 6 4 1 30 10 Sources: For IBRD, IDA, and Social Fund data—World Bank database; for LICUS Trust Fund, Institutional Development Fund, and Post-Conflict Fund data—LICUS Unit, OPCS. Note: Capacity development is defined broadly to include Bank interventions dealing with, for example, state capacity development, accountability, and private sector development. Capacity- development financing here includes financing for all Bank-supported projects that were either free-standing capacity-development projects or where capacity development accounted for at least 80 percent of the project cost. The list of capacity-development projects includes those financed through IBRD, Global Environment Facility (GEF), Special Financing (country- specific trust funds), IDA, LICUS Trust Fund, Post-Conflict Fund (PCF), and Institutional Development Fund (IDF)—they were identified by IEG and cleared by the LICUS Unit and are pre- sented in appendix K. noted that country programs generally do not has not been effective in Continuing weaknesses in address—systematically and in an integrated developing capacity, be- the design and way—the issue of countries’ ability to build cause it was provided capacity. In the majority of country programs, with few Afghan counter- implementation of capacity-building support remains fragmented— parts, resulting in little capacity-development work designed and managed operation-by-operation. transfer of knowledge. suggest that the Bank’s Fiduciary and other ESW products still involve Government capacity re- clients mainly in organizational tasks and data mains weak nearly four effectiveness will remain collection, and only to a limited extent in data years after the formation limited in this area. analysis, report writing, and dissemination. of the first government. Despite the severity of governance and The “buying of capacity” through massive institutional problems in LICUS and limited past technical assistance has not delivered capacity success in these areas, the Bank has continued development, and some evidence suggests that it to rely on traditional approaches to capacity detracted from this objective. The amount of development. In Tajikistan, the Bank identified technical assistance provided to date is well highly relevant areas for capacity development, beyond the country’s absorptive capacity. Massive but the approach used a Project Implementation technical assistance has Unit (PIU). The PIU was distant from the govern- not only meant wasted Fresh approaches for ment and had little impact on broader govern- resources that could have enhancing capacity and mental capacity development. The recent been used more produc- accountability are Country Partnership Strategy in Tajikistan aims tively elsewhere, but also to link interventions to various levels of govern- a diversion of scarce needed. ment, but there is little implementation experi- institutional capacity toward lower-priority tasks. ence so far. In Timor-Leste, while some capacity Some government officials also pointed out that development has been accomplished in the open-ended consultant contracts create perverse health sector, capacity development in other key incentives to provide unnecessary technical areas (education, agriculture, and growth) has assistance. been limited. In the Stakeholder Survey, Bank respondents IEG’s fieldwork for Afghanistan found that the were more positive than other donor respon- Bank has provided a significant amount of dents in their views on the extent to which the technical assistance (over $135 million in the Bank’s technical assistance has achieved its past three years). This was useful in the early intended results. Despite that, 45 percent of stages to start critical government functions, but Bank respondents said that the results were 23 E N G A G I N G W I T H F R A G I L E S T AT E S achieved to only a slight extent or not at all, tion and contributed to enhancement of the compared with 65 percent of other donor government’s capacity to design and implement respondents and 47 percent of in-country transport investments—the lack of a specialized respondents (appendix Z). PIU did not result in major delays or complica- In the absence of tions in implementation.8 But even in that Reliance on massive capacity development in country, this institutional arrangement was an technical assistance has the central administra- exception rather than the general practice. not helped to develop tion, the Bank’s capacity Similarly, in Afghanistan, the capacity development efforts in development effort through the Priority capacity. Timor-Leste have been Restructuring and Reform process, which “inevitably ad hoc and uneven” (Schiavo-Campo aimed directly at ministries and their incentive 2003), lacking a strategic approach. While structures rather than at creating PIUs, is an capacity development was not within the Bank’s example of a promising approach (box 2.2). mandate according to the agreement made at Nevertheless, capacity development efforts in the Tokyo meeting, which entrusted the United the country at large still face significant Nations Transitional Administration in East hurdles. Timor (UNTAET) with this responsibility, the In Cambodia, a number of recent Bank reports Bank took on the objective of developing assessing capacity development from 2000–05 capacity at the community level and setting up a similarly concluded that achievements were local governance structure through the extremely limited because of a combination of Community Empowerment Projects. governance problems and poor donor coordina- As noted by a 2004 European Union evaluation, tion9 and the insufficient level of resources given the importance of capacity building, there (particularly human resources in the field) should have been discussion and agreement provided by the Bank in the early years of the CAS between the Bank and UNTAET to address the period. Cambodia’s 2000 CAS called for sector- issue jointly (IEG 2006c). The assumption that wide approaches (SWAps) to reduce the transac- capacity would be developed in the government tion costs of multiple projects and multiple PIUs through on-the-job transfer of expertise from and to develop capacity by supporting govern- international advisors and training was flawed in ment programs in a joint effort financed by an environment of very weak country capacity. several donors, and implemented by the respon- The international advisors ended up focusing on sible government agency. project implementation and had little time for Two years after the end of the CAS period, no ensuring the transfer of capacity (IEG 2006c). SWAp had yet been approved.10 Several of the In general, the Bank has had more success in IDA-funded operations more directly geared to contributing to long-term capacity when it has capacity development also did not take place worked through governments rather than (Public Sector Reform, Legal Reform) or through PIUs (IEG 2000) or when a longer-term encountered substantial problems (Forestry plan for the transfer of relevant functions from Learning and Innovation Loan, which was rated the PIU to government unsatisfactory, and the Economic Capacity The contribution to long- agencies has been Building Project, which was rated unsatisfactory term capacity has been instituted and executed but upgraded to satisfactory in June 2005). greater when working from early days (IEG Although the LICUS Initiative has sought to 2002, 2005b). increase the Bank’s emphasis on improving through government For example, in the governance in LICUS, the Bank has yet to agencies or when making Bank’s Road Mainte- address sufficiently the basic governance an early effort to transfer nance Program in Lao problems that plague some of its own projects. PDR—which relied on Recently, the Bank’s Integrity Department functions from the PIU to the relevant ministry to investigated seven projects in Cambodia, and government agencies. lead program prepara- problems such as misuse of funds, misprocure- 24 EFFECTIVENESS OF THE BANK’S LICUS APPROACH Box 2.2: Afghanistan Priority Restructuring and Reform Process: Developing State Capacity through the Direct Restructuring of the Civil Service The Priority Restructuring and Reform process (PRR) in Afghanistan with a program of “lateral entry” that brought qualified Afghans allows individual ministries and government organizations to in the country or greater region into the civil service on a con- award higher salaries (up to $300 a month compared with current tract basis, as well as with a recruitment program that provided monthly salaries of about $30) in return for implementing a reform even higher salaries to attract highly qualified expatriate Afghans program as approved by the Civil Service Commission. Units con- living in Western countries ($800–1,500 a month for lateral entry cerned are required to define their objectives, plans for rational- and $4,000–7,000 for Afghan expatriates). ization/downsizing, job specifications, selection criteria for each The PRR is expected to build a pyramid structure in the civil ser- position, and a competitive process of recruitment. vice, placing civil servants at the base, and successively fewer This approach seeks to draw out existing skills in the civil ser- lateral-entry and expatriate Afghans toward the top. Foreign con- vice, while weeding out unqualified recruits previously brought sultants to develop capacity would only be recruited once nec- into the government. In 2004, the PRR process was supplemented essary Afghan staff was in place. Source: Fieldwork undertaken for this review in Afghanistan. ment, fraud, collusion, and corruption were determine them in countries with weak overall found in certain contracts in each of the projects capacity (in the balance between immediate and (World Bank 2006c). longer-term needs, or different levels of The Bank’s operational approaches do not administration or government functions) would sufficiently address issues of instrument choice be beneficial for strategy development. Entry for capacity development, or discuss which points will necessarily differ across LICUS, and instruments—technical assistance, SWaps, the Bank has not sufficiently developed capacity development as part of investment operational approaches to distinguish between projects, or a totally new instrument—are situations where it is justified in developing state effective under which LICUS situations. The capacity directly, where it is justified in doing so Bank also has not defined its capacity- indirectly through non-state actors, and where development strategy to achieve its state- the capacity of non-state actors themselves building objective, and has thus far insufficiently needs to be developed. monitored its work in the area of capacity The 2005 LICUS Operational approaches development and governance. approach made state Furthermore, the operational approaches building its central ob- need to be further have not been differentiated enough to fit the jective, despite the Bank’s differentiated to fit the varying institutional environments of different weak record in capacity varying institutional LICUS. In some LICUS, the state is exclusionary development and gover- or represents only the interests of particular nance. In taking on the environments of different groups. How does the Bank intend to approach complex and more LICUS. state building in these circumstances? Similarly, ambitious state-building what would its approach be in LICUS where agenda, the Bank has the responsibility to indicate engaging with non-state actors may not be what this agenda does and does not include, feasible given strong negative reactions of the demonstrate that it has identified its role and government to such contact? The complexities comparative advantage in state building, and of addressing state building in LICUS are develop operational approaches to ensure a outlined in box 2.3. higher chance of success than in the past. To address capacity constraints effectively, a Some other donors have similarly adopted distillation of experience that would indicate the the state-building terminology. The Bank is right entry points, as well as approaches to working with them to articulate a state-building 25 E N G A G I N G W I T H F R A G I L E S T AT E S Box 2.3: The Complexities of State Building in LICUS: Is the Bank Ready? In Sudan, both the government and the Sudan People’s Liberation Movement welcome the focus on capacity development implicit in the Bank’s activities, including in the LICUS Trust Fund. But the Bank faces tough choices. For example: • Whose capacity should be built, particularly in North Sudan? • To what extent should the Bank strengthen and transform existing structures and chains of command? • To what extent should it circumvent or replace structures considered too heavily influenced by political patronage interests? • What is the moral responsibility of the Bank in the difficult circumstances presented by the genocide in Darfur? In Papua New Guinea, the Bank faces another set of tough questions: • Is the concept of a “central” government applicable? • Does the country need more decentralization, especially if the center is weak? • Should the Bank’s capacity development efforts reinforce clan structures or supplant them? In Timor-Leste, some key questions are: • How far should donors go in promoting plans for long-term development and local capacity development for a country with- out a sovereign government (as was the case in Timor-Leste before 2002)? • How can foreign expertise be delivered in a way that promotes local capacity development? Sources: For Sudan, fieldwork undertaken for this review, IEG 2006c; for Papua New Guinea, background work undertaken for this review; for Timor-Leste, IEG 2006c. agenda and to identify its own role within it, and Building stronger state institutions and the Bank has introduced some relevant capacity- governance requires social transformations, development measures (such as leadership including those of civil society and in the support, anti-corruption work, political relationship between the state and civil society. economy of reform, development policy While the Bank’s 2002 LICUS approach noted the operations, transitional results matrixes, and importance of these issues and its 2005 approach demand-side measures). These measures do reaffirmed their importance, the Bank has yet to not, however, amount to the jump in thinking develop specific guidance on, for example, what and approach required to address effectively the the appropriate balance is between state and tough capacity development and governance non-state capacity development in different challenges in LICUS. LICUS situations and how it can be achieved A deeper approach to capacity development effectively. Country strategies and the Bank’s and governance than what we see in the Bank’s assistance in the field also have yet to be 2005 LICUS Update and LICUS country strate- adequately informed by such considerations. gies seems warranted. While state fragility is Leadership training will bring limited often associated with weak institutions and poor benefits unless it is complemented by measures governance, the more operationally relevant that simultaneously foster a broader political question is why such problems exist in LICUS. debate and discourse, including one stimulated Any approach that does by the media. The inherently political nature of The state-building agenda not sufficiently address such activities cannot be ignored. The Bank needs to be unpacked this deeper question is needs to be explicit about what aspects of the and the Bank’s specific unlikely to help LICUS problem it will include in its assistance program, transition out of their as well as which donors it will work with to role and comparative fragility in a sustainable ensure that its own efforts are adequately advantage identified. manner. complemented and supported. 26 EFFECTIVENESS OF THE BANK’S LICUS APPROACH Furthermore, while the Bank’s approach to relevant to matching the state’s role to its state building emphasizes the development of capability in LICUS (Grindle 2004). It basic systems of public administration, public emphasizes selectivity in a world in which all finance, and macroeconomic management in good things cannot be pursued at once. LICUS, their prioritization will be key, but Instead, the task is to determine what is remains inadequately addressed thus far. essential and what is not, what should come Finally, with respect to social service delivery, first and what should follow, what can be guidance is lacking on how much and to whom achieved in the short term and what can only to deliver these services. Given the inevitable be achieved over the longer term, and what is resource and capacity constraints in LICUS, feasible and what is not. how will the Bank strike a balance between IEG comments on the Bank’s Action Plan to catering to elite interests (which may be implement the recommendations of the necessary to stabilize power) and catering to Bank’s 2005 Task Force on Capacity Develop- those of the poor (which will be necessary to ment in Africa have also identified several areas equitably improve living conditions and avoid for further development of the Bank’s negative spillovers)? approach (box 2.4). In terms of measuring The question remains whether the Bank governance and accountability performance, should have adopted state building as a central the Bank’s 2006 Global Monitoring Report LICUS objective without first fully understand- identifies specific CPIA indicators;11 Kaufmann, ing what it entails and how it can be achieved, Kraay, and Mastruzzi (KKZ) indicators; Doing especially given the Bank’s traditional Business 12 and Investment Climate Survey weakness in the area of capacity development indicators;13 Public Expenditure and Financial and governance. Does the Bank have convinc- Accountability (PEFA) indicators; 14 Global ing approaches to bring about accountable Integrity Index indicators;15 Polity indicators;16 governance? Where demand for capacity and Transparency International indicators as development is low (as is likely to be the case good measures.17 in LICUS), what approaches will the Bank Resource-rich LICUS pose special problems adopt? How will the Bank ensure that alterna- of accountability and rent-seeking. While tive delivery mechanisms do not detract from country strategy design has emphasized issues the state-building agenda? of governance in natural resource management The focus on state building in LICUS also in recent years, the Bank’s implementation raises questions about the adequacy of staff arrangements have been inadequate. For with the relevant public sector management example, in the Central African Republic, skills. In addition, the Bank has yet to necessary actions for the forestry and mining develop an appropriate set of performance sectors are outlined in Bank’s country strategy, indicators against which state-building but it is unclear how they will be implemented, outcomes can be measured. And finally, the or what happens if they are not implemented choice of the term “state building” may itself (appendix L). be inappropriate given its political and ideological connotations. Support Simple and Feasible Entry-Level The Bank’s World Development Report Reforms 1997: The State in a Changing World recommended matching the state’s role to its Stated approach capability, while at the same time raising state The 2002 LICUS Task Force Report emphasized capability by reinvigorating public institutions. the importance of a highly focused reform The Bank has not given enough attention to agenda in LICUS. It explained that this would developing approaches that address both these consist of two or three reforms that are aspects of state capacity in LICUS. The idea of important in economic terms and are likely to “good enough governance” would seem to be result in a rapid and substantial payoff, but that 27 E N G A G I N G W I T H F R A G I L E S T AT E S Box 2.4: Areas for Further Development in the Bank’s Approach to Capacity Development The Bank’s Africa Action Plan (World Bank 2005i) needs to in- ing block, a long list of substantive areas—ranging from public ex- dicate how the Bank will act on the broad and general obser- penditure management and health service delivery to empower- vations of the Bank’s 2005 Africa Capacity Development Report. ing the press and parliaments. The capacity development record The Action Plan should aim to answer the following specific to date in some of these areas is weak, so it will be important for questions: the Action Plan to identify the areas of the Bank’s comparative ad- How will the CAS be improved to reflect capacity building as vantage and what will be done to achieve better results. a core country program objective? The report’s first building block How can lending instruments be made more effective? What for the Bank is to use the CAS better. What changes will be made actual activities—training, technical assistance, and other to ensure that capacity-building objectives are clearly articulated interventions—have proved effective for building capacity within in the CAS results matrix and tracked during CAS implementation? lending operations? The Task Force recommends a continued How will the Bank and borrowers take account of the sectoral shift to programmatic lending. Although there are potential ad- characteristics that affect capacity needs? The health sector, for vantages to programmatic lending, this instrument has not auto- example, has different capacity issues than the roads sector. What matically led to better results. How should programmatic loans be roles will networks and sector boards play in developing sector- designed to establish clear objectives and effective actions for ca- specific analysis and benchmarking? A single corporate focal pacity building? point, as proposed by the 2005 Africa Capacity Development Task How should World Bank Institute (WBI) activities change to Force, is unlikely to be able to provide deep enough sectoral support country programs more effectively? The 2005 Africa Ca- knowledge to meet these needs. pacity Development Report calls for a “more focused” working re- In which areas will the Bank concentrate, and how will it sup- lationship for WBI in country programs, but describes a WBI role port capacity building more effectively than in the past? The 2005 that appears quite similar to its present one. What changes are Africa Capacity Development Report proposes, as its second build- planned? Source: IEG comments on World Bank 2005i. Note: The 2005 Africa Capacity Development Report refers to the report of the Bank’s 2005 Task Force on Capacity Development in Africa (World Bank 2005a). are also feasible in sociopolitical terms, countries. It also has often contributed to the tending to unite a broad coalition for reform. delivery of significant amounts of physical The 2005 Fragile States Report noted that infrastructure. political, security, and development linkages Helping to move a country through the are particularly important in fragile state immediate post-conflict reconstruction phase contexts. and into the development phase presents major challenges, especially for institution building Implementation experience and employment creation (for example, mass Implementation progress on this principle has protests broke out in Dili, Timor-Leste, first in been modest, with good progress on some 2002 and again in mid-2005, partly as a result of reform aspects (such as contributing to people seeking jobs), where the Bank has been macroeconomic stability, including controlling less successful. Areas that are less technocratic inflation or delivery of physical infrastructure), and where the cultural content of institutions is but less progress on others (such as prioritizing greater—governance, corruption, conflict of and sequencing reforms, building institutions, interest between public and private interests— or strengthening governance). have proven to be tough. The Bank has generally contributed to IEG’s fieldwork in Timor-Leste found that macroeconomic stability, including the control there should have been a more deliberate of inflation, through, for example, currency and process of transition from the immediate post- banking reforms, especially in post-conflict conflict reconstruction phase to the develop- 28 EFFECTIVENESS OF THE BANK’S LICUS APPROACH ment phase, and that longer-term develop- As the examples of the Central African ment challenges should have been more Republic and Haiti show, all sorts of obstacles thoroughly considered. A more timely CAS may make the follow-through on a successful (originally scheduled for 2002 but not LICUS engagement difficult. The Central African completed until 2005) might have helped Republic is now faced with a potentially stimulate discussion about a relevant develop- disastrous budget crisis ment strategy. just at the moment The Bank has succeeded While the Trust Fund for East Timor (TFET) when its political suc- in contributing to provided much-needed immediate reconstruc- cess needs to be backed macroeconomic stability, tion support, and did so reasonably quickly, it up on the economic has not been as successful in policy develop- side. In Haiti, the donor especially in post-conflict ment in some important areas (education, community seems to situations. agriculture, economic growth). Although the have given inadequate process surrounding the Bank’s three budget attention to ensuring a minimum level of support loans, the Transitional Support security. In both cases, a good initial result of the Programs (TSPI-III), drew government and LICUS Initiative is now at risk of being donor attention to the broad policy agenda, diminished. including the issue of expenditure manage- The Bank needs to improve its effectiveness ment, it involved several hundred individual in the period that follows post-conflict actions and insufficiently focused on the core reconstruction, when easy reforms have been issues of institutional capacity and the future exhausted and structural change (in institutions direction of the economy. The general use of or governance, for example) is needed. In some Development Policy Loans (DPLs) and other ways the Bank appears forms of budget support in LICUS is discussed to have moved from Staying engaged is only a in appendix M. emergency reconstruc- IEG’s fieldwork for the Central African tion to development means to an end and Republic also raises the question of whether the without discussion of needs to be followed up LICUS approach and its instruments are the process and the with an effective reform adequate for meeting all the challenges facing implications for Bank the Bank’s strategy in the next, post-political strategy in post-conflict agenda. transition phase of its engagement with the situations. Box 2.5 country. That challenge is to build on the presents three lessons from the Bank’s post- success of the election process and use it to conflict experience. consolidate and amplify the reform effort. The Bank’s approach in post-conflict LICUS While the Bank has played an important early also has other shortcomings. For example, engagement role in several LICUS, staying immediately following the cessation of conflict, engaged is only a means to an end. In some international donors, including the Bank, have instances, strategic disengagement—with the often committed large amounts of aid coupled exception of in-house analytical work—may be with overly ambitious agendas. This has frequently needed, at least for periods of time, especially created high expectations among the population when involvement with the Bank is seen as and led to disillusionment when expectations inappropriately giving legitimacy to the LICUS remained unfulfilled and few tangible improve- government or when it dampens internal ments are seen in day-to- pressure for reform. Where engagement is day living. Effectiveness needs to appropriate, effective follow-up through a clear This disillusionment improve in the phase and relevant reform agenda will be important, has been further aggra- following reconstruction, or else the early successes of engagement may vated by the perception be short-lived and may contribute little to the of a “foreign footprint” when structural change is achievement of CAS objectives. created by the infusion of needed. 29 E N G A G I N G W I T H F R A G I L E S T AT E S Box 2.5: Three Lessons from Post-Conflict Countries Domestic political processes, however slow, are necessary to This requires the early formation of the basic public institutions, prevent a relapse into conflict. The hard work of building a adequate salaries for staff (and hence the imperative of donor strong domestic base and engaging civil society in an organized budget support in the first years of post-conflict assistance), train- manner is unavoidable, but highly time consuming. ing of staff, and making good use of existing institutions and indi- Often the post-conflict period sees a continuation of strife, but by viduals. External capacity will have to be bought, but should be nonviolent means. The principal instrument for the continuation of the contracted only with sunset provisions to maintain the incentives conflict can be economic. Constructing artificial barriers to flows of to develop local capacity. commerce and transport; severing of normal economic linkages Economic instruments (aid, policy advice, technical assis- across ethnic groups; refusal to harmonize taxes and customs; reg- tance) work best when kept in line with the absorptive capacity ulations that impede the creation of a common economic and trade of the country and with the willingness and the appetite of repre- space; and fracturing of institutions or mechanisms that process sentative governments to reform. This calls for scaling down the economic conflict (courts, arbitration mechanisms, chambers of ambitions of rapid state building. commerce) have all been used to keep differences alive. Where the Early expectations in a number of post-conflict countries of a organizers of separateness have access to quasi-state resources, and very large physical reconstruction program—initially through the means to extract the allegiance of the population through force donor grants and rapidly thereafter by the country—leading to self- or loyalty, creating a coherent national unit will be a long process. sufficiency within a decade or so have often been vitiated by se- Institutional development and state capacity formation need to curity problems and inadequate local engagement in the vision. start from the first days of a post-conflict program. A post-conflict Donor involvement has turned out to be of greater duration than society’s capacity to use aid efficiently is low, yet both donor and initially thought, with disbursements stretched out and much more country expectations of the peace dividend are high. The need to modest in the initial years than planned. reinforce capacity is thus a high priority. Source: Mitra 2004. large amounts of international funds and person- countries (as opposed to the “highly focused nel in the country. According to a 2005 Reuters reform agenda” recommended by the 2002 estimate, up to 60 cents of each dollar of an aid LICUS Task Force) do not augur well for effective- project in Afghanistan goes into overhead, includ- ness. Nicolas van de Walle (2005) notes, ing payment to donor staff (Francois and Sud “although [the] LICUS [Initiative] voices all of the 2006). Better communication is critical to lower right concerns for ownership and selectivity, it is expectations to realistic levels and is something terribly vague about how the Bank will avoid the that the Bank needs to invest in. past pitfalls in this strategy’s implementation” Speed should not be prioritized over the (Van de Walle 2005, p. 80). achievement of longer-term objectives, especially While donors must strive for collective donor when the objectives relate to difficult institutional selectivity, this is far from being achieved, as issues that require a learning-by-doing approach Afghanistan’s donor-endorsed reform agenda by stakeholders. An excessive focus on speed in and Haiti’s ICF (discussed below) show. However, the initial phases may compromise laying the even if this collective donor selectivity is not groundwork necessary immediately achieved, the Bank itself needs to A large infusion of for sustainable future ensure focus and selectivity in its own assistance external funds increases outcomes. program, based on its core competences. Such Given the limited Bank selectivity has been increasing in recent the “foreign footprint” in capacity in LICUS, the years, but remains a challenge, as the example of the country and can Bank’s currently broad São Tomé and Principe (discussed below) seems cause resentment. reform agendas in several to suggest. 30 EFFECTIVENESS OF THE BANK’S LICUS APPROACH In Afghanistan, the reforms covered by dents and a third of Appropriate sequencing of donors are wide-ranging, show lack of other donor respon- reforms is key to ensuring sufficient priority, and have led to 120 pieces of dents said that the Bank pending legislation. These reforms, dealing supported a focused selectivity while avoiding with virtually every economic and social aspect reform agenda consist- partial solutions. of the country, need to be carefully prioritized ing of key actions and and sequenced, but donors have yet to do so. reforms in the LICUS country to a slight extent In Haiti, the ICF, which is meant to guide or not at all. international assistance and cooperation with Haiti through September 2006, covers practi- Explore Innovative Mechanisms cally all basic state functions, ranging from for Social Service Delivery security, to national dialogue, to economic governance, to economic recovery, to basic Stated approach services. Individually, all these areas seem The 2002 LICUS Task Force Report emphasized important, but together they add up to a the importance of exploring alternative formidable program. mechanisms for social service delivery. The With respect to the Bank’s own assistance report argues that the strategy for improving program, São Tomé and Principe is an example basic social outcomes is to supplement weak where the Bank was far too ambitious in central government delivery by strengthening relation to the resources allocated to the multiple alternative channels. Compared with country, with the result that many of the CAS the 2002 LICUS Task Force Report, the 2005 objectives were not achieved or were only Fragile States Report plays down the importance partially achieved. of alternative service-delivery mechanisms. Beyond selectivity in CASs, it is critical to ensure that actual reform agendas in the field are focused and well-prioritized. While it is Box 2.6: Timor-Leste: Excessive Optimism, difficult to be selective in complex LICUS Impatience, and Partial Solutions? environments where reforms are needed in virtually every area, greater effort must be From 1999 to 2002, the United Nations (UN) lent enthusiastic support made to prioritize and sequence reforms to to an ambitious nation-building project in Timor-Leste. The project, avoid overtaxing limited capacity, while at the however, did not work out as many had envisioned. An economic up- same time rejecting partial solutions. tick during the three years of UN rule in Timor-Leste collapsed after many In Timor-Leste, donors may have pulled out too of the foreign advisors departed. Recent donor reports state that little quickly, without sufficiently dealing with the headway has been made in improving basic services in the country. country’s pressing capacity needs (box 2.6). In A Timorese country official expressed concern that, below the min- Haiti, development assistance has greatly fluctu- isterial level, the country lacked people with adequate experience to fill ated over the years. The country has gone through essential jobs and run things on a day-to-day basis. He complained that several “feast or famine” cycles in its relations with “we have ministers but no middle managers” and that a good deal of the the donor community. This may have been nation’s mess is the result of actions taken by the foreign donors. avoided had various donors better coordinated the The Timorese government has asserted that more than half the for- sequencing of their aid. eign assistance to their country was spent on salaries and consultancy Overall, the Bank has been overly optimistic fees for the foreign advisors. The country official explained that, in about what it can achieve in LICUS, as indicated essence, the foreigners were too impatient. They came, spread their by some wide-ranging country strategy money around, and left. “They all had a time frame—one year, two years, objectives and the mostly unsatisfactory or four years,” the country official said. “You can’t build a country from noth- moderately unsatisfactory outcome ratings ing in that amount of time.” given by in IEG in its CAS Completion Report Source: Perlez 2006. Quotes from Sidonio Freitas, Senior Manager, Timor Sea Designated (CASCR) Reviews for LICUS. In the Stakeholder Authority. Survey, a fifth of in-country and Bank respon- 31 E N G A G I N G W I T H F R A G I L E S T AT E S Implementation experience seen as an NGO-driven agenda, disconnected Alternative service delivery mechanisms have from the pressing concerns of state building succeeded in the quick and effective delivery of and enhancing central government capacity. substantial infrastructure, but there is less The 2005 CAS has a very limited discussion of evidence of success in meeting other objectives, CDD, and instead relies on developing state such as empowering communities or develop- capacity for service delivery. More than eight ing government capacity. Implementation years after the end of the civil war, donors have progress on this princi- made little headway in developing government Alternative delivery ple has thus been capacity. mechanisms have modest. In the Stakeholder Survey, the majority of in- Alternative service country and Bank respondents said that the use provided substantial delivery mechanisms of nongovernmental or semi-autonomous infrastructure. have included social arrangements made a small positive contribu- funds, community- tion to both service delivery and the develop- driven development (CDD) projects, and ment of long-term government capacity in projects with nongovernmental organization LICUS. About 40 percent of other donor respon- (NGO) or private sector involvement. A number dents said that the contribution to service of these projects, such as the Angola Social delivery was large, while 40 percent also said Action Fund, Cambodia Social Fund, and Timor- that there was no contribution to development Leste Community Empowerment Projects, have of long-term government capacity (figure 2.3). succeeded in quickly disbursing significant When alternative delivery mechanisms are amounts of resources and delivering substantial used, there needs to be a clear transition plan amounts of infrastructure. But other benefits for moving the functions back to the govern- have been more elusive. ment. In Afghanistan, more than three years In Timor-Leste, the breakdown of institutions, after the start of post-conflict reconstruction, poor governance, widespread suffering, and none of the donors has a strategy for doing so. massive displacement of the population put Alternative delivery mechanisms may not be pressure on the donor community to respond feasible in some LICUS. For instance, in Lao PDR, speedily. The Bank met the challenge through where there is a single-party system and a very three Community Empowerment Projects (CEP-I, limited role for civil society, alternative II, and III) that quickly transferred resources to mechanisms would make unlikely candidates for communities and delivered massive amounts of service delivery. Similarly, in Zimbabwe, the infrastructure. But speed came at the cost of the government’s sensitivity to the Bank engaging other project objectives, particularly community with nongovernmental actors limits the extent to empowerment and the development of local which the Bank can follow this principle. In institutions. CEP-established project councils, Timor-Leste, an otherwise democratic and open which bypassed local traditional leaders, were government has begun to adopt an increasingly seen by communities primarily as conduits for hostile attitude toward civil society organiza- channeling donor money tions. The Bank has recently attempted to But they have had less and were not able to take understand how to engage with civil society in success in empowering on the larger role LICUS effectively (appendix N). communities or envisaged for them as development agents in Work Closely with Other Donors developing government their communities (IEG capacity. 2006c). Stated approach In Tajikistan, while The 2002 LICUS Task Force Report noted that the 2003 CAS built strongly on the CDD partnership with other agencies is central to Bank approach, the CASCR acknowledged that CDD Group activities, particularly in LICUS. The 2005 did not take root as planned because it was Fragile States Report emphasizes higher levels of 32 EFFECTIVENESS OF THE BANK’S LICUS APPROACH Figure 2.3: Most Respondents Said That the Use of Nongovernmental or Semi-Autonomous Arrangements Has Made a Small Positive Contribution To Service Delivery in LICUS 80 60 Percent 40 20 0 Large positive Small positive No contribution Negative World Bank [N = 190] Donors [N = 26] In-country [N = 21] To the Development of Long-Term Government Capacity in LICUS 80 60 Percent 40 20 0 Large positive Small positive No contribution Negative World Bank [N = 189] Donors [N = 26] In-country [N = 21] Source: Appendix Z (Stakeholder Survey results). Note: N indicates the number of valid responses. partnership, such as donor alignment, policy number of LICUS, there Alternative delivery coherence, and harmonization, and also are still instances of basic mechanisms may not be highlights the importance of addressing capacity disagreements among constraints through donor collaboration. donors on critical strate- feasible in some LICUS. gies, and implementa- Implementation experience tion follow-through on agreed policies has The quality of donor coordination, with strong typically been weak.18 Implementation progress Bank participation, has been substantial at the on donor coordination in LICUS is thus rated international policy level, as exemplified by the medium overall. recent donor agreement on the 12 OECD-DAC principles of engagement. But the quality of Donor coordination at the international policy level. donor coordination at the country level has been The Bank has long been active in international medium to low. While country-level policy agree- policy debates on fragile states, often playing a ments among donors have been increasing in a leading role as co-chair of donor events and co- 33 E N G A G I N G W I T H F R A G I L E S T AT E S author of joint policy papers (appendix Y). An builds on and complements earlier joint work by important element in the Bank’s drive for the UNDP and the Bank’s Conflict Prevention partnership and cooperation at the international and Reconstruction Unit on multilateral needs policy level is the Fragile States Group (FSG).19 assessments in post-conflict situations (UNDG, The FSG, supported by the DAC Secretariat in UNDP , and World Bank 2004). the OECD, is motivated by the understanding The Bank and UN have also collaborated on that all donors recognized the importance of developing the Joint Assessment Missions more emphasis on fragile states, the need to stay (JAM) tool, and donors have recently completed engaged and find ways of working effectively, joint country strategies in Cambodia, Nigeria, and the importance of acting together in these Somalia, and Togo, and they are under way for difficult circumstances at about the same time. the Central African Republic and the Democra- The FSG’s work is uniformly well regarded tic Republic of Congo. Furthermore, the Bank among donors.20 Most believe it has been instru- has collaborated with the UN Department of mental in elevating the issue of fragile states to Peace-keeping on a joint staff training program the international level and in laying a foundation and is developing a joint state-building program for coordinated action.21 The best-known result with UNDP. of this is the recently promulgated principles of In Timor-Leste, while significant problems international engagement in fragile states between some donors emerged later (for (appendix E) (OECD 2005c). example, between the Bank and the UN, as The strength of donor coordination will be discussed below), the immediate donor tested during implementation of the 12 OECD- response following the referendum for DAC principles through pilots. This process is independence was well executed. A JAM of likely to present several coordination challenges. experts from five donor countries, the European Many solutions have been identified to address Commission, UN agencies, the Asian Develop- these challenges in a concept note prepared by ment Bank (ADB), and the World Bank visited the FSG (OECD 2005b). However, implementing Timor-Leste in late 1999. A subsequent donors’ these solutions will re- meeting in Tokyo endorsed the establishment of Implementation of the quire significant effort. two multidonor trust funds, one of which was OECD-DAC principles For example, compara- under the trusteeship of the Bank with the ADB ble actions by 12 donors as co-implementer. It moved quickly to commit needs to be monitored. in 10 pilot countries that funds to cover virtually every sector in its sphere represent a wide range of of responsibility. views will have to be agreed and translated into In a number of instances, however, Bank-UN concrete actions.22 Close interaction with relations have encountered significant prob- ongoing security and humanitarian missions will lems. In Timor-Leste, the Bank and the UN could be needed. The likelihood that the pilots will not agree on a common approach to community result in effective ways of implementing the development. The UN favored close alignment design embodied in the principles is far from with the emerging district administration, while certain and should be monitored. the Bank argued for bypassing the weak district From the outset, the LICUS Initiative has level. The end result, as noted by the joint sought to build relevant partnerships with key government–civil society study of development players—UN agencies such as the UN Develop- projects, was that the Bank’s community ment Program (UNDP) and UN Development empowerment program developed in parallel Group (UNDG),23 and some bilateral donors, with UN local government grant agreements. In such as the United Kingdom and Australia. The Sudan, the Bank and UN worked on the JAM Bank has had a number of successful partner- process without any formal agreement between ships with the UN at the policy level. The recent them. UNDG-Bank Operational Note on Transitional Among the Bank’s bilateral partnerships, that Results Matrices (World Bank and UNDP 2005) with the DFID stands out, especially because of 34 EFFECTIVENESS OF THE BANK’S LICUS APPROACH agreement on many aspects of how develop- modernization from The Bank has made a ment in fragile states might be undertaken. the Bank, USAID, the strong effort to Some donors view DFID as providing the European Union (EU), intellectual leadership on fragile states, while and DFID. The ADB has coordinate with the UN the Bank provides a systematic approach. The spread its technical from the outset. partnership with Australia is also strong. assistance for capacity Cooperation is growing with several other development (about $15.2 million) over 18 bilateral donors, including Canada, Denmark, tasks in various ministries. Similarly, large France, the Netherlands, Norway, and the United missions can overload the government, and in States. the absence of early warning and sustained follow-up can be counterproductive. Donor coordination at the country level. The quality In Angola, several development partners of donor coordination experience at the country noted that the Bank could have done more to level has been medium to low. While there are a foster partnership and coordination, and the number of examples of good aid coordination at Bank has only lately emphasized this aspect. The the country level, in far too many cases coordi- Bank was perceived as giving priority to nation is unsatisfactory. Donors have pointed establishing its own credentials with the govern- out that the Bank’s current tendency—to get ment at the expense of a joint approach and as approval first up the management line (often at giving insufficient attention to the crucial details headquarters), and then to seek collaboration of partnership, increasing the risk of duplication with other donors, asking them to harmonize of effort. with the Bank—is against the spirit of donor In semistructured It has similarly made a coordination. interviews, some com- strong effort to plained that the Bank Examples of good country-level aid coordination. In was unwilling to listen to coordinate with Liberia, close donor coordination resulted in donors with many years bilaterals. agreement on the Governance and Economic of experience in Angola Management Assistance Program (GEMAP) and that partners were being informed ex post between the National Transitional Government rather than consulted ex ante. Incoming and various bilateral and multilateral donors. missions from the Bank did not seek close The GEMAP was developed to help improve collaboration with other partners, who then financial and fiscal administration, transparency, concluded that decision making in the Bank was and accountability in Liberia and received the strongly centralized in Washington. Although recognition and support of the UN Security the difficult overall environment in Angola Council. created many challenges, the Bank’s particular Despite a poor earlier donor coordination engagement contributed to donors and the experience, Cambodia has made progress government inadequately coordinating the recently. The 2005 CAS is the product of close policy dialogue relating to governance and coordination between donors, with large parts transparency issues. of it common to the ADB, DFID, and the UN As noted earlier, country-level donor coordi- development system. nation in Timor-Leste experienced significant problems. The experience of Timor-Leste Examples of weak country-level donor coordi- highlights the critical nation. It is not uncommon to find multiple importance of coordina- The Bank needs to donors providing capacity development to the tion among donors replicate its success at the same organization using different procedures, supporting reconstruc- international policy level further burdening overloaded ministries. In Af- tion in a newly inde- ghanistan, for example, the Ministry of Finance pendent post-conflict by raising its game at the receives technical assistance for customs country. While the inter- country level. 35 E N G A G I N G W I T H F R A G I L E S T AT E S national community was able to raise a large ative advantage over other donors in all (or even amount of resources for development of Timor- many) aspects of capacity development, the Leste, the general lack of coordination between identification of capacity development as an area the major players in the country worked to the for donor coordination in the 2002 LICUS Task detriment of both the donors and the country Force Report was useful. Subsequent LICUS and led to less effective and efficient utilization documents, including the 2005 Fragile States of resources (IEG 2006c). Report, do not elaborate on specific areas of the IEG’s Stakeholder Survey did not specifically Bank’s comparative advantage and how donor ask respondents about country-level donor coordination for capacity building could be coordination. With respect to overall donor effectively pursued. From an operational coordination, the majority of in-country respon- perspective, this is an important omission, dents said that the Bank has pursued collabora- especially given that state building is now a tion with other donors to a large or moderate central focus of the LICUS Initiative. extent. Other donor respondents were about Borrower governments could contribute to, equally divided among the large, moderate, and and learn from, the debates and discussions slight choices. The majority of all respondents among donor agencies, but their participation said that there has been a positive change in the so far has been scant. Representatives from Bank’s effectiveness in pursuing donor collabo- recipient countries (or from key NGOs) are ration, comparing the period before and after rarely drawn into contributing to and learning the adoption of the LICUS Initiative (figures 2.4 from the debates and discussions among donor and 2.5). agencies and associated institutions. Indeed, the Capacity building was identified as an area for discussion in forums such as the FSG seems donor coordination in the 2002 LICUS Task Force quite removed and does not adequately reflect Report, but the Bank has not provided much of country circumstances, although the recent an operational approach to achieve it effectively. piloting of the principles of engagement is likely Given that the Bank is not likely to have compar- to address this problem, at least in part. Figure 2.4: The Majority of Bank Respondents Said the Bank Has Pursued Collaboration with Donors to a Large Extent; the Majority of In-Country Respondents Said It Has Done So to a Large or Moderate Extent Source: Appendix Z (Stakeholder Survey results). Note: N indicates the number of valid responses. 36 EFFECTIVENESS OF THE BANK’S LICUS APPROACH Figure 2.5: The Majority of Respondents Noted a Positive Change in the Bank’s Effectiveness with Donor Collaboration Source: Appendix Z (Stakeholder Survey results). Note: N indicates the number of valid responses. Increased inclusion of credible participants from overall objective, policy Borrower governments fragile states in forums such as the FSG can help coherence is unlikely. have participated little in to ground discussions in LICUS country circum- Varying donor visions in stances further, thus producing more Afghanistan were also donor agency debates. customized strategies for varying fragile reported by country environments. officials (box 2.7). The Bank’s donor coordina- Donors may not have invested enough in tion efforts and modalities are insufficiently confidence-building measures among borrower informed by the objectives of the different governments before introducing fully coordi- players in a country. Donor coordination, nated or joint strategies. In Papua New Guinea, however, is a form of collective action and the government did not take kindly to the requires that other donors also improve their proposed joint strategy by the three biggest outreach to the Bank and subordinate bilateral donors (Australia, World Bank, and Asian agendas to agreed multilateral objectives. Development Bank), fearing this to be an Coordination is not only important among attempt by donors to “gang up.” different multilateral and bilateral donor The Bank’s approach has not fully recognized agencies, but it is also a vital issue within the the differing motivations of donors for engaging Bank itself. Bank projects in different sectors in with LICUS. Although the broad concept of the same LICUS country still often work in fragility is widely understood and accepted, the parallel and do not tap synergies—an example is countries identified by donors as fragile vary. the Community Empowerment and Agricultural The motivations for supporting fragile states Projects in Timor-Leste (IEG 2006c). range from security, to aid effectiveness, to A side effect of the Bank’s decentralization equitable development, to poverty reduction, to to country offices has state building, to peace building and conflict been the concentration Gaining government prevention. of country knowledge confidence is necessary In Afghanistan and Tajikistan, IEG’s fieldwork among local staff and its revealed that major donors did not subscribe to inadequate dissemina- for effective donor a single clear objective. Without a common tion across the country coordination. 37 E N G A G I N G W I T H F R A G I L E S T AT E S Box 2.7: Afghanistan: Lack of a Common Vision term vision for recovery” (World Bank 2005h, among Donors Works against Effective Action p. 7). The 2005 LICUS Update did not, however, spell out the indicators that would be used for Interviewer: The American government and international community such measurement. have said Afghanistan is at the center of the war on terrorism and have poured in billions of dollars. Do they have a vision? Implementation experience The LICUS Initiative’s intended emphasis on Ali Jalali: They all have different visions. You see this clearly in the de- monitoring and evaluation has not yet made its velopment of the security sector. The main pillars of reform—army, way into country strategies, and the focus on police, justice, counter-narcotics, and disarmament—are intercon- results monitoring across LICUS remains negligi- nected, but they were each supported by one “lead nation” from the ble. An assessment of 16 strategy documents in G-7 group, and their approaches can be very different. For example, even LICUS24 carried out as part of this review found if you built a very good police force, the criminal justice sector being that in only 5 countries does the Bank’s strategy developed very weakly by the Italians wouldn’t support it. When you build on a clear articulation of expected arrest a suspect, the police can legally hold him for 24 hours, and then outcomes of the financed activities, including he goes to the judicial sector. Often the suspects buy their way out. clearly defined strategic objectives, well- Source: Interview with Ali Jalali, former interior minister in the Karzai government, Afghanistan. measured baselines, a clear definition of Conducted by Marc Kaufman, The Washington Post, May 28, 2006. outcomes from the Bank’s interventions, reasonable timelines of outcomes, and objective and monitorable milestones and indicators. In The Bank’s donor team, especially those the Stakeholder Survey, a quarter of in-country based in Washington. respondents, 35 percent of Bank respondents, coordination efforts and Addressing the prob- and 42 percent of other donor respondents said modalities need to be lems of coordination that the Bank has defined clear and monitorable informed by the across the departments indicators to measure “success” in LICUS only to of the Bank (such as a slight extent or not at all (appendix Z). objectives of the different among those dealing Two main problems can be identified in players in a country. with public sector effective monitoring and evaluation at the management, conflict country-strategy level. One is insufficient clarity prevention and reconstruction, LICUS, capacity and measurability of expected results. The development, and research) is particularly Transitional Support Strategy in Afghanistan and important in LICUS, where problems are the ISN in Papua New Guinea, for example, lack complex and widespread, and often require a results matrix, and progress indicators are multisectoral solutions. either not defined or not quantified. In Sudan, capacity development is one of the Measure and Monitor Results main themes of the Bank’s Country Reengage- ment Note, but indicators against which Stated approach progress would be assessed are not specified, The 2002 LICUS Task Force emphasized that the making assessment difficult. In Zimbabwe, the Bank’s programs in LICUS should identify Results Summary Matrix in the ISN is presented expected outcomes and indicators to measure in general terms, such as “enhanced knowledge success. The 2005 LICUS Update recognized base” or “enhanced in-country partnerships,” that the “logical corollary of a central focus on with nonspecific indicators such as “progress peace-building and state-building in the Bank’s toward” and “improved response and imple- assistance strategy for fragile states is that short- mentation capacity,” which are impossible to term results measurement should also monitor properly.25 emphasize these dimensions, while continuing The second problem is insufficient selectivity to focus on growth, poverty reduction, and the and prioritization of objectives and indicators. Millennium Development Goals within the long- Few Bank strategies are adequately prioritized, 38 EFFECTIVENESS OF THE BANK’S LICUS APPROACH which hinders effective monitoring. For respondents said that Monitoring and example, while the Liberian Results-Focused the Bank’s overall evaluation have yet to be Transitional Framework Matrix made some program has made a contributions in monitoring, it is almost 40 positive, though small, sufficiently incorporated pages long—hardly the simple planning tool contribution to develop- into country strategies. envisioned by the 2005 Senior-Level Forum on ment in LICUS, and that Development Effectiveness in Fragile States. without Bank support there would have been Transitional results matrixes can potentially less development in the country (figures prove to be useful tools for donor coordination, 2.6–2.10). prioritizing actions, and monitoring country- Given that familiarity with the LICUS ap- level progress. While it is too soon to judge their proach was low among in-country respondents ultimate impact, some examples point to the and other donor respondents, but also among need for closer attention to their implementa- Bank respondents working on LICUS (appendix tion. For example, in Haiti a comprehensive Z), the LICUS approach cannot be assumed to transitional results matrix was prepared, but by have also made a small positive contribution to May 2005 (some 10 months after its adoption), development in LICUS. the reporting system on matrix results had not The percentage of Monitoring and yet started functioning. That made it difficult to projects rated satisfactory evaluation are essential address problems encountered and to assess on the Bank’s Develop- actual implementation. ment Objective (DO) to learning by doing. Monitoring and evaluation are a necessity in ratings for the active LICUS. First, the Bank, like other donors, is still LICUS portfolio increased from 89 percent in fiscal learning what approaches work in LICUS 2000–02 to 91 percent in fiscal 2003–05, and the contexts. Therefore, closely monitoring experi- difference in the percentage of projects rated ences to draw lessons is critical, and learning satisfactory between the non-LICUS LIC and the and sharing needs to become a more LICUS portfolios declined from 5 percentage prominent feature of LICUS work. Second, points during fiscal 2000–02 to 2 percentage given that progress is often slow in these points during fiscal 2003–05 (appendix P). countries, it is important to reassess continu- However, on average, 27 percent of the ally whether the program is on course to projects in the fiscal 2003–05 active LICUS achieve the desired outcomes. Third, a portfolio were at risk of not meeting their constantly changing and volatile LICUS en- development objectives. This is a marginal vironment, where the progress is often nonlin- improvement from 28 percent in fiscal 2000–02. ear, means that program adaptation is essential. Year-on-year, the percentage of projects in the Closely tracking performance will help active LICUS portfolio at risk of not meeting determine when and what kind of adaptation is their development objectives rose by one necessary. percentage point, from 32 percent in fiscal 2002 to 33 percent in 2003, but then declined to 27 Overall LICUS Approach percent in 2004 and 23 percent in 2005. The This section presents the limited available realism rating26 for the LICUS is especially low, at aggregate data relating to the effectiveness of 57 percent for fiscal 2003–05, compared with 80 the overall LICUS approach. The data sources percent for non-LICUS LICs (appendix P). are IEG’s Stakeholder Survey, Bank ratings for The realism of the Bank’s DO rating is an issue. active projects, QAG ratings for realism, IEG An in-depth assessment ratings for closed projects, and IEG CASCR of a sample of projects Bank support has made a Reviews. Trends in the KKZ governance indica- conducted by QAG for small positive tor are also noted. the 2005 Annual Report The results of the Stakeholder Survey show on Portfolio Performance contribution to that the majority of in-country, Bank, and donor (ARPP) found 22 percent development in LICUS. 39 E N G A G I N G W I T H F R A G I L E S T AT E S On average, a fourth of of the portfolio at risk of LICUS Initiative, IEG’s Bank performance not meeting its develop- ratings for closed projects in LICUS show an the fiscal 2003–05 LICUS ment objectives, while improvement from 65 percent satisfactory in portfolio is at risk of not assessments by the Re- fiscal 2000–02 to 72 percent satisfactory in fiscal meeting objectives. gions rate only about 16 2003–05. The 2003–05 figures for Bank percent of projects as at performance in LICUS are more-or-less similar risk. On quality-at-entry, the ARPP found that the to those for Bank performance in non-LICUS Africa Region, especially in low CPIA countries, can LICs (appendix Q). improve quality further through simpler, more IEG’s CASCR Review ratings for country focused project design. The ARPP concluded that strategy outcomes have generally been in the special attention is needed to improve candor and unsatisfactory range, indicating that the realism of project performance ratings during objectives of the Bank’s assistance programs in supervision and to address risky projects through LICUS have been consistently underachieved. more aggressive project restructuring and downsiz- Of the four available CASCRs reviewed by IEG ing when needed. that covered at least part of the second period Only one LICUS project approved since the since the LICUS Initiative began, three were start of the LICUS Initia- rated moderately unsatisfactory or unsatisfac- There has been a tive had closed and been tory, and one was rated moderately satisfactory narrowing gap in the evaluated by IEG by fiscal (only one of these CASCRs—rated moderately percentage of projects 2005. However, some of satisfactory—was for a CAS period fully within rated satisfactory by IEG the projects that had the initiative’s tenure) (appendix R). closed during fiscal This underachievement of objectives was on outcome between 2000–05 were active sometimes the product of overambitious Bank LICUS and non-LICUS during the period since objectives (leading to a scaling down of LICs. the beginning of LICUS objectives). But it was also partly a result of Initiative, and their super- inadequate Bank effort or inappropriate input, vision may have benefited from the initiative. as suggested by the mixed implementation IEG outcome ratings for closed projects in experience documented in this review and in LICUS show an improving trend over time. The IEG’s CASCR Reviews (thus requiring scaling up percentage of closed projects rated satisfactory of effort). on outcome in LICUS improved from 55 percent IEG’s CASCR Review for one country (of the in fiscal 2000–02 to 68 percent (13 percentage four countries for which such reviews, covering at points) in fiscal 2003–05. By comparison, the least part of the period since the LICUS Initiative percentage of closed projects rated satisfactory began, are available) found that the Bank focused on outcome in non-LICUS LICs improved from inadequately on project design and implementa- 74 percent to 76 percent (two percentage tion issues. For example, the emphasis on the points), respectively, for the two time periods. social sectors came at the expense of physical Year-on-year, the percentage of closed LICUS infrastructure, potentially constraining private projects rated satisfactory on outcome by IEG sector growth; social sector projects did not increased from 50 percent in fiscal 2002, prior to appropriately reflect the division of responsibility the LICUS Initiative, to 58 percent in fiscal 2003, in the country’s federal system; and the approach 65 percent in fiscal 2004, and 82 percent in fiscal to community development needed to be more 2005. The corresponding numbers for projects cognizant of fiduciary and capacity issues and in non-LICUS LICs consistent across interventions. IEG’s CASCR IEG ratings for LICUS ranged from 70 to 79 Review for another country found that Bank CASs completed thus far percent (appendix Q). implementation was weak, with inadequate While these results supervision and follow-up in many cases. have mostly been apply mainly to projects No IEG Country Assistance Evaluations have unsatisfactory. approved before the been done thus far for LICUS country programs 40 EFFECTIVENESS OF THE BANK’S LICUS APPROACH Perceptions of the Effectiveness of the LICUS Approach Figure 2.6: The Majority of Stakeholder Figure 2.7: The Majority of Stakeholder Respondents Said the Bank’s Overall Respondents Said Development Would Have Program Made a Small Positive Been Smaller without Bank Support Contribution to Development of LICUS Source: Appendix Z (Stakeholder Survey results). Source: Appendix Z (Stakeholder Survey results). Note: N indicates the number of valid responses. Note: N indicates the number of valid responses. Figure 2.8: The Majority of World Bank Figure 2.9: The Majority of Stakeholder Respondents Said the Bank’s Contribution Respondents Said World Bank Lending and to Development Was Greater Than That Grant Support to LICUS Has Achieved Its of Other Donors Intended Results to a Moderate or Slight Extent Source: Appendix Z (Stakeholder Survey results). Source: Appendix Z (Stakeholder Survey results). Note: N indicates the number of valid responses. Note: N indicates the number of valid responses. 41 E N G A G I N G W I T H F R A G I L E S T AT E S Figure 2.10: Comparing the Pre-LICUS and Post-LICUS Initiative Periods, the Majority of Stakeholder Respondents Said There Is Improvement in the Effectiveness of the Bank’s Lending and Nonlending Support Lending Nonlending 75 75 50 50 Percent 25 25 0 0 Large positive Small positive No change Negative Large positive Small positive No change Negative World Bank Donors In-country Source: Appendix Z (Stakeholder Survey results). Note: Number of valid responses ranges from 33 to 35 for other donors and 20 to 21 for in-country stakeholders and is 248 for the Bamk. covering the period since the LICUS Initiative The first part of this chapter and table 2.3 show began (appendix R). The Bank needs to that the Bank’s LICUS Initiative has been more determine what it is likely to be able to achieve effective with respect to some principles or specific in the varying LICUS business model groups, aspects of them (staying engaged, supporting set realistic objectives, and be held accountable macroeconomic reforms, delivering physical for the achievement of those objectives. infrastructure through alternative mechanisms, While not fully attrib- and coordinating with other donors at the interna- IEG ratings for LICUS utable to the Bank, there tional policy level) than others (supporting the CASs completed thus far is a deteriorating trend transition from immediate post-conflict re- in KKZ’s governance construction to development, contributing to cap- have generally been indicator for LICUS in acity development and governance, ensuring unsatisfactory. the period since the selectivity and prioritization in reforms, translating LICUS Initiative began.27 political understanding into country strategies, The deterioration is similar in LICUS and non- and donor coordination at the country level). LICUS LICs, but for LICUS, the decline is from The second part of the chapter provides a already low levels (chapter 1, figure 1.7). patchwork of aggregate data. The Stakeholder Survey indicates a small positive contribution to Conclusion development of the Bank’s overall program in The implementation experience across the core LICUS—a view that refers to Bank support country-level LICUS principles has been mixed. generally, and not to the LICUS approach per se. 42 EFFECTIVENESS OF THE BANK’S LICUS APPROACH Table 2.3: Implementation Experience with the Core Country-Level LICUS Principles LICUS principle Implementation experience rating Stay engaged Substantial Anchor strategies in stronger sociopolitical analysis Medium • Political understanding • Medium-substantial • Internalizing political understanding in strategy design and implementation • Medium-low Promote domestic demand and capacity for positive change Low Support simple and feasible entry-level reforms Medium-low • Macroeconomic reforms • Substantial • Delivery of physical infrastructure • Substantial • Transition from the immediate post-conflict reconstruction phase to the development phase • Low • Selectivity and prioritization • Low Explore innovative mechanisms for social service delivery Medium Donor collaboration Medium • At international policy level • Substantial • At country level • Medium-low Measure and monitor resultsa Low Source: Fieldwork and thematic background analysis undertaken for this review. a. Not specifically mentioned as a separate core principle by the Bank, but included by IEG because it is pivotal to the Bank’s learning-by-doing LICUS agenda. 43 Chapter 3: Evaluation Highlights • The Bank continues to rely almost exclusively on the CPIA to iden- tify LICUS, although the CPIA does not sufficiently capture some key aspects of state fragility and conflict. • The recently introduced LICUS business models are likely to permit a more tailored response to different groups of LICUS. • But the operational guidance contained in the business models needs to be sharpened and the extent of operational usefulness of the business models tested through implementation. • The Bank needs to review its aid-allocation criteria in light of its ob- jectives for LICUS and ensure that LICUS are not under- or over-aided. 3 Operational Utility of the LICUS Identification, Classification, and Aid-Allocation System his chapter assesses the operational utility of three aspects of the LICUS T approach: identification of LICUS, classification of LICUS into business models, and the aid-allocation system for LICUS. Operational utility is assessed against the objectives of the LICUS Initiative. LICUS Identification modation of political Despite the increased dissent and of conflict, Distinguishing LICUS from other low-income such as political insta- focus on state building countries bility and security or and peace building, the The Bank’s 2002 LICUS Initiative was motivated susceptibility to con- criteria to identify LICUS by general aid effectiveness concerns within the flict.1 The Bank has rec- have not been refined. Bank. The 2005 Fragile States Report focused ognized that security- the initiative on state-building and peace- related variables are missing from the CPIA: building objectives. “The CPIA . . . does not measure the reach of Despite this focus, the criteria used to service provision and administrative control identify LICUS were not refined to capture these across geographical territory and it devotes aspects sufficiently. The CPIA, on which the greater weight to the economic, administra- Bank relies almost exclusively to identify LICUS, tive, and service delivery functions of the state has its advantages. Most important, because it is than to institutions dealing with security and based on policy performance (not outcomes), rule of law” (World Bank 2005h, p. 7). the CPIA has the conceptual advantage of • Second, the CPIA gives equal weight to all its reflecting more recent policy situations, constituent elements, although some of them whereas outcomes may be the result of and may have much more bearing on state build- capture past policy situations. However, there ing and peace building than others. It could are also several shortcomings: be argued, for example, that improvements in the efficiency of resource mobilization or in the • First, the CPIA fails to capture sufficiently some equity of public resource use should take key aspects of state fragility, such as accom- precedence over some macroeconomic indi- 45 E N G A G I N G W I T H F R A G I L E S T AT E S cators in the CPIA if state building is a key rize them into smaller groupings, as the Bank objective. has recently done using business models. The • Third, there is up to a 24-month lag between heterogeneity of LICUS was recognized by the the period being measured by the CPIA rating 2002 LICUS Task Force, which identified six and the time when the rating actually informs categories of LICUS: policy-poor but resource- policy decisions. This makes it more difficult rich; exceptionally weak government capacity; to identify—in a timely and effective way— government/donor lack of consensus; limita- policy improvements or deteriorations in LICUS tions on engagement; countries emerging from that can guide resource allocation. conflict; and countries in early stages of a • Fourth, the confidentiality surrounding the domestically generated reform process. The CPIA (removed in June 2006), and consequent 2003 Implementation Overview Report (World perceptions of the lack of transparency and ob- Bank 2003a, pp. 3–4) also differentiated its jectivity, have not helped the dialogue among guidance by type of LICUS—for example, donors (who have had to devise proxies such countries in weak transition, countries with as the CPIA-quintile-based definition of fragile no progress or deterioration, post-conflict states, since quintiles are publicly available, countries, and more stable and active countries. but not individual scores2) or among country This classification evolved into a more sys- clients (many of whom question their CPIA tematic, fourfold typology of business models in status relative to that of others). the 2005 Fragile States Report: deterioration, prolonged political crisis or impasse, post- A stronger approach to the identification of conflict or political transition, and gradual LICUS will require an analytical framework that improvement (appendix D). These business more explicitly focuses on the objectives of the models are based on the extent of consensus LICUS Initiative. Given the Bank’s state-building between donors and government on develop- and peace-building objectives and the shortcom- ment strategy and the pace and direction of ings of the CPIA, the Bank will need to reexamine change. the appropriateness of the CPIA criterion to The first two types of LICUS (those experi- identify LICUS, and supplement it as needed.3 encing deterioration and those facing prolonged Donors and researchers have come up with political crisis or impasse) represent countries different lists of difficult countries, using different where there is little consensus between donors definitions (Foreign Policy 2005; Van de Walle and government on development strategy. The 2005). To the extent that the Bank’s current list of other two (those that are post-conflict or in LICUS misses some political transition and those experiencing The criteria used to relevant countries, the gradual improvement) represent countries with identify LICUS need to effectiveness of the such consensus. The pace and direction of derive from the objectives Bank’s assistance is change are then used to classify LICUS within reduced. At the same each of these two groups for a total of four of the LICUS Initiative. time, to the extent that business models. These business models are the Bank’s current list likely to permit a more tailored response to includes some countries not fully relevant to its different groups of LICUS but have yet to be fully objectives, Bank resources that could be used to developed. address those objectives are not. Currently, for instance, the operational guidance on state capacity and accountability LICUS Classification contained in each of the business models is broad and insufficiently customized to the Differentiating within the LICUS group of institutional characteristics of countries that fall countries into various business models. For example, it LICUS are a highly diverse group (see chapter states: “focus on transparency, dialogue and 1), and it is useful for policy purposes to catego- maintaining institutional capital to facilitate 46 O P E R AT I O N A L U T I L I T Y O F T H E L I C U S I D E N T I F I C AT I O N , C L A S S I F I C AT I O N , A N D A I D - A L L O C AT I O N S Y S T E M eventual turnaround” (deterioration business selectivity of the PBA The business models are model); “focus on institutional analysis, system has increased dialogue and counterpart training” (prolonged over the years5 and likely to permit a more crisis or impasse business model); “support for fewer IDA funds have tailored response to a broad state-building agenda, through institu- been available for different groups of LICUS. tion building and, where appropriate, develop- countries with weaker ment policy operations with robust oversight policies, institutions, and governance. This has mechanisms and sector programs” (post- raised the question of whether LICUS are receiv- conflict or political transition business model); ing appropriate amounts of IDA funding. and “development policy operations, where Adjustments to the PBA have resulted in appropriate and restricted in scope, supported increased IDA financing, including some post- by sector and capacity-building projects and conflict LICUS and LICUS experiencing political with strong oversight mechanisms” (gradual transitions (box 3.1). Indeed, during fiscal improvement business model). 2003–05, post-conflict LICUS received a large Further refinement of the business models by share of the IDA financ- more explicitly factoring in differences in ing to LICUS, averaging Refining the business capacity to perform core state functions (such $8.1 per capita annually, as resource generation, resource allocation, compared with $1.5 per models by more explicitly basic social service and infrastructure provision, capita in non-post- factoring in differences in and political accommodation of dissent and conflict LICUS. capacity would help the security) is needed to enable the Bank to better All seven post-conflict Bank to better address its reflect the institutional situations of different LICUS received higher groups of LICUS in its response, and thereby to per capita IDA financing, state-building objective. meet its state-building objective better. For even when compared example, the Bank’s institutional response in with the average for non-LICUS LICs (figure 3.1). political-transition LICUS, where state capacity Yet it remains far from clear whether the current to perform some or all core functions is lacking, levels of IDA ensure that LICUS are not under- or will have to be different from that in political- over-aided. transition LICUS with capable states. The aid-allocation issue has once again come The experience emerging from the to the fore with some research that questions the implementation of the Bank’s differentiated empirical evidence for the positive link between business models needs to be systematically policies and aid effective- monitored and will comprise the ultimate test of ness6 (which underlies The extent of operational the operational relevance of the business the PBA), and other models. Implementation data should be used to research that argues that usefulness of the business ascertain how much value the business models aid can be effective in models needs to be tested add over the CAS-driven country-by-country promoting sustainable through implementation. approach. policy turnarounds in failing states by building and strengthening the Aid-Allocation System for LICUS preconditions for reform or by enhancing the Twenty-three of the 25 LICUS are IDA-only chances that the reform will be sustained once it is countries for which IDA financing has histori- set in place (Chauvet and Collier 2004). The latter cally been allocated based on the Performance- research finds that potential returns from aid to Based Allocation (PBA) system. Implicit in the LICUS can be extraordinarily high, even though PBA system is the assumption that aid is more the risks of failure are substantial (Chauvet and effective in environments with good policies, Collier 2005). For its part, the Bank has yet to institutions, and governance, with the CPIA address the aid-allocation issue for LICUS in a way rating used to determine institutional quality that reflects its objectives for these countries and across developing countries.4 The policy ensures that LICUS are not under- or over-aided. 47 E N G A G I N G W I T H F R A G I L E S T AT E S Figure 3.1: Highly Variable per Capita IDA and Trust Fund Financing across LICUS during Fiscal 2003–05 Source: Trust Funds and World Bank databases. Box 3.1: Adjustments to IDA’s Performance-Based Allocation System That Affect LICUS • Agreement under IDA 12 authorizing special pre-arrears IDA countries in the aftermath of major natural disasters in clearance allocations to aid-eligible countries coming out of cases where the existing allocation would not allow for a suf- active conflict and in the process of normalizing IDA relations. ficient response. • Agreement under IDA 13 to provide exceptional allocations • Agreement under IDA 13 (continued under IDA 14) to have a to countries emerging from severe conflict in support of their special provision for regional integration projects. Up to SDR recovery and in recognition of exceptional need. 300 million of such projects yearly are envisioned under IDA 14. • Agreement by the IDA deputies during the IDA 13 Mid-Term • Agreement during IDA 14 that additional allocations may be Review to stretch out the phasing of the special post-conflict provided on a one-time basis to countries in the process of allocations to fit the cycle of absorptive capacity of receiv- reengaging with IDA after a prolonged period of inactivity on ing countries, while maintaining the same total allocation. the basis of a strong transition plan with concerted donor sup- • Agreement (since 1985) to provide exceptional access to port. The exception is to be used only after all other options IDA resources to small island economies, which have per have been exhausted and is not intended to last for more than capita incomes above the IDA eligibility cut-off but have no two years, with a possible additional year, subject to strong or very limited creditworthiness, which limits or precludes ac- performance. cess to IBRD borrowing. • Agreement during IDA 14 also included exceptional IDA fi- • Agreement under IDA 13 to provide additional allocations to nancing for natural disaster response and regional projects. Sources: IDA 2002, 2004, 2005. 48 O P E R AT I O N A L U T I L I T Y O F T H E L I C U S I D E N T I F I C AT I O N , C L A S S I F I C AT I O N , A N D A I D - A L L O C AT I O N S Y S T E M Early findings from the pilot implementation as develop criteria that Trust funds have played of the 12 Principles of International Engagement enable it to determine an important role in in Fragile States (OECD 2005d) show that there assistance volumes that is a group of countries that receives low aid flows reflect its objectives for supplementing IDA in relation to need and governance indicators, LICUS and ensure that financing, but have been compared with other countries with similar these countries are not highly variable across governance indicators.7 Overall, eight coun- under- or over-aided.8 tries—Burundi, the Central African Republic, While it does not nec- LICUS. Chad, Guinea-Bissau, Niger, Sierra Leone, essarily follow that more Tajikistan, and Togo—show the greatest im- should be provided to LICUS, the Bank needs to balances and appear to attract relatively little make a strategic assessment of the appropriate donor attention. form and level of financial engagement in LICUS. Although trust funds have been an important Earlier discussions have focused only on supplement to IDA financing, these funds, for the “more” or “less” aid but have not established most part, were concentrated in a few LICUS that “how much more” or “how much less.” already benefit from post-conflict IDA financing Whether and to what extent the Bank’s aid- (for example, Afghanistan, the Democratic allocation criteria should be based on factors Republic of Congo) (figure 3.1). During fiscal other than policy performance—such as levels 2003–05, average annual trust fund financing per of other donor assistance, assessment of capita in post-conflict LICUS was $5.6, compared potential risks and rewards, and regional and with $0.2 in non-post-conflict LICUS. global spillovers—needs to be examined, In addition to the introduction of exceptional keeping in mind that aid is limited and trade- post-conflict IDA financing, the Bank has made offs will have to be made. While the aid alloca- several other adjustments to the PBA system. tion issue goes beyond the LICUS Initiative, it Some key questions that need to be addressed remains an issue of crucial importance for the include the appropriate number and size of achievement of the Bank’s objectives in adjustments to the PBA; the basis for specific LICUS. adjustments, including the robustness of the Beyond its own financing, the Bank needs to high-risk/high-reward argument; and the help address the gap in the international aid countries that should or should not receive architecture in relation to the aggregate alloca- exceptional treatment. tion for fragile states. Elements of a strategy to address this would include strengthened efforts Conclusion at coordinated donor planning in which the The Bank needs to conduct a technical review of Bank would have a role (and for which the the cumulative effect of the various adjustments OECD Watch List on Fragile States may be a to the PBA system on financing to LICUS, as well starting point) (OECD 2005e). 49 Chapter 4: Evaluation Highlights • Substantial progress has been made in expanding analytical work and in developing guidance notes on specific topics. • Progress on human resource reforms remains unsatisfactory, and LICUS do not yet consistently attract staff capable of effectively addressing the difficult LICUS situations. • Learning by doing requires much more active and ongoing stock taking and knowledge sharing than currently takes place and needs to be a more prominent feature of LICUS work. • The Bank needs to ensure a receptive institutional environment and management support for staff to feel at ease sharing negative experiences. • Confidentiality and learning by doing are conflicting objectives. • There is significant confusion between the roles of the LICUS and Conflict Prevention and Reconstruction Units. 4 The Bank’s Internal Support for LICUS Work he 2002 LICUS Task Force Report identified four areas in the Bank’s in- T ternal support for LICUS work that needed to be addressed to en- hance effectiveness in LICUS: analytical work, staffing and incentives, operational policies and procedures, and management attention and opera- tional guidance. The report emphasized the need for an internal • The Bank increased ESW budgets have culture shift at the Bank to enable implementa- budgets for ESW and increased at the tion of the LICUS approach (World Bank 2003a, technical assistance to p. 4). The Bank’s 2005 Fragile States Report LICUS at the aggregate aggregate level. reinforced the need for such a shift, supporting level during fiscal all four change areas identified by the 2002 2003–05 compared with fiscal 2000–02. LICUS Task Force Report. This chapter assesses • Resources for ESW and technical assistance the Bank’s progress in addressing the four areas have not increased in six LICUS. of change. • While the overall quality of analytical work in LICUS has improved, process aspects still need Analytical Work attention—the preparation of analytical work is insufficiently undertaken in coordination Stated Bank approach with governments and donors, and this has ad- The first area of change advocated in the 2002 versely affected its policy influence. LICUS Task Force Report called for increased analytical work and changing the overall balance With respect to the overall balance between knowledge and finance to be more between knowledge and finance, while this heavily weighted toward the former. balance still favors finance, even in fiscal 2003–05, the change compared with fiscal Implementation experience 2000–02 is in the right direction for adminis- Implementation progress on support for analyt- trative budgets, but only marginally so for ical work has been substantial and was discussed lending. This statement is based on defini- in chapter 2 under “Stay Engaged.” The main tions of “knowledge” and “finance” presented conclusions are as follows: in the note to table 4.1. 51 E N G A G I N G W I T H F R A G I L E S T AT E S Table 4.1: Overall Balance Between Knowledge and Finance for Administrative Budgets and Lending Knowledge (US$ million) Finance (US$ million) Knowledge: Finance Fiscal Fiscal Percent Fiscal Fiscal Percent Fiscal Fiscal 2000–02 2003–05 change 2000–02 2003–05 change 2000–02 2003–05 Lending 215 397 84 2,265 3,740 65 0.09 0.11 Trust funds 774 403 –48 384 1,571 309 2.02 0.26 Administrative budget 23 48 109 80 113 41 0.29 0.42 Total 1,012 848 –16 2,729 5,424 99 0.37 0.16 Source: World Bank database. Note: Lending for “knowledge” comprises lending for IDA, IBRD, and special financing projects that were either freestanding capacity development projects or where capacity develop- ment accounted for at least 80 percent of the project cost. Lending for “finance” comprises all other IDA, IBRD, and special financing projects. Trust funds for “knowledge” comprise trust funds for technical assistance, and trust funds for “finance” comprise all other trust funds. Administrative budgets for “knowledge” comprise the administrative budget for ESW and tech- nical assistance, and administrative budgets for “finance” comprise the rest of the administrative budget. Staffing and Incentives • The development of field postings in LICUS as more attractive to staff by including a 5 percent Stated Bank approach premium on top of the existing hardship al- The 2002 LICUS Task Force Report pointed to lowance for working in LICUS, more generous the need to ensure high-quality staff in LICUS rest and relaxation allowance in some LICUS and noted the importance of providing the right (such as Afghanistan), and allowing sector spe- incentives to encourage staff to work on LICUS. cialists based in neighboring countries to visit the LICUS country rather than live in it (as in Implementation experience the case of Tajikistan). The forthcoming Strengthening the Organiza- tional Response to Fragile States paper currently There has been little follow-up to these initial under way is welcome, even if late. Yet progress attempts. The proposed human resource on staffing and incentive working group was not formed until May 2005. Progress on staffing and issues remains unsatis- While the accelerated development program has incentives remains factory three years into found some management support, there are unsatisfactory three years the LICUS Initiative. concerns within human resources about the The Bank made initial implications of establishing yet another separate into the LICUS Initiative. attempts to address the program for staff when human resource reforms staffing problem in LICUS through three initiatives: aim to simplify and streamline. Furthermore, semistructured interviews • The identification of elements of a strategy to found that the majority of staff had not heard of address the staffing problem in LICUS (ap- any specific changes in human resource policy pendix S) with the intention that they would with respect to working in LICUS, while several be further addressed in more detail by a human commented that small changes, such as the 5 resources working group percent premium or generous rest and • Introduction of an accelerated development relaxation policies, were insufficient to make a program to provide a bridge between young difference. professional and higher-level staff positions In the Stakeholder Survey, the majority of with a specific LICUS focus by giving staff di- Bank respondents said that there has been no rect exposure to LICUS contexts and issues change when working in LICUS with respect to while guaranteeing their re-entry into head- the following human resource matters—overall quarters at a position consistent with their ca- career prospects, overall financial compensa- reer trajectory prior to working on LICUS tion, realism in expectations by Bank manage- 52 THE BANK’S INTERNAL SUPPORT FOR LICUS WORK ment about what can be accomplished, level of Understanding of Staffing numbers and support from Bank management, and efforts country circumstances is quality are not an issue made by the Bank to ensure personal security often best achieved and safety (figure 4.1). through substantial field in high-profile LICUS, but While staffing numbers and quality are less of presence, though that they are in other LICUS. an issue in some high-profile post-conflict alone is not enough. countries, finding specialist staff to work on Internalizing analysis throughout all involved other LICUS, especially in field offices, remains a Bank units, and applying its lessons to all interven- problem. In 2005, about 70 percent of LICUS did tions, is equally important. not have a professional specialist in the field, In Cambodia, for example, the Bank’s field compared with about 25 percent for non-LICUS presence has significantly improved understand- LICs.1 The Bank’s recent move to state building ing of the political situation, but discussions with and peace building as central objectives would country team members and other stakeholders also suggest the need to ensure adequate staff suggest that this knowledge may still be highly with public sector management skills and staff concentrated among a few managers and staff who are comfortable seeking and using (if not (mostly in the country office and Bangkok hub), necessarily producing) political knowledge in with relatively limited decision making. dissemination to the Field presence alone In the Stakeholder Survey only about half of broader country team. is insufficient for Bank respondents said that their colleagues The issue appears to working in LICUS are to a large extent competent. have shifted from a adequately addressing While 63 percent of other donor respondents partial understanding of country circumstances. said that Bank staff who work in LICUS are the political realities of competent to a large extent, only 33 percent of Cambodia to one of where this knowledge is in-country respondents agreed (appendix Z). located within the Bank’s country team and how Figure 4.1: Change in Various Factors When Working on LICUS Source: Appendix Z (Stakeholder Survey results). Note: Number of valid responses ranges from 213 to 238. The question in the survey did not differentiate between staff who had worked on a LICUS country and those who had worked on a non-LICUS country in their previous assignment. 53 E N G A G I N G W I T H F R A G I L E S T AT E S it is used to guide decision making in strategy delivery in difficult environments, tends to inflate and program implementation. The concentra- expectations of staff impact. In a LICUS country tion of in-depth country knowledge among a this is problematic because of the very strong few staff implies that only some Bank activities likelihood that programs will not work out as and interventions benefit. In general, greater planned. Several staff noted the importance of knowledge transfer is needed between donor working on both a non-LICUS and a LICUS country offices and headquarters-based country country so that they could rely on a stream of and sector staff. outputs delivered in the former to boost their Effective donor Effective donor co- overall performance assessment. ordination by the Bank Clarity among Bank staff about how success is coordination depends, in requires having the right measured and interpreted is especially part, on the personalities kind of staff involved in important with respect to the Bank’s relatively of field staff. the country. In semi- harder-to-measure state-building and peace- structured interviews, building agendas. To avoid creating unrealistic several donors emphasized that coordination is expectations and discouraging staff from unusually susceptible to the strengths and the working toward these huge agendas, it will be foibles of the individuals involved. More critical to establish realistic interim benchmarks appropriate training for staff posted to difficult for state-building and peace-building outcomes field assignments and improved incentives for which Bank staff will be accountable. How within the Bank that encourage staff to collabo- will the Bank measure success in these areas and rate with other donors might ameliorate these for what actions and results will staff be idiosyncratic risks.2 rewarded? Bank staff expressed Similarly, staff needs clear signals about how Staff evaluations focus on concern about the much risk it is reasonable to take in LICUS. the delivery of outputs or performance criteria for Given that these are high-risk countries, will products, while change in staff working on LICUS, there be fewer penalties for failure? How will and in particular how the success be judged in staff evaluations and career LICUS is often process Bank is interpreting development—how much will the achievement oriented and “success” in these of small, incremental steps be rewarded? The incremental. countries. The fact that answers to these important issues will influence LICUS are, by definition, staff behavior. often “off-track” or making only modest headway against zero or first-generation reforms Operational Policies and Procedures leaves staff wondering how best to present their contribution to Bank outputs. Stated Bank approach In semistructured interviews, Bank staff The third area of change identified by the 2002 noted that change is often highly incremental LICUS Task Force Report was for the Bank to and process oriented in LICUS. They stated that clarify further, disseminate, and revise its it is sometimes difficult to ascribe to specific operational policies and procedures for LICUS interventions. In con- work to enable a faster and more effective To ensure the desired staff trast, staff evaluations response in LICUS. behavior, the Bank needs are centered on the delivery of tangible out- Implementation experience to communicate to staff puts or products. Progress on adapting relevant Bank operational how success will be Such a strong em- policies and procedures to the special circum- measured and what phasis on the delivery of stances of LICUS has been slow and small. No outputs, as opposed to specific operational policies in the Bank govern constitutes a reasonable the processes necessary LICUS work, and existing operational policies do level of risk-taking. to bring about effective not provide adequate guidance for working in 54 THE BANK’S INTERNAL SUPPORT FOR LICUS WORK the full range of LICUS environments. The more appropriate. Pend- The LICUS concept has LICUS concept has been superimposed onto the ing finalization, it is not been superimposed onto Bank’s existing set of Operational Policies and clear if the conceptual Bank Procedures (OPs/BPs). problems in applying the Bank’s existing Appendix T summarizes the main OPs and OP/BP 8.50 to LICUS policies with no specific BPs identified by the LICUS Unit as particularly conditions will be ade- policy governing LICUS relevant for LICUS. quately addressed (box Bank respondents point to a host of difficul- 4.2). work. ties in working with existing OPs/BPs in LICUS. Procurement procedures are considered too One of the main difficulties with these OPs/BPs cumbersome in most LICUS. The Bank’s official is the assumption that there is sufficient capacity stance on procurement is that nothing in the in the recipient government/administration to policy framework governing procurement engage in the Bank’s investment lending prevents the Bank from procedures. working effectively in Some recent changes in For example, although OP/BP 8.503 allows LICUS, as long as the operational policies have staff to speed up the project preparation process Bank’s activities are in emergency situations, it requires staff to refer transparent, legitimate, increased flexibility, but back to standard Bank OPs for the remaining and accountable. On other changes have been phases of the project, including project paper, this would appear pending for a long time. implementation. The expectation is that a to be largely true. There normal way of doing business is possible, largely are provisions for projects executed by the Bank premised on the idea of a single event or disaster and by third parties in the event that recipient that still leaves government institutions intact. execution is not possible. There are provisions In many LICUS, and particularly in those for local competitive bidding, national competi- emerging from conflict, such an assumption is tive bidding, and single sourcing of goods and rarely tenable and can lead to undue delays (box services if there is no alternative or if alternatives 4.1). The result is that emergency operations would be too high risk. designed under OP/BP 8.50 are prepared quickly However, staff noted but can take several months to become effective. difficulties in finding the Procurement procedures Some recent revisions to OPs are relevant and right procurement solu- are seen by Bank staff as effective overall, but others have been slow in tion in specific LICUS coming. The Bank has undertaken or is and noted that procure- too cumbersome in most undertaking important revisions to some ment staff were not LICUS. OPs/BPs, such as OP/BP 6.04 and OP/BP 8.50. The Bank has revised OP/BP 6.0 on Bank financing to allow for Bank financing of recurrent costs, local Box 4.1: OP/BP 8.50: A Major Source of Delays expenditures, and local taxes and duties, provid- in Liberia ing much-needed flexibility given weak state capacity and the low tax/gross domestic product Bank staff on the Liberia country team noted major difficulties in draw- (GDP) ratio in most LICUS. Bank staff working ing on OP/BP 8.50 to prepare project proposals under the newly cre- on Tajikistan noted a major change in flexibility ated Liberia Trust Fund. The presence of a transitional government in as a result of revisions to OP/BP 6.0. Work contin- Liberia made it very difficult to engage through a framework largely de- ues on revisions to OP/BP 8.50, which have been signed with more capable, IDA-eligible countries in mind. under way since late 2003. The result was protracted dealings among different Bank depart- The proposed revisions to OP 8.50 will ensure ments, causing serious delays. The high internal transaction costs ap- that emergency procedures are not applied to the peared to contradict the principle of “emergency” and of the need for more unstable and unpredictable LICUS environ- speed and flexibility in a LICUS context. ments, where speed may not be appropriate and Source: Interview with the Bank’s Liberia Country Team, 2005. a more deliberate approach would, in fact, be 55 E N G A G I N G W I T H F R A G I L E S T AT E S Box 4.2: Conceptual Problems in Applying holder Survey, more than 60 percent of Bank OP/BP 8.50 to LICUS Environments respondents said that the Bank’s procurement procedures are not adapted or are only adapted OP 8.50 on emergency recovery assistance, currently also applicable to a slight extent to the low-capacity or higher- to post-conflict reconstruction, focuses mainly on events of short du- risk environment of LICUS (appendix Z). ration, such as earthquakes, floods, or hurricanes, that do not affect Procurement problems have typically led to institutions and that require a rapid response to rebuild physical in- significant delays in Bank operations in LICUS. In frastructure. Civil conflicts, in contrast, are of a protracted duration and the Central African Republic, it took more than destroy the social fabric of a country. Their causes typically go back six months to deliver to the government three in time and result in situations that require long-term development ef- 4 × 4 vehicles badly needed to cut down on forts. Furthermore, unlike natural disasters, civil conflicts require major blatant smuggling abuses, while millions of efforts in dealing with institutional frameworks and macroeconomic dollars in forgone customs receipts were lost conditions. each month. Similarly, in countries where the UN Source: IEG 1998. is a major executing agent, staff point to a huge amount of time spent clarifying whose procure- ment and financial management procedures always aware of the flexibility provided for in the apply, leading to a loss of crucial time in respond- current procurement guidelines. Several ing to changing circumstances in the country. operational staff noted high levels of risk The Bank has begun to discuss the need for aversion among procurement staff with respect staff training on procurement options in LICUS. to LICUS. The perception among staff is that That would involve both procurement staff and project designs or grant proposals are expected program team members. It would also consider to conform with generic low-income country or the possibility of carving out a core group of IDA standards rather than being adapted to LICUS (those with limited Bank country the different risk-reward presence) in which the UN is encouraged to take Because corruption is a contexts of LICUS. over procurement altogether on behalf of the real danger in LICUS, Box 4.3 provides Bank. examples of procure- Discussions on both of these topics are at an procurement should be ment problems faced in early stage, and no timetable has been set. The closely watched. LICUS. In the Stake- merits of these arrangements will need to be assessed. The key is to find procurement solutions that do not hamper the Bank’s Box 4.3: Examples of Procurement Problems operational work in LICUS, while at the same in LICUS time ensuring that the Bank’s fiduciary standards are not compromised. In Tajikistan, the threshold for international competitive bidding is con- Compared with some other portfolios in the sidered too low, given the weak interest from foreign bidders. Bank, quality-at-entry of safeguard compliance In Liberia, Bank execution for procurement is used widely because of in LICUS projects has been relatively better (88 the risk environment, but Bank execution does not allow for certain types percent moderately satisfactory or better), of purchases, especially of equipment. Using a third party for execution though it is still short of the zero tolerance (such as the UN Office of Project Services) allows the Bank to get around policy. this, but at a cost that, in a budget-constrained program, diverts scarce Safeguard compliance during implementation resources from other uses. is much weaker (37 percent moderately satisfac- Bank staff note a lack of lesson sharing across countries and regions tory or better) than compliance at entry and is on procurement issues, resulting in a great deal of “reinventing the similar to some other portfolios. Safeguard wheel,” particularly in arrangements with other development partners such compliance during implementation warrants as multidonor trust funds. attention by the Bank and the borrower and Source: Bank staff interviews. could be challenging, given the weak capacity environment in LICUS (appendix U). 56 THE BANK’S INTERNAL SUPPORT FOR LICUS WORK Other Bank operational procedures are also covering a larger, more The key is to find in need of adaptation. In the Stakeholder Survey, “successful,” or higher- procurement solutions half of Bank respondents said that the Bank’s profile country. Coun- project preparation is adapted to the low- tries that are LICUS but that do not hamper the capacity or higher-risk environment of LICUS to of low international Bank’s operational work a large or moderate extent. The other half said interest tend to lose out, in LICUS, while at the that it is only so to a slight degree or to no both to better perform- extent. Bank respondents were also roughly ers and to front-burner same time ensuring that equally divided on the extent to which the LICUS. In the Stake- the Bank’s fiduciary Bank’s project supervision is adapted to the holder Survey, about 40 standards are not environment of LICUS. With regard to both the percent of Bank respon- Bank’s financial management procedures and dents said the Bank’s compromised. the Bank’s legal framework, about 60 percent of lending and nonlending Bank respondents said they are not at all support to LICUS has only slightly or not at all adapted or only slightly adapted to the LICUS attracted adequate management attention or environment (appendix Z). involvement (appendix Z). IEG’s fieldwork for this review noted apprecia- Management Attention and Operational tion from field staff for the role played by the LICUS Guidance Unit. In particular, they appreciated its advocacy and strategic role. Several Bank staff referred to the Stated Bank approach importance of the unit in promoting the LICUS The fourth area of change identified by the 2002 agenda with external partners. However, a number LICUS Task Force Report was the need for a of other staff were unclear about what exactly the more balanced approach to LICUS country LICUS Unit did. programs, underpinned by enhanced institu- The majority of Bank respondents in the tional support and management attention. The Stakeholder Survey said the Bank’s LICUS Unit has report also identified the need for further clarify- been effective to a large or moderate extent, with ing and disseminating good practices for LICUS. regard to providing access to trust funds as well as substantive support for country strategy develop- Implementation experience ment and implementa- Progress on ensuring that LICUS managers have tion. However, the ma- Country directors are less access to the Bank’s senior management has jority also said that the likely to pay attention to been substantial, but this has yet to be translated Bank’s LICUS Unit has into adequate management attention that yields been only slightly or not at LICUS if they are also clear improvements in human resource policies all effective in providing covering a larger, more and incentives to undertake LICUS work. The substantive support for “successful,” or higher- introduction of quarterly LICUS meetings with projects, providing sub- profile country. Regional vice presidents and country directors, stantive support for chaired by the managing director, has helped research or analytical work, unlocking procedural spotlight specific countries and helped staff or policy difficulties at headquarters, and facilitating navigate technical and procedural hurdles when donor collaboration and harmonization (figure 4.2 they are brought to light. and appendix Z).5 Sector staff (that is, sector Staff also noted greater attention to LICUS directors, sector mana- issues at the Regional vice presidential level and gers, sector economists, Operational staff from OPCS. In Zimbabwe, the country team felt and sector specialists) appreciate the role played that the LICUS Unit had ensured high-level senior were statistically signifi- by the LICUS Unit, but the management attention to a difficult situation. cantly more likely to The attention of individual country directors report greater effec- extent of knowledge is more limited, especially if they are also tiveness of the LICUS about its role varies. 57 E N G A G I N G W I T H F R A G I L E S T AT E S Figure 4.2: Bank Respondents’ Views on Effectiveness of LICUS Unit Source: Appendix Z (Stakeholder Survey results). Note: Number of valid responses ranges from 162 to 169. Seventy-five percent of Bank headquarters staff, staff with less than two years of experience, or staff not working full time on LICUS did not respond to these questions. Unit for all categories, compared with other donor processes, and ensuring coherence staff. between external policy debates and ongoing Staff capacity within the LICUS Unit is a country work. potential concern, especially with regard to the Guidance notes on specific topics have been provision of substantive support. With a very prepared for staff working on LICUS, although small staff ,6 the amount of support that the only recently. Few interviewees mentioned any LICUS Unit can provide is inevitably con- of these guidance notes directly. Based on strained. While the unit’s staff work hard to country experience, the LICUS Unit has distilled respond and to be consistent across Regions guidance on a number of important issues and country teams—and this is appreciated by (appendix V) and has fed it into both operational field-level staff—gaps and inconsistencies arise advice to country teams and broader external because of the sheer breadth and complexity of policy debates. the agenda. However, country-level staff, in semi- Unit staff themselves note particular difficul- structured interviews for this review, said that the ties in keeping up with lack of country knowledge sometimes constrains Guidance notes have the regular review of technical input from the unit. They also noted been prepared on specific country strategies and that advice from the unit has tended to be in topics, but the topics are other operational sup- response to specific problems rather than being port to the Regions, comprehensive, strategic, and systematic. not widely known among fielding staff at short In some key areas, the Bank’s operational staff. notice to support multi- approach is clearly lacking—for example, on state 58 THE BANK’S INTERNAL SUPPORT FOR LICUS WORK building and peace building. What actions and impasse and deteriorat- The Bank’s operational reforms will effectively contribute to which ing governance business aspects of state building? As noted in chapter 3, models. In Papua New approach is lacking in the Bank’s record on capacity development has Guinea, the Bank has some key areas. generally been weak. Unless the Bank provides stayed engaged, but it is fresh guidance on effective approaches to state not clear what the engagement is achieving. The building and strengthening accountability in Bank’s country team expressed concern about LICUS—where governance and institutional where the country is heading and how the Bank challenges are greatest—the Bank’s state-building can contribute. The implicit objective seems to efforts are likely to meet with little success. be simply to “stay engaged” while continuing to Other areas where the Bank needs to further think about possible courses of action. develop its operational approach include priori- Australia too—one of Papua New Guinea’s tizing and sequencing reforms; at the same time, major donors—seems unsure of the best way the Bank should avoid partial solutions. The forward, seemingly reverting back to the Bank needs to deliver services quickly without previously tried and failed government capacity- harming long-term government capacity development approaches of the 1980s. Lack of development; address trade-offs between foster- donor coordination and widespread confusion ing political reconciliation with development of concerning the best course of action to promote effective and legitimate local institutions and a sustained and effective development agenda translate political understanding into country has left the Bank somewhat inactive and ineffec- strategy; prevent conflict; and address linkages tive in Papua New Guinea; that has also left it between politics, security, and development. struggling to define a coherent operational The Bank has recently taken an important initial approach, given that everything seems by and step with respect to providing staff guidance on large to have failed in the past. the political, security, and development nexus The Bank has recognized that, as with other from the Bank’s perspective by developing a international partners, it is still learning what framework (World Bank 2005e). works in fragile contexts (World Bank 2005e, The balance of the Bank’s recent operational p. vii). In cases where the Bank is not ready to guidance on LICUS is tilted more toward what produce a guidance instruments should be used than on outlining note, it will be critical to The Bank needs to invest actual operational approaches with respect to ensure more active and more in actively what needs to be done differently and how. ongoing learning. This LICUS country teams would also benefit from underlines the need for monitoring and more narrative-based guidance, of the kind the strong monitoring evaluating LICUS presented in chapter 2 of this review, and and evaluation of ex- experiences. through short, problem-oriented notes rather periences. than more formal guidance notes, which are The Bank is not likely to be able to rely on often too condensed and devoid of sufficient client monitoring and evaluation systems in these country context. countries. It therefore needs to invest adequately While the Bank’s approach in post-conflict in monitoring and evaluating the performance of LICUS has been articulated more clearly, a its LICUS support and to distill emerging experi- number of shortcomings in the approach need ence continually. The implementation narratives to be addressed. The approach needs to be presented in chapter 2 of this review illustrate developed further—for example, to guide the how this may be done. transition and development phases that follow Sharing experiences of what is working and the immediate post-conflict reconstruction what is not in different LICUS situations can phase. foster learning. Though the Bank has done The Bank also still needs to develop a more some sharing of lessons through its LICUS effective approach for the political crisis or Learning Group Seminar Series, much more 59 E N G A G I N G W I T H F R A G I L E S T AT E S attention is needed to intensify the systematic sions about which business model(s) pertain to sharing and dissemination of emerging LICUS specific LICUS will be critical to ensure construc- experiences—both those of the Bank and those tive debates about effective solutions among of other donors. Creating a more receptive donors and country clients, as well as to ensure institutional environment and ensuring manage- ongoing refinement of the business models ment support for sharing negative experiences themselves. The resistance of some govern- will be key. So far, the Bank seems willing mainly ments to being classified as LICUS or fragile to share positive examples, as in its recent LICUS states, and possible future resistance to being reports (World Bank 2005e, 2005h). classified into one or more business model(s), Familiarity with the LICUS approach is low underlines the need for more effective among in-country respondents and other donor communication of the Bank’s objectives and respondents, but also among Bank respondents approach in this area. Finding more neutral or working on LICUS. In objective terms to describe both the overall Widespread lack of the Stakeholder Survey category of LICUS and the different business awareness of the Bank’s almost a quarter of Bank model groups could also help. But confidential- respondents said that ity can only hinder learning and consensus LICUS approach raises they are familiar with the building. questions about the Bank’s LICUS approach There is confusion about the role of the utility of the approach in only to a slight extent or LICUS Unit relative to the Conflict Prevention not at all (appendix Z). and Reconstruction (CPR) Unit. While Bank staff informing country In Lao PDR, the 2005 commented on the high quality of staff in both strategies. CAS does not mention units, they also expressed significant confusion the country as a LICUS over their roles. Both units deal with re- country, and neither the government nor other engagement issues in post-conflict countries, donors were aware that the concept existed or and both provide technical advice and support that their country was part of this group. to country transition processes. The confusion Several Bank staff working in Lao PDR were arises when country teams are looking for also unaware that the country fell in this specific guidance on procedural and policy category; they were aware that Myanmar did matters. (despite the fact that all these countries come Several staff noted the advantage of going to under the same Bank country management the LICUS Unit because of its strategic location unit). The rather widespread lack of awareness in OPCS, which can help attract senior manage- of the Bank’s LICUS classification and ment attention more easily. Others who have approach inside and outside the Bank does been working in post-conflict countries for some raise questions about the extent of its time noted the technical knowledge of the CPR influence so far. Unit and its larger staff complement. That The confidentiality of the LICUS list and knowledge enables the unit to provide more criteria (removed in June 2006 through the operational support. Bank’s disclosure of the country-level CPIA Of concern to staff are the practical questions ratings) has not helped foster active learning of which unit to turn to for specific types of and sharing of experiences around the LICUS advice and what kinds of support could be approach. Some countries (Nigeria, Tajikistan, expected from each unit. In semistructured Zimbabwe) have vehemently opposed their interviews conducted for this review, Bank staff classification into the LICUS category—and in Afghanistan and Haiti said they were unclear insufficient transparency in assigning CPIA about the relative roles of the LICUS and the CPR ratings has not helped the dialogue. Units, and there was a general feeling that there Going forward, open- is considerable duplication. Confidentiality hinders ness during country Staff expressed confusion about why the learning and sharing. strategy design discus- LICUS and CPR Units were undertaking 60 THE BANK’S INTERNAL SUPPORT FOR LICUS WORK seemingly parallel work on Poverty Reduction for a turnaround in The significant Strategy Papers (PRSPs). They were also LICUS. The WBI has duplication and overlap confused about the distinctions among Post- been important in some Conflict Multilateral Needs Assessments, JAMs, LICUS, but not in others. between the CPR and and Transitional Results Matrices. Appendix W provides an LICUS Units needs to be Bank staff also noted that there is confusion overview of WBI activi- resolved. among external partners and field-level counter- ties in LICUS. parts over which of the two units to engage. This While working on such countries often confusion is aggravated at times by inadequate requires an opportunistic approach and is likely collaboration and communication between the to be better when led by country teams, the two units. limited availability of core WBI resources for the In the Stakeholder Survey, about two-thirds vast majority of LICUS has constrained both the of Bank respondents saw some problem with scale and scope of WBI’s work. A notable the current organizational arrangement: 37 exception is Sudan, where WBI has been a driving percent said that there is some duplication force from the very early days of reengagement. between the support of the Bank’s LICUS Unit In the Stakeholder Survey, more than two- and that of the CPR Unit; 15 percent said that thirds of Bank respondents said that the techni- there is a lot of duplication, and 12 percent that cal input from WBI was slightly or not at all there is even a conflict or contradiction sufficient or timely. Sixty percent said it was (appendix Z). One Bank staff member remarked slightly or not at all of good quality (figure 4.3). that the CPR Unit and the LICUS Unit are the The Development Economics Vice Presi- “most conflictual units around.” dency (DEC) is involved in some highly relevant Given that conflict and state fragility influence LICUS research—for example, on political each other, the Bank needs to consider whether economy, terrorism, drugs, and conflict it would be more effective to combine the two (appendix X). DEC was one of the leading voices units into one. The Bank aims to address the in the early aid effectiveness debates, yet this has overlap between these units through an not been kept up more recently. ongoing study of the Bank’s Regional and sector Furthermore, DEC’s LICUS work has primarily mappings. Issues to consider in determining the focused on conflict and has given relatively little right organizational arrangement are the attention to other forms of state fragility. The key following: question of how to break the low-performance trap still needs further examination. In the • Ensuring that there is no duplication or frag- Stakeholder Survey, more than 70 percent of mentation of support to country teams and that Bank respondents said that the technical input they have easy access to expert technical ad- from DEC was slightly or not at all sufficient or vice with no conflicting messages timely, and 54 percent said it was slightly or not • Ensuring an efficient use of the Bank’s ad- at all of good quality (figure 4.4).7 ministrative budget and managerial resources • Determining the need for a continued central Conclusion strategic role with respect to both the internal There has been substantial progress in increas- and external agenda for LICUS/fragile states ing analytical work in a number of LICUS and in and related donor processes ensuring that LICUS managers have access to • Assessing the need for continued operational the Bank’s senior management. However, support from central teams. progress has been slow and slight in adapting OPs/BPs to the high-risk, low-capacity circum- The 2002 LICUS Task Force Report noted a stances of LICUS. It has also been slow in potentially important role for the World Bank ensuring sufficient management attention, Institute (WBI) because of the importance of which yields clear improvements in staffing knowledge sharing and capacity development numbers, staffing quality, and incentives. 61 E N G A G I N G W I T H F R A G I L E S T AT E S There is confusion about the role of the organizational arrangement. Finally, the Bank LICUS Unit relative to the CPR Unit, with two- has yet to benefit fully from the contributions thirds of Bank respondents in the Stakeholder DEC and WBI could potentially make to LICUS Survey noting some problem with the current work. Figure 4.3: World Bank Institute’s Technical Input Source: Appendix Z (Stakeholder Survey results). Note: Number of valid responses ranges from 158 to 169. Figure 4.4: Development Economics Vice Presidency’s Technical Input Source: Appendix Z (Stakeholder Survey results). Note: Number of valid responses ranges from 123 to 132. 62 5 Conclusions, Lessons, and Recommendations his chapter summarizes the conclusions of this review, distills lessons T of experience relevant both for the Bank and other donors, and then presents the recommendations. Conclusions following: (i) expanding analytical work by de- linking administrative budgets for ESW and Early successes technical assistance from lending volumes; (ii) There have been several notable early successes using ISNs, which allow for the design of strate- with regard to the LICUS principles. The Bank’s gies that cover a shorter period to accommodate operational readiness to support LICUS has the volatile LICUS conditions; (iii) developing improved since the LICUS Initiative began and guidance notes on specific topics; (iv) providing the Bank has engaged with a number of these access to LICUS managers to the Bank’s senior countries from the early days of peace or politi- management; and (v) introducing the LICUS cal transition. The Bank has also contributed to Trust Fund to finance countries in non-accrual macroeconomic stability and to the delivery of (for which the Bank previously lacked an instru- significant amounts of physical infrastructure, ment). Results from the Stakeholder Survey especially in post-conflict LICUS. indicate a small positive contribution to Substantial progress has been made in donor development of the Bank’s overall program in coordination at the international policy level, LICUS—a view that refers to Bank support exemplified by the recent agreement to the 12 generally, and not to the LICUS approach per se. OECD-DAC principles of international engage- ment by a wide spectrum of donors, including Challenges the Bank. The Bank has often played a leading The Bank’s initial engagement with a number of role as co-chair of international donor events LICUS has not been adequately followed up by a and coauthor of jointly undertaken policy focused and well-sequenced reform agenda. papers. The Bank’s recently introduced business Furthermore, the Bank has not yet sufficiently models, which differentiate between different internalized political understanding in strategy types of LICUS, are likely to permit a more design and implementation. The Bank also tailored response to these countries. needs to strengthen the quality of its country- The Bank’s internal support for LICUS work level coordination with other donors, especially has also progressed in several areas, notably the with respect to implementation follow-through. 63 E N G A G I N G W I T H F R A G I L E S T AT E S In addition, the Bank has made one of its raised the question of whether LICUS are receiv- traditional areas of weakness (capacity develop- ing appropriate amounts of IDA funding. Adjust- ment and governance) a central part of its focus ments to the PBA have resulted in increased IDA by adopting the more complex state-building financing, including to some post-conflict LICUS objective. This new emphasis requires that the and LICUS experiencing political transitions. Yet Bank identify more effectively its comparative it remains far from clear whether the current advantage; improve performance, including levels of IDA ensure that LICUS are not under- or through the development of innovative over-aided. approaches; and identify partners that can The aid-allocation issue has once again come complement its work to ensure the achievement to the fore with some research questioning the of intended outcomes. Finally, the choice of the empirical evidence for the positive link between term “state-building” may itself be inappropri- policies and aid effectiveness (which underlies ate, given its political and ideological the PBA). Other research argues that aid can be connotations. effective in promoting failing states’ sustainable The Bank needs to develop its operational policy turnarounds by building and strengthen- approaches in LICUS, especially for the deterio- ing the preconditions for reform or by enhanc- ration and prolonged crisis or impasse business ing the chances that the reform will be sustained models. Further refinement of the business once it is set in place. The latter research finds models by more explicitly factoring in differ- that potential returns from aid to LICUS can be ences in the capacity to perform core state extraordinarily high, even though the risks of functions (for example, resource generation, failure are substantial. For its part, the Bank has resource allocation, basic social service and yet to address the aid-allocation issue for LICUS infrastructure provision, and political accommo- in a way that reflects its objectives for these dation of dissent and security) is also needed. countries and ensures that LICUS are not under- That will enable the Bank to achieve a better fit or over-aided. between its operational approaches and the The Bank’s internal support for LICUS work varying institutional environments of different has progressed little on critical human resource LICUS. reforms relating to staffing numbers, staffing The Bank’s work on post-conflict countries quality, and incentives to undertake LICUS work. predates the LICUS approach, and the In IEG’s Stakeholder Survey, the majority of corresponding business model for post-conflict Bank respondents said that there has been no LICUS is articulated more clearly than the other change when working in LICUS with respect to business models. However, it has shortcomings the following human resource matters: overall and needs to be further developed to guide the career prospects, overall financial compensa- transition and development phases that follow tion, realism in expectations by Bank manage- the immediate post-conflict reconstruction ment about what can be accomplished, level of phase. support from Bank management, and efforts Furthermore, while the Bank has given more made by the Bank to ensure personal security attention to conflict prevention, there is limited and safety. knowledge about the effectiveness of those The Bank, as other donors, is still learning efforts. The Bank’s role and comparative what approaches work in LICUS contexts. advantage in this area have yet to be clearly Therefore, closely monitoring experiences in established, especially as conflict prevention LICUS to draw lessons is critical, and learning requires the Bank to give greater attention to and sharing needs to become a more the root causes of conflict. prominent feature of LICUS work. Although the The policy selectivity of the PBA system has Bank has developed guidance notes that distill increased over the years, and fewer IDA funds lessons of experience in specific areas (such as have been available for countries with weaker development policy loans), there is need for policies, institutions, and governance. This has much more active and ongoing stock-taking 64 C O N C L U S I O N S , L E S S O N S , A N D R E C O M M E N D AT I O N S and sharing of experiences among those LICUS engagement working on LICUS. The implementation narratives presented in Staying engaged is only a means to an end and needs chapter 2 of this review illustrate the kind of to be quickly followed by a clear and relevant reform stock-taking that could prove useful for LICUS. A agenda in LICUS. In the absence of a clear and new series of informal field-oriented discussion relevant reform agenda, early successes of notes that strive for timely lesson learning would engagement may be short lived and contribute also be useful. To achieve an open exchange of little to the achievement of country strategy negative experiences, a more receptive institu- objectives. The examples of the Central African tional environment that ensures management Republic and Haiti show that various obstacles support for the sharing of negative experiences may make the follow-up to a successful initial will be key. So far, the Bank seems much more LICUS engagement difficult. willing to share positive examples, as in its For example, as political successes were recent LICUS reports (World Bank 2005e, insufficiently backed up on the economic side in 2005h). The LICUS Unit’s learning-by-doing the Central African Republic, the government is objective would be much better served by giving now faced with a potentially disastrous budget adequate attention to both positive and negative crisis. In Haiti, the donor community seems to experiences. have given inadequate attention to ensuring a There is significant duplication of and con- minimum level of security. In both cases, good fusion over the roles and responsibilities of the initial results of the LICUS Initiative are now at LICUS Unit and the CPR Unit, which need to be risk of being diminished. resolved. Of concern to staff were the practical In certain instances, strategic disengagement— questions of which unit to turn to for specific with the exception of in-house analytical work— types of advice and what kinds of support to may be needed, at least for periods of time, expect from each unit. In IEG’s Stakeholder especially when involvement with the Bank is seen Survey, about two-thirds of Bank respondents as inappropriately giving legitimacy to the LICUS saw some problem with the current organiza- government or when such involvement dampens tional arrangement: 37 percent said that there is internal pressure for reform, and thus potentially some duplication between the support of the hinders the emergence of conditions needed to Bank’s LICUS Unit and that of the CPR Unit; 15 bring about serious and sustainable political percent said that there is a lot of duplication, reform. and 12 percent that there is even a conflict or In the deterioration and prolonged crisis or contradiction. impasse business models, where there is often little consensus between donors and govern- Lessons of Experience for the Bank ment on development strategy, engagement and Other Donors needs to include policy dialogue aimed at Several lessons emerge from this review’s assess- creating an opening for reform, while simultane- ment of the Bank’s experience in implementing ously working on a reform agenda should a the core principles of the LICUS approach. Many window of opportunity appear. of the issues covered under these lessons were In the post-conflict or political transition and noted as areas in need of improvement in the gradual improvement business models, engage- 2002 LICUS Task Force Report, such as the need ment will need to have more technical content to anchor strategies in stronger sociopolitical and a stronger focus on implementing the analysis or support highly focused reform reform agenda, given the existence of greater agendas. They were also emphasized in the reform consensus between donors and govern- Bank’s 2005 LICUS reports. The lessons derive ment. The Bank’s guidance for prolonged from the Bank’s own implementation experience conflict or political impasse countries states, but may also be useful in guiding other donor “Relatively noncontroversial development assistance in LICUS. issues may provide an entry point for construc- 65 E N G A G I N G W I T H F R A G I L E S T AT E S tive dialogue between the parties to a conflict.” Political understanding and its use in country For deteriorating governance countries, the strategy Bank’s guidance states that the Bank should provide “input on specific economic issues Commissioning and consuming—not necessarily which are important for mediation efforts and producing—good political analysis is critical for LICUS may serve as a way to restart dialogue” (World donors. The objective of a country team should be Bank 2005e). to commission or consume (not necessarily produce) analysis that is directly relevant to and Country ownership and absorptive capacity usable in the development of a strategy. In LICUS constraints apply as much to knowledge products as situations, especially in environments where speed to financial products. The involvement of country is of the essence, donors need to ensure that counterparts in the Bank’s analytical work existing political analysis is mined before commis- remains limited to administrative aspects, with sioning a new analysis. much less country-client participation in select- For example, in Lao PDR, the Bank ing topics and undertaking analysis. This effectively tapped existing political analysis thereby reduces national buy-in. Yet the involve- and invited a political scientist who had ment of country counterparts is key to ensuring published extensively on the country to make client ownership and improving impact of a presentation to the country team on politics analytical work. and reform in the country. This allowed for the In Tajikistan, the lack of government involve- preparation of an independent summary of ment in the selection and preparation of the relevant political analysis (tailored to the Bank’s analytical work limited the government’s needs of the donor community in general and interest in the results, which subsequently the Bank in particular) and its dissemination hindered effective implementation. In Angola, to a relevant group of Bank staff and other senior government officials saw some Bank-led donors. It avoided the higher costs of prepar- analytical work (for instance, the recent Country ing a “Bank” analysis, as well as potential Economic Memorandum) as an imposition of tension with the government because it Bank views on internal affairs, leading to limited allowed the Bank to avoid getting bogged ownership and capacity development. Without down in some of the sensitivities surrounding country ownership, the chance of analytical the analysis. For the Bank, the acquisition of work influencing government policy is small. existing knowledge as well as its dissemination LICUS governments’ absorptive capacity proved more important and effective in Lao constraints in using analytical work may also PDR than knowledge creation. limit possible knowledge transfer. The Angolan government, for instance, endorsed the Bank’s The main focus of donor efforts needs to be on helping ISN but expressed concern regarding the staff internalize political analysis in strategy design amount of foreseen analytical and advisory and implementation. Though the Bank has activities. This has raised doubt about whether conducted or had access to good political analysis the government would fully use analytical in some LICUS, it has inadequately reflected such products. The absorptive capacity of the govern- analysis in its strategy. For example, the interim ment is severely limited, and analytical and strategy in Papua New Guinea contains a good advisory activities undertaken mostly by the discussion of the political system and recognizes Bank risk straining relations with the govern- problems such as clan loyalties, political patron- ment, regardless of their technical quality. age, corruption, and lack of capacity. Yet the In Cambodia, plans for analytical and advisory strategy treats these problems as technical in services in the 2005 CAS—totaling 30 tasks to be nature and does not adequately use them to completed over fiscal 2005–07—appear overly underpin the overall approach. ambitious, considering the country’s limited Specific types of political analysis that can institutional capacity. help strategy design are as follows: 66 C O N C L U S I O N S , L E S S O N S , A N D R E C O M M E N D AT I O N S • Political risk analysis, which can help make the economic recovery, to basic services. Individu- decision whether the Bank should engage in ally, all these areas seem important, but together a certain country and, if so, how to engage they add up to a formidable agenda. • Structural analysis, which can help identify With respect to the Bank’s own assistance major characteristics of the political situation program, São Tomé and Principe is an example that will affect the Bank’s work, no matter what where the Bank was far too ambitious in relation specific strategy is chosen to the resources allocated to the country, with • Analysis of day-to-day politics, which can help the result that many of the CAS objectives were assess the distribution of power among dif- not achieved or were only partially achieved. ferent political forces in the capital and in the Beyond selectivity in CASs, it is critical to Regions, or even localities—which needs to go ensure that actual reform agendas in the field beyond the political gossip about who is up and are focused and well prioritized. The lack of who is down in the capital selectivity and prioritization in the reform • Analysis of the history of reform in the coun- agendas raises questions of effectiveness, try as well as neighboring countries, which especially given the limited capacity in LICUS. can help avoid past reform failures, such as While it is difficult to be selective in a country failed privatization attempts that may have cre- where many things need fixing urgently, the ated a strong backlash in the past. appropriate sequencing of reforms is key to ensuring that limited LICUS capacity is not Focused reform agenda overtaxed and that partial solutions are avoided. Well-sequenced reforms spanning a sufficient In complex LICUS environments, where virtually number of years, along with donor commitment every sector requires reform, appropriate sequenc- to see them through, will be essential. In Timor- ing of reforms and sufficient time to implement them Leste, donors may have pulled out too quickly, are crucial for achieving results without overwhelm- without sufficiently dealing with the country’s ing limited LICUS capacity. Donors must strive for pressing capacity needs. In Haiti, development collective donor selectivity, yet this is far from assistance has fluctuated greatly, with the being achieved, as the examples of Afghanistan’s country having gone through several feast-or- donor-endorsed reform agenda and Haiti’s ICF famine cycles in its relations with the donor (presented below) indicate. However, even if community. This could have been avoided if collective donor selectivity is not immediately various donors had better timed and sequenced achieved, the Bank itself needs to ensure focus their aid. and selectivity in its own assistance program, based on its core competences. Such Bank Capacity development in post-conflict LICUS selectivity has been increasing in recent years but remains a challenge, as the example of São Capacity development and governance programs Tomé and Principe suggests. need to start early, even in post-conflict LICUS. In Afghanistan, the reforms covered by Immediately following the cessation of conflict, donors are wide ranging, show lack of sufficient the international donor community tends to priority, and have led to 120 pieces of pending focus its assistance on physical reconstruction. legislation. These reforms, dealing with virtually Because capacity to use aid effectively in post- every economic and social aspect of the country, conflict LICUS is low and governance is often need to be carefully prioritized and sequenced. poor, the focus from the beginning also needs to Donors have yet to do this. be on the development of capacity and improve- In Haiti, the ICF was meant to guide interna- ment of governance, not merely on reconstruc- tional assistance and cooperation through tion of physical infrastructure. This may require September 2006 and covers practically all basic the creation or strengthening of public institu- state functions, ranging from security, to tions, civil service reform, and use of local expert- national dialogue, to economic governance, to ise. If foreign experts are brought in to provide 67 E N G A G I N G W I T H F R A G I L E S T AT E S technical assistance, this must not compromise (such as among Bank departments dealing with the long-term development of local capacity. public sector management, conflict prevention and reconstruction, LICUS, capacity develop- Donor coordination ment, and research departments) is particularly important in LICUS, where problems are Donor coordination cannot succeed without a complex and widespread and often require common vision and purpose among donors—when multisectoral solutions. donor objectives cannot be fully harmonized, it is important that they at least be complementary. The Results measurement and monitoring Bank’s approach has not fully recognized the differing motivations of donors for engaging Monitoring and evaluation are at least as important with LICUS. Although the broad concept of in LICUS as they are in any other country. Monitor- fragility is widely understood and accepted, the ing and evaluation are crucial in LICUS for countries identified by donors as fragile vary. several reasons. First, the Bank, like other Motivations for supporting fragile states range donors, is still learning what approaches work in from security, to aid effectiveness, to equitable LICUS contexts. Therefore, closely monitoring development, to poverty reduction, to state experiences to draw lessons is critical, and building, to conflict prevention. learning and sharing needs to become a more In both Afghanistan and Tajikistan, IEG’s prominent feature of LICUS work. fieldwork found that major donors did not Second, given that progress is often slow in subscribe to a single clear objective. Without a these countries, it is important to reassess common overall objective, policy coherence is continually whether the program is on course unlikely. to achieve the desired outcomes. Third, a The Bank’s donor coordination efforts and constantly changing and volatile LICUS modalities are insufficiently informed by the environment, where progress is often nonlin- objectives of the different players in a country. ear, means that program adaptation is That said, donor coordination is a form of collec- essential—closely tracking performance will tive action, requiring that other donors similarly help determine when and what kind of adapta- improve their outreach to the Bank and subordi- tion is necessary. Effective learning by doing to nate bilateral agendas to agreed multilateral improve the Bank’s future effectiveness in objectives. LICUS can only happen with strong monitoring and evaluation. Coordination needs to begin within each donor The Bank has stated that state building and agency. Coordination is not only important peace building should be the goals by which to among multilateral and bilateral donor agencies, measure the LICUS Initiative’s success. but it is also a vital issue within each donor However, it has yet to identify performance agency. Projects in different sectors of the same indicators by which this can be done, or country often work in parallel and fail to tap yardsticks against which performance may be synergies, as was the case, for example, with the measured. Where change is often more process Bank’s Community Empowerment and Agricul- oriented—especially in the deterioration and tural projects in Timor-Leste. prolonged crisis or impasse business models— A side effect of the Bank’s decentralization to outputs and outcomes that may be expected in country offices has been the concentration of the other business models may not be appropri- country knowledge among local staff and its ate yardsticks of success. Objectives should be inadequate dissemination across the country appropriate to particular LICUS contexts, which team, especially to those based in Washington. would in turn determine yardsticks and ensure Addressing the problems of coordination across that the bar of success is set at an appropriate the various departments of donor agencies height. 68 C O N C L U S I O N S , L E S S O N S , A N D R E C O M M E N D AT I O N S Improving internal organizational support for because of the lack of lower levels—led to LICUS work unnecessary tensions. Donor decisions regard- ing the number of staff in each LICUS should Field presence alone is insufficient for effective reflect the extent and nature of intended country strategy implementation. It needs to be engagement, considering respective donor’s complemented by adequate communication between objectives in those countries. field and headquarters donor agency staff, as well as Apart from the absolute numbers, field office an adequate number of field staff with the appropri- staff also need to have sufficient authority to ate authority and skills. Understanding of country ensure that not every decision has to be circumstances is often best achieved through approved by headquarters. Effective field substantial field presence, though that alone is presence requires having the right kind of staff not enough. Internalizing analysis of the country involved in the country. conditions throughout all involved donor In semistructured interviews, several donors agency departments, and applying its lessons to emphasized that coordination is unusually all interventions, is equally important. susceptible to the strengths and the foibles of In Cambodia, for example, the Bank’s field the individuals involved. More appropriate presence has significantly improved understand- training for staff posted to difficult field assign- ing of the political situation. Discussions with ments and improved incentives within the Bank country team members and other stakeholders, that encourage staff to collaborate with other however, suggest that this knowledge may still donors might ameliorate these idiosyncratic be highly concentrated among a few managers risks. and staff (mostly in the country office and In the deterioration business model, where Bangkok hub), with relatively limited dissemina- there might be a breakdown of dialogue with the tion to the broader country team. government, donor agency staff will need strong The issue appears to have shifted from a diplomatic and persuasion skills to ensure that partial understanding of the political realities of the door remains open for a dialogue with the Cambodia to one of where this knowledge is government, while simultaneously mobilizing located within the Bank’s country team and how nongovernmental groups, including civil society. it is used to guide decision making in strategy In the prolonged crisis or impasse business and program implementation. The concentra- model, where problems are chronic or there is tion of in-depth country knowledge among a political stalemate, the necessary staff skills will few staff implies that only some Bank activities include immense patience as well as creativity, and interventions benefit. In general, greater with constant innovation relating to ways of knowledge transfer is needed between donor breaking the persisting logjam. In the post- country offices and their headquarters-based conflict or political transition business model, country and sector staff. the necessary staff skills will include specific Despite the cost, field offices need to be technical knowledge of how to develop sound adequately staffed if they are to engage effectively economic systems, institutions, and key with clients. In Angola, the initially small group of infrastructure. field staff faced a multiplicity of tasks, from strate- Staff should also possess the ability to act gic dialogue with government and donors to quickly and decisively in these environments, logistics such as moving the office to new before the optimism following peace dissipates. premises. The situation was made more difficult Staff needs to help guard against these by the lack of operational-level staff in the field countries’ falling back into conflict. As these office who could, in consultation with ministry situations often attract massive international aid, staff, prepare the ground before high-level donor staff needs strong coordination and meetings between the ministers and the Bank. sequencing skills to organize both the develop- Moving issues to the top too quickly— ment partners and their activities. 69 E N G A G I N G W I T H F R A G I L E S T AT E S In the gradual improvement business model, and how, and what results should be expected. the primary skill needed is the ability to provide The communication strategy in the gradual customized technical assistance and work hand improvement business model will need to be in hand with a client that is already reforming. more informational, presenting relevant cross- country and cross-sectoral experiences. Sharing experiences—both positive and negative— Immediately following the cessation of is essential for learning, but doing so effectively conflict, international donors, including the requires a receptive institutional environment and Bank, have often committed large amounts of management support. Sharing experiences of what aid coupled with overly ambitious agendas. This is working and what is not in different LICUS has frequently created high expectations among situations can foster learning. Learning is the population and led to subsequent disillu- especially important in LICUS work, because the sionment when expectations have remained donor community is continuing to grapple with unfulfilled and day-to-day living has seen few the question of how best to assist these tangible improvements. Avoiding overambitious challenging countries. While the Bank has agendas and providing better communication to shared some lessons through its LICUS Learning lower expectations to realistic levels are critical; Group Seminar Series, much more attention is the Bank needs to invest in this effort. needed to intensify the systematic stock-taking and dissemination of emerging LICUS Better operational guidance is needed for tailoring experiences—those of both the Bank and other donor approaches to the special conditions of LICUS. donors, and both positive and negative. The LICUS Initiative has raised awareness of the Creating a more receptive institutional need to act differently in LICUS, but the Bank environment and ensuring management and other donors have yet to identify precisely support for the sharing of negative experiences how to do this. The extent to which donor will be key. So far, the Bank seems mainly willing approaches to LICUS need to, and can, to share positive examples. efficiently address the causes—not just symptoms—of countries becoming or remain- Effective communication is essential to ensure ing characterized as LICUS also need greater country acceptance of donor approaches for LICUS attention. Solutions that view causes as givens and to temper unrealistic country expectations about may miss all-important contextual factors. what can be achieved, especially immediately after Donor operational guidance must ensure that the cessation of conflict. Better communication of areas outside the comparative advantage of donors’ objectives and approaches for LICUS particular donors be left to others, while their will be needed to ensure country buy-in and to own work both adequately factors in the work prevent disillusionment among stakeholders done by others and complements it. about what can be achieved in a specific period The Bank’s deterioration and prolonged of time. crisis or impasse business models, and the In the Bank’s deterioration and prolonged transition and development phases that follow crisis or impasse business models, where the the immediate reconstruction phase in the post- economic and social situation is for the most conflict or political transition business model, part worsening or stagnant, the communication pose some of the biggest challenges for the strategy would need to disseminate actively the donor community. These are areas in which benefits of reform both to the government and there has been relatively little innovative to civil society. thinking. In the Bank’s post-conflict or political transi- Issues for which operational guidance is tion business model, to prevent disillusionment particularly needed include ways to prioritize from unrealistic expectations, the communica- and sequence reforms, while avoiding partial tion strategy should target the entire population solutions; ways to deliver services quickly, and be explicit about what donors will do, when, without harming long-term government 70 C O N C L U S I O N S , L E S S O N S , A N D R E C O M M E N D AT I O N S capacity development; ways to foster political system on aid volumes to LICUS. Aid-allocation reconciliation, while also contributing to criteria that reflect the Bank’s objectives in effective and legitimate governance; ways to LICUS and ensure that these countries are not internalize political understanding within under- or over-aided need to be developed. country strategy design and implementation; Whether and to what extent the criteria should and ways to address linkages between politics, be based on factors other than policy per- security, and development effectively. formance (such as levels of other donor assis- The balance of the Bank’s recent guidance on tance, assessment of potential risks and rewards, LICUS is tilted more toward what instruments and regional and global spillovers) needs to should be used than on an outline of actual be examined, keeping in mind that aid is lim- operational approaches for what needs to be ited and trade-offs will have to be made. done differently, and how. LICUS country teams would also benefit from more narrative-based • Strengthen internal Bank support for LICUS work guidance, of the kind presented in chapter 2 of over the next three years. this review, and through short, problem-oriented Two aspects of internal Bank support need notes rather than only more formal guidance attention. First, staffing numbers, skills, and in- notes, which are often too condensed and centives for working on LICUS need to be pri- devoid of sufficient country context. oritized. Ensuring adequate incentives to attract qualified staff—both at headquarters and in Recommendations field offices—will require giving clear signals of what is deemed to be success in LICUS, what • Clarify the scope and content of the Bank’s state- outcomes staff will be held accountable for, building agenda, and strengthen the design and how much risk it is reasonable to take, how fail- delivery of capacity development and governance ure will be judged, and how overall performance support in LICUS. evaluation ratings and staff career development Given its weak record on capacity devel- will take these into account. opment and governance, as well as its focus on As in Olympic diving, where the scoring the more ambitious and complex state-building system factors in both the technical perfec- objective in LICUS, the Bank needs to clarify tion and the difficulty of the dive, staff per- its areas of comparative advantage in relation formance in LICUS should be similarly judged to other donors. The Bank needs to adopt in- by assigning due weight to the extent of chal- novative approaches that ensure better ca- lenges presented by varying LICUS environ- pacity and governance outcomes. Innovative ments. Signaling the importance of LICUS work approaches need to be developed for achiev- throughout the management hierarchy will ing a better fit between the Bank’s interventions also be required. and the capacity of LICUS to perform core Apart from incentives, the Bank needs to en- state functions; ensuring implementation of fo- sure that staff working on LICUS have relevant cused and well-sequenced interventions in skills, are capable of seeking and using politi- LICUS environments, where virtually every as- cal knowledge, and are willing and able to work pect of capacity and governance may need sig- in interdisciplinary teams. Current plans to ad- nificant improvement; and effectively mon- dress these issues in the forthcoming Strength- itoring capacity and governance outcomes. ening the Organizational Response to Fragile States paper are welcome, even if late. • Develop aid-allocation criteria for LICUS that en- More systematic thinking is needed on sure that these countries are not under- or over- staffing decisions for LICUS within the con- aided. text of the Bank’s overall staffing, recognizing The Bank needs to conduct a technical re- that assigning more and better-qualified staff view of the cumulative effect of the various ad- to work on LICUS would likely mean trade-offs justments to the performance-based allocation for other Bank country teams. Trade-offs to 71 E N G A G I N G W I T H F R A G I L E S T AT E S benefit LICUS may or may not be justified, de- the business models, needs to be independently pending on the Bank’s objectives for LICUS, as evaluated after three years, when sufficient ex- well as other Bank clients’ need for assistance. perience on the outcomes of the approach will Second, the organizational structure for be available. At that time, it should be possible LICUS and conflict work needs to be stream- to address the more fundamental question of lined. The Bank needs to ensure an efficient or- whether and to what extent Bank assistance can ganizational arrangement that removes effectively support sustainable state building. duplication and fragmentation of support be- Continued Bank support for the LICUS category tween the LICUS and CPR Units. and approach should be based on the findings of that reassessment. • Reassess the value added by the LICUS approach after three years. The value of the LICUS category and ap- proach, including the operational usefulness of 72 APPENDIXES APPENDIX A: DEFINITIONS AND DATA SOURCES Term Definition and data sources AAA Analytical and advisory activities (AAA) is an umbrella term for several product lines. Examples of AAA include economic and sector work, technical assistance, donor and aid coordination, research services, the World Development Report, and impact evaluation. Source: http://intranet.worldbank.org/WBSITE/INTRANET/UNITS/INTOPCS/INTDELIV ERYMGMT/ 0,,contentMDK:20267395-menuPK:764245~pagePK:64137152~piPK:64136883~the SitePK:388672,00.html#1 Administrative budget Administrative budget refers to the allocations made to each vice presidential unit and then to country units based on country benchmarks for resource allocation, seen as a function of the supervision needs, lending requirements, and AAA requirements. Lending requirements were dropped from the formula in fiscal 2004 (thus disconnecting the administrative budget from lending). The elements of AAA requirements in the setting of country benchmarks are related to population size, poverty, and countries that are relatively new members of the Bank. Source: http://intranet.worldbank.org/WBSITE/INTRANET/UNITS/INTRMAy0„content MDK:20336533-menuPK:422043~pagePK:64088751~piPK:64087868~theSitePK:330235- ROLE:(ALS),00.html Analytical work Analytical work is synonymous with AAA. See AAA above. Africa Catalytic Growth Fund The Africa Catalytic Growth Fund (ACGF) was launched in March 2006 to provide rapid, targeted support to countries with credible programs to accelerate growth, poverty reduction, and attainment of the Millennium Development Goals (MDGs). The group of countries that ACGF aims to serve is almost entirely separate from the group served by the LICUS Trust Fund grants. In some cases, the “transforming countries” could be countries where the LICUS Trust Fund has already pre- pared the groundwork, but that are now ready to graduate, shifting from post-conflict or LICUS status to a more substantial scale of funding based on clearer evidence of government reforms. Source: ACGF Board Report-Africa Catalytic Growth Fund http://siteresources.world bank.org/INTAFRCATGROFUND/Resources/board_report.pdf Conflict-affected countries Conflict-affected countries are countries that have recently experienced, are experiencing, or are widely regarded as at risk of experiencing violent conflict. These countries are identified by the Regional vice presidencies. The conflict-affected LICUS are: Afghanistan, Angola, Burundi, Cambodia, Central African Republic, Comoros, Democratic Republic of Congo, Republic of Congo, Guinea-Bissau, Haiti, Kosovo, Liberia, Myanmar, Nigeria, Solomon Islands, Somalia, Sudan, Tajikistan, and Timor-Leste. Source: World Bank data. 75 E N G A G I N G W I T H F R A G I L E S T AT E S Term Definition and data sources CPIA The Country Policy and Institutional Assessment (CPIA) assesses the quality of a country’s pres- ent policy and institutional framework, particularly how conducive that framework is to fos- tering poverty reduction, sustainable growth, and the effective use of development assistance. The CPIA consists (as of 2004) of a set of 16 criteria representing the policy and institutional dimensions of an effective poverty reduction and growth strategy, which are divided into four clusters: economic management, structural policies, policies for social inclusion/equity, and public sector management and institutions. For each criterion, countries are rated on a scale of 1 (low) to 6 (high). A rating of 1 corresponds to very weak performance, and a 6 rating to very strong performance. Intermediate scores of 1.5, 2.5, 3.5, 4.5, and 5.5 may also be given. Source: OPCS, ”Country Policy and Institutional Assessments: 2004 Assessment Questionnaire.” ESW Economic and sector work (ESW) supports World Bank country operations by adapting research to specific projects or circumstances. Source: http://intranet.worldbank.org/WBSITE/INTRANET/UNITS/INTOPCS/INTDELIV ERYMGMT/0„contentMDK:20267395-menuPK:764245-pagePK:64137152-piPK:64136883-the SitePK:388672,00.html#1 HDI The Human Development Index (HDI) is a composite index that measures average achieve- ment in three basic dimensions of human development—a long and healthy life (measured by life expectancy at birth), knowledge (measured by adult literacy rate and gross enrollment ratio), and a decent standard of living (measured by GDP per capita). Source: http://hdr.undp.org/reports/global/2005/pdf/HDR05_HDI.pdf HIPC The Heavily Indebted Poor Countries (HIPC) Initiative was launched in 1996 and modified in 1999 to create a framework for all creditors, including multilateral creditors, to provide debt relief to the world’s poorest and most heavily indebted countries, and thereby reduce the con- straint on economic growth and poverty reduction imposed by the debt build-up in these coun- tries. The HIPC decision point is the date at which a heavily indebted poor country with a record of good performance under adjustment programs supported by the International Monetary Fund (IMF) and the World Bank commits to undertake additional reforms and to develop and im- plement a poverty reduction strategy. The HIPC completion point is the date at which the coun- try successfully completes the key structural reforms agreed at the decision point, including the development and implementation of its poverty reduction strategy. The country then re- ceives the bulk of debt relief under the HIPC Initiative without any further policy conditions. The HIPC Initiative currently identifies 40 countries, including 12 LICUS, as potentially eligi- ble to receive debt relief. The 12 LICUS are: Burundi, Central African Republic, Comoros, De- mocratic Republic of Congo, Republic of Congo, Guinea-Bissau, Haiti, Liberia, São Tomé and Principe, Somalia, Sudan, and Togo. None of the 12 LICUS is at HIPC completion point. Sources: http.7/intranet.worldbank.oig/WBSITE/INTRANET/UNITS/INTPREMNET/INTDEBT DEPT/0„content; World Bank and IMF 2006a. 76 A P P E N D I X A : D E F I N I T I O N S A N D D AT A S O U R C E S Term Definition and data sources IDA PC IDA Post-Conflict Exceptional Financing (IDAPC) is based on Post-Conflict Progress Indicators (PCPIs). ISN Interim Strategy Note (ISN) is the umbrella term for Transitional Support Strategies (TSSs) and Country Re-engagement Notes (CRNs). When country circumstances are not conducive to a normal Country Assistance Strategy (CAS) approach, the Bank may prepare an ISN. An ISN may also be used for countries affected by or emerging from conflict or in countries in which the Bank has not recently been engaged. The ISN may be put in place for a period of up to 24 months and may be renewed for additional periods with the endorsement of the executive di- rectors. An ISN normally includes a discussion of the country context, including the legal con- text, the history of the Bank’s involvement in the country, and the roles of regional and international partners. It establishes immediate priority assistance objectives along with a pro- posed program of assistance to meet these objectives. An ISN also includes an assessment of risks, strategies for entry and exit, and contingency responses to a reversal of progress, es- pecially renewed conflict. Finally, it describes benchmarks and performance monitoring indi- cators for assessing progress, and a schedule for periodic consultations with the Board. The term “Watching Brief “refers to a phase in World Bank engagement in conflict-affected countries. A country is determined to be in the Watching Brief phase when conflict is ongo- ing and prevents the Bank from continued assistance or other business. The Watching Brief was introduced in 1996 as a way of maintaining constructive engagement with countries where the Bank might otherwise have been absent. KKZ Kaufmann, Kraay, Mastruzzi (KKZ) indicators reflect the statistical compilation of responses on the quality of governance given by a large number of enterprise, citizen, and expert survey respondents in industrial and developing countries, as reported by a number of survey insti- tutes, think tanks, nongovernmental organizations, and international organizations (drawn from 37 separate data sources constructed by 31 different organizations, including the CPIA). The KKZ scale ranges from –2.5 to +2.5. Source: http://www.worldbank.org/wbi/governance/data.html Lending Lending refers to commitments for projects financed by (i) IDA (credits, exceptional IDA post- conflict grants, guarantees, and other grants); (ii) IBRD; or (iii) Special Financing. Other grants under IDA include: debt vulnerability, poorest countries, natural disasters, and HIV/AIDS. LIC Sixty-two countries were classified as low-income countries (LICs) by the 2005 World De- velopment Indicators. The list includes the 25 LICUS, 35 non-LICUS LICs, and Chad and Sierra Leone. Source: World Bank 2005j. 77 E N G A G I N G W I T H F R A G I L E S T AT E S Term Definition and data sources LICUS The Bank classifies a country as LICUS if it is a low-income country (falling within the thresh- old of IDA eligibility) scoring = 3.0 on both the overall and governance CPIA averages, plus low- income countries without CPIA data (including Afghanistan, Liberia, Myanmar, Somalia, Timor-Leste, and the territory of Kosovo). The fiscal 2005 LICUS list of countries (created using the fiscal 2004 Gross National Income [GNI] threshold of $865 or less per capita and 2003 CPIA ratings) includes 25 countries: Afghanistan, Angola, Burundi, Cambodia, Central African Republic, Comoros, Democratic Republic of Congo, Republic of Congo, Guinea-Bissau, Haiti, Kosovo, Lao People’s Democratic Republic, Liberia, Myanmar, Nigeria, Papua New Guinea, São Tomé and Principe, Solomon Islands, Somalia, Sudan, Tajikistan, Timor-Leste, Togo, Uzbekistan, and Zim- babwe. Source: LICUS Unit, World Bank. LICUS Trust Fund The Bank created the LICUS Trust Fund on January 15, 2004. The fund targets the most marginalized LICUS in non-accrual that cannot use IDA funds for basic reforms or capacity building. Under exceptional circumstances, it also permits a Bank contribution to an agreed mul- tidonor strategy in an active IDA LICUS, where existing IDA funds are inappropriate for this pur- pose and executive directors have endorsed such a cofinancing program in the country strategy document. LICUS classified as post-conflict by the Resource Mobilization Department (FRM) and that are eligible for IDA post-conflict financing are not eligible for support under the trust fund. Activities eligible for financing under the LICUS Trust Fund include capacity building to support governance reform and strengthening social service delivery, including the fight against HIV/AIDS. To date, beneficiaries of the LICUS Trust Fund have been Central African Republic, Comoros, Haiti, Liberia, Somalia, Sudan, Togo, and Zimbabwe. A total of $23.8 million has been committed through 37 LICUS Trust Funds in 8 non-accrual LICUS since the establishment of the trust fund. Source: http://imagebank.worldbank.org/servlet/WDSContentServer/IW3P/IB/2004/10/07/ 000090341_ 20041007095849/Rendered/PDF/275350GLB0rev.pdf Non-accrual status Loans to, or guaranteed by, a sovereign are placed in non-accrual status when the oldest pay- ment arrears are six months overdue—that is, when the second consecutive payment is missed on the loans with the oldest arrears. In order to be eligible for new loans, the sover- eign concerned must clear all payment arrears in full. Once all arrears are cleared, all loans to, or guaranteed by, the sovereign are generally restored to accrual status. Source: http://intranet.worldbank.org/WBSITE/INTRANET/UNITS/SRM/0„contentMDK: 20669963-menuPK:1748372-pagePK:67677-piPK:64094917~theSitePK:134920,00.html Non-LICUS LICs Non-LICUS LICs are low-income countries, excluding those that were classified as “core” or “severe” LICUS in fiscal 2003–05. This excludes from LICs the 25 LICUS (mentioned above), Chad, and Sierra Leone. Chad and Sierra Leone were excluded because they were not classified as LICUS in fiscal 2005, although Chad was classified as LICUS in fiscal 2003 and Sierra Leone was classified as LICUS in fiscal 2003 and 2004. The non-LICUS LICs thus include: Bangladesh, Benin, Bhutan, Burkina Faso, Cameroon, Côte d’lvoire, Eritrea, Ethiopia, The Gambia, Ghana, Guinea, India, Kenya, Democratic People’s Republic of Korea, Kyrgyz Republic, Lesotho, Mada- gascar, Malawi, Mali, Mauritania, Moldova, Mongolia, Mozambique, Nepal, Nicaragua, Niger, Pakistan, Rwanda, Senegal, Tanzania, Uganda, Vietnam, Republic of Yemen, and Zambia. Sources: World Bank 2005j and LICUS Unit, World Bank. 78 A P P E N D I X A : D E F I N I T I O N S A N D D AT A S O U R C E S Term Definition and data sources ODA Official development assistance (ODA) comprises grants or loans to developing countries and territories on the OECD Development Assistance Committee list of aid recipients that are un- dertaken by the official sector with promotion of economic development and welfare as the main objective and at concessional financial terms (if a loan, having a grant element of at least 25 percent). Technical cooperation is included. Grants, loans, and credits for military purposes are excluded. Also excluded is aid to more advanced developing and transition countries, as determined by DAC. Source: http://sima.worldbank.org/gmis/mdg/UNDG%20documeiTLfinal.pdf Post-conflict countries Post-conflict countries are a subset of conflict-affected countries. They are identified based on Post-Conflict Progress Indicators (PCPI) for purposes of determining exceptional IDA grants. Eligible countries receive up to 4 years of full post-conflict allocations. In years 5, 6, and 7, allocations are reduced by 25 percent, 50 percent, and 75 percent, respectively, of the excess above the CPIA norm. By year 8, they return to CPIA-based allocations. The LICUS that became eligible for these exceptional loans are: Guinea-Bissau (fiscal 2001), Democratic Republic of Congo (fiscal 2002), Republic of Congo (fiscal 2002), Afghanistan (fiscal 2003), Angola (fiscal 2003), Burundi (fiscal 2003), and Timor-Leste (fiscal 2003). Source: LICUS Unit, OPCS, Financial Resource Mobilization (FRM), World Bank. PCF The Post-Conflict Fund (PCF) was established in 1997 and became a trust fund in 1999, eligi- ble to receive contributions from donors. The PCF supports research, planning, piloting, and analysis of ground-breaking activities through funding to governments and partner organiza- tions in the forefront of this work, including nongovernmental organizations, United Nations agencies, transitional authorities, governmental institutions, and civil society groups. The main focus has widened from that of mainly rebuilding infrastructure to promoting economic recovery, creating effective and accountable institutions, assisting vulnerable groups, working to im- prove health and education services, supporting community-driven reconstruction processes, and demobilizing and returning ex-combatants and displaced people to their communities. As of July 2005, PCF beneficiaries were Afghanistan, Burundi, Cambodia, Comoros, Democratic Republic of Congo, Republic of Congo, Haiti, Kosovo, Nigeria, Somalia, Sudan, Tajikistan, and Timor-Leste. Source: http:/Aveb.worldbank.orgMBSITBEXTERNAL/TOPICS/EXTSOCIALDEVELOPMENT/ EXTCPR/0,,menuPK:407746-pagePK:1’49018-piPK:149093-theSitePK:407740,00.html; http://siteresources.worldbank.org/INTCPR/21’4578-1115615449417/20698452PCFAnnual Report05.pdf PCPI The Post-Conflict Performance Indicators (PCPI) ratings framework is designed to measure change in countries that are eligible for exceptional post-conflict allocations from IDA. The scale of the PCPI is 1–6 (low to high). A rating of 1 is equal to a situation of ongoing or re-ignited con- flict, and therefore no positive change; a rating of 6 indicates very strong performance, which roughly equals a rating of 4 on the CPIA scale, and would be very unusual for any country under the PCPI framework. Source: http://siteresources.worldbank.org/INTCPRi1090479-1115613025365/20482305/Post- Conflict+Performance+lndicators,+2004-05.pdf 79 E N G A G I N G W I T H F R A G I L E S T AT E S Term Definition and data sources Special Financing Country-specific trust funds. TA Technical assistance (TA) supports external clients to implement reforms or strengthen insti- tutions. For example, a Bank ESW report may recommend that a government establish a uni- fied body to regulate the power and water sectors, while a follow-on Bank technical assistance activity assists the government in developing a draft act for the establishment of such a mul- tisectoral regulator. Source: http://intranet.worldbank.org/WBSITE/INTRANET/UNITS/INTOPCS/INTDELIV ERYMGMT/ 0,,contentMDK:20267395~menuPK:764245~pagePK:64137152~piPK:64136883~the- SitePK:388672,00.html#1 TRM The Transitional Results Matrix (TRM), also referred to as a Transitional Calendar or Results- Focused Transitional Framework (RFTF), is a planning, coordination, and management tool de- veloped by the UNDG and the World Bank that national stakeholders and donors can use to prioritize actions necessary to achieve a successful transition in fragile states. The TRM helps launch a poverty-reduction strategy (PRS) approach in these environments, either as an early framework to lay the groundwork for a PRS or, later, as a way to operationalize a PRS in low-capacity countries. TRMs are organized by clusters and sectors that are key to the re- covery process and include the following: (i) strategic objective or goal, (ii) baseline (or cur- rent situation), (iii) time intervals for actions and priority outputs, (iv) targets and monitoring indicators, and (iv) agencies/units responsible for implementation of each action. Source: World Bank and UNDP 2005. Trust funds Financial and administrative arrangements between the World Bank and external donors under which donors entrust funds to the Bank to finance specific development-related activ- ities. Formal legal agreements with donors designate the Bank as trustee and define the terms and conditions for use of the funds. Donors include many Bank member countries, the private sector, foundations, and nongovernmental organizations, including the World Bank Group. The top 10 trust fund programs in LICUS are HIPC, Afghanistan Reconstruction Trust Fund (ARTF), Single-Purpose Trust Funds (SPTF), Trust Fund for East Timor–Bank Executed (ETBK), Japan So- cial Development Fund (JSDF), Trust Fund for East Timor–ADB Executed (ET-ADB), Global En- vironment Facility–IBRD as Implementing Agency (GEFIA), Debt Service Trust Fund (DS), LICUS, and Institutional Development Fund (IDF). Source: http://intranet.worldbank.oip/WBSITE/INTRANET/UNITS/INTCFP/0„contentlVIDK: 20153435-menuPK:323877~pagePK:64060698~piPK:64060705~theSitePK:299971,OO.html 80 APPENDIX B: LICUS, FISCAL 2003–06 Core/ severe Fiscal Fiscal Fiscal Fiscal LICUS in all 2003 2004 2005 2006 4 years Country Conflict status C = Core; S = Severe; M = Marginal Y = Yes Africa Angola Post-conflict and conflict-affected C S S C Y Burundi Post-conflict and conflict-affected C C C C Y Cameroon Non-conflict-affected M M — — — Central African Republica Conflict-affected M S S S — Chad Non-conflict-affected C M M M — Comoros Conflict-affected M C C S — Congo, Democratic Republic of Post-conflict and conflict-affected C C C C Y Congo, Republic of Post-conflict and conflict-affected M C C C — Côte d’lvoirea Conflict-affected — — M C — Equatorial Guinea Non-conflict-affected — S — — — Eritrea Conflict-affected — — M C — Gambia, The Non-conflict-affected — M M M — Guinea Non-conflict-affected — M M C — Guinea-Bissau Post-conflict and conflict-affected C S C C Y Liberiaa Conflict-affected C S S S Y Niger Non-conflict-affected — M M — — Nigeria Conflict-affected M C C C — São Tomé and Principe Non-conflict-affected — C C M — Sierra-Leone Conflict-affected C C M M — a Somalia Conflict-affected C S S S Y Sudana Conflict-affected C S S C Y a Togo Non-conflict-affected M S C C — Zimbabwea Non-conflict-affected C S S S Y East Asia and Pacific Cambodia Conflict-affected M M C C — Lao People’s Democratic Republic Non-conflict-affected C C C C Y Papua New Guinea Non-conflict-affected M C C M — Solomon Islands Conflict-affected — S S C — Timor-Leste Post-conflict and conflict-affected — C C C — Tonga Non-conflict-affected — — — M — Vanuatu Non-conflict-affected — — — C — (Continues on the following page.) 81 E N G A G I N G W I T H F R A G I L E S T AT E S Core/ severe Fiscal Fiscal Fiscal Fiscal LICUS in all 2003 2004 2005 2006 4 years Country Conflict status C = Core; S = Severe; M = Marginal Y = Yes Europe and Central Asia Kosovo Conflict-affected — C C C — Kyrgyz Republic Non-conflict-affected M — — — — Tajikistan Conflict-affected C C C M — Uzbekistan Non-conflict-affected C S C M — Latin America and the Caribbean Haiti Conflict-affected C S S C Y Middle East and North Africa Djibouti Non-conflict-affected C — — M — West Bank and Gaza Conflict-affected — — — C — Yemen Non-conflict-affected M — — — — South Asia Afghanistan Post-conflict and conflict-affected C S S S Y Myanmara Conflict-affected C S S S Y Total (core and severe only)b 17 26 25 26 12 Sources: For LICUS list of countries—OPCS, World Bank. For list of countries in non-accrual—World Bank 2005g. Note: S indicates LICUS classified as “severe” (an overall and governance CPIA of 2.5 or less); C indicates LICUS classified as “core” (an overall and governance CPIA of 2.6–3.0); and M indicates LICUS classified as “marginal” (an overall and governance CPIA of 3.2). The criteria for identifying LICUS have undergone modification over time. a. Countries in non-accrual. b. Marginal LICUS score on the edge of what is considered LICUS and are identified by the Bank only for monitoring purposes (hence, the total includes only “core” and “severe” LICUS, not “marginal” LICUS). 82 APPENDIX C: PROGRESS ON MILLENNIUM DEVELOPMENT GOALS Most LICUS are unlikely to achieve the Millen- meet three of the six targets. Countries that are nium Development Goal (MDG) targets by 2015. doing relatively better are Angola, Cambodia, Six LICUS—Burundi, the Democratic Republic and Lao People’s Democratic Republic. These of Congo, Liberia, Nigeria, Papua New Guinea, countries will be able to achieve three of the six and Sudan—are unlikely to meet four or more MDG targets if they continue to progress accord- of the six (for which the data is available) MDG ing to the past trend. While this analysis is targets. Another seven LICUS—the Republic of simplistic, it does shed light on the challenge Congo, Haiti, Myanmar, Tajikistan, Togo, the World Bank and other donors face in helping Uzbekistan, and Zimbabwe—are unlikely to the LICUS to achieve the MDGs. Improving, Goal Target Indicator Likely Possible but unlikely Unlikely Goal 1: Eradicate Reduce by half Population below Angola, Cambodia, Myanmar Comoros, Solomon Burundi, extreme poverty the proportion of minimum level of Central African Islands,a Democratic and hunger people who suffer dietary energy Republic, Republic Zimbabwe, LIC Republic of from hunger consumption (%) of Congo, Haiti, Congo, Liberia, Lao People’s Nigeria, Sudan, Democratic Tajikistan, Togo, Republic Uzbekistan Goal 2: Achieve Ensure that all Primary Cambodia, Lao Comoros, Republic of Congo, Burundi, Demo- universal primary boys and girls completion rate, People’s Demo- Uzbekistanc Myanmar, Papua cratic Republic of education complete a full total (% of cratic Republic, New Guinea, Congo, Guinea- course of primary relevant age Liberia,b São Sudan, Nigeria,b Bissau, schooling group) Tomé and Principe, LIC Zimbabwe Solomon Islands, Tajikistan, Togo Goal 3: Promote Eliminate gender Ratio of girls to Angola, Myanmar, Cambodia, Nigeria Burundi, gender equality disparity in pri- boys in primary Sudan, Uzbekistan, Comoros, Lao Republic of and empower mary and second- and secondary LIC People’s Demo- Congo, Liberia, women ary education, education (%) cratic Republic, Tajikistan preferably by 2005, Papua New and at all levels Guinea, Togo, by 2015 Zimbabwec (Continues on the following page.) 83 E N G A G I N G W I T H F R A G I L E S T AT E S Improving, Goal Target Indicator Likely Possible but unlikely Unlikely Goal 4: Reduce Reduce by two- Under-5 mortality Lao People’s Comoros, Solomon Guinea-Bissau, Angola, Burundi, child mortality thirds the mor- rate (per 1,000) Democratic Islands Haiti, Myanmar, Cambodia, Cen- tality rate among Republic Nigeria, Papua tral African Re- children under New Guinea, public, Democra- five Sudan, Tajikistan, tic Republic of Timor-Leste, Togo, Congo, Republic Uzbekistan, LIC of Congo, Liberia, São Tomé and Principe, Somalia, Zimbabwe Goal 5: Improve Reduce by Maternal maternal health three-quarters mortality ratio the maternal mortality rate Goal 6: Combat Halt and begin Prevalence of Burundi, Lao People’s Angola, Haiti, HIV/AIDS, to reverse the HIV, total (per- Cambodia, Repub- Democratic Liberia, Sudan, malaria, and spread of cent of popu- lic of Congo, Republic,d Myanmar, Papua other diseases HIV/AIDS lation aged Nigeria, Democratic New Guinea 15–49) Zimbabwe, Togo Republic of Congo,d Central African Republic,d Uzbekistan,d LIC d Goal 7: Ensure Reduce by half Access to an Angola, Burundi, Nigeria Democratic Papua New environmental the proportion improved water Central African Republic of Guinea, Togo, sustainability of people without source (percent Republic, Congo, Haiti, Uzbekistan sustainable of population) Comoros, Liberia, Sudan access to safe Myanmar, drinking water Zimbabwe, LIC Source: World Bank 2005j. Note: LIC refers to low-income countries with a Gross National Income (GNI) per capita of $825 or less in 2004. (i) The calculations are based on past trends between two points, the earliest taken from the period 1990–94 and the other from 1997 to 2003. Then it was determined how long it would take a country to achieve the MDG based on the growth rate between the two points (in a few cases extrapolations are based on data from 2000 and 2003, for example, Haiti Target 1). Countries that would achieve the MDG based on the past trend on or before 2015 were considered “likely”; countries that were close to the target by 2015, or close to the target in 2003, regardless of past trends, were considered “possible”; countries that are moving in the right direction, but are unlikely to achieve the target by 2015, were considered “improving, but unlikely”; and countries that have made no progress or are moving in the reverse direction were considered “unlikely.” (ii) No data available on Afghanistan and Kosovo; LICs include all countries with a Gross National Income (GNI) per capita of $825 or less in 2004. a. Because data were lacking on the indicator, an alternative indicator—prevalence of underweight in children (under five years of age) was used. b. Because data were lacking on the indicator, an alternative indicator—net primary enrollment ratio (% of relevant age group)—was used. c. “Possible” if declining trend controlled. d. Stagnant growth rates for prevalence of HIV, total (% of population aged 15–49). 84 APPENDIX D: FOUR LICUS BUSINESS MODELS Deterioration • Public administration, service delivery, and • Interim Strategy Note, focusing on stemming economic development to address areas with decline in governance and social services, and crime, insecurity, or conflict. contributing in economic and development areas to multidonor conflict-prevention ef- Prolonged Crisis or Impasse forts. Limited new financing; focus on portfo- • Interim strategy note, focusing on maintaining lio restructuring. operational readiness for reengagement and • Increased use of CDD, private sector, NGO, and providing economic inputs to early peace or ring-fenced mechanisms (including service de- reconciliation dialogue. livery and local economic development in areas • Small grant-based finance, aiming at local eco- of insecurity). nomic development and protection of human • State capacity and accountability: focus on capital, generally through nongovernmental transparency, dialogue, and maintaining insti- recipients (including service delivery and local tutional capital to facilitate eventual economic development in areas of insecurity). turnaround. • Capacity and accountability: focus on institu- • Contribute to community-level conflict pre- tional analysis, dialogue, and counterpart training. vention and to multidonor efforts for peace- • Use of socioeconomic issues for restoration of building or governance reform at a national dialogue/identification of entry points for level. change. Post-Conflict or Political Transition Gradual Improvement • Interim Strategy Note, focusing on rebuilding • Country Assistance Strategy, focusing on build- state capacity and accountability and deliver- ing state capacity and accountability, achieving ing rapid, visible development results in sup- selective development results, and boosting port of peace building. support for reform currents, supported by • Exceptional IDA allocation. moderate IDA allocation. • Joint needs assessment/recovery planning that • Activities to boost domestic reform currents, links political, security, economic, and social including leadership support, communications recovery. initiatives, training, and capacity building. • State capacity and accountability: support for • State capacity and accountability: development a broad state-building agenda through insti- policy operations (where appropriate and re- tution building and, where appropriate, de- stricted in volume), supported by sector and velopment policy operations with robust capacity-building projects and with strong over- oversight mechanisms and sector programs sight mechanisms. Asymmetric reforms. (including transitional projects that work • Public administration, service delivery, and through CDD or NGO mechanisms). Leader- economic development to address areas with ship and civil society support. crime, insecurity, or conflict. Source: World Bank 2005e. 85 APPENDIX E: SUMMARY OF OECD-DAC PRINCIPLES FOR INTERNATIONAL ENGAGEMENT IN FRAGILE STATES • Take context as the starting point. Conduct po- • Promote coherence between donor agencies. In- litical analysis above and beyond quantitative volve those responsible for security, political, and indicators of governance, institutional strength, economic affairs as well as those responsible for or conflict. development aid and humanitarian assistance. • Move from reaction to prevention. Share and re- • Agree on practical coordination mechanisms be- spond to risk analysis, address the root causes tween international actors. Include upstream of state fragility, and strengthen the capacity of analysis, joint assessments, shared strategies, regional organizations. coordination of political engagement, joint of- • Focus on state building as the long-term vision. fices, multidonor trust funds, and common Strengthen the capacity of state structures to reporting frameworks. perform core functions. Help ensure the legit- • Do no harm. Avoid activities that undermine na- imacy and accountability of those structures tional institution building, such as bypassing and their ability to provide an enabling envi- budget processes or setting high salaries for ronment for strong economic performance. local staff. • Align with local priorities. Acknowledge and ac- • Mix and sequence instruments. Use both state re- cept priorities where governments demon- current financing and nongovernmental de- strate the political will to foster their countries’ livery to fit different contexts. development; where donor and government • Act fast . . . and with flexibility at short notice consensus is lacking, seek wider consultations when opportunities occur . . . and partial or shadow alignment. • . . . but stay engaged long enough to give success • Recognize the political-security-development a chance. Capacity development in core insti- nexus. Support national reformers in devel- tutions will take at least 10 years. oping unified planning frameworks for politi- • Avoid pockets of exclusion. Address “aid or- cal, security, humanitarian, economic, and phans” and coordinate with donors to pre- development activities at the country level. vent excessive aid volatility. Sources: OECD 2005c, http://www.oecd.org/dataoecd/59/55/34700989.pdf. 87 APPENDIX F: FOUR BANK THEMES BASED ON THE OECD-DAC PRINCIPLES • Building state capacity and accountability. A long- difficult political environments mean that frag- term focus on state capacity and accountabil- mented international dialogue or donor pro- ity is critical if these countries are ever to find grams are unlikely to deliver results. a durable exit from crisis. • Institutional flexibility and responsiveness. Donor • Peace, security, and development linkages. Polit- organizational responses must be calibrated ical, security, and development linkages are to the specific needs of the countries, acting particularly important in fragile states. faster and more flexibly, staying engaged for the • Donor coordination for results. Particularly close long term, and coordinating to address prob- partnerships between international actors are lems of aid orphans and donor-driven aid needed, because low counterpart capacity and volatility. Source: World Bank 2005e. 89 APPENDIX G: FISCAL 2003–05 COUNTRY STRATEGY DOCUMENTS FOR FISCAL 2005 LICUS Fiscal year Date Country Strategy document 2003 Jul-02 Kosovo Transitional Support Strategy Feb-03 Afghanistan Transitional Support Strategy Feb-03 Tajikistan Country Assistance Strategy Mar-03 Angola Transitional Support Strategy Apr-03 Somalia Country Reengagement Note Jun-03 Sudan Country Reengagement Note Jan-03 Haiti Country Reengagement Note 2004 Aug-03 Republic of Congo Transitional Support Strategy Jan-04 Democratic Republic of Congo Transitional Support Strategy Mar-04 Liberia Country Reengagement Note Apr-04 Kosovo Transitional Support Strategy 2005 Jul-04 Central African Republic Country Reengagement Note Nov-04 Togo Country Reengagement Note Dec-04 Haiti Transitional Support Strategy Jan-05 Angola Interim Strategy Note Mar-05 Lao People’s Democratic Republic Country Assistance Strategy Mar-05 Papua New Guinea Strategy Note Mar-05 Zimbabwe Interim Strategy Note Apr-05 Burundi Transitional Support Strategy Apr-05 Cambodia Country Assistance Strategy May-05 São Tomé and Principe Country Assistance Strategy May-05 Solomon Islands (Pacific Islands) Country engagement Jun-05 Nigeria Country Assistance Strategy Jun-05 Timor-Leste Country Assistance Strategy Total number of strategies, fiscal 2003–05: 24 Source: World Bank database. 91 APPENDIX H: IMPLICATIONS OF THE HIPC INITIATIVE FOR LICUS Background work was carried out on the • It may be more difficult for LICUS to establish a implications of the Enhanced HIPC Initiative for macroeconomic record than it is for other countries. LICUS as part of this review.1 The main findings This is not to imply that entry requirements should are presented below. be relaxed, but rather that a concerted effort should be made to help LICUS meet HIPC requirements. • LICUS have lagged behind in establishing a record to qualify for HIPC debt relief, but a large number The weak capacity and volatility in LICUS will of these countries could potentially benefit from make it more difficult for them to establish a debt relief if they fulfill requirements for eligibil- policy record before HIPC sunset in 2006, ity before the sunset clause expires at end 2006. suggesting that the Bank needs to pay more attention to supporting policy measures Of the 25 LICUS, none has reached the necessary for LICUS to meet HIPC decision point completion point, five are between the decision and completion point requirements. This is and completion points2 (Burundi, Democratic particularly important in light of the Multilateral Republic of Congo, Guinea-Bissau, Republic of Debt Relief Initiative for 100 percent debt cancel- Congo, and São Tomé and Principe), and seven lation for completion point HIPC countries, are potentially eligible for the HIPC Initiative which will be implemented in July 2006 (World (Central African Republic, Comoros, Haiti, Bank 2006e). A 2003 IEG evaluation of the HIPC Liberia, Somalia, Sudan, and Togo).3 Three of Initiative found that in the case of the “millen- these potentially eligible LICUS (Central African nium rush” countries that qualified in late 2000, Republic, Haiti, and Togo) satisfy the policy the relaxation of eligibility requirements raises performance criterion for eligibility under the the risk of not achieving HIPC objectives, given Enhanced HIPC Initiative. The Central African that the majority of these countries experienced Republic and Haiti are moving toward qualifica- policy slippages after reaching their decision tion for decision point under HIPC, because points and are ceteris paribus less likely to they have Emergency Post-Conflict Assistance achieve good development results (IEG 2003a). arrangements and are preparing their PRSP and Similarly, Collier (2005) argues that “debt relief is I-PRSP, respectively. Togo has expressed its an aid modality that, unless carefully managed, willingness to seek support for its programs as comes closest to turning aid into oil.”4 soon as the security condition stabilizes and has recently prepared an I-PRSP (Interim Poverty • Changes under IDA 14 linking grant financing with Reduction Strategy Paper), although the paper debt distress are a welcome development, par- has not yet been submitted to the Boards of the ticularly for LICUS, but long-term debt sustainabil- Bretton Woods Institutions. The other four ity depends on sustained improvements in policy. potentially eligible LICUS (Comoros, Liberia, Somalia, and Sudan) have not had an IMF- or During the IDA 14 discussions it was agreed IDA-supported program since 1995, which they that debt sustainability will be the basis for the would need before end-2006 in order to be allocation of grants to IDA-only countries in IDA eligible for relief under the enhanced HIPC. 14, so that the share of grants in total IDA financ- 93 E N G A G I N G W I T H F R A G I L E S T AT E S ing will emerge from a country-by-country and the grant allocation mechanism are analysis of the risk of debt distress (IDA 2005, important steps in helping LICUS relieve their p. 25). The Joint Bank-Fund debt sustainability debt burden. However, as suggested by a recent framework, which links the risk of debt distress IEG Evaluation Update on HIPC (IEG 2006b), to the quality of policies and institutions in low- debt reduction alone is not a sufficient instru- income countries,5 was endorsed as the analyti- ment to affect the multiple drivers of debt cal underpinning for the link between debt sustainability.6 Sustained improvements in sustainability and grant eligibility (IDA 2005, export diversification, fiscal management, and p. 25). This move is particularly important for public debt management are also needed (IEG debt sustainability in LICUS, since under the 2006b). Moreover, as suggested by Collier new framework, LICUS qualify for 100 percent (2005), debt relief faces a potentially severe time grant financing. In addition, an exception to the consistency problem—that is, once debts are debt-distress grant eligibility criterion was made cancelled, there is no incentive for the govern- for Kosovo and Timor-Leste, which are thus ment to abide by any continuing conditions. made eligible for grants. Sustained improvements in policies in LICUS Both the extension of the HIPC sunset clause will be crucial for debt sustainability. 94 APPENDIX I: BREAKDOWN BY COUNTRY: LENDING AND ADMINISTRATIVE BUDGET—TOTAL AND FOR ANALYTICAL WORK Table I.1: Total Dollar (US$) Amounts: Lending and Administrative Budget (total and for analytical work) Lending Administrative Analytical work (million US$) budget (‘000 US$) (‘000 US$) Fiscal Fiscal Fiscal Fiscal Fiscal Fiscal Country 2000–02 2003–05 2000–02 2003–05 2000–02 2003–05 Afghanistan 100 793 4,190 19,992 1,094 7,693 Angola 33 176 2,886 5,791 537 1,987 Burundi 131 223 2,942 4,270 385 860 Democratic Republic of Congo 500 1,332 4,621 11,422 956 2,990 Guinea-Bissau 51 14 2,193 2,658 197 494 Republic of Congo 90 110 1,934 4,407 582 922 Timor-Leste 112 15 6,146 5,915 2,401 1,934 Post-conflict 2,664 54,455 16,879 Cambodia 135 167 9,624 13,937 1,700 4,073 Central African Republic 45 0 1,328 1,878 98 758 Comoros 17 13 2,143 1,759 137 211 Haiti 0 75 1,788 3,623 202 1,661 Kosovo 63 30 7,240 5,921 2,302 1,946 Lao People’s Democratic Republic 87 136 7,525 17,598 638 2,065 Liberia 0 0 386 2,069 169 1,239 Myanmar 0 0 320 322 320 272 Nigeria 682 881 19,432 22,755 7,229 5,352 Papua New Guinea 190 0 5,098 5,426 1,144 2,183 São Tomé and Principe 10 12 1,538 2,107 457 448 Solomon Islands 4 0 823 639 37 290 Somalia 0 0 37 1,084 25 831 Sudan 0 0 908 5,354 778 3,900 Tajikistan 122 59 6,060 11,605 748 3,578 Togo 0 0 2,656 1,729 443 1,017 Uzbekistan 105 100 6,341 7,530 1,518 2,118 Zimbabwe 5 0 5,506 1,332 1,245 855 Non-post-conflict 1,473 106,668 32,899 LICUS 2,480 4,137 103,665 161,123 25,342 49,778 Non-LICUS LICs 18,557 20,400 379,941 449,637 69,088 109,947 Non-LICUS LICs (excluding India) 12,011 14,569 318,967 380,754 56,040 89,723 Source: World Bank database. 95 E N G A G I N G W I T H F R A G I L E S T AT E S Table I.2: Per Capita Annual Amounts: Lending and Administrative Budget (total and for analytical work) Lending Administrative Analytical work (US$) budget (US$) (US$) Fiscal Fiscal Fiscal Fiscal Fiscal Fiscal Country 2000–02 2003–05 2000–02 2003–05 2000–02 2003–05 Afghanistan 1.2 9.2 0.051 0.232 0.013 0.089 Angola 0.9 4.3 0.075 0.143 0.014 0.049 Burundi 6.3 10.3 0.141 0.198 0.018 0.040 Democratic Republic of Congo 3.3 8.4 0.031 0.072 0.006 0.019 Guinea-Bissau 12.1 3.1 0.520 0.595 0.047 0.111 Republic of Congo 8.4 9.8 0.181 0.391 0.055 0.082 Timor-Leste 46.6 5.7 2.560 2.248 1.000 0.735 Post-conflict 8.2 0.167 0.052 Cambodia 3.5 4.2 0.248 0.347 0.044 0.101 Central African Republic 4.0 0.0 0.118 0.161 0.009 0.065 Comoros 10.1 7.2 1.249 0.977 0.080 0.117 Haiti 0.0 3.0 0.073 0.143 0.008 0.066 Kosovo 8.7 4.2 1.006 0.822 0.320 0.270 Lao People’s Democratic Republic 5.3 8.0 0.464 1.036 0.039 0.122 Liberia 0.0 0.0 0.040 0.204 0.018 0.122 Myanmar 0.0 0.0 0.002 0.002 0.002 0.002 Nigeria 1.7 2.2 0.050 0.056 0.019 0.013 Papua New Guinea 12.0 0.0 0.323 0.329 0.073 0.132 São Tomé and Principe 22.1 25.4 3.393 4.462 1.008 0.949 Solomon Islands 3.1 0.0 0.636 0.466 0.029 0.212 Somalia 0.0 0.0 0.001 0.038 0.001 0.029 Sudan 0.0 0.0 0.009 0.053 0.008 0.039 Tajikistan 6.5 3.1 0.324 0.614 0.040 0.194 Togo 0.0 0.0 0.190 0.119 0.032 0.070 Uzbekistan 1.4 1.3 0.085 0.098 0.020 0.028 Zimbabwe 0.1 0.0 0.143 0.034 0.032 0.022 Non-post-conflict 1.5 0.110 0.034 LICUS 2.0 3.2 0.084 0.124 0.020 0.038 Non-LICUS LICs 3.4 3.6 0.070 0.080 0.013 0.020 Non-LICUS LICs (excluding India) 5.2 6.0 0.138 0.158 0.024 0.037 Source: World Bank database. 96 APPENDIX J: POLITICAL ANALYSIS RELEVANT FOR STRATEGY DEVELOPMENT IN LICUS The objective of a country team should be to sectors. Structural analysis focuses on the commission or consume (not necessarily weakness of a particular state, leading to the produce) analysis that is directly relevant to and development of strategies that avoid undermin- usable in the development of a strategy. Many ing it further. types and layers of political analysis are useful Analysis of day-to-day politics needs to go for this purpose. Some illustrations are outlined beyond the political gossip about who is up and below. who is down in the capital, but also needs to Political risk analysis underlies the decision look at the distribution of power among differ- of whether the Bank should engage in a certain ent political forces in the capital and in the country and, if so, how it should engage. This regions, or even localities. Such analysis is analysis can provide an overall evaluation of the crucial in deciding, for example, whether a more level of stability and instability in the country and centralized or decentralized approach to reform suggest the different scenarios that might in a particular sector is desirable. Depending on unfold. Political risk analysis needs to be the situation, decentralization may lead to repeated regularly in LICUS. policies that are more responsive to local needs Structural analysis seeks to identify major or, for example, put even more power in the characteristics of the political situation that will hands of warlords. affect the work of the Bank, no matter what the Analysis of the history of reform in the specific strategy chosen. For example, structural country, and often in neighboring ones, is analysis focuses on the existence of major ethnic critical for designing an effective strategy. or religious conflicts in the country and the Botched privatization attempts that created a distribution of various groups over the national strong backlash in the recent past, for example, territory that may transform a policy that would would suggest that further privatization should work well in a homogeneous environment into not be part of the Bank strategy immediately, but one that discriminates against a particular a compromise solution might be sought group. This analysis can be carried out at the instead—for example, commercialization rather national, regional, or local level, or for particular than privatization of utilities. Source: Background work undertaken by Marina Ottaway for this review, 2005. 97 APPENDIX K: CAPACITY DEVELOPMENT PROJECTS IN LICUS Over fiscal 2000–05, 206 projects were approved countries emerging from conflict—many of which in the 25 LICUS—148 IDA, 10 IBRD, 38 Special are LICUS—through research, planning, piloting, Financing, and 10 project-related Global Environ- and analysis of ground-breaking activities. ment Facility (GEF) grants. Of these 206 projects, Given the large number of trust funds, the 30 were free-standing capacity development, analysis was restricted to the LICUS Trust Fund, institutional strengthening, or technical PCF, and IDF. Eighteen grants under the LICUS assistance projects, or were projects in which Trust Fund and 30 grants under PCF were free- such components added up to 80 percent of the standing capacity development, institutional total project cost (see table below). strengthening, or technical assistance grants, or Similarly, over fiscal 2000–05, a large number of were grants in which such components added up grants were approved under different trust fund to 80 percent of the total cost (see table below). programs; for example, 29 grants were approved The list of LICUS Trust Fund capacity- under LICUS Trust Funds and 56 under Institu- development grants was provided by the Bank’s tional Development Funds (IDFs). The Post- LICUS Unit and the list of PCF capacity- Conflict Fund (PCF) is not a trust fund program, development grants was provided by the Bank’s but since 1999 it has been eligible to receive Social Development Department. All IDF grants contributions from donors. PCF supports were deemed to be capacity-development grants. Country Project/program title Instrument Afghanistan Emergency Public Administration Project IDA Programmatic Support for Institution Building IDA Second Emergency Public Administration Project IDA Public Administration Capacity Building Project IDA Enhancing Knowledge and Partnerships PCF Reconstruction Strategy for Afghanistan with Afghan and Other Stakeholder Participation PCF Afghanistan Priority Sectors Support Program, and Launch Package for Community Empowerment Program PCF Launch Package for Community Empowerment Program PCF Afghan Female Teacher in-Service Training in Peshwar (Pakistan) PCF Teacher Training Programs for Afghan Refugees PCF Balochistan Refugee Teacher Training Project PCF Angola Economic Management Technical Assistance IDA Burundi Planning for Burundi’s Future: Building Leadership Capacity PCF Ex-Combatants Assistance (BEAP) PCF (PREVCONB) Program for Prevention of Conflict in Burundi PCF (Continues on the following page.) 99 E N G A G I N G W I T H F R A G I L E S T AT E S Country Project/program title Instrument Cambodia Land Management and Administration Project IDA Rural Investment and Local Governance IDA Economic and Private Sector Capacity Building Project IDA Cambodian Center for Conflict Resolution—Capacity Development Program PCF Central African Republic Policy Support Project IDA Public Financial Management and Education Sector LTF Public Financial Management and Governance LTF Strategic Leadership Seminar for Central African Republic LTF Comoros Support to the Comorian Transition Process LTF Support to the Comorian Transition Process LTF Support to the Comorian Transition Process - Leadership Seminars LTF Anjouan Professional Integration of Militia PCF Transitional Support to Comoros Economic Management PCF Reintegration of Young Militias in Anjouan PCF Democratic Republic of Congo Pilot Post-Conflict Rapid Assessment of Living Conditions and Infrastructure PCF Demobilization & Rehabilitation Program: Preparatory Phase, Social Reintegration Program - Pilot Activities PCF Guinea Bissau Private Sector Rehabilitation and Development IDA Haiti Economic Governance Reform Operation IDA Governance Technical Assistance Grant IDA Support for Economic Governance Reform LTF Building Institutional Capacity & Strengthening Provision of School Feeding Program LTF Haiti Disaster Risk Management Pilot LTF Kosovo Economic Policy/Public Expenditure Management IDA Energy Sector Technical Assistance 2 IDA Business Environment Technical Assistance IDA Energy Sector Technical Assistance 3 IDA Kosovo Youth Development Grant PCF Energy Sector Technical Assistance SF Private Sector Development Technical Assistance SF Lao People’s Democratic Republic Financial Management Capacity Building Credit IDA Liberia Liberia: Community Empowerment Project LTF Liberia: Support for Economic Management & Development Strategy LTF Public Financial Management LTF Reactivation of the Forestry Sector and Forest Management LTF Support to Donor Coordination (RIMCO) LTF Regional Support for Reconstruction of the Liberian Legal System LTF Nigeria Local Empowerment and Environmental Management Project GEF Economic Management Capacity Building IDA State Governance and Capacity Building TAL (fiscal 2005) IDA Economic Reform and Governance Project IDA Papua New Guinea Governance Promotion Adjustment Loan IBRD 100 A P P E N D I X K : C A PA C I T Y D E V E L O P M E N T P R O J E C T S I N L I C U S Country Project/program title Instrument Republic of Congo Transparency and Governance Capacity Building IDA Emergency Support for Integration of Ex-Combatants and Unemployed Youth into Agricultural Sector (FAO) PCF São Tomé and Principe Public Resource Management IDA Public Resource Management Technical Assistance IDA Capacity Building and Technical Assistance IDA Somalia Capacity Building for Somali Planners LTF Livestock Project: Puntland Pastoralists Program LTF Sudan Capacity Building for Development in Post-Conflict Sudan LTF Expanded Watching Brief - Part 1 PCF Expanded Watching Brief - Part 2 PCF Nuba Mountains Project PCF Tajikistan Empowering Women: Socioeconomic Development in Post-Conflict - 1st Phase PCF Women’s Empowerment and Socio-economic Development - 2nd Phase PCF Timor-Leste Transition Support Program SF Administrative Services Capacity Building Project PCF Post-Conflict Reconstruction and Reintegration Program - Part 1 PCF Post-Conflict Reconstruction and Reintegration Program - Part 2 PCF Post-Conflict Reconstruction and Reintegration Program - Part 3 PCF Support to Poor Widows: Widows, War and Welfare PCF Capacity Building Assistance & Development Project PCF Leadership Capacity Building for Economic Development (LED) PCF Economic Institutions for Capacity Building SF 101 APPENDIX L: ACCOUNTABILITY IN THE MANAGEMENT OF NATURAL RESOURCES IN LICUS Background work on accountability in the Bank’s annual audit. In the Republic of Congo, management of natural resources in LICUS was publication of the annual audit of accounts of done as part of this review. The main findings the national oil company was among the post- are presented below. conflict performance indicators (PCPI). The Bank has emphasized issues of governance in Implementation arrangements, however, are natural resource management in country strategies inadequate. in recent years. In the Central African Republic, necessary actions Of the 25 LICUS, 7—Angola, the Central African for the forestry and mining sectors are outlined in Republic, the Democratic Republic of Congo, the the Bank’s country strategy, but it is unclear how Republic of Congo, Nigeria, Papua New Guinea, they will be implemented, or what happens if and Togo—were identified as “extractive indus- they are not implemented. In Papua New Guinea, tries–dependent countries” in the IEG 2005 eval- similarly, the Bank’s Interim Strategy (2005) uation Extractive Industries and Sustainable mentions that better management of revenues Development (IEG 2003b). The LICUS Task Force from the extractive industries sector is a priority, report pointed out high levels of opportunistic and that the Bank will support the International behavior in LICUS, especially in natural resource Finance Corporation’s (IFC’s) investment in the extraction activities, and recommended that mining sector through advisory work (World measures to improve governance and intensify Bank 2005l, p. 27). At the same time, the Interim scrutiny over the uses of natural resource rents be Strategy Note lacks details on its engagement, among the high priorities. A review of the Bank’s benchmarks, milestones, or other monitoring strategies in the seven resource-rich (extractive indicators against which progress could be industries–dependent) LICUS listed above effectively measured. suggests that issues of governance in natural Furthermore, emphasis on governance in resource management have indeed been natural resource management is not LICUS- emphasized in recent years. In most cases, this specific; instead, it is part of the overall trend takes the form of analytical work on governance in within the Bank to base strategies more strongly the natural resource sectors1 or general financial on governance considerations. The IEG evalua- management and fiduciary studies.2 tion of extractive industries suggests, with regard to the evolution of Bank’s policy and role Natural resource management is also included in in extractive industries, that “in the latter part of progress indicators and triggers. the 1990s, there was an increased focus on In Angola, financial support beyond the Transi- reform and deregulation programs in an effort tional Support Strategy was contingent on (i) to further good governance as a central element publication of all government tax revenues and in the improvement of country economic (ii) completion of the Oil Diagnostic Study, and performance” (IEG 2003b, p. 61). Given the movement to the Central Bank of all oil particularly weak governance environment in revenues (except those earmarked to service oil- LICUS, and the abundant evidence of the backed debt) and their inclusion in the Central negative impact of natural resource windfalls, 103 E N G A G I N G W I T H F R A G I L E S T AT E S additional attention and more focused tee with members from civil society and Parlia- approaches may be required. Even in the Chad- ment, and a Future Generations Fund in the Cameroon Oil Pipeline Project, where the Bank amount of 10 percent of oil revenues, these applied some of the strongest safeguards, provisions proved to be insufficient. The in- including a revenue management law designed dependent oversight committee proved to be to ensure that earnings from oil are directed understaffed and did not have sufficient informa- toward poverty reduction, an oversight commit- tion from the government and Exxon Mobil. 104 APPENDIX M: DEVELOPMENT POLICY LENDING The LICUS Initiative suggested that while ment Credit experienced a one-year delay). development policy lending (DPL) is not always A review of adjustment operations approved appropriate in all fragile-state contexts, it could during fiscal 2002–04 and evaluated by IEG1 be under two business models—the post- suggests a direct relationship between outcomes conflict or political transition and gradual and institutional quality (table M.1). While unsatis- improvement business models. When success- factory outcomes are few,2 they tend to be identi- ful, DPL can potentially deliver larger, country- fied with countries that have lower CPIA ratings. wide benefits by stabilizing government during a While the experience of adjustment operations transition, alleviating liquidity pressures in a approved during fiscal 2002–05 is similar to that of cash-strapped environment, supporting institu- investment projects, there is a notable difference tion building, and fostering harmonized donor in borrower performance (tables M.2 and M.3). support for a focused set of policy and institu- The stronger link with CPIA in adjustment tional actions. According to the OPCS Note on operations can be explained in part by their Development Policy Operations in Fragile States heavier reliance on budgetary and financial (World Bank 2005f), where revenue collections management procedures of partner countries and are weak, the stability of state institutions and agreement on overall development objectives. improvements in service delivery will require Similarly, a review of recent adjustment and budgetary support, as well as a rapid donor investment lending in LICUS3 suggests that response, in order to maintain momentum. investment lending has fared somewhat better in DPL was introduced in 2004 and its use in overall outcome attainability and institutional LICUS has so far been minimal—two approved development impact. While 44 percent of adjust- DPL operations (in Lao PDR) and nine more in ment operations (4 out of 9) resulted in unsatis- the pipeline for fiscal 2005–06. Adjustment factory results,4 similar outcomes are found in lending, which DPL replaces, has also been only 18 percent (4 out of 22) of investment limited, with only nine operations approved projects. during fiscal 2002–05. This limited experience reveals better outcomes associated with post- conflict transitions (Democratic Republic of Table M.1: Mean CPIA, by IEG Outcome Ratings Congo, Kosovo, Timor-Leste); government commitment was an important success factor. At Standard the same time, the Financial Management Adjust- Outcome mean Mean deviation Obs. ment Credit in Lao PDR faced weak compliance Highly satisfactory 3.93 0.41 4 and government resistance to reforms, which Satisfactory 3.71 0.43 41 produced unsatisfactory outcomes. In design, programmatic single-tranche operations have Marginally satisfactory 3.67 0.41 30 also performed better than multiple-tranche Marginally unsatisfactory 3.55 0.07 2 loans by avoiding second-tranche release delays Unsatisfactory 3.25 0.30 6 caused by difficulties in fulfilling release Total 3.67 0.43 83 conditions (the Financial Management Adjust- Source: IEG and World Bank databases. 105 E N G A G I N G W I T H F R A G I L E S T AT E S Table M.2: Performance and CPIA in Adjustment countries with a good track record, strong Lending (fiscal 2002–05) ownership of the reform program, a reasonably sound policy and institutional framework, and Overall borrower Mean Standard commitment and sufficient capacity to allocate performance CPIA deviation Obs. resources effectively and in accordance with Highly satisfactory 4.00 0.26 3 development priorities.” Indeed, most of the Satisfactory 3.69 0.44 71 Poverty Reduction Support Credits (PRSCs) to Unsatisfactory 3.43 0.17 9 date have gone to countries in the top two Total 3.67 0.43 83 quintiles of the CPIA distribution. Therefore, careful consideration of the appropriateness of Source: IEG and World Bank database. DPL in countries with no obvious political or post-conflict turnaround and weaker govern- ment ownership and reform consensus will be Table M.3: Performance and CPIA in Investment particularly important. Lending (fiscal 2002–05) This does not necessarily imply, however, resorting to free-standing investment projects. Overall borrower Mean Standard Approaches such as SWAps may also be consid- performance CPIA deviation Obs. ered as they too address the limitations of Satisfactory 3.78 0.35 9 fragmented project approaches and provide Unsatisfactory 3.76 0.66 7 benefits similar to budget support operations, Total 3.77 0.49 16 while allowing for additional safeguards through Source: IEG and World Bank database. the use of various financing modalities (budget support, pooled and project financing) within a Recent experience suggests that while there common program, as well as common policy may be a prima facie argument for providing dialogue and joint monitoring against one set of budget support in post-conflict countries, their targets and indicators.5 At the same time, SWAps higher dependence on institutional quality and may not be an approach of choice, given their good borrower performance will require a more long-term view, when the goal is alleviating cautious approach when considering DPL in short-term cash needs, for instance. The choice LICUS, as compared with non-LICUS, as well as of assistance modalities will be a complex one, careful design and additional monitoring. As particularly in fragile environments. The pros suggested by Koeberle and Stavreski (2005), and cons of different options should be weighed “budget support is most appropriate for (box M.1) in light of country conditions. Box M.1. Projects versus Budget Support: Pros and Cons Projects can facilitate implementation and monitoring, both in alignment of donor funds with the government’s budget cycle and terms of the Bank’s ability to ensure quick project implementa- the often off-budget aid flow that limits the predictability of aid flows. tion and to collect the necessary data to report on project Benefits associated with budget support and SWAps include: (i) progress. increased predictability of funds; (ii) greater efficiency of budgetary Common criticisms of the project approach include: (i) frag- programming and spending; (iii) capacity development; (iv) greater mented environment that is not conducive to the formulation of a uni- ownership on part of the government; and (v) being in line with cur- fied long-term reform program by the government; (ii) parallel rent Bank strategy, as embodied in the Comprehensive Development implementation mechanisms that fail to facilitate, or even undermine, Framework. longer-term institutional development; (iii) increased transaction In very poor governance environments, the Bank may find it dif- costs associated with duplication of effort necessary to meet different ficult to monitor and control the use of resources provided through procedural requirements and multiple donor missions; and (iv) mis- budget support and SWAps. 106 APPENDIX N: THE CHALLENGES OF ENGAGING CIVIL SOCIETY ORGANIZATIONS IN LICUS From its Angola, Guinea-Bissau, and Togo case affected and fragile states, because it gave the studies, the World Bank report on engaging civil organizations limited opportunity to develop society organizations (World Bank 2005d) capacity, specialization, strategic planning, and concluded that: long-term investments in beneficiary communi- ties. Competition for scarce resources made • In Angola, extensive donor presence during the CSOs donor-driven, with accountability focused conflict led to a significant yet uncoordinated upward to donors rather than downward to rise of civil society organizations (CSOs) dom- citizens. The report’s main recommendation inated by high-capacity international NGOs. was for donors to shift from the project-by- • In Guinea-Bissau, several NGOs support the project approach of supporting CSOs to a more CSOs that were created by the citizens to coun- sustained engagement, with less ad hoc project teract a weak state and other problems, but do funding and one-time training events and more so project by project, lacking the resources systematic cooperation and commitment, and capacity to build institutions and ensure including partnering and funding the long-term sustainability. institutional development of CSOs. • In Togo, neither government nor civil society To understand the challenges of working with is able to provide minimal social services be- civil society organizations in LICUS better, the cause of a repressive state and drastic donor Participation and Civic Engagement Group and cutbacks, enabling fraudulent NGOs to take CPR Unit are piloting a Civil Society Assessment advantage of poor communities. Tool. On May 25, 2006, the Bank and Inter Action hosted a joint workshop on CSOs in fragile The study found that financing CSOs project states. The results from the workshop are by project was especially problematic in the expected to provide input to the OECD-DAC rapidly changing environments of conflict- work on service delivery in fragile states. Sources: World Bank 2004c, 2005d. 107 APPENDIX O: THE WORLD BANK’S ECONOMIC AND SECTOR WORK IN LICUS There has been an increase in the number of core diagnostic ESW product (minimum or economic and sector work (ESW) products in otherwise) over fiscal 2003–05: Afghanistan, the fiscal 2003–05 compared with fiscal 2000–02 for Central African Republic, Comoros, Haiti, both LICUS and non-LICUS LICs. While the Liberia, Myanmar, the Solomon Islands, and increase in the number of ESW products was 60 Zimbabwe. Overall, countries with 3 or more percent in non-LICUS LICs, it was 166 percent in core diagnostic reports have increased from 2 to LICUS (table O.1). 10 (5 times) among LICUS, compared with an The 2002 LICUS Task Force Report noted that a increase from 8 to 20 (2.5 times) among non- minimum set of good practice ESW should be LICUS LICs (table O.2). feasible, even in countries where there is no or little While the administrative budget for ESW in government interest. This “minimum set” of ESW LICUS has more than doubled since the LICUS includes core diagnostic ESW such as Development Initiative, one-fourth or more of LICUS do not Policy Reviews (DPRs), Poverty Assessments (PAs), have any ESW being conducted in Sector Boards Integrative Fiduciary Assessments (IFAs), and such as Education; Environment; Health, Institutional and Governance Reviews (IGRs).1 Nutrition, and Population; Social Development; Country Financial Accountability Assessments Social Protection; Transport; Urban Develop- (CFAAs) and Country Procurement Assessment ment; and Water Supply and Sanitation (table Reports (CPARs) can be integrated into the IGR O.3). This lack of ESW in important Sector where a separate exercise may be difficult. Boards in several LICUS raises some questions Yet there are some LICUS without a single about the effectiveness of future Bank assistance. Table O.1: ESW Products LICUS (25) Non-LICUS (34) Product Fiscal 2000–02 Fiscal 2003–05 Fiscal 2000–02 Fiscal 2003–05 Core diagnostic reports 17 43 67 112 Other diagnostic reports 3 29 17 105 Advisory reports 22 52 126 158 Not assigned 23 49 141 188 Total 65 173 351 563 Source: World Bank database. 109 E N G A G I N G W I T H F R A G I L E S T AT E S Table O.2: Core Diagnostic ESW Reports by Country Country Fiscal 2000–02 Fiscal 2003–05 Post-conflict LICUS 13 Afghanistan Angola CFAA, CPAR, IFA Burundi CFAA, CPAR Democratic Republic of Congo CFAA, CPAR, PER Guinea-Bissau PER Republic of Congo PER Timor-Leste CFAA, CEM CPAR, PA, PER Non-post-conflict LICUS 30 (29) Cambodia PA CFAA, CPAR, PER Central African Republic Comoros Haiti CPAR Kosovo PA CEM, CFAA, PER, PA Lao People’s Democratic Republic CFAA, CPAR, PER CEM Liberia Myanmar CEM Nigeria CFAA, CPAR, PER, IGRa CFAA, CPAR, PA Papua New Guinea PA CPAR, PA, PER São Tomé and Principe CEM,PA Solomon Islands Somalia CEM Sudan CEM (2) Tajikistan PA, CEM CFAA, CPAR, PA, PER Togo CPAR, DPR, PA Uzbekistan CPAR CEM, CFAA, PA, PER Zimbabwe LICUS (25) 16 43 (42) Non-LICUS LICs (34) 66 (56) 111(99) Source: World Bank database. Note: CEM = Country Economic Memorandum, CFAA = Country Financial Accountability Assessment, CPAR = Country Procurement Assessment Report, DPR = Development Policy Review, IFA = Integrative Fiduciary Assessment, IGR = Institutional and Governance Review, PA = Poverty Assessment, PER = Public Expenditure Review. The IGR has also been included in the list of LICUS core diagnostic reports because the 2002 LICUS Task Force report identified it as an essential piece of ESW for LICUS. 110 APPENDIX O: THE WORLD BANK’S ECONOMIC AND SECTOR WORK IN LICUS Table O.3: Number of Countries Covered by a Sector Board’s ESW Product, Fiscal 2003–05 Sector Board LICUS Non-LICUS Economic policy 15 29 Education 6 18 Energy and mining 8 12 Environment 3 11 Financial management 8 25 Financial sector 6 23 Gender and development 4 12 Global information/communications technology 0 2 Health, nutrition, and population 5 20 Operational services 2 3 Poverty reduction 9 23 Private sector development 11 24 Procurement 10 21 Project finance and guarantees 0 1 Public sector governance 10 20 Rural sector 10 16 Social development 3 10 Social protection 6 10 Transport 1 9 Urban development 1 9 Water supply and sanitation 0 11 Source: World Bank database. 111 APPENDIX P: PERFORMANCE OF ACTIVE PROJECTS Projects in 25 LICUS Evaluated by QAG Composition of the LICUS Portfolio Quality Assessment Group (QAG) assessments Over fiscal 2000–05, the Bank had 104–137 show a decline in quality at entry for projects in active projects per year in the 25 LICUS. Over LICUS assessed in fiscal 2000–03. Quality of the same period, the Bank had 465–510 active supervision, however, shows a marked improve- projects per year in the non-LICUS LICs. Table ment from a low of 61 percent before fiscal 2000 to P.2 illustrates the percentage of projects rated as 85 percent for fiscal 2000–03 (table P .1).1 The problems on development objectives and percentage of projects in LICUS rated satisfactory implementation progress and the percentage of for quality at entry and quality of supervision for projects and commitments “at risk” for the fiscal 2000–03 are comparable to the percentage of active portfolio for each year during fiscal projects rated satisfactory in non-LICUS LICs. 2000–05 and the average for two time periods: However, projects in non-LICUS LICs show an fiscal 2000–02 and 2003–05. improvement in both ratings over time, while projects in LICUS show a decline in quality at entry. Table P.1: QAG Ratings for Active Projects Quality at entry Quality of supervision Number of Percent Number of Percent Time period projects satisfactory projects satisfactory LICUS Fiscal 1997–99 12 92 36 61 Fiscal 2000–03 30 84 13 85 Non-LICUS LICs Fiscal 1997–99 89 79 212 70 Fiscal 2000–03 76 84 90 84 Source: World Bank database. 113 E N G A G I N G W I T H F R A G I L E S T AT E S Table P.2: Project Performance of the Active Portfolio Average Fiscal year Fiscal Fiscal 2000 2001 2002 2003 2004 2005 2000–02 2003–05 Post-conflict LICUS Number of projects 39 50 57 49 Net commitments 1,569 2,942 2,840 2,450 Development objective (% problem) 10 0 7 5 Implementation progress (% problem) 13 0 7 6 At risk (%) 41 20 21 26 Realism (%) 44 0 42 32 Non-post-conflict LICUS Number of projects 84 74 80 79 Net commitments 2,200 2,369 2,829 2,466 Development objective (% problem) 11 12 10 11 Implementation progress (% problem) 20 22 15 19 At risk (%) 30 31 24 28 Realism (%) 68 78 68 72 LICUS Number of projects 105 105 117 123 124 137 109 128 Net commitments 2,510 2,220 3,098 3,790 5,340 5,471 2,609 4,867 Development objective (% problem) 14 12 7 11 7 9 11 9 Implementation progress (% problem) 13 12 10 18 13 12 12 14 At risk (%) 27 26 32 33 27 23 28 27 Realism (%) 71 63 32 59 55 58 53 57 Non-LICUS LICs Number of projects 517 521 500 497 488 481 513 489 Net commitments 32,873 34,762 34,130 34,267 33,697 33,529 33,922 33,831 Development objective (% problem) 6 5 5 7 7 8 6 7 Implementation progress (% problem) 10 9 11 9 11 13 10 11 At risk (%) 15 12 18 14 16 19 15 16 Realism (%) 78 87 67 81 85 76 76 80 Source: World Bank database. 114 APPENDIX Q: PERFORMANCE OF CLOSED PROJECTS AND LESSONS IN LICUS This appendix first presents the trends in project Composition of the 2005 Closed LICUS performance in the 25 countries categorized as Projects Evaluated by IEG LICUS by the Bank in fiscal 2005 based on ICR IEG evaluated 1,672 closed projects from fiscal Reviews conducted by IEG for 129 projects that 2000 to June of fiscal 2006. This evaluated cohort closed over fiscal 2000–05. All ICR Reviews in includes 129 projects approved in the 25 LICUS IEG’s ICR Review and Tracking Database for and 529 projects approved in non-LICUS LICs projects in each of the 25 LICUS that were (the approval years are given in figure Q.1). In evaluated by July 2005 were also assessed for nominal net commitment terms, the LICUS implementation experience (107 total). The cohort covers $3.3 billion and the non-LICUS most frequently noted significant outcomes, LIC cohort covers $31.6 billion. Table Q.1 shortcomings, and lessons from these projects illustrates the IEG ratings for the exiting LICUS are presented in tables Q.2–Q.4. and non-LICUS cohorts. Figure Q.1: Approval Years of Evaluated Projects 80 Number of evaluated projects 60 40 20 0 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 Fiscal year LICUS Non-LICUS LICs Source: World Bank database. 115 E N G A G I N G W I T H F R A G I L E S T AT E S Only two projects that were approved after the the improving trend in outcome ratings in LICUS Initiative had been evaluated by IEG as of LICUS over fiscal 2002–04 is due to improved June 2006. Both were rated satisfactory on Bank performance. Ratings for Bank perform- outcome. Project performance of the LICUS ance were found to be significantly correlated cohort (approved prior to the initiative, but (positively) to outcome ratings. exited after it began) has shown an improving The net disconnect has been higher for the trend, from 58 percent for projects exiting in fiscal LICUS cohort than for non-LICUS LICs for all 2003 to 82 percent for projects exiting in fiscal years over fiscal 2000–05, except for 2003 (table 2005. In contrast, the percentage of projects rated Q.l). The net disconnect has, however, declined satisfactory on outcome for the non-LICUS LICs over time for both the LICUS and non-LICUS increased from 70 percent in fiscal 2003 to 77 LICs and was about 6 percent for LICUS and 4 percent in 2005 (figure Q.2, table Q.l). percent for non-LICUS LICs for projects exiting QAG has argued in its fiscal 2004 ARPP that in fiscal 2005. Figure Q.2. Percentage of Projects Rated Satisfactory on Outcome by IEG 100 80 Percent satisfactory 60 40 20 0 2000 2001 2002 2003 2004 2005 [L = 23; NL = 89] [L = 22; NL = 84] [L = 24; NL = 100] [L = 19; NL = 83] [L = 24; NL = 92] [L = 17; NL = 81] Exit fiscal year LICUS (L) Non-LICUS (NL) LICs Source: World Bank database. 116 APPENDIX Q: PERFORMANCE OF CLOSED PROJECTS AND LESSONS IN LICUS Table Q.1: Performance of Projects That Exited and Were Evaluated by IEG Between Fiscal 2000 and 2005 for LICUS and Non-LICUS LICs Exit fiscal year 2000 2001 2002 2003 2004 2005 2000–02 2003–05 LICUS cohort Number of projects 23 22 24 19 24 17 69 60 Net commitments 751 669 468 695 266 420 1,888 1,381 Outcome (% satisfactory) 61 55 50 58 65 82 55 68 Sustainability (% likely) 41 38 32 44 43 67 37 50 Institutional development impact (% substantial) 35 18 14 32 22 59 22 36 Bank overall performance (% satisfactory) 83 59 54 79 63 76 65 72 Borrower overall performance (% satisfactory) 61 50 33 53 67 76 48 65 Net disconnect (%) 17 27 23 11 9 6 22 8 Non-LICUS LICs Number of projects 89 84 100 83 92 81 273 256 Net commitments 4,774 4,412 6,496 4,927 5,625 5,347 15,682 15,899 Outcome (% satisfactory) 71 77 75 70 79 77 74 76 Sustainability (% likely) 63 72 75 70 77 80 70 76 Institutional development impact (% substantial) 42 49 40 46 53 63 43 54 Bank overall performance (% satisfactory) 73 74 75 72 75 78 74 75 Borrower overall performance (% satisfactory) 64 77 74 70 75 72 72 72 Net disconnect (%) 11 17 20 20 8 4 16 11 Source: World Bank database. Note: (i) Exit fiscal year denotes the year in which the project leaves the World Bank’s active portfolio, normally at the end of disbursements—percents exclude projects not rated. (ii) The data for fiscal 2005 exits represent a partial lending sample and reflect all IEG project evaluations through June 2006. The processing of the remainder of the fiscal 2005 exits is on- going and is expected to be completed by the end of fiscal 2006. 117 E N G A G I N G W I T H F R A G I L E S T AT E S Table Q.2: Outcomes of Closed Projects in LICUS Outcome Examples 1 Increased amounts of physical The Lao PDR Southern Province Rural Electrification Project connected 51,805 provincial house- infrastructure constructed or holds (exceeding the target of 50,000) through grid extension, and the GEF-supported off-grid rehabilitated (schools, health component provided electricity to 6,097 households (32% greater than the target of 4,600), facilities, roads, power grids, mainly through solar home systems and microhydropower. It thereby achieved an electrification water and drainage works) ratio in the project provinces of 42%, exceeding significantly the appraisal target of 20%. Angola’s Social Action Project supported the construction of significant amounts of physical infrastructure: 232 schools; 66 health clinics; 338 water and sanitation facilities; 38 productive and 9 economic subprojects. The output was simple but efficient and cheaper than that funded by other organizations. 2 Improved quality of and access While Uzbekistan’s First Health Project experienced difficulties with some of its components, its to social services objective of improving the quality and cost effectiveness of primary health care services was substantially achieved through the construction, consolidation, and rehabilitation of rural medical centers (SVPs); the upgrading of services (clinical, primary and preventive care, child health ser- vices, reproductive health, emergency care, and the provision of drugs, medical supplies, logisti- cal support); and health promotion, including communications equipment, technical assistance, and training. Rehabilitation and equipment of SVPs improved the availability of key primary health care services, with the population’s appreciation of these services (proxy for quality) evi- dent in the dramatic increases in use of services offered (prenatal services, vaccination rates) as well as in the results of a survey. Despite data inconsistencies and difficulty in attributing outcomes solely to this project, Timor- Leste’s Health Sector Rehabilitation and Development Project laid the groundwork for strength- ening the quality and quantity of basic primary health care at the district level. The project’s objective to provide high-priority primary care via contracted NGOs, improve the supply and logistics of essential drugs, rehabilitate and equip health centers, and strengthen administrative/ technical capacity at district and central levels was substantially achieved. Outpatient utilization rates were very encouraging (0.75 visits per capita in 2000 versus 2.13 visits in 2004; target was 2.5 visits), indicating a growing appreciation and trust of government health services by the population as well as the greater availability of health centers. 3 Increased community The Comoros Pilot Agricultural Services Project was restructured at midterm, adding the third participation and new objective of reinforcing the capacity of local communities and producer groups. While neither of the two original objectives was fully achieved, results from demand-driven productive investments showed significantly increased revenue-generating capacity of small farmers and financing productive investments increased producers’ incomes by at least 25 percent. Sixty-one producers’ organizations were established as legal entities, and members received training and are fully functioning; 60 private extension agents were trained and 58 subprojects were approved and financed (116 percent of the target), involving about 1,000 farmers in various crop and live- stock production initiatives. The actual cost of the project was $2.1 million, making this a cost- efficient learning exercise. By financing activities to carry out participatory rural appraisals (PRAs), the objective of Cambo- dia’s Northeast Village Development Project—to introduce a decentralized, participatory poverty- 118 APPENDIX Q: PERFORMANCE OF CLOSED PROJECTS AND LESSONS IN LICUS Outcome Examples focused rural development planning system, starting at the village level with the formation of Village Development Committees (VDCs)—was substantially achieved. The targeting process was satisfactory, VDCs were elected in 120 targeted villages, village-level PRAs were completed, and village action plans were formulated according to the priorities of the villagers. Training was provided to each community and their VDCs, and operation and maintenance com- mittees were organized in participating villages. The objective to gain experience in managing such programs needed by the Cambodian government was also substantially achieved. Technical guidelines on subproject implementation and operations and maintenance were developed, tested, and revised during project implementation, and lessons learned were disseminated through national and provincial-level workshops. 4 Advances in institutional Cambodia’s Disease Control and Health Development Project resulted in enormous strides in development planning, budgeting, and elaboration of specific implementation strategies in all three national disease programs. Health management agreements were set up in all 11 provinces and have become the basis for a realignment of the health system, with national centers responsible for technical direction and strategy and provinces for managing implementation. Substantial capacity building in management and technical areas and effective leadership elevated the National AIDS Office from “a collection of small and scattered donor-supported pilot schemes to a cohesive national program” (ICR) within the Ministry of Health that could spearhead the national response with complete national ownership. Substantial investment in monitoring and evaluation through surveys, surveillances, and outreach programs provided a foundation for and commitment to evidence-based decision making. Tajikistan’s Institution Building Technical Assistance Project helped the government develop a legal basis for privatization. The project conducted training in privatization procedures and had substantial progress in privatizing small-scale enterprises (95 percent privatized) as well as medium- and large-scale firms (95 percent corporatized and 30 percent privatized). A plan for privatization of the cotton processing and marketing organization was prepared, technical assis- tance was provided for privatization of 22 cotton ginneries, and MOA was provided with the required legal framework to initiate the farm restructuring program, including land access rights, transfer of these rights, and implementation of farm restructuring. The project also saw: a new banking law implemented, more efficient payment clearing, training in implementing new pru- dential regulations, a new accounting system, and on-site supervision of banks. Significant progress was made in privatization, the legal basis for private property, and the skills of officials working in these areas. The banking sector was strengthened as a result of a new banking law, better bank supervision, and a payment-clearing system that reduced clearing time from four days to one. Twenty-two state-owned cotton ginneries were prepared for privatization, and a large number of enterprises were privatized. 5 Increased economic stabilization Guinea-Bissau’s Economic Management Credit supported the introduction of prior authorization and improved financial of expenditure commitments by the Ministry of Economy and Finance and partial integration of management recurrent and capital budgets; financed audits of public expenditure procedures identifying (Continues on the following page.) 119 E N G A G I N G W I T H F R A G I L E S T AT E S Table Q.2: Outcomes of Closed Projects in LICUS (continued) Outcome Examples actions to strengthen budgetary management; trained staff in West African Economic and Monetary Union (WAEMU) procedures; and enacted a comprehensive tax reform consistent with WAEMU countries. The country’s economic management improved by taking several steps: increasing the budgetary revenue/GDP ratio by about 5 percentage points to 15.4 percent of GDP during the same period; improving the current primary balance/GDP ratio by almost 4 percent to 5.5 percent of GDP from 1993 to 1997; liquidating or placing under tender 17 public enterprises and transferring 7 to private management; facilitating accession to the WAEMU; improving technical skills of civil servants in key ministries; and improving the informational and financial management of the civil service. The Democratic Republic of Congo’s Economic Recovery Credit aimed to support economic stabi- lization and structural reforms to lay the basis for recovery within the l-PRSP strategy. The 2003 budget was adopted with improved estimates for public and capital expenditures, the budgeting process was streamlined, communications on fiscal data between the Treasury and the Central Bank were improved, and expenditure tracking systems were created to trace spending to ulti- mate beneficiaries, particularly to assess poverty reduction expenditures. The independence of the Central Bank was confirmed by a new charter; an audit of the operations of the Central Bank was completed; new legislation was prepared for financial institutions; audits and strategies were completed to determine the liquidation, privatization, or restructuring of several public and private banks; and a financial sector strategy was adopted. With assistance from other donors, the project permitted the DRC to reestablish relations with the international donor community and regain its creditworthiness, due to the clearance of arrears to the Bank and the IMF. Source: ICR Reviews from the ICR Review and Tracking Database. 120 APPENDIX Q: PERFORMANCE OF CLOSED PROJECTS AND LESSONS IN LICUS Table Q.3: Shortcomings of Closed Projects in LICUS Shortcoming Examples 1 Weak or irrelevant monitoring In Timor-Leste’s Community and Local Governance Project, there was inadequate tracking of proj- and evaluation ect outputs, with certain basic output indicators identified by the project—such as number of O/M committees formed—remaining unmonitored. While the project emphasized gender consid- erations in its design, it failed to prepare gender-disaggregated information concerning the project’s beneficiaries. Another aim of the project was to reduce poverty, with a subcomponent set to specifically measure poverty impact, yet no such measurement was undertaken. One of the four revised project objectives in Togo’s Lome Urban Development Project sought to alleviate urban poverty. The project design assumed that the beneficiaries would primarily be the urban poor, yet it failed to include indicators to monitor and measure the impact on the poor, resulting in insufficient analysis of whether and to what extent the objective had been achieved. 2 Insufficient understanding of In the Comoros Emergency Economic Recovery Credit, although the Bank had identified the risk the political environment: project that the reconciliation process could stall or be reversed, the measure to guard against this risk too complex/ambitious for local was only modest. By only requiring stakeholders to express their commitments to the reconcilia- circumstances and extent of tion process ex ante, the Bank seemed to have seriously underestimated the extent of mistrust political resistance and political disagreements between the varying levels of government. This lack of under- standing on the Bank’s end further spurred on political instability, resulting in limited achieve- ments of the credit’s objectives. The Bank overestimated the borrower’s ability to carry out Nigeria’s Primary Education Project in deteriorating economic and social circumstances. The large-scale cascade model—whereby terti- ary institutions would train trainers, who would train education officers, who would train teachers— that was created for teacher training on textbook use and student assessment was too complex and proved impossible to implement in the context of the Nigerian situation. Certain regional initia- tives linked to the project could not be carried out because of the highly centralized nature of Nigeria’s military government, and extensive governance problems undermined project implementation. 3 Unclear/inappropriate project The design of São Tomé and Principe’s Health and Education Project was flawed. Baseline indica- design, procedures, or poverty tors and quantifiable objectives were not established, and the planned interventions were not targeting at appraisal clearly linked to stated objectives. The health infrastructure investments, which represented more than half of total project costs (the exact percent is not given in the ICR), were highly ineffi- cient. Constructed drainage canals collapsed during the project’s first five years, resulting in their total replacement and project extension. Drainage was, however, not even the correct intervention—as the vector involved prefers small accumulations of clean water—and the num- ber of reported cases of malaria increased approximately 40 percent from 1995 to 2000. Despite the project’s heavy emphasis on infrastructure, infrastructure specialists were not included on Bank supervisory teams until September 2000. Tajikistan’s Post-Conflict Reconstruction Project had an ambiguous project objective, which was treated differently in different project-related documents. The legal agreement cited the objec- tive as addressing specific post-conflict reconstruction needs in order to restore assets and productivity. According to the MOP and Bank ICR, the objective was to implement the Peace Agreement. And in the borrower ICR, the objective was stated as providing assistance and creat- ing favorable conditions for economic growth in the project area. (Continues on the following page.) 121 E N G A G I N G W I T H F R A G I L E S T AT E S Table Q.3: Shortcomings of Closed Projects in LICUS (continued) Shortcoming Examples 4 Procurement problems caused The borrower did not comply with IDA procurement guidelines in Haiti’s Road Maintenance and by weak ownership, insufficient Rehabilitation Project, which resulted in the formal declaration of 19 contracts as misprocured training on Bank procedures, and the suspension of disbursements from the Credit. political interference, and delays In Angola’s Lobito Benguela Urban Environmental Rehabilitation Project, key project management in government formation and procurement decisions were regularly deferred for months because of inadequate communi- cation between the project’s management, implementing entities, and IDA. 5 Overestimation of government Papua New Guinea’s Emergency El Nino Drought Response Project overestimated the country’s and local support, capacity, and institutional capability as well as the commitment to adopt participatory principles. The Bank commitment to project worked on the false assumption that participating provinces would have the recurrent financial implementation resources to support the project activities. The project’s components demanded provincial and district authorities to adopt a more participatory approach to subproject selection and management, yet the two provinces involved demonstrated little eagerness or institutional capacity to do this. Unexpected on the Bank side, despite the clearly distinct roles of the federal and state govern- ments, Nigeria’s federal government was unwilling to involve local communities in the design of the Small Towns Water Project and failed to devolve ownership to local government and commu- nities. The federal government’s unwillingness to pass on completed facilities inhibited the creation of local agreements to operate and manage them. In opposition to the design’s inten- tions, the project further deepened the gulf between the local and federal governments, increasing the mistrust of the former. 6 Delays in implementation Project implementation for Nigeria’s Water Rehabilitation Project was slow from the start. The and audits first civil works contract was not awarded until approximately 30 months after the date of proj- ect effectiveness. This was caused by persistent delays in the design and preparation of bidding documents. While planned in the project design, Kosovo’s Emergency Farm Reconstruction Project did not become effective in time for the first cropping season after the 1999 conflict. The project experi- enced delays in establishing effective institutional arrangements for implementing the project between the three principal parties (IDA, UNMIK, and FAO). 7 Difficulty in recruiting counterpart The impact of training on MOH capacity and service quality remained limited in Guinea-Bissau’s staff and heavy dependence on Social Sector Project, as most of those trained under the project left government service because expatriates because of lack of of the war and low pay relative to donors, NGOs, and other countries. Some staff quit after train- incentives for locals, causing ing when they did not receive expected promotions. Lao PDR’s District Upland Development and high attrition rates and little Conservation Project experienced difficulties in its education initiative because of Department of institutional memory Education tardiness in providing adequate incentives to non-formal education workers in the villages. Little capacity was built in the district and provincial agriculture offices, and there was low com- mitment because of the lack of incentives to cooperate. As a result, the project had problems recruiting provincial and district officers for relocation to the isolated villages and applying existing expertise where it was most needed. Source: ICR Reviews from the ICR Review and Tracking Database. 122 APPENDIX Q: PERFORMANCE OF CLOSED PROJECTS AND LESSONS IN LICUS Table Q.4: Lessons from Closed Projects in LICUS 1. Projects must be especially flexible in an evolving context of fragile and changing circumstances, with continuous reappraisal to see whether they are still practicable and subsequent restructuring to respond appropriately to new conditions. Conducting rigor- ous social and economic evaluations can help make important midcourse changes in project design and implementation. 2. In countries with uncertain economic performance and fragile institutions, the Bank should be particularly vigilant in creating ob- jectives realistically calibrated and focused, taking into account the stability of the political system, degree of administrative capacity, and extent to which the government owns the project’s objectives. While this may mean the Bank expends more resources over a longer period to achieve the end result, a series of limited successes is better than attempting to attain all desirable goals at once, with all the attendant risks. This is particularly the case when the appraisal team is faced with impending elections with uncertain results. 3. Roles and responsibilities should be clearly articulated when different units are in charge of administration and execution to mini- mize conflict and disagreements over the use of funds and execution of contracts. 4. Especially in risky circumstances, projects should contain minimal conditions of effectiveness, and conditions that establish satis- factory accounting and financial management systems should be formulated as conditions of Board presentation. 5. Before project effectiveness, the Bank should make an intensive effort to identify clear benchmarks and indicators that are easily measurable by the implementation agencies to make it easy to assess whether implementation is working well; monitoring indica- tors should reflect incremental stages of achievement and be adapted as necessary during project implementation. 6. The Bank’s sustained support is critical to achieving overall development impact and can contribute to developing a strong working relationship with local authorities while attracting other donors to the area. 7. Before agreeing to hire project directors, who may have networks of connections and obligations that conflict with their project-related obligations and may be hard to remove, the Bank should ascertain that these directors can be replaced in their role easily and quickly (even if they retain their position in the public sector). 8. Extensive training of local staff in the Bank’s procurement policies and financial management procedures should be planned in the project’s design and conducted before start-up to build project implementation capacity and ensure timely disbursement in low-capacity environments. Procurement irregularities can be eliminated if procurement audits are done after the first year of project implementation. 9. Rather than one large project spanning multiple sectors in a country with limited implementation capacity, multiple small loans can be useful instruments to introduce significant reforms on a minor scale in difficult countries and enable learning before scaling up good-practice outcomes. 10. Human Resources should not limit its focus to training; issues such as incentives, career development, public/private partnership, and decentralization also need to be addressed. Source: ICR Reviews from the ICR Review and Tracking Database. 123 APPENDIX R: COUNTRY ASSISTANCE EVALUATIONS AND CAS COMPLETION REPORT REVIEWS Five Country Assistance Evaluations (CAEs) are • Inadequate assessment of priorities/ available for the 25 LICUS. They assess the pre- timing. In Zimbabwe, the Public Expenditure LICUS Initiative period (1986–2002) and rate all Reviews at the end of 1995 came too late to in- but one country program as moderately unsatis- form the design of the Structural Adjustment factory or unsatisfactory (see table R.1). The Credits, and although many analytical products main reasons are for the unsatisfactory ratings were completed during the 1990s, there was are as follows: no substantive analytical work on poverty. In Papua New Guinea, the Bank’s attention was • Poor assessment of political and gover- inconsistent, with a period of intense activity nance constraints. In Haiti, the Bank’s ob- followed by inactivity. In Cambodia, projects in jectives were consistent with major economic the areas of agricultural and rural develop- problems, but relevance was limited by the ment were not immediately supported despite failure to give highest priority to resolving the their importance. political and governance problems that un- dermined economic development. The poor Country Assistance Strategy Completion assessment of political constraints has resulted Report (CASCR) Reviews in Fiscal 2005 in excessive optimism on the Bank’s part, for LICUS example, in Papua New Guinea. In Zimbabwe Of the four IEG CASCR Reviews available thus far during 1998–2000, when there were clear signs for LICUS, and that covered at least part of the that the Bank’s strategy was not working, the period since the start of the LICUS Initiative, Bank continued to appraise and approve new three were rated moderately unsatisfactory or projects, as well as negotiate the third Struc- unsatisfactory and one was rated moderately tural Adjustment Credit, with poor results. satisfactory. Table R.1: Country Assistance Evaluations for Fiscal 2005 LICUS Institutional development Country CAE date Period Outcome Sustainability impact Cambodia 11/16/2000 1992–99 Moderately satisfactory Uncertain Substantial Haiti 02/12/2002 1986–2001 Unsatisfactory/highly unsatisfactory Unlikely Negligible Papua New Guinea 03/06/2000 1989–99 Unsatisfactory Modest Uncertain Solomon Islands 03/31/2005 1992–2002 Moderately unsatisfactory Unlikely Negligible Zimbabwe 05/21/2004 1990–2000 Unsatisfactory Unlikely Negligible Note: The CAE and the ratings are for the Pacific member countries group, and not for the Solomon Islands in particular. 125 APPENDIX S: HUMAN RESOURCE PROPOSALS FOR IMPROVING STAFFING IN LICUS Staff deployment proposals Progress Technical promotion criteria to level GH and the managerial Generic criteria for GH technical positions under “qualifications, knowl- selection criteria to include “demonstrated ability to understand edge, experience and competencies” revised in 2004 as follows: Prior work the challenges of the poorest countries and to work effectively experience should also include (i) work with significant impact in an LIC in such environments.” Announced for implementation in 2003, or (ii) a country office assignment. until 2006 this element is to be considered in the promotion criteria. After 2006, demonstration of this competency is proposed as mandatory for promotion. Sector Boards to ensure planned rotation and secondment of Strategic staffing reviews by Sector Boards to assess the health of the senior and experienced task team leaders into supporting roles internal and external candidate pipeline, plan for cross-vice presidential for LICs where needs are not met through the vacancy unit assignments, plan for (re)-entry into the sector of selected staff, iden- management system. tify knowledge and skill gaps among sector staff, and anticipate vacan- cies plus plan for targeted external recruitment to strengthen skills, experience, and/or diversity of the sector. No specific mention of identi- fying staffing priorities for LICUS/LICs. The Bank will examine ways of lessening the difficulties of The Review of Overseas Assignment Benefits points to the need to make safety and travel experienced by staff working on LICs. some improvements in the benefits package, particularly in mitigating increased living costs and alleviating hardship conditions. A stronger differentiation of cost of living allowances is suggested to provide a better relation to type of posting, staff salary, grade profile, and family size. Meanwhile, changes to the locality premium are proposed to reflect better the hardship conditions and the shift to assigning more senior staff to overseas assignments. A locality premium of 5 percent for duty stations in LICUS with hardship premiums of 20 percent to 25 percent is also introduced, effective April 1, 2005. Rest and recreation travel for difficult locations is re-introduced. Introduction of special collaboration between research units No evidence. (DEC and anchors) and staff working on the poorest countries is proposed so that staff get an opportunity to participate in cutting-edge research and analytical work with the special “spotlight” it provides, including publication of their work. Tangible actions would be supported by “softer” recognition No evidence. programs, including spotlighting LIC teams in corporate events, meetings with senior management and the Board, and consider- ation of introducing specific recognition in the President’s Award for Excellence Program. Source: Interviews with the Bank’s Human Resource and other staff, 2005. 127 APPENDIX T: OPERATIONAL POLICIES AND BANK PROCEDURES RELEVANT TO LICUS OP/BP Revised/updated Coverage OP/BP 2.30 December 2003 Covers countries vulnerable to conflict, in conflict, or in transition from conflict. Emphasizes the Development Bank’s focus on reconstruction and development, the importance of working with and through Cooperation and other development partners, especially the UN, and providing exceptional financial assistance Conflict to help countries emerging from conflict meet their transitional financial needs in a timely manner (OP 2.30 establishes the case for exceptional post-conflict assistance under IDA 13). OP/BP 8.50 Revision under way Currently covers emergencies defined as “extraordinary event of limited duration, such as war, Emergency civil disturbance, or natural disasters.” The current objectives of emergency recovery assistance Recovery are “to restore assets and production levels in the disrupted economy.” Revisions to this OP/BP Assistance that are under way include expanding coverage of the OP/BP to all events that (i) have caused, (ii) are likely to cause in the absence of immediate preventive action, or (iii) periodically cause a rapid and major adverse economic and social impact, which requires an urgent response from the government. Revisions to the objectives of economic recovery loans include (i) rebuilding and restoring physical assets; (ii) restoring production and economic activities; (iii) preserving human, institutional, and/or social capital; (iv) restoring social activities; (v) preserving or restor- ing essential services; and/or (vi) supporting preventive measures designed to mitigate or avert the effects of anticipated imminent or future emergencies. OP/BP 7.30 July 2001 Covers operations in countries where a de facto government comes into or remains in power De facto by means not provided for in the country’s constitution. Ensures continued engagement as long Governments as the Bank is satisfied that the government: is in effective control of the country; recognizes the country’s past international obligations; is willing and able to assume all its predecessor’s obligations and ensure continued implementation of Bank loans. New operations require an assessment of the financial/legal exposure of the Bank, the number of countries that have recognized the government, and the position of international organizations toward the de facto government. OP/BP 6.00 August 2004 Applies to projects in countries for which the Bank has established country financing parame- Bank Financing ters. Allows for Bank loan proceeds to finance expenditures (including cost sharing, recurrent costs, local costs, and taxes and duties) necessary to meet the development objectives of operations supported by the loan. Source: World Bank database. 129 APPENDIX U: SAFEGUARDS IN LICUS Background work was undertaken on safeguard issues requiring attention during project execu- compliance in Bank projects in LICUS as part of tion. For the remaining 19 projects, safeguards this review. The methodology, overall findings, compliance during implementation was rated and conclusions are presented below. moderately satisfactory and better for only 37 percent of cases (compared with 35 percent for Methodology the CDD portfolio), indicating cause for concern. The Bank approved 217 projects in LICUS over • Overall Rating. The overall rating was a fiscal 2000–05, of which 184 were assigned an weighted average of the entry and implemen- Environmental Assessment (EA) category in tation ratings, taking into account the time accordance with OP 4.01.1 A random sample of from Board approval and the seriousness of the 25 projects was drawn for analysis, stratified by safeguard issues being addressed. Overall, 84 EA category, age,2 and project size.3 For each percent of projects were rated moderately sat- sample project, a desk review was made of at isfactory or better. least the following documents: Project Appraisal • Grants versus Loans. Some of the LICUS have Document; Integrated Safeguards Data Sheet; circumstances that make them ineligible for Environmental Impact Assessment (EIA)/ regular borrowing from the IBRD or IDA, and Resettlement Action Plan/Indigenous Peoples’ projects in these countries are funded through Development Plan (if used); Project Status grants (from IDA, trust funds, GEF). Grants Reports (PSRs)/supervision Aide Memoire; and (or grant/loan combinations) financed 6 of the ICRs (for completed projects). In selected cases, 25 sample projects. However, the hypothesis legal documents and/or Project Implementation that grant-funded projects are reviewed less dili- Manuals were also consulted. gently for compliance with safeguard policies was not supported; in fact, 100 percent of proj- Overall Findings ects with grants were moderately satisfactory or better on the overall rating compared with • Quality of Safeguard Compliance at Entry. 78 percent for IDA credits. Overall quality at entry with respect to safe- • Large versus Small Projects. Larger projects ap- guard policy compliance for the 25 randomly pear to have better safeguard compliance (92 selected projects in LICUS and approved be- percent moderately satisfactory or better on the tween fiscal 2000 and 2005 was rated at 88 overall rating) than smaller ones (72 percent), percent moderately satisfactory or better, which possibly because the larger projects are in the shows that safeguards compliance is relatively larger countries, which have somewhat better good (compared with 70 percent for the institutional capacity. community-driven development portfolio), though still somewhat short of the “zero tol- Conclusions erance” policy supposedly in effect. While quality at entry needs some improvement, • Quality of Safeguard Compliance during Im- safeguard compliance during implementation plementation. For implementation, six Category warrants much greater attention by the Bank C projects were not rated, as they did not have and borrowers. Despite the 2001 changes to the 131 E N G A G I N G W I T H F R A G I L E S T AT E S format of the PSR (now called ISR), which reporting and should assess the level of training should encourage complete reporting on needed for effective monitoring of safeguards. The safeguard issues, reporting on safeguards during provisions of emergency lending, which push more implementation remains sparse and inadequate. of the analysis into the implementation phase, do In the great majority of cases, little or nothing is pose some dangers in light of the generally poor said on the implementation of agreed-on monitoring of compliance in this phase. Given the safeguard measures or on any unforeseen weak implementation capacity in most of the problems. Looking beyond the PSRs to mission LICUS, particular attention needs to be given to the Aides Memoire or Back-to-Office reports does design, implementation and oversight of institu- little to modify this finding. tional strengthening, capacity development, and For LICUS projects under implementation, Bank monitoring and evaluation systems, with respect to management should focus on improved safeguard safeguards. 132 APPENDIX V: KEY GUIDANCE NOTES FOR LICUS Guidelines Coverage Status Fragile States: Good This paper draws together lessons from country strategy development and implementation in fragile Released Practices in Country states. The principles and approaches presented in the paper are not intended to be prescriptive, December Assistance Strategies but rather to provide a basic framework and menu of tools to facilitate sharing lessons between 2005 countries and regions. Good Practice Note This note provides guidance to task teams in applying development policy lending in the context of Released for Development Policy fragile states. It argues that development policy lending is appropriate in LICUS that exhibit gradual June Lending in Fragile States improvement or those transitioning from conflict/political crisis, but not in LICUS that are experi- 2005 encing deteriorating governance or prolonged conflict/political crisis. Guidelines for CRNs lay out a short-term strategy for countries where the World Bank is actively reengaging Released Preparation, Review beyond the scope of a Watching Brief, but where it has not yet completed the analytical work and mid-2003 and Clearance of LICUS dialogue necessary to formulate a full assistance strategy; or where the conditions are not con- Country Reengagement ducive to a normal TSS or CAS approach. The CRN will normally be followed within one year by an Notes (CRNs) update, a TSS, or a CAS. An Operational Note on In settings that do not allow for a full PRSP process, the TRM provides a ”quick-and-dirty” tool Released Transitional Results with which to identify key priorities, measure early results, provide a framework in which to embed January Matrices (TRM)—Using assistance programs and capacity-building initiatives, and function as a vehicle for donor coordination 2005 Results-Based Frame- in challenging situations—ranging from abundant resources and high expectations (Timor-Leste), to works in Fragile States little money and a legacy of mistrust (Central African Republic). Fragile States: Early The note provides a preliminary framework for thinking about early warning indicators of state Draft, Warning Frameworks fragility. It argues that conflict is not a good predictor of state failure and that an alternative December and Indicators approach would be to examine a state’s four core functions as the basic framework for understand- 2004 ing state fragility and state failure: (i) resource generation; (ii) management and governance; (iii) accommodation of political dissent and maintenance of security; and (iv) provision of basic social services and infrastructure. A two-stage functional analysis could potentially yield an effective early warning framework: a risk assessment of a state’s structural weakness and a monitoring of short-term conjunctural events likely to precipitate failure. It suggests that, going forward, the focus should be on developing reliable short-term conjunctural indicators for each of the four functions because existing long-term structural indicators are already well developed. Synthesis Note on Prepared in collaboration with the Poverty Reduction and Economic Management Network, the note Released Leadership Workshops summarizes the range of leadership activities undertaken in six countries, drawing on these experi- December ences to identify enabling factors and critical choices for consideration when designing leadership 2004 interventions in LICUS. Separate leadership notes have been prepared for each of the six countries. Source: Guidance Notes, LICUS Unit, World Bank. 133 APPENDIX W: OVERVIEW OF THE WORLD BANK INSTITUTE’S LICUS ACTIVITIES The World Bank Institute (WBI) has steadily WBI is helping to strengthen local training increased its activities in the three-year period, institutes, which will play a critical role in 2002–05, and has included work in more than 30 helping to build the professionalism of the LICUS in governance, public finance, education, country’s revenue management institutions, environment, health, and monitoring and ranging from the finance ministry to independ- evaluation. WBI’s initiatives have focused mainly ent oversight and auditing bodies. In Lao PDR, on rebuilding and strengthening basic WBI is working with the national civil servants’ economic, social, institutional, and governance training institute, which has the objective of policies, with particular attention to building building capacity at the central and provincial capacity for better governance and improved levels to implement priority reforms under the service delivery. Weak capacity has presented PRSP. In addition to training trainers and considerable challenges in these fragile states, updating their curricula in public financial and World Bank country teams have turned to management, economic development, and WBI to engage in capacity development activi- project analysis, WBI has also facilitated ties to support the implementation of key twinning arrangements between Lao PDR and reforms. international institutions such as the Korea WBI is working with the government of Haiti, Development Institute. for example, to carry out a countrywide Other WBI work in LICUS has included leader- governance diagnostic survey as an input into ship training, thematic learning programs, and the country’s plan to develop a comprehensive technical assistance. WBI has jointly organized anticorruption strategy. A similar survey has with the country team a series of leadership been completed in Guinea, where WBI is events in the Central African Republic aimed at helping the government develop a long-term engaging multiple stakeholders in short-term governance strategy. As part of the Bank’s Africa action planning to implement the agreed priori- Action Plan, WBI has committed to working with ties under the PRSP, and engaging national leaders up to 10 African countries in devising strategies in an experience-sharing and peer-learning event for improved governance. This will include with fellow leaders who have themselves diagnostic surveys in Burundi, the Democratic managed similar situations. Similarly, two leader- Republic of Congo, and Niger, among other ship events for senior government officials in countries. Tajikistan have been organized to expose leaders WBI is also working with local training to international experience and best practices in institutes to strengthen their capacity to scale promoting key reforms. A third workshop will up and support the implementation of key focus on issues of public sector management and development objectives. In Chad, for example, administrative/public expenditure reform. Source: WBI staff, World Bank. 135 APPENDIX X: OVERVIEW OF THE DEVELOPMENT ECONOMICS VICE PRESIDENCY’S LICUS ACTIVITIES Development Economics Vice Presidency • Propensity to Civil Disobedience and the Prob- (DEC) work on sources of conflict and ability of an Armed Struggle in Niger Delta Re- responses to post-conflict situations is ongoing gion of Nigeria across the research complex and involves • Neighboring States, Conflict, and Instability significant collaboration with the International • Systems of Violence in Post-Conflict Societies Peace Research Institute in Oslo and with • Alternative Measures and Estimates of Peace- Oxford University. A large part of the research Building Success has been funded by the Norwegian govern- • Disarming Fears of Diversity: Ethnic Hetero- ment. Current work touches on a number of geneity and State Militarization, 1988–2002 topics. A sample of the papers under way is highlighted here. Democracy and Conflict Types and Aspects of Conflict • Moral Hazard, Adverse Selection, and Power Sharing • Transitional Justice and Sustainable Peace • Democratic Jihad? Military Intervention and • Manifestations of Violence: Civil Wars, Coups, Democracy and Others • Credible Commitment and Insurgency in • Population Size, Concentration, and Civil War. Democracies and Autocracies A Geographically Disaggregated Analysis • Walking the Tightrope: Extending the Fran- • What Is Civil War? chise in the Presence of Political Competition • Political Institutions, Horizontal Inequalities, Fragile States and Peace Building and Civil Conflict • Interim Institutions and the Development Macro- and Microeconomic Policy Process: Strategies for Pro-Poor Judicial Re- Choices in Post-Conflict Countries form in Cambodia • Beyond Fractionalization: Mapping Ethnicity • Rwandan Crop Failure and Rural Coping onto Nationalist Insurgencies Mechanisms • Military Expenditure in Post-Conflict Societies • Post-Conflict Capital Flight and Return • Post-Conflict Risks of Conflict Resumption • Are Non-Poor Households Always Less Vul- • Post-Conflict Risks nerable? The Case of Households Exposed to • The Long-Term Legacy of the Khmer Rouge Protracted Civil War in Southern Sudan Period in Cambodia • The Demand for Money around the End of • Poverty, Social Divisions, and Conflict in Nepal Civil Wars • DDR and Optimal Aid Allocation in Post-Conflict • Scaling-Up Aid, Real Exchange Rate, and Catch- Countries up Growth in Post-Conflict Countries • Cheap Guns, More War? The Economics of • The Aftermath of Civil Wars: An Event-Study Small Arms Approach to Post-Conflict Transitions Source: DEC staff, World Bank. 137 APPENDIX Y: DONOR RELATIONSHIPS Calendar of major events and key policy papers on fragile states, 2002–06 Key: Joint reports World Bank is member EC - Governance and AsDB - Approach to World Bank SIDA - Promoting World Bank of the working group co-chaired OECD-DAC peace and authored development weakly performing for the OECD-DAC paper - Meetings communication developing member LAP 4th Meeting - security through Fragile States - Good Alignment and Harmonization LAP name Practice in Country discusses difficult countries in Fragile States development partnerships (Feb 2004) changed to FSG cooperation Assistance Strateies Policy (Dec 2004) (May 2005) (Jan 2006) notes (Oct 2003) (Oct 2005) UN High-Level Panel - A More Secure World: OECD-DAC - Development Our Shared Draft principles cooperation in UNDG-ECHA - DFID - Why We endorsed and Responsibility, Need to Work World Bank difficult partnerships Joint report on Recommends pilot exercise CIDA - Guidelines for co-chaired (Dec 2002) transitional Effectively in launched effective development Informal Donors Creation of Peace- Fragile States issues Building Commission OECD HLM cooperation in Meeting on fragile states (Feb 2004) (Jan 2005) (Mar 2005) fragile states (Dec 2004) (Mar 2006) (Nov 2005) 2002 2003 2004 2005 2006 UNDG-ECHA - The World Bank co-chaired Conflict Analysis AFD - Taking action workshop on working Framework USAID - Fragile to assist fragile OECD - DAC members for development in (Nov 2004) States Strategy actors and societies ranked fragile difficult partnerships. (Jan 2005) (Sep 2005) states work as a Recommends top priority at SLM establishments World Bank (Dec 2005) of the LAP World Bank Agreement to co-authored (Oct 2002) with UNDG and UNDP: co-authored develop Principles with UNDP: An for Good World Bank co-chaired HLM discussed AusAID - Statement to Practical Guide to International Multilateral Needs Operational Note workshop with UNDP progress made on parliament stressed the World Bank on Transitional Engagement in entitled: Rebuilding Post- fragile states Assessments in need for engagement co-chaired meeting Post-Conflict Results Matrices Fragile States at Conflict Societies: Lessons (Apr 2006) with poor performing that formed the (Jan 2005) SLF, cochaired by from a Decade of states Situations the World Bank OECD - DAC LAP (Aug 2004) Global Experience (Sep 2002) (Oct 2003) (Jan 2005) (Sep 2005) Source: LICUS Unit, OPCS, World Bank. Note: AFD = Agence Francaise de Developpement (French Development Agency), AsDB = Asian Development Bank, AusAID = Australian Agency for International Development, DFlD = U.K. Department for International Development, EC = European Commission, FSG = Fragile States Group, HLM = High-level meeting, LAP = Learning and Advisory Process (changed name to Fragile States Group in May 2005), OECD-DAC = Organisation for Economic Co-operation and Development, Development Assistance Committee, SIDA = Swedish International Devel- opment Agency, SLF = Senior-level forum, UN-ECHA = United Nations Executive Committee on Humanitarian Assistance, UNDG = United Nations Development Group, UNDP = United Nations Development Programme, USAID = United States Agency for International Development. 139 APPENDIX Z: STAKEHOLDER PERCEPTIONS BASED ON SURVEY RESULTS As part of this review, three groups of stakehold- ers were surveyed—Bank staff (including both Table Z.1: Response Rate those in the field and at headquarters), other donors (including international donors and Open Total Response international NGOs), and in-country stakehold- surveysa sample rate (%) ers (including government officials, local NGO In-country 24 146 16.4 staff, academics/researchers, and private sector Donors 49 141 34.8 individuals). The aim of the survey was to elicit Bank staff 382 1,237 30.9 stakeholder views on the relevance and effective- a. Number of stakeholders who started the survey but who may not necessarily have completed it, ness of the Bank’s LICUS approach. The Bank although most did. staff survey was administered to 1,237 Bank staff working on the 25 LICUS. Bank staff included and 16 percent (24) of in-country stakeholders country staff, sector staff, and network anchor responded to the survey. A few adjustments staff as covered in the standard distribution lists were made to the final data for the analysis: (i) if of staff working on LICUS. Other donor staff and local NGOs attempted the donor survey, they in-country stakeholders were identified by were moved to the in-country stakeholder country directors of each of the 25 LICUS. The group; (ii) stakeholders that attempted only the survey was sent to all 141 other donor staff and profile section or less were dropped from the 146 in-country stakeholders identified by the analysis. The two adjustments resulted in 328 respective country directors. If Bank staff and Bank staff, 43 donor, and 25 in-country donor staff worked on more than one LICUS stakeholder surveys. Tables Z.2 and Z.3 present country, they were instructed to answer the the details by each question for each survey for the LICUS which they most focused on. stakeholder group. The response rate calcula- Table Z.1 presents the response rates for each tions in this review drop all missing entries, stakeholder group. Thirty-one percent of Bank “not applicable” entries, and “do not know” staff (382), 35 percent (49) of other donor staff, entries from the denominator. 141 E N G A G I N G W I T H F R A G I L E S T AT E S Table Z.2: Responses by Stakeholder Group (total number of respondents) World In- Question Response Bank Donors country I. Respondent Profile (not reported here) II. Your Views about the Effectiveness of World Bank Support Q1 To what extent do you think World Bank lending and grant support through projects or programs to the LICUS country you most focus on/your country Q1.a Has been timely? To a large extent 109 4 3 To a moderate extent 118 21 10 To a slight extent 50 10 6 Not at all 22 5 2 Do not know 24 2 2 Missing 5 1 2 Q1.b Has had an influence on government policies? To a large extent 66 7 7 To a moderate extent 126 18 8 To a slight extent 86 12 6 Not at all 23 5 1 Do not know 19 0 1 Missing 8 1 2 Q1.c Has been coordinated with other donor support? To a large extent 132 6 4 To a moderate extent 105 18 9 To a slight extent 56 14 8 Not at all 10 4 0 Do not know 16 0 2 Missing 9 1 2 Q1.d Has achieved its intended results? To a large extent 32 2 1 To a moderate extent 144 12 9 To a slight extent 82 20 11 Not at all 13 4 1 Do not know 49 4 2 Missing 8 1 1 Q2 To what extent do you think World Bank nonlending support through analytical work to the LICUS country you most focus on/your country Q2.a Has been timely? To a large extent 92 9 6 To a moderate extent 134 24 9 To a slight extent 47 4 5 Not at all 15 1 1 Do not know 33 4 2 Missing 7 1 2 Q2.b Has had an influence on government policies? To a large extent 52 6 6 To a moderate extent 123 17 8 To a slight extent 86 13 6 142 A P P E N D I X Z : S T A K E H O L D E R P E R C E P T I O N S B A S E D O N S U R V E Y R E S U LT S World In- Question Response Bank Donors country Not at all 24 4 2 Do not know 33 2 0 Missing 10 1 3 Q2.c Has been coordinated with other donor support? To a large extent 93 8 2 To a moderate extent 105 16 9 To a slight extent 79 10 7 Not at all 9 7 1 Do not know 31 1 4 Missing 11 1 2 Q2.d Has achieved its intended results? To a large extent 32 3 3 To a moderate extent 121 12 7 To a slight extent 93 20 11 Not at all 16 3 0 Do not know 52 4 3 Missing 14 1 1 Q2.e Has had an influence on the Bank’s own To a large extent 105 10 6 assistance strategy for the country? To a moderate extent 130 15 8 To a slight extent 32 6 3 Not at all 15 2 0 Do not know 33 9 6 Missing 13 1 2 Q3 To what extent do you think World Bank nonlending support through policy dialogue to the LICUS country you most focus on/your country Q3.a Has been timely? To a large extent 103 9 3 To a moderate extent 128 20 10 To a slight extent 45 8 6 Not at all 10 2 2 Do not know 26 3 1 Missing 16 1 3 Q3.b Has had an influence on government policies? To a large extent 52 6 3 To a moderate extent 136 17 11 To a slight extent 80 12 5 Not at all 17 6 2 Do not know 22 1 2 Missing 21 1 2 Q3.c Has been coordinated with other donor support? To a large extent 98 10 0 To a moderat extent 122 14 9 To a slight extent 55 13 9 Not at all 9 3 1 (Continues on the following page.) 143 E N G A G I N G W I T H F R A G I L E S T AT E S Table Z.2: Responses by Stakeholder Group (total number of respondents) (continued) World In- Question Response Bank Donors country Do not know 25 2 4 Missing 19 1 2 Q3.d Has achieved its intended results? To a large extent 24 1 1 To a moderate extent 134 18 7 To a slight extent 89 17 7 Not at all 16 2 2 Do not know 45 3 5 Missing 20 2 3 Q3.e Has had an influence on the Bank’s own assistance strategy To a large extent 96 9 3 for the country? To a moderate extent 123 16 7 To a slight extent 47 5 7 Not at all 12 4 1 Do not know 28 8 5 Missing 22 1 2 Q4 To what extent do you think World Bank non-lending support through technical assistance to the LICUS country you most focus on/your country Q4.a Has been timely? To a large extent 74 4 6 To a moderate extent 129 23 7 To a slight extent 53 5 6 Not at all 14 4 2 Do not know 34 6 1 Missing 24 1 3 Q4.b Has had an influence on government policies? To a large extent 36 5 4 To a moderate extent 117 12 8 To a slight extent 97 16 8 Not at all 21 6 0 Do not know 32 3 3 Missing 25 1 2 Q4.c Has been coordinated with other donor support? To a large extent 72 4 0 To a moderate extent 103 19 7 To a slight extent 82 12 12 Not at all 8 4 0 Do not know 36 3 4 Missing 27 1 2 Q4.d Has achieved its intended results? To a large extent 30 1 2 To a moderate extent 113 12 8 To a slight extent 96 19 9 Not at all 17 5 0 Do not know 47 5 4 Missing 25 1 2 144 A P P E N D I X Z : S T A K E H O L D E R P E R C E P T I O N S B A S E D O N S U R V E Y R E S U LT S World In- Question Response Bank Donors country Q4.e Has had an influence on the Bank’s own assistance To a large extent 53 4 3 strategy for the country? To a moderate extent 130 18 9 To a slight extent 66 5 3 Not at all 16 3 0 Do not know 36 11 7 Missing 27 2 3 Q5 To what extent has the World Bank’s work in the LICUS country you To a large extent 113 11 6 most focus on/your country been adequately grounded in an To a moderate extent 135 18 12 understanding of the country’s politics? To a slight extent 47 10 4 Not at all 10 3 1 Do not know 10 0 0 Missing 13 1 2 Q6 To what extent has the World Bank supported a focused reform To a large extent 100 13 8 agenda consisting of key actions and reforms in the LICUS country To a moderate extent 127 13 9 you most focus on/your country? To a slight extent 53 10 2 Not at all 10 3 2 Do not know 16 3 1 Missing 22 1 3 Q7 In general, what contribution have the reforms supported by the Large positive contribution 60 6 4 World Bank made to development in the LICUS country you most Small positive contribution 202 27 16 focus on/your country? No contribution 16 3 1 Small negative contribution 2 1 0 Large negative contribution 0 1 0 Do not know 27 4 1 Missing 21 1 3 Q8 If the World Bank has used nongovernmental or semi-autonomous Large positive effect 51 10 7 arrangements to deliver services, what effect have they had on Small positive effect 115 11 13 service delivery in the LICUS country you most focus on/your country? No effect 15 3 0 Small negative effect 7 2 1 Large negative effect 2 0 0 Do not know 55 9 1 Not Applicable 54 6 1 Missing 29 2 2 Q9 What effect have the nongovernmental or semi-autonomous Large positive effect 25 5 3 arrangements supported by the World Bank had on the development Small positive effect 97 10 16 of long-term government capacity in the LICUS country you No effect 48 10 1 most focus on? Small negative effect 12 1 1 Large negative effect 7 0 0 Do not know 56 9 1 Missing 83 8 3 (Continues on the following page.) 145 E N G A G I N G W I T H F R A G I L E S T AT E S Table Z.2: Responses by Stakeholder Group (total number of respondents) (continued) World In- Question Response Bank Donors country Q10 Overall, what contribution has the World Bank’s assistance made Large positive contribution 54 5 6 in helping to develop long-term government capacity in the LICUS Small positive contribution 199 26 14 country you most focus on/your country? No contribution 20 8 2 Small negative contribution 3 0 0 Large negative contribution 0 1 0 Do not know 23 1 1 Missing 29 2 2 Q11 To what extent has the World Bank adequately pursued collaboration To a large extent 147 11 4 with other donors in the LICUS country you most focus on/your country? To a moderate extent 109 15 13 To a slight extent 30 10 3 Not at all 0 3 0 Do not know 15 1 3 Missing 27 3 2 Q12 To what extent has the World Bank adequately pursued collaboration To a large extent 116 9 7 with international partners in the diplomatic, peace-building, and peace- To a moderate extent 85 8 10 keeping areas in the LICUS country you most focus on/your country? To a slight extent 42 9 1 Not at all 9 1 0 Do not know 30 5 4 Not applicable 17 5 1 Missing 29 6 2 Q13 To what extent has the World Bank clearly defined what constitutes To a large extent 45 2 3 “success” in the LICUS country you most focus on/your country? To a moderate extent 144 16 13 To a slight extent 57 7 3 Not at all 29 8 1 Do not know 25 5 3 Missing 28 5 2 Q14 To what extent has the World Bank defined clear and monitorable To a large extent 48 7 5 indicators to measure “success” in the LICUS country you most To a moderate extent 128 11 10 focus on/your country? To a slight extent 74 8 5 Not at all 19 5 0 Do not know 24 9 2 Missing 35 3 3 Q15 With what frequency has progress toward “success” been monitored Frequently or twice a year 101 11 9 by the World Bank in the LICUS country you most focus on/your country? Once a year 93 12 6 Once in 2 years 30 2 2 Never 19 5 1 Don’t know 23 6 2 NA since no indicator 4 2 1 New project 3 1 0 Adhoc 4 1 0 146 A P P E N D I X Z : S T A K E H O L D E R P E R C E P T I O N S B A S E D O N S U R V E Y R E S U LT S World In- Question Response Bank Donors country Based on project agreement 1 0 0 Missing 50 3 4 III. Your Views about the Relevance and Evolution of the World Bank’s Approach Q1 In the last three years, to what extent has the World Bank’s approach To a large extent 110 13 2 to development been relevant to the key issues facing the LICUS To a moderate extent 115 14 12 country you most focus on/your country? To a slight extent 50 8 3 Not at all 5 3 0 Do not know 14 0 2 Not applicable 2 2 1 Missing 32 3 5 Q2 Still thinking about the last three years, what change has there been in the World Bank’s approach to development in the LICUS country you most focus on/your country Q2.a Effectiveness of lending and grant support (through Large positive change 58 5 4 projects or programs) Small positive change 143 16 14 No change 42 10 3 Small negative change 4 2 0 Large negative change 1 0 0 Do not know 26 5 0 Missing 54 5 4 Q2.b Effectiveness of nonlending support (through analytical work, Large positive change 60 10 4 policy dialogue, and technical assistance) Small positive change 139 17 14 No change 45 6 1 Small negative change 4 1 0 Large negative change 0 1 1 Do not know 27 4 2 Missing 53 4 3 Q2.c Grounding of World Bank work in an understanding of the Large positive change 70 11 8 country’s politics Small positive change 114 16 7 No change 64 7 3 Small negative change 6 2 1 Large negative change 0 1 0 Do not know 22 2 2 Missing 52 4 4 Q2.d Support for a more focused reform agenda consisting of Large positive change 67 10 7 key actions and reforms Small positive change 118 14 9 No change 66 8 4 Small negative change 3 2 0 Large negative change 1 1 0 (Continues on the following page.) 147 E N G A G I N G W I T H F R A G I L E S T AT E S Table Z.2: Responses by Stakeholder Group (total number of respondents) (continued) World In- Question Response Bank Donors country Do not know 21 3 1 Missing 52 5 4 Q2.e Attention to building long-term government capacity Large positive change 60 5 6 Small positive change 124 19 12 No change 67 9 3 Small negative change 6 1 0 Do not know 18 4 1 Missing 53 5 3 Q2.f Effectiveness in pursuing donor collaboration by the World Bank Large positive change 97 12 5 Small positive change 120 14 11 No change 33 6 1 Small negative change 1 1 0 Large negative change 1 0 0 Do not know 23 5 3 Missing 53 5 5 IV. Your Knowledge of the World Bank’s LICUS Approach Q1 To what extent are you familiar with the World Bank’s LICUS approach? To a large extent 82 7 3 To a moderate extent 132 16 10 To a slight extent 55 11 6 Not at all 15 4 2 Missing 44 5 4 V. Your Overall Impressions Q1 In your opinion, to what extent are World Bank staff who work on To a large extent 132 22 7 the LICUS country you most focus on/your country competent? To a moderate extent 99 8 11 To a slight extent 27 3 3 Not at all 3 2 0 Do not know 19 3 0 Missing 48 5 4 Q2 What contribution has the World Bank’s field office made in Large positive contribution 150 14 5 supporting the development of the LICUS country you most focus Small positive contribution 101 16 16 on/your country? No contribution 11 4 0 Small negative contribution 4 2 0 Large negative contribution 0 1 0 Do not know 14 1 0 Missing 48 5 4 148 A P P E N D I X Z : S T A K E H O L D E R P E R C E P T I O N S B A S E D O N S U R V E Y R E S U LT S World In- Question Response Bank Donors country Q3 What contribution have the visiting World Bank missions made in Large positive contribution 123 9 10 supporting the development of the LICUS country you most focus Small positive contribution 118 19 10 on/your country? No contribution 6 4 1 Small negative contribution 5 2 0 Do not know 25 4 1 Missing 51 5 3 Q4 What contribution has the World Bank’s overall program made in Large positive contribution 83 11 7 supporting the development of the LICUS country you most Small positive contribution 159 20 13 focus on/your country? No contribution 14 1 2 Small negative contribution 1 1 0 Large negative contribution 0 1 0 Do not know 17 4 0 Missing 54 5 3 Q5 In comparison with other donors, has the contribution of the Greater 148 14 5 World Bank’s overall program in supporting the development Equal 64 12 9 of the LICUS country you most focus on/your country been: Smaller 30 10 4 Do not know 24 2 4 Missing 62 5 3 Q6 Without World Bank support, do you think the development of the Greater 2 1 0 LICUS country you most focus on/your country would have been: Equal 32 8 5 Smaller 205 27 14 Do not know 26 2 3 Missing 63 5 3 Total number of respondents 328 43 25 149 E N G A G I N G W I T H F R A G I L E S T AT E S Table Z.3: Responses of the World Bank Staff on Internal World Bank Support for LICUS Work Question Response World Bank (N = 328) Q1 To what extent do you think the Bank’s operational policies and procedures listed below are adapted to the low capacity or higher risk environment of the LICUS country you most focus on? Q1.a Project preparation To a large extent 40 To a moderate extent 86 To a slight extent 66 Not at all 51 Don’t know 20 Missing 65 Q1.b Project supervision To a large extent 43 To a moderate extent 77 To a slight extent 70 Not at all 46 Don’t know 24 Missing 68 Q1.c Procurement procedures To a large extent 27 To a moderate extent 61 To a slight extent 75 Not at all 67 Don’t know 30 Missing 68 Q1.d Financial management procedures To a large extent 32 To a moderate extent 66 To a slight extent 79 Not at all 50 Don’t know 34 Missing 67 Q1.e Legal framework To a large extent 28 To a moderate extent 62 To a slight extent 72 Not at all 54 Don’t know 43 Missing 69 Q2 Is the overall level of Bank Budget (BB) available to the LICUS More than adequate 11 country you most focus on adequate given the opportunities and Adequate 68 challenges facing the country? Less than adequate 159 Don’t know 21 Missing 69 150 A P P E N D I X Z : S T A K E H O L D E R P E R C E P T I O N S B A S E D O N S U R V E Y R E S U LT S Question Response World Bank (N = 328) Q3 To what extent has the Bank’s lending and grant support (through projects or programs) in the country you most focus on been given adequate Bank budget and senior management attention for the following: Q3.a Been given adequate Bank Budget (BB) relative to other priorities To a large extent 45 To a moderate extent 77 To a slight extent 69 Not at all 45 Don’t know 25 Missing 67 Q3.b Attracted adequate senior management attention To a large extent 66 or involvement To a moderate extent 81 To a slight extent 75 Not at all 18 Don’t know 19 Missing 69 Q4 To what extent has the Bank’s non-lending support (through analytical work, policy dialogue, and technical assistance) in the country you most focus on been given adequate Bank budget and senior management attention for the following: Q4.a Been given adequate Bank Budget (BB) relative to other priorities To a large extent 34 To a moderate extent 83 To a slight extent 74 Not at all 34 Don’t know 30 Missing 73 Q4.b Attracted adequate senior management attention or involvement To a large extent 44 To a moderate extent 84 To a slight extent 75 Not at all 20 Don’t know 32 Missing 73 Q5 When working on a LICUS country, what has been your experience in each of the following human resource matters? Have the following improved, remained the same, or worsened: Q5.a Your overall career prospects in the Bank (promotions, obtaining Improved 41 good jobs in the future, etc.) Remained the same 134 Worsened 38 Don’t know 49 Missing 66 (Continues on the following page.) 151 E N G A G I N G W I T H F R A G I L E S T AT E S Table Z.3: Responses of the World Bank Staff on Internal World Bank Support for LICUS Work (continued) Question Response World Bank (N = 328) Q5.b Your overall financial compensation (salary increases, Improved 23 hardship allowances, etc.) Remained the same 186 Worsened 21 Don’t know 32 Missing 66 Q5.c Realism in expectations by Bank Management about what can Improved 41 be accomplished Remained the same 149 Worsened 39 Don’t know 32 Missing 67 Q5.d Level of support from Bank Management Improved 50 Remained the same 155 Worsened 33 Don’t know 22 Missing 68 Q5.e Efforts made by the Bank to ensure your personal security and safety Improved 91 Remained the same 126 Worsened 20 Don’t know 24 Missing 67 Q6 To what extent has the Bank’s LICUS Unit been effective with regard to: Q6.a Providing access to Trust Funds for LICUS To a large extent 40 To a moderate extent 57 To a slight extent 37 Not at all 28 Don’t know 93 Missing 73 Q6.b Providing substantive support for country strategy development To a large extent 35 and implementation To a moderate extent 55 To a slight extent 48 Not at all 29 Don’t know 87 Missing 74 Q6.c Providing substantive support for projects To a large extent 19 To a moderate extent 55 To a slight extent 46 Not at all 44 Don’t know 89 Missing 75 152 A P P E N D I X Z : S T A K E H O L D E R P E R C E P T I O N S B A S E D O N S U R V E Y R E S U LT S Question Response World Bank (N = 328) Q6.d Providing substantive support for research or analytical work To a large extent 20 To a moderate extent 52 To a slight extent 53 Not at all 44 Don’t know 85 Missing 74 Q6.e Unlocking procedural or policy difficulties at Headquarters To a large extent 13 To a moderate extent 47 To a slight extent 52 Not at all 50 Don’t know 91 Missing 75 Q6.f Facilitating donor collaboration and harmonization advice To a large extent 33 To a moderate extent 42 To a slight extent 45 Not at all 43 Don’t know 92 Missing 73 Q6.g Getting visibility and support from Senior Management To a large extent 25 To a moderate extent 55 To a slight extent 42 Not at all 38 Don’t know 91 Missing 77 Q7 To what extent is the technical input you get from WBI: Q7.a Sufficient To a large extent 17 To a moderate extent 37 To a slight extent 38 Not at all 77 Don’t know 82 Missing 77 Q7.b Of good quality To a large extent 23 To a moderate extent 40 To a slight extent 32 Not at all 64 Don’t know 90 Missing 79 Q7.c Timely To a large extent 20 To a moderate extent 34 To a slight extent 34 (Continues on the following page.) 153 E N G A G I N G W I T H F R A G I L E S T AT E S Table Z.3: Responses of the World Bank Staff on Internal World Bank Support for LICUS Work (continued) Question Response World Bank (N = 328) Not at all 70 Don’t know 93 Missing 77 Q8 To what extent is the technical input you get from DEC: Q8.a Sufficient To a large extent 10 To a moderate extent 26 To a slight extent 34 Not at all 62 Don’t know 111 Missing 85 Q8.b Of good quality To a large extent 20 To a moderate extent 38 To a slight extent 20 Not at all 47 Don’t know 117 Missing 86 Q8.c Timely To a large extent 10 To a moderate extent 21 To a slight extent 31 Not at all 61 Don’t know 116 Missing 89 Q9 Does the support of the Bank’s LICUS Unit complement, duplicate, Complementary support 33 or conflict with that of the Conflict Prevention and Reconstruction Unit? Some duplication 34 Lot of duplication 14 Conflicting/contradiction 11 Don’t know 156 Missing 80 154 APPENDIX AA: STATEMENT BY THE EXTERNAL ADVISORY PANEL The Panel welcomes this rich and thought- However, in our language, effectiveness in provoking report and the opportunity to share the more limited sense of whether the Bank has some of its impressions. The subject—how to been doing what it says it wishes to do and manage support by the donor community to whether this can be done better is worth LICUS—is of major importance, given the examining now, as is the question of the number of fragile states, the hardships endured relevance of the formal determinants of LICUS by their inhabitants, and the spillovers to and of their performance. Addressing these neighbors, as well as the fact that in certain questions rigorously is essential to assess later, instances such states may form a breeding when adequate data are available, whether the ground for terrorism. approach chosen delivers acceptable outcomes The Bank and other members of the interna- in the use of scarce development resources. tional donor community have grappled for The Bank has made commendable progress several years now with the question of how to in its engagement with LICUS and in the help LICUS emerge from their frequently performance of closed projects (see chapter 2 desperate situations. Given the defining charac- and appendix Q). However, the donor teristics of LICUS, weakness of governance, community has shifted the goal posts for institutions, and policies, and the outcome of intervention with the relatively recent, intensi- earlier research and experience that financial fied, and explicit focus on state building and, assistance against such a background tends to where relevant, conflict prevention. This shift is be ineffective, it was clear that useful engage- logical in the context of the problems posed by ment with these countries would require a new LICUS. The Panel agrees with IEG, however, that framework. The Bank is to be commended for the Bank needs to undertake major efforts to fit having played and for continuing to play a in with the new focus. leading role in developing such a framework. While the narrowing of the focus to state and The Panel was impressed by the methodology peace building should induce the Bank to move of the IEG report. It believes that the right away from overly broad reform agendas, which “do questions have been asked and that the combina- not augur well for effectiveness,” the Bank’s tion of analysis, common sense, and the underpin- effectiveness in the area of governance and capacity ning of findings by wide-flung surveys has resulted building needs to be improved. IEG is right to in highly relevant lessons and recommendations. recommend that the Bank spell out concrete strate- To no small extent this is also thanks to interaction gies and policies for this purpose. That, at the with management that has clearly been fruitful. country level, strategies need to be underpinned While one may argue in general with a rush to by internalized socio/political analysis may appear evaluate before the necessary data are available, self-evident, but in practice proves to be difficult. in this case an evaluation with a carefully Without such analysis, Bank engagement as well as restricted scope is very useful. The report is right that of other donors runs the risk of being ineffec- to point out that the question of ultimate tive and wasteful of resources. Without wishing to effectiveness of Bank interventions cannot yet be attribute responsibility, the recent experience in addressed. Timor-Leste appears to illustrate the point. 155 E N G A G I N G W I T H F R A G I L E S T AT E S IEG also rightly stresses that capacity building criteria and indicators, monitoring and evaluation must be a major part of state-building programs will not have at its disposal the toeholds needed and that the Bank’s track record indicates a need for learning adequately from experience and for to strengthen the design and delivery thereof. timely adjustment of country strategies. The lesson that country ownership and The Panel agrees with the lessons drawn on absorptive-capacity constraints apply as much to how to improve the Bank’s internal organization knowledge products as to financial products to meet the challenges posed by LICUS more does not make the challenge any easier. The effectively. Criteria for successful performance Panel is convinced that unless weaknesses in of staff in LICUS, where the traditional criteria state and capacity building are overcome, future only partly apply, need to be elaborated. Also, outcomes will be disappointing, distorting IEG’s point is well taken that the selection of judgments on the usefulness of multilateral and people for work on LICUS must take account of bilateral donor support to LICUS. their willingness and ability to communicate and The joint responsibility of donors in the areas collaborate effectively inside the Bank and with of state building and conflict prevention and other donors and the recipients. across the range of issues involved in supporting The Panel has high regard for how the Bank LICUS once again leads to an obvious lesson: the has immersed itself in the challenging and risky need for donors to coordinate to provide more area of support for LICUS. It welcomes the effective support jointly and severally. And once positive interaction between practice and evalua- again the simple lesson is difficult to translate tion, as evinced in the present report. In the into systematic practice at the country level. Yet, Panel’s view, IEG’s comments are balanced and its as IEG’s report brings out, the failure to do so recommendations sensible. Implementing them can mean the difference between a whole that is will not be easy, but is necessary to improve the larger or smaller than the sum of the parts, effectiveness of Bank support to LICUS, as well as between effective and ineffective support. that of other donors. We would be surprised if The Panel agrees with IEG on the importance of further progress based on inescapable realities further work on criteria by which to identify LICUS does not materialize. Such progress is all the more and on the need for a break-down by business necessary because the tipping point between models. Similarly, performance indicators require success and failure with equal effort lies much elaboration in order to determine the kinds and closer to failure in LICUS than in other countries. amounts of support to be given. Post-conflict Adoption of the eminently practical lessons and LICUS are already treated very differently from the recommendations of IEG can shift the tipping others, and have proven to be fertile recipients of point onto more favorable terrain. The possibility certain kinds of financial aid. Careful specification of emergence from extreme fragility of the state could also strengthen decision making vis-à-vis and the associated misery of its inhabitants will be resource-rich countries. Moreover, without such greatly enhanced. Olayinka Creighton-Randall Former Coordinator of the Campaign for Good Governance Freetown, Sierra Leone John Githongo Senior Associate Member St. Antony’s College Oxford, United Kingdom Gunnar Sørbø Director Chr. Michelsen Institute Bergen, Norway Pieter Stek Former Executive Director of the World Bank Group Ohain, Belgium 156 APPENDIX BB: MANAGEMENT RESPONSE Introduction Key Issues of Agreement and Divergence Management welcomes IEG’s review of the This Management Response first outlines the effectiveness of Bank support to Low-Income areas in which management agrees with the Countries Under Stress (LICUS) and is grateful review and then discusses areas in which for the review team’s effort to incorporate many management believes that IEG has missed of management’s comments in the review. As critical factors or could have given a fuller the review indicates, fragile states represent a account of efforts the Bank is already making. critical challenge for the Bank and other development actors and make up an increas- A. Areas of agreement ingly significant segment of the Bank’s portfolio. Management agrees with many of the review’s The review provides useful analysis of a wide findings, and the review serves as a powerful, range of issues and contributes substance to the targeted, and well-timed renewal of these Bank’s understanding of difficult engagements arguments. Indeed, many of the areas are in fragile states. Management also notes that already part of an ongoing work program. early conclusions of the IEG evaluation were useful in feeding in to the Good Practice Note Improved institutional response. Management on Country Assistance Strategies in fragile states concurs with IEG’s diagnosis of the various (World Bank 2005e) and would like to thank the institutional bottlenecks where the Bank needs IEG team for its close cooperation in this regard. to redouble its efforts to restructure for a better As the review notes, the Bank has played a performance. Increased field presence in fragile leading role in global policy development on states, better incentives and skills development fragile states. In pointing to several areas for for staff, and improved surge capacity are all future improvement, the review reinforces critical challenges that the review correctly important messages for the Bank’s engagement highlights; these were areas of attention at the in fragile states that were set out in the LICUS January 2006 Board discussions. Since then, the Update (World Bank 2005h), which the Board Fragile States Group has been addressing this endorsed on January 17, 2006. Management set of issues in the review “Strengthening the therefore agrees with IEG on many of the princi- Organizational Response to Fragile States,” to be ples and ideas raised in the review, some of completed in fiscal 2007. The review examines which have been the subject of Fragile States: issues of particular importance, including Good Practices for Country Assistance Strate- achieving the right level of field presence gies and other guidance and good practice notes through incentives for staff, and the organiza- issued by OPCS. (By way of illustration, Attach- tion and capacity necessary to support the ment A provides a matrix showing key issues needs of country teams. Likewise, IEG has raised in the review and guidance on that issue rightly identified some of the Bank’s procedures that has been provided in one of the notes as barriers to rapid responses. Right-fitting aid issued by the Fragile States Group [formerly the allocations to ensure positive, not perverse, LICUS Unit] in OPCS.) incentives to countries is a complex and Bank- 157 E N G A G I N G W I T H F R A G I L E S T AT E S wide challenge. And while management concurs B. Areas of divergence with IEG that LICUS face the problem of too Overall, the review brings out less clearly the little or too much aid, it also would draw positive trends in performance data and some attention to the considerable progress made interesting and innovative approaches that the since the LICUS Initiative first raised these LICUS Initiative has encouraged and supported. concerns (see Management Action Record). Three areas of particular concern are the review’s presentation of performance data and Differentiated approach. Management welcomes country examples; its discussion of state IEG’s emphasis on the particular challenges that building, governance, and capacity develop- countries in “deterioration” or “prolonged ment; and its assessment of selectivity and crisis/political impasse” and post-conflict prioritization, results measurement, and in- countries transitioning from the immediate country donor collaboration. post-conflict phase pose for the Bank and other donors. The Bank’s Fragile States Group is now 1. Performance data and country examples working with partners in OECD-DAC’s Fragile Management notes that the review tends to bring States Group to take a more detailed look at out a relatively negative side of the picture; for differentiating approaches by business models, example, the summary makes use of 18 country as the LICUS Update explains. Recent experi- examples, 17 of which are negative. While manage- ence in such countries has highlighted the need ment welcomes identification of weak spots where to articulate common messages from the they exist, it would note that many Bank country international community and focus both on teams have also innovated and found successful national unity and accountability within the modes of engagement that others can learn from. state-building agenda and on longer-term Fragile States: Good Practices in Country capacity building. Assistance Strategies (World Bank 2005e) records 25 of these cases, ranging from the use of nontra- Learning space. Management also welcomes the ditional partnerships to secure a robust economic review’s emphasis on lesson learning as a critical intervention in Liberia to a results-focused strategy part of the Fragile States Group’s role. Regular in Tajikistan (World Bank 2005e), but they are not LICUS learning seminars, often cohosted by SDV reflected in IEG’s review. or PREM, provide a forum for informal exchange among practitioners and outside experts across a Project performance data. While the country cases balanced agenda of themes and country cases. dominate, the summary discussion downplays data The Fragile States Group has also organized on project performance. When unbundled, the more targeted events with country teams to data provided in the review for project perform- address in real time specific issues of interest— ance and at-risk projects reveal a positive year-on- as recent examples, a review of political economy year trend, both absolutely and vis-à-vis non-LICUS factors in Ethiopia and Sudan and a discussion of low-income countries.2 In 2005 projects in LICUS development policy operations in deteriorating actually achieved higher levels of performance than governance situations. In addition, the Fragile projects in non-LICUS low-income countries, a States Group has regularly produced Good testament to the efforts of country teams working Practice Notes, most recently on harmonization under difficult conditions. Management views this and alignment.1 Management concurs that there as a positive trend that should be supported and is increasing demand for operational guidance sustained as a real step to more effective engage- for field actors and innovative approaches that ment. However, the summary makes only cursory have a proven track record for capacity develop- mention of it, despite the fact that the LICUS ment; however, it notes that substantial Update (World Bank 2005h) identified project operational guidance and learning activities have performance as one of the key indicators of the already been provided, and more are in the work Bank’s performance in LICUS (see section below plan articulated in the LICUS Update. on results measurement). 158 APPENDIX BB: MANAGEMENT RESPONSE Data clarity. The review also gives the impres- (principles 3 and 11) that state building should sion that it is basing judgments on some be a focus and that this in turn requires a long- implementation evaluations that relate to term commitment to capacity development. operations begun before the initiative.3 Manage- Among other groups in the Bank, the PREM ment recognizes that three years does not allow public sector governance team has undertaken for much data to be gathered; however, the use critical analysis of the failings of past capacity of pre-initiative data and country examples development efforts that bolster, not could have been more clearly separated from undermine, the rationale for state building: newer data. Sidebar texts do little to clarify this confusion; for example, one reads that “IEG By the early 1990s the realization began to ratings for LICUS CASs completed thus far have dawn that policies themselves were built mostly been unsatisfactory” (chapter 2) despite on an underpinning of social, political, the fact, noted in the text, that of these only one and state institutions and that weaknesses was for a period fully postdating the initiative. in this institutional foundation could undercut the economic policy reform 2. State building, governance, and capacity agenda in three ways: by short-circuiting development efforts at policy change, by failing to Management agrees that the state-building provide a robust platform of credibility agenda addresses the critical areas of capacity and and conflict resolution for market activity, governance head on. However, the review and by being unable to provide comple- repeats the formula that the Bank “has made a mentary physical and social infrastruc- traditional area of weakness [governance and tures (Levy and Kpundeh 2004). capacity development] a central part of its focus” and often couples this idea with the concept of Thus many experts see state building as a overambition, which it further ties to the areas of response to the failure of past development initia- selectivity and results measurement. In fragile tives to see the bigger context in which the techno- states, governance and capacity are central to cratic policy reforms advocated by the Bank and longer-term stabilization and development and others would inevitably fail. The fragile states work require early and sustained engagement from the has attempted to place these approaches within a international community. Management agrees broader political context to ensure greater impact, that in these countries the state-building agenda ownership, and sustainability; emphasize Bank is an enormous challenge for governments and staff ’s knowledge of basic administrative systems; other stakeholders, as well as for the international and balance invisible and visible results to maintain community. The fact that these are hard goals to political momentum for reform.4 achieve does not mean the Bank should not make them central to the agenda. The very reason for Capacity development and governance. The adopting a state-building agenda that puts these review is relatively dismissive of the Bank’s record issues front and center is that in the past the on capacity development and governance— international community has been too ready to which have become areas of increasing focus ignore the task of making state institutions more under the LICUS Initiative. It is beyond the scope effective and accountable to their people, of this response to address this claim; however, it focusing instead on delivering quick fixes should be noted at least that the record is more through parallel and unsustainable structures. nuanced. The Sector Strategy Implementation Update on Public Sector Governance concludes, Importance of state building. In adopting this among other things, that the overall quality of stance the Fragile States Group is in line both economic and sector work (89 percent) and with international partners and with other parts country analytic and advisory activities (97 of the Bank. The OECD-DAC Principles of Good percent) for projects on public sector governance International Engagement in Fragile States state is significantly higher than Bank averages (84 159 E N G A G I N G W I T H F R A G I L E S T AT E S percent and 91 percent, respectively), and that emphasis on state building that encourages within the public sector governance portfolio, the development actors to take a comprehensive success rate of public financial management perspective on the context while taking action operations was approximately 84 percent, regard- along a “critical path” of feasible reforms. The less of the larger governance environment (World review highlights one CAS that was not particu- Bank 2005e, 2005k). While there are no easy larly selective—that of São Tomé and Principe, solutions in the field of state building, manage- which covers a period straddling the initiative’s ment notes that there is a role for the Bank to inception. Management would highlight that identify catalytic entry points for reform where several country teams have more recently the Bank also has a comparative advantage. adopted innovative approaches with tighter prioritization and sequencing, both in the 3. Selectivity and prioritization, results measurement, Bank’s work with national counterparts on and in-country donor collaboration overarching recovery strategies and in the As the review notes, prioritization and sequencing, Bank’s own CAS processes. The Central African donor collaboration, and effective monitoring are Republic, Tajikistan, Cambodia, Liberia, Nigeria, all critical components of a successful reform Togo, and Zimbabwe are examples of interim agenda in a fragile state. The review marks the strategy notes or CASs that have adopted a implementation experience in all three areas as low. conscious LICUS approach with strong selectiv- Management would highlight three responses. ity and prioritization. Donor coordination. Regarding donor coordina- Results measurement. Management concurs that tion, the review tends to downplay the achieve- results measurement must be a critical element ments of the Bank that management feels has been of the LICUS approach. At a central level the highly innovative in terms of instruments Fragile States Group in OPCS has focused on deployed, such as the transitional results matrix, CPIA and IEG data on project performance, the LICUS Trust Fund, and country examples of both of which have improved consistently since joint strategies. The LICUS Trust Fund mandates the initiative began (the LICUS Update [World multidonor approaches; the Transitional Results Bank 2005h] provides comprehensive report- Matrices used in the Liberian, Haitian, Timorese, ing). At the country level, results measurement and Central African Republic transitions all is as per Bank norms and depends critically on supported strong coordination at the country level the availability of budget and staffing to support among actors within a government-owned matrix. identification of results in national development Joint country strategies have been completed in plans and Bank assistance strategies. These are Cambodia, Togo, Somalia, and Nigeria and are issues that the Fragile States Group is address- under way for the Democratic Republic of Congo ing both through its strategic advice to country and the Central African Republic (a proportion that teams on selectivity, reform sequencing, and is at least as high as that for joint strategies in non- results focus and through its strategic staffing LICUS low-income countries). In addition, the report—which advocates for more dedicated Comprehensive Development Framework report staff posted in the field, where identification of notes that “improved coordination among key results for the national recovery strategy external partners around the TRM is providing a and the Bank’s strategy needs to occur. Manage- basis for strengthening government leadership of ment concurs with the view that the identifica- development assistance coordination” (World tion of indicators to monitor progress against Bank 2005c). (Attachment A highlights where the peace building is at an early stage in the interna- Bank has provided guidance on selectivity and tional community and is committed to working prioritization issues.) with other actors at the OECD-DAC and the United Nations Peace-Building Commission Selectivity and prioritization. Selectivity and to further a common sense of progress prioritization are logical corollaries of an measurement. 160 APPENDIX BB: MANAGEMENT RESPONSE Management Action Record IEG recommendation Management response Clarify the scope and content of the Bank’s state-building Partly done, partly ongoing agenda and strengthen the design and delivery of ca- pacity development and governance support in LICUS. Given its weak record on capacity development and governance, In January 2006 Fragile States: Good Practice in Country As- as well as its focus now on the more ambitious and complex state- sistance Strategies (World Bank 2005e) was discussed with building objective in LICUS, the Bank needs to clarify its areas and endorsed by Board. The paper gives more detailed and dif- of comparative advantage vis-à-vis other donors and adopt in- ferentiated guidance on country strategy and operations than novative approaches that ensure better capacity and gover- other agencies have given to date, clarifying both the Bank’s com- nance outcomes. Innovative approaches need to be developed parative advantage within the sphere of state building (“core eco- to achieve a better fit between the Bank’s interventions and the nomic and development competences”) and setting out innovative capacity of a LICUS to perform core state functions; ensuring im- practices that can have a positive effect on capacity development. plementation of focused and well-sequenced interventions in LICUS environments, where virtually every aspect of capacity and The Fragile States Group (formerly the LICUS Unit) is responsi- governance may need significant improvement; and effectively ble for disseminating the good practice to support country team monitoring capacity and governance outcomes. application of the lessons in the implementation of their programs in fragile states. This work is ongoing and integrated, as a pri- ority, into the work program for fiscal 2007—see the LICUS Up- date also endorsed by Board in 2006 (p. 9). In addition, the Fragile States Group will roll out, during fiscal 2007, a program of learning activities based on examples of in- novative approaches taken from the Good Practice Note. The Bank has organized two key state-building events. One in September 2005 convened a group of national reformers in New York from post-conflict situations for two days of facilitated discussions on state building. The second, in January 2006, convened a mixture of academics and policymakers to discuss state building with Bank staff in a one-day learning session. Both sessions provided a forum for intensive debate on core state func- tions, ways to match assistance with capacity, and how inter- national organizations should engage. The September session has resulted in a joint work plan with the UNDP on state building that includes country-level work, pol- icy research, and thematic workshops for lesson learning and dis- cussion. Funding is in place and activities are under way, including research on peace agreements and state building, post-conflict planning processes and state building, as well as country work involving Sudan. 161 E N G A G I N G W I T H F R A G I L E S T AT E S IEG recommendation Management response This work, which supports considerable clarification of the defi- nition, objectives, and division of labor of state building, will also feed into OECD-DAC’s planned work around state building as part of the Fragile States Group work stream and will help the Bank play a role in shaping this agenda. The activities of OECD-DAC’s work stream are to be defined in the first half of fiscal 2007 and activities to begin in the second half of fiscal 2007. By the end of calendar year 2008 the work is expected to have helped support the development of policy clarification on many of these issues. The forthcoming (first half of fiscal 2007) publication Aid that Works: Successful Development in Fragile States contributes some practical insights into project-level approaches for frag- ile states. It explores in particular the role of local-level gover- nance institutions, the potential for complementarity between short-term results and long-term capacity development, and the importance of “bringing the state back in.” Management agrees to support the conclusion of these activi- ties according to their respective schedules. In management’s view, these actions will provide a substantive response to the recommendations made. Develop aid-allocation criteria for LICUS that ensure that Ongoing/Not agreed these countries are not under- or over-aided. The Bank needs to conduct a technical review of the cumulative Management believes that the current IDA allocation system re- effect of the various adjustments to the performance-based al- flects fairly on the one hand the consensus in the larger devel- location system on aid volumes to LICUS. Aid-allocation crite- opment community that a performance-based system is needed ria that reflect the Bank’s objectives in LICUS and ensure that to steer scarce resources where they are most likely to allevi- these countries are not under- or over-aided need to be devel- ate poverty most effectively, and on the other hand the IDA oped. Whether and to what extent the criteria should be based donors’ specific views on how much allocations should be in- on factors other than policy performance (such as levels of other creased in stronger-performing countries. Accordingly, the weaker donor assistance, assessment of potential risks and rewards, and performers—broadly constituting the LICUS group—receive regional and global spillovers) needs to be examined, keeping smaller allocations per capita. Within this broad framework in mind that aid is limited and trade-offs will have to be made. there is already a recognition of some of the points raised by IEG, as evidenced by special allocations for (a) countries coming out of severe conflict, (b) qualifying regional projects, and (c) in ex- ceptional cases countries reengaging with IDA. IDA donors no doubt will continue to raise points about the performance-based allocation system. One point relates to the role of governance. A technical note on this area is being pre- pared for discussion during the IDA14 Mid-Term Review in 162 APPENDIX BB: MANAGEMENT RESPONSE IEG recommendation Management response November. In the meantime, management continues to see broad support for the current approach described above and does not think that reopening basic allocation questions would be helpful in preserving the broad policy consensus that should underpin a strong IDA15 replenishment. Strengthen internal Bank support for LICUS work over the Ongoing/Agreed in part next three years. This is being addressed through the review Strengthening the Two aspects of internal Bank support need attention. First, Organizational Response to Fragile States, now in final draft and staffing numbers, skills, and incentives for working on LICUS need to be completed in fiscal 2007. The review examines issues of to be prioritized. Ensuring adequate incentives to attract quali- particular importance, including achieving the right level of field fied staff—both at headquarters and in field offices—to work presence through incentives for staff and the organization and on LICUS will require giving clear signals of what is deemed to capacity necessary to support the needs of country teams. Dis- be success in LICUS, what outcomes staff will be held ac- cussions are under way with all the Regions on how to strengthen countable for, how much risk it is reasonable to take, how fail- the field presence in fragile states. The Fragile States Group will ure will be judged, and how overall performance evaluation update management and the Board with recommendations and ratings and staff career development will take these into account. their attendant cost estimates in fiscal 2007. Following discus- As in Olympic diving, where the scoring system factors in both sion of this paper, OPCS will also take steps to strengthen the the technical perfection as well as the difficulty of the dive, staff Bank’s surge capacity, staff guidance, and training in fragile performance in LICUS should be similarly judged by assigning states, in line with the IEG recommendations. due weight to the extent of challenges presented by varying LICUS environments. Signaling the importance of LICUS work through- Work is now under way to develop a comprehensive program of out the management hierarchy will also be required. critical skills training based on the assessed need for staff in field offices. This training program—which management considers Apart from incentives, the Bank needs to ensure that staff work- to be the response to this recommendation—is to be rolled out ing on LICUS have relevant skills, such as in public sector man- in fiscal 2007 and will include modules for basic public admin- agement, are capable of seeking and using political knowledge, istration reforms, including the budget function, as well as gam- and are willing and able to work in interdisciplinary teams. Cur- ing scenarios to test Bank staff response skills in complex or rapid rent plans to address these issues in the forthcoming Strength- transitions. ening the Organizational Response to Fragile States are welcome, even if late. More systematic thinking is needed on staffing de- Discussion of the overlap of roles and responsibilities between cisions for LICUS within the context of the Bank’s overall staffing, the Fragile States Group and CPR in SDV hides the useful col- recognizing that assigning more and better-qualified staff to laboration that is taking place between the two teams. Notably, work on LICUS would likely mean trade-offs for other Bank these include joint management of the LICUS Trust Fund grants, country teams. Trade-offs to benefit LICUS may or may not be collaboration on post-conflict needs assessment work, and peer- justified depending on the Bank’s objectives for LICUS as well review functions. Management needs to ensure that all its pri- as other Bank clients’ need for assistance. orities on fragile states are covered and cannot commit to a change in structure. Second, the organizational structure for LICUS and conflict work needs to be streamlined. The Bank needs to ensure an efficient organizational arrangement that removes duplication and frag- mentation of support between the LICUS Unit and the Conflict Prevention and Reconstruction Unit. 163 E N G A G I N G W I T H F R A G I L E S T AT E S IEG recommendation Management response Reassess the value added of the LICUS approach after Agreed three years. The value of the LICUS category and approach, including the op- erational usefulness of the business models, needs to be inde- pendently evaluated after three years, when sufficient experience on the outcomes of the approach will be available. At that time, it should be possible to address the more fundamental question of whether and to what extent Bank assistance can effectively support sustainable state building. Continued Bank support for the LICUS category and approach should be based on the find- ings of that reassessment. 164 APPENDIX BB: MANAGEMENT RESPONSE Attachment A. Review Recommendations Already Covered in Bank Guidance IEG review Reference to policy notes Executive Summary The review focuses on selectivity and prioritization of reform efforts in fragile states. For example: p. xxx: The review notes the importance of selectivity and pri- Selectivity and practicality of approach is also a pillar of the Bank’s core guidance on oritization of reform agendas and within the Bank’s CASs, stat- fragile states. For example, the Good Practice Note on Transitional Results Matrices ing: “[E]ven if collective donor selectivity is not immediately (TRM GPN) notes five “core principles for developing TRMs”—the first two of which achieved, the Bank itself needs to ensure focus and selec- are “Simple” and “Selective,” noting both the strong forces against selectivity in frag- tivity in its own assistance program….” ile states and the risks faced by an overambitious reform strategy. and The TRM GPN also states that “the desirable end result [of developing a TRM] is a ma- trix that focuses on a few key reform goals that will generate visible results and p. xxviii: The review emphasizes the need for donors, in- strengthen a platform for further reform and reconstruction.” cluding the Bank, to understand that “in the absence of a clear and relevant reform agenda, early successes of engagement In addition, the Good Practice Note Fragile States: Good Practice in Country Assistance may be short lived and contribute little to the achievement Strategies (CAS GPN) clearly sets out criteria for prioritization of reforms: “Building on of country strategy objectives.” the zero generation reform approach laid out by the LICUS task force, parameters used to determine priorities in different fragile states have included: (i) actions necessary to lock in promising reforms or lay the basis for future improvements in state delivery; (ii) actions necessary to prevent potential instability; (iii) actions necessary to build popu- lar momentum for reform by generating visible results.” In addition, the CAS GPN notes, “Efforts to build state capacity and accountability in all fragile states will tend to put particular emphasis on the prioritization needed to continue improvement in state performance or prevent failure of key functions” (p. 3, para 11). p. xviii: The review states that “In the deterioration and prolonged The CAS GPN notes “restarting dialogue” as one of the priorities for prolonged crisis crisis or impasse business models, given that there is often lit- countries, stating that “in some situations of prolonged conflict or political impasse, rel- tle consensus between donors and government on development atively noncontroversial development issues may provide an entry point for constructive strategy, engagement needs to include policy dialogue aimed dialogue between the parties to a conflict.” In deteriorating governance countries, the at creating an opening for reform, and simultaneously work on CAS GPN notes that the Bank can provide “input on specific economic issues which are a reform agenda should a window of opportunity arise.” important for mediation efforts and may serve as a way to restart dialogue.” p. xxx: The review notes that “since capacity to use aid ef- The CAS GPN notes the importance of early capacity-building efforts: “In close collab- fectively in post-conflict LICUS is low and governance is oration with the International Monetary Fund, the Bank plays a key role in rebuilding ca- often poor, the focus from day one also needs to be on the pacity on economic policy, public financial management systems, and civil service development of capacity and improvement of governance, not reform or strengthening. It is therefore critical that the Bank is involved in the immedi- just physical infrastructure.” ate post-conflict period (and indeed prior to this), when many critical decisions on the size, scope, and parameters o f public administration will be taken. Assistance in this area may include policy dialogue, analytical work, capacity building and support to re- current expenditures: since state institutions are often new or extremely weak, much stronger knowledge of basic public financial and administrative systems is needed than 165 E N G A G I N G W I T H F R A G I L E S T AT E S IEG review Reference to policy notes in the Bank’s regular IDA clients. The Bank is also engaging more closely with leader- ship capacity building in the early stages of post-conflict transitions, in recognition that leadership that is new to peace-time government may require exceptional support to make this transition successfully.” The TRM GPN states: “As important as the early and visible delivery of tangible bene- fits can be, there are other much less visible actions that must be initiated early on, even though their benefits will not be felt for some time. Strategic and planning efforts must not be delayed; sector visioning, strategy development and policy formulation, defini- tion of institutional capacity needs and planning for the associated capacity develop- ment” (p. 7). Chapter 2 p. 21: The review states: “Critical to the Bank’s effectiveness The CAS GPN emphasizes the importance of political understanding: “The Bank should is its ability to reflect sound political analysis in its strategy continue to encourage country teams to incorporate analysis of the political economy… adequately. This has been an area of weakness in the Bank.” in both CAS/ISN processes and upstream project preparation” (p. 8). p. 26: The review claims: (a) “Building stronger state insti- The CAS GPN states, “It is therefore important that institution-building initiatives avoid tutions and governance are not merely technocratic processes purely technocratic approaches, devoting considerable attention to the process of de- involving the state, but…” cision making and implementation, and to well-designed participation and widespread communication of reform initiatives” (p. 5). (b) “…requires social transformations including those of The CAS GPN also highlights the importance of demand-side reform: “A vibrant civil so- civil society and the relationship between the state and civil ciety and private sector are critical for effective governance: indeed, without a strong society. Bank approaches need to be adequately informed by private sector to generate jobs, incomes and tax revenues, or without popular and civil such considerations.” society demand for accountable services, public sector reforms are unlikely to be sus- tained. Assistance for ‘state-building’ therefore includes support for private sector and civil society development, in all fragile state contexts” (pp. 5–6). Chapter 4 p. 61: The review claims: “Other areas where the Bank needs The CAS GPN lays out for the first time a coherent framework for addressing the polit- to further develop its operational approach include …address ical, security, and development nexus from the Bank’s perspective: linkages between politics, security, and development….” “Moving forward, there is justification to extend successful country experiences in link- ing development and peace building to a deeper and more systematic consideration of these linkages in the Bank’s operational engagement. Recognizing the need for peace- building to be nationally driven and the constraints posed by the Bank’s mandate and expertise, an emphasis on responsiveness to requests from national counterparts for sup- port; maintaining a focus on the Bank’s core economic and development competences; and partnerships with other institutions should be the underlying principles of assistance in this area. In particular, experience from country programs indicates that: • Political economy and conflict analysis are important to inform the selection and se- quencing of priorities for country assistance strategies, as well as project design is- sues. The Bank should continue to encourage country teams to incorporate analysis 166 APPENDIX BB: MANAGEMENT RESPONSE IEG review Reference to policy notes of the political economy and conflict dynamics in both CAS/ISN processes and up- stream project preparation. • The Bank plays an important role in supporting various cross-cutting development processes where peace building may emerge as a priority such as post-conflict needs assessments, recovery plans and results frameworks, PRSPs, public expen- diture and governance assessments, multidonor budget support operations, multi- sector/multidonor trust funds and donor-coordination processes. These processes are by their nature integrative: precluding peace and security issues and institutions from consideration, or placing them on a separate track, creates the real risk of diminishing their importance, missing opportunities for synergy, or ignoring factors which may undermine longer-term development outcomes. The Bank’s role in engaging with po- litical and security sector institutions should focus on its core economic and devel- opment competences (such as generic development planning or public finance capacity building), developing as appropriate partnerships with other donors or in- stitutions that have expertise in specialized technical reform or capacity building in the peace and security areas. • While retaining an emphasis on the Bank’s core economic and development activities, there is scope to increase emphasis on peace-building goals. Peace building is a valid goal to use in country assistance strategies, where sustaining a fragile peace, preventing escalation of conflict or addressing crime and violence which constrain the welfare and development opportunities of the poor have emerged as national priorities. Ac- tivities which contribute to peace-building goals are not only those which directly touch on the security sector, such as demobilization and reintegration of ex-combatants. All economic and development activities infrastructure, human and social sector devel- opment, economic management, private sector and agricultural recovery, etc. can po- tentially be selected or designed to contribute to peace-building goals. Bank assistance strategies and programs can also include the development of partner- ships with other donors and national counterparts which combine respective technical capacities to support peace-building priorities. For example, the Bank may work with the UN (or other institutions taking the lead on political governance and peace building, in- cluding civil society organizations) to provide economic inputs or training to the parties to peace and national reconciliation talks; constitutional reform processes; or economic and development training to political parties and parliamentarians, provided that in all cases this dialogue is nonpartisan and part of a multidonor effort. This approach allows the Bank to make a more systematic contribution to the evolving international partnership for peace-building. It acknowledges that the Bank is still learn- ing about the linkages between peace-building and development; signals a respect for the mandate and expertise of other international institutions; and recognizes that close partnerships are needed” (pp. 7–9). 167 APPENDIX CC: CHAIRMAN’S SUMMARY: COMMITTEE ON DEVELOPMENT EFFECTIVENESS (CODE) On July 26, 2006 the Committee on Develop- selectivity and prioritization required in donor ment Effectiveness (CODE) considered the and Bank reform agendas; (ii) improve Bank report World Bank Support to Low-Income effectiveness in fragile states in deterioration Countries Under Stress: An IEG Review and the and prolonged crisis or impasse; (iii) improve draft Management Response. The statement by the Bank’s donor coordination at the country the External Advisory Panel on the IEG review level to match its strong coordination at the was circulated as background document. international policy level; (iv) clarify the Bank’s central goals in fragile states, state and peace Background. The Bank outlined its approach to building; (v) finalize and implement critical Low-Income Countries Under Stress (LICUS) in human resource reforms (for example, staffing, 2002. In January 2006 the Board considered the incentives); and (vi) take stock of the various LICUS Update, together with the staff guidance adjustments made over the years to the note Fragile States: Good Practices in Country Performance-Based Allocation System (PBA), Assistance Strategies. The LICUS Update called whose cumulative effect on financing for fragile for: (i) increased attention to peace- and state- states is not clear. The four IEG recommenda- building goals in fragile state assistance strate- tions were: (i) clarify the scope and content of gies; (ii) stronger partnership with other organi- the Bank’s state-building agenda and strengthen zations; and (iii) stronger Bank organizational the design and delivery of capacity development response. In January 2006, the Board also and governance support; (ii) develop aid- supported the replenishment of the LICUS Trust allocation criteria for LICUS that ensures that Fund created in 2004. The LICUS Trust Fund is countries are not under- or over-aided; (iii) the only fund that can provide significant strengthen internal Bank support for LICUS assistance to recovering countries in non- work over the next three years; and (iv) reassess accrual status with the Bank, although there are the value added of the LICUS approach after other trust funds to support LICUS, including three years. the Post-Conflict Fund. In the past year, the Board discussed several LICUS country Draft Management Response. Management assistance strategies, including for Afghanistan, welcomed the IEG review, noting that many Kosovo, Tajikistan, Timor-Leste, and Sudan points reinforced the key messages in the LICUS (update), and a group of Executive Directors Update, and echoed the issues considered in visited the Central African Republic. Fragile States: Good Practices for Country Assistance Strategies. It noted that the prelimi- Main Findings and Recommendations of the IEG nary conclusions of the IEG evaluation had been Review. IEG found that the LICUS Initiative has particularly useful in helping staff refine the increased Bank attention to these countries, but fragile state business models presented to Board implementation experience to date has been in January. However, management found that the mixed, although it is too early to assess full IEG review could have been more balanced in outcomes. It noted that significant challenges reflecting positive trends in the performance data remained, including the need to: (i) increase and country examples; discussing state building, 169 E N G A G I N G W I T H F R A G I L E S T AT E S governance, and capacity building; and assessing ing the LICUS Initiative, speakers’ comments selectivity and prioritization, results measure- focused on the following: role and comparative ment, and in-country donor collaboration. advantage of the Bank especially in peace and Country team speakers presented a range of state building; possible refinements of the examples of application of fragile state existing business models and use of instruments; approaches and international partnerships in need to strengthen the knowledge base; aid- country strategy and operations, as well as allocation mechanism and possible need for its underscoring the importance of senior manage- adjustment; measurement of results; and donor ment attention and staffing issues. Management coordination. There was consensus regarding emphasized the newness of the LICUS Initiative the importance of strengthening internal whereby the Bank is learning by doing and the support for LICUS (for example, staffing, need to level the expectation given the difficulties incentives, and organizational structures). faced in fragile states and the high-risk–high- reward nature of work. It elaborated on Bank Next Steps. The IEG review (including the efforts to address much of the IEG findings, Management Response and CODE Chairman’s providing country examples. Management Summary) will be disclosed in September 2006, partially agreed to IEG recommendation (iii) and in absence of a request for a full Board discus- noted its ongoing work on strengthening the sion. There was agreement to reassess the value organizational response to fragile states. Manage- added of the LICUS approach after three years, ment disagreed with IEG recommendation (ii), as recommended by IEG. and believed the current IDA allocation system The main issues raised during the meeting fairly reflects the consensus in the larger develop- were the following: ment community and the IDA donors on the need for a PBA system to ensure aid effectiveness. General Comments. Several speakers noted that it was too early to assess the outcome and Overall Conclusions. The Committee welcomed cautioned against drawing hasty conclusions; the opportunity to discuss the IEG review and they viewed the report as more about learning the draft Management Response. While it may than an assessment. Others were disappointed be too early to draw definitive conclusions about about poor funding for and slow and regionally outcomes, given the complexity of issues faced variable implementation of the Board-endorsed in fragile states, speakers considered it impera- LICUS guidelines, although a number of tive to learn from experience on a frequent speakers also highlighted the risk and ongoing basis. The IEG review was commended uncertainty of fragile states, the challenge of for being informative and incisive, raising critical producing results, and the narrow difference issues, and promoting substantive and construc- between success and failure in difficult country tive dialogue acknowledged by both IEG and environments. Staff commented on the tradeoffs management. CODE also appreciated the between speed, good governance, and capacity presentation of country experiences by development in providing support to fragile operational staff. states and on the need to better address them. A The Committee strongly supported contin- member asked about the prioritization and ued Bank engagement in fragile states, and sequencing of the IEG recommendations, while several speakers expressed appreciation for the another speaker requested IEG to review the dedicated staff working in difficult environments. messages included in the summary of the report, While encouraged by the preliminary findings to make sure that they match the analysis. IEG and early successes of the Bank’s engagement, considered clarifying and monitoring of the members agreed with IEG that there is little state-building agenda as most important, room for complacency. Emphasizing the followed by making the resource allocation importance of “raising the game” in implement- more systematic, and addressing internal 170 A P P E N D I X C C : C H A I R M A N ’ S S U M M A R Y: C O M M I T T E E O N D E V E L O P M E N T E F F E C T I V E N E S S ( C O D E ) organizational issues. A speaker expressed successes in this area, particularly in-country appreciation for the Norwegian Aid Agency’s donor partnerships and capacity development cooperation and support to the preparation of in public finance systems, than implied in the the IEG review. IEG review, and provided examples. The Bank’s Role. Speakers strongly endorsed the Instruments of Support. A number of speakers Bank’s continued engagement in LICUS and the suggested refining the existing business model, focus on state building. They had questions tailoring the approach to varied country about the comparative advantage of the Bank, situations within the LICUS group, and effective the scope and content of the Bank’s state- and use of various Bank Group instruments, peace-building agenda, conflict prevention, especially the analytical and advisory activities. promotion of macroeconomic stability, and Some members also emphasized the capacity development and governance support. importance of the knowledge base of LICUS, In this connection, one speaker observed that particularly the analysis of political economy, the fundamental issue is the alignment between drawing on existing information, involving local the security and development agenda. Some stakeholders to build country ownership, and other speaker noted that these countries face outsourcing as necessarily. The need to periodic setbacks, such as Timor-Leste, which strengthen the quality and relevance of analytic was referred to in the IEG review and the work and of sharing of experiences was Statement by the External Advisory Panel. This emphasized. Management emphasized that the speaker viewed that while lessons should be Bank is adjusting its support and use of instru- learned from such crises, setbacks should be ments to match the changing country context. considered a normal part of engagement in It also assured the Committee of its efforts to fragile situations, and not necessarily an indica- work with countries and donors in delivering tion of failure of donor assistance, including of economic sector work and technical the Bank. Where countries are able to rapidly assistance, building on available informa- recover within their constitutional structures tion—a point that had also been emphasized and without descending into state failure, this in the LICUS Update. Management responded institutional resilience demonstrates a positive in affirmative to a member’s question about result of international investment in institution whether the new Operational Policy on building. Some members cautioned against Emergency Lending will address the procure- overly optimistic expectations, especially the ment and financial management issues faced fiduciary aspects and absorptive capacity. Other in fragile states; the OP will be accompanied by speakers stressed selectivity and prioritization, appropriate guidelines. addressing gender issues and continuous efforts in monitoring, evaluation, and measuring Classification of Countries. Some members sought results. more transparency in classifying countries as Management considered state capacity and LICUS or fragile states. A few of them proposed accountability as core issues faced by fragile introducing a criterion to define “fragility” of a states. It referred back to the LICUS Update of country, while one sought more involvement of January 2006, where it clarified the basis of the partner government in the classification of Bank’s engagement in state and peace building countries as fragile states. The Chair requested (based on country ownership, the Bank’s core IEG to review the use of term “LICUS” in its economic and development competencies, and report in view of the recent broad preference to partnerships with other donors to address refer to these countries as “fragile states.” A peace, security, and development linkages in speaker proposed monitoring “countries at an integrated manner). Management also risk,” reporting annually to the Board. Manage- stressed that there have been more Bank ment acknowledged that the Bank does not 171 E N G A G I N G W I T H F R A G I L E S T AT E S have a strong system of analyzing the risks of maintaining focus on governance. Manage- fragility, which it was prepared to examine ment assured CODE that the issues of time lag of further. CPIA and resource allocation for turnaround situations were on its radar screen, but there Aid-Allocation System. Diverse views were was no easy solution given the careful due expressed with respect to IEG’s recommenda- process required for CPIA. Speakers looked tion to develop aid-allocation criteria for LICUS forward to the IDA 14 Mid-Term Review, which to ensure these countries are not under- or over- was expected to address some of these issues. aided. While some speakers suggested clarifying whether LICUS are under- or over-aided, taking Partnership. Members commented on strength- into consideration the countries’ absorptive ening donor coordination, assessing effective- capacity, others observed that this matter is ness of partnerships, and ensuring the Bank’s beyond the Bank’s control, given the importance country assistance complements that of the UN of other donor allocations. On the issue of and other donors. Staff provided examples of absorptive capacity, management mentioned donor coordination in Sudan, Liberia, and the that two changes in IDA 14 had been made to Central African Republic; speakers appreciated address the needs of fragile state: (i) stretch the hearing about the improvements in this area. In resource allocations for post-conflict countries, response to the interest expressed in UN-Bank based on research that indicates improvements cooperation, management said there have in absorptive capacity three to four years after been improvements, and in countries such as the end of conflict; and (ii) introduce an Democratic Republic of Congo, Timor-Leste, exceptional provision for countries newly re- Liberia, and Haiti, the UN brings their expertise engaging with the international community. It in political governance and security sector also echoed a member’s emphasis on need for reform, while the Bank contributes to sustained and predictable financial support for economic recovery and public finance and fragile states, noting that some of these countries civil service reforms under an integrated (for example, Timor-Leste) do face periodic strategy or results framework . setbacks, and the Bank and donors need to be prepared to stand by these countries Internal Coordination. Many speakers asked about throughout. roles and responsibilities, overlaps, and the pros Many speakers thought the PBA mechanism and cons of merging the LICUS Unit and the could be adjusted or fine tuned. However, a Conflict Prevention and Reconstruction (CPR) member wanted to maintain the current system Unit. A few of them noted the confusion among while others cautioned against major adjust- external partners and the need to strengthen ments and allocation criteria based on factors the case for maintaining two separate units. other than performance. The need for a clearly Some also wondered about the implications, if articulated and defined framework for allocation any, of the new Sustainable Development was also stressed. One member requested Network (SDN). Other questions related to the management views on IEG findings about the link between the LICUS and other Bank initia- “patched-up” nature of the current aid-allocation tives (for example, Africa capacity building) and system for fragile states. The limitations of the the effective use of Trust Funds, which could be CPIA were discussed, including the need to react assessed. Management responded that the CPR quickly to quick turnaround or sharp deteriora- Unit in the Social Development Department is tion of the country situation. Management a technical unit that works with Regions, while agreed with IEG that at each round of IDA the LICUS Unit was established to reinforce replenishment, some adjustments are made to support for fragile stages through enhanced the allocation system. Accordingly, there should cross-sectoral coordination. It had a more be a periodic review of it, including ways to positive view about the collaboration between simplify and enhance transparency, while the two units than IEG, but also accepted the 172 A P P E N D I X C C : C H A I R M A N ’ S S U M M A R Y: C O M M I T T E E O N D E V E L O P M E N T E F F E C T I V E N E S S ( C O D E ) need to further consider the IEG recommenda- elaborated on recent efforts to differentiate tion in the context of the recently constituted incentives between service in LICUS and other SDN. Management expected the new SDN countries by introducing better locality would enhance the synergy between social, premium, hazard pay, and R & R; improving environment, and infrastructure to support reentry guarantees; establishing LICUS service CPR work, as well as better linkage with as a criterion for technical promotion at level Hazard Risk Management Team (also under H; and accommodating family needs. At the SDN), in places where conflict and natural same time, management acknowledged that disasters converge, such as in Aceh. more needed to be done, and Operations Policy and Country Services and Human Resource Staffing. Speakers stressed the need for change Departments, together with the Regions, were in organizational culture and improved deploy- working to further improve the incentive ment of internal resources to support fragile structure for LICUS assignments. A member states, commenting on issues related to deploy- asked management to commit to a timetable for ment of experienced staff in the field, the setting presenting concrete proposals to address up of supportive incentive systems (for example, internal organizational issues, to which manage- promotion, family support, special benefits), ment responded that draft paper on strength- and more transparency regarding allocation of ening the organizational response to fragile staff resources for LICUS. Management states is expected to be ready later in 2006. Pietro Veglio Chairman 173 ENDNOTES Chapter 1 ernance CPIA of 2.6–3.0; and marginal LICUS have an 1. The events of September 11 prompted the overall and governance CPIA of 3.2. Bank to look anew at its mission and mandate. Pres- 8. While fragile states continue to be a tightly de- ident James Wolfensohn was quick to articulate that fined group, the Bank has recognized that fragility is the poverty reduction mission was more important not clear cut and has pointed out that higher-income than ever, because “failed states” with territory out- countries facing the aftermath of conflict, genocide, or side the control of a recognized and reputable gov- social instability (such as the Balkans), more strongly ernment offered fertile soil on which terrorism could performing countries facing rising conflict risks (for ex- thrive (World Bank 2004c). ample, Nepal), and strongly performing states facing 2. Chauvet and Collier (2004) estimate that the eco- fragility in particular subnational regions (as in India, nomic cost (cost in terms of growth) for a country that the Philippines) have found elements of the donor de- starts out as LICUS and has likely prospects of a turn- bates on fragile states useful (World Bank 2005h). around averages 4.6 times its initial GDP, and the eco- 9. The 25 countries classified as LICUS by the Bank nomic cost to the typical neighbor is 3.4 times its in fiscal 2005 had a population of 432 million in 2003. initial GDP. The population figures would increase if countries clas- 3. See, for example, World Bank 2002, 2003a, and sified as LICUS in other fiscal years are also included. In- UNCTAD 2000. The weak past performance of Bank come data are available for 8 of the 25 LICUS and are operations in LICUS is also demonstrated by IEG proj- for different years. Social indicators are birth-weighted ect, Country Assistance Evaluation, and CAS Com- averages for 23 of the 25 LICUS for which these data are pletion Report Review ratings (appendixes Q and R). available. 4. Both DFID and OECD-DAC have identified frag- 10. Although 36 percent of total lending went to ile states as countries in the bottom two quintiles of 18 non-post-conflict LICUS, the lending within this the Bank’s CPIA, as well as those “not rated” on the group was concentrated in a few countries (Nigeria, CPIA. One difference with the Bank, which uses the 60 percent; Cambodia, 11 percent; Lao PDR, 9 percent; CPIA (see appendix A for definition) rating for Public and Uzbekistan, 7 percent) (appendix I). Of the 18 Sector Management and Institutions in addition to the non-post-conflict LICUS, 7 were in non-accrual (indi- overall CPIA rating in defining LICUS, is that DFID and cated in table 1.1). OECD-DAC only use the latter. 11. The administrative budget is more evenly dis- 5. The fiscal 2005 list of LICUS was created using tributed than lending across the LICUS group. How- the fiscal 2004 Gross National Income (GNI) thresh- ever, the variation within each of the two LICUS groups old of $865 or less per capita. The overall CPIA rat- (post-conflict LICUS and non-post-conflict LICUS) is ing is used first as a filter, and then the CPIA rating for higher for the administrative budget than for lending Public Sector Management and Institutions is con- (appendix I). sidered. Appendix A provides the definition of CPIA. 12. Donors agreed to the principles of international 6. In fiscal 2005, countries without CPIA ratings engagement in January 2005 at the Senior-Level Forum were Afghanistan, Liberia, Myanmar, Somalia, Timor- on Aid Effectiveness in Fragile States, co-sponsored by Leste, and the territory of Kosovo. the Bank, OECD-DAC, the European Community, and the 7. Severe LICUS have an overall and governance United Nations Development Program (OECD 2005c). CPIA of 2.5 or less; core LICUS have an overall and gov- http://www.oecd.org/dataoecd/59/55/34700989.pdf. 175 E N G A G I N G W I T H F R A G I L E S T AT E S 13.The term capacity building is used in this re- 9. Including the fiscal 2005 CAS, the CAS Comple- view only when discussing a document that specifically tion Report, and the report to the 2004 Consultative used the term. In all other instances, the term capacity Group Meeting. development is used as the Bank is increasingly using 10. A SWAp supporting public financial manage- this term. ment reform is scheduled for early fiscal 2006. 14. This includes the Multi-country Demobiliza- 11. CPIA 12—Property Rights and Rule-based Gov- tion and Reintegration Program, which accounts for ernance; CPIA 13—Quality of Budgetary and Financial $500 million and covers Angola, Burundi, the Central Management; CPIA 15—Quality of Public Ad- African Republic, and the Democratic Republic of ministration; CPIA cluster D—Public Sector Manage- Congo. ment and Institutions; average CPIA cluster A–C—Economic Management, Structural Policies, Chapter 2 Policies for Social Inclusion/Equity. 1. Interim Strategy Note (ISN) is the umbrella term 12. Doing Business: http://www.doingbusiness.org/ for Transitional Support Strategies and Country Reen- 13. Investment Climate Survey: http://ire gagement Notes. When a normal Country Assistance search.worldbank.org/ics/jsp/index.jsp Strategy approach is not conducive because of coun- 14. Public Expenditure and Financial Accountabil- try circumstances, the Bank may prepare an ISN. For ity: http://www.pefa.org/ details, see “Definitions and Data Sources” in appen- 15. Global Integrity Index: www.globalintegrity.org dix A. 16. Polity: http://www.cidcm.umd.edu/ 2. If India were included, the per capita lending for 17. Transparency International: http://www.trans non-LICUS LICs would be $3.4 in fiscal 2000–02, and parency.org/ $3.6 in 2003–05. 18. Management feels that the evaluation does not 3. The World Bank’s policy places all IBRD loans and take sufficient account of commitment to country- IDA credits to a country in non-accrual status if pay- level donor coordination by the World Bank in frag- ment on any loan or credit is overdue by more than ile states and would draw attention to the joint Country six months. Assistance Strategies completed in Nigeria, Cambodia, 4. Kosovo and Timor-Leste were exceptions to the Somalia, and Togo and under planning in the Demo- debt-distress grant eligibility criterion, and are thus el- cratic Republic of Congo and the Central African Re- igible for grants (see IDA 2005). public; joint Transitional Results Matrixes in Liberia, 5. The ratio of administrative budget to lending Sudan, the Central African Republic, and Haiti; and declined in LICUS (from 0.042 in fiscal 2000–02 to multidonor trust funds and harmonized budget sup- 0.039 in fiscal 2003–05), indicating a lower adminis- port in Afghanistan, Timor Leste, and Sudan and now trative budget for each dollar lent in LICUS in fiscal under planning in the Central African Republic. 2003–05. While the ratio for non-LICUS LICs was 19. Formerly known as the Joint Learning and Ad- lower (0.022) than for LICUS (0.039) in fiscal visory Process on Difficult Partnerships. 2003–05, the ratio for non-LICUS LICs increased 20. The results of a survey for 2004 among mem- (from 0.020 in fiscal 2000–02 to 0.022 in fiscal bers of the Development Assistance Committee rated 2003–05) (table 2.1). the fragile states partnership among the top five net- 6. According to surveys for 79 percent of Haitians, works for quality. On a scale of 0 (unsatisfactory) to radio is the main source of information; for 13 percent, 4 (outstanding), the survey found the quality of the it is word-of-mouth; for 10 percent, it is TV; and for partnership’s work to be 3.4 and its impact 3.08 (3.5 only 4 percent, it is newspapers. and 3.2 in 2003). 7. In 2001, the name of the Post-conflict Unit was 21. Interviews with respondents in the United changed to the Conflict Prevention and Reconstruc- States, the Netherlands, and France. tion (CPR) Unit. The CPR Unit has increased conflict 22. Each pilot country will be managed by a single analysis that examines the causes of conflict, but con- or pair of donors. For example, the pilot program in tinuing work is needed in this area. Sudan will be managed by Norway and the program 8. World Bank. Project No. P064821, PID (2000); ICR in Somalia will be managed jointly by the World Bank (2005). and the United Kingdom. 176 ENDNOTES 23. The UNDG brings together operational agen- and Dollar and Pritchett (1998), which claim to have es- cies in the UN system working on development at the tablished empirically a positive relationship between country level. measures of policy and institutional quality and the ef- 24. The strategy documents reviewed were: fectiveness of aid in bringing about poverty reduction. Afghanistan TSS (2003), Angola ISN (2005), Cambo- 5. During fiscal 1993–95, for every $1 per capita dia CAS (2005), Central African Republic CRN (2004), lent to IDA borrowers overall, about $1.20 was allocated Republic of Congo TSS (2003), Democratic Republic to the top CPIA-quintile performers and about $0.85 per of Congo TSS (2004), Haiti TSS (2004), Kosovo TSS capita to the lowest quintile. But by fiscal 1998–2000, (2004), Liberia RFTF Revision (2005), Nigeria DflD/WB the spread had widened to $2.10 versus $0.60 per Joint Strategy (2005), Papua New Guinea (2005), So- capita. By the time the LICUS Initiative was formulated malia CRN (2004), Sudan JAM Synthesis Framework in 2002, the relationship between aid and governance (2005), Tajikistan CAS (2005), Timor-Leste CAS (2005), had strengthened to the point that a standard deviation and Zimbabwe ISN (2005). increment on the CPIA translated into nearly 100 per- 25. Zimbabwe fieldwork undertaken for this re- cent more assistance (Dollar and Levin 2004). The link view, IEG. between IDA commitments and the IDA Country Per- 26. QAG defines its realism index as the ratio of formance Ratings continued improving throughout problem projects to projects at risk. fiscal 2002–05 (IDA’s commitments, disbursements, 27. There is an improving trend in the overall CPIA and funding for fiscal 2003–05). rating and the CPIA rating for the Public Sector Man- 6. For instance, Beynon (2003), Lensink and White agement and Institutions cluster over the same period. (2001), Dalgaard and Hansen (2001), Hansen and However, the deteriorating trend in the KKZ indica- Tarp (2001), Guillamont and Chauvet (2001), East- tors may be a more robust result, because the KKZ in- erly, Levine, and Roodman (2003), Roodman (2004), dicators are a statistical compilation based on data from and Rajan and Subramanian (2005). several organizations (including the Bank’s CPIA), 7. Their governance indicators are only marginally while the CPIA ratings are based on assessments by worse (and in some cases following an improving Bank staff only. Furthermore, the improvement in trend) than in other countries receiving more aid. the overall CPIA rating and the CPIA rating for the Pub- 8. A technical review paper on governance in the lic Sector Management and Institutions cluster in fis- PBA system is under way at the request of IDA deputies cal 2004 are at least partially explained by the for discussion during the IDA 14 Mid-Term Review in refinement undertaken by the Bank to the bottom of November 2006. the CPIA spectrum in fiscal 2004. In some countries, 100 percent of the improvement in the CPIA rating for Chapter 4 the Public Sector Management and Institutions clus- 1. Refers to Grade E and above staff working on ed- ter results from this refinement. ucation and training, energy, environment, forest/tree crops, health/nutrition/population, highway, irriga- Chapter 3 tion, industry, power, private sector development, 1. The security and reconciliation cluster of the public sector development, rural development, sen- Post-Conflict Progress Indicator (PCPI; appendix A), ior management, social development, social protec- which covers public security, reconciliation, and dis- tion, transportation, urban, and water/sanitation. armament/demobilization, and reintegration, would 2. A few donors suggested that Bank staff would be an example of peace-building variables for post- benefit from the UNDP training course for Resident conflict countries. Representatives. 2. For example, DFID/DAC. 3. For complete text of OP/BP 8.50, see http:// 3. Kanbur (2005), for instance, proposes enriching web.worldbank.org/WBSITE/EXTERNAL/TOPICS/EXTS the CPIA formula by including measures of the rate OCIALDEVELOPMENT/EXTCPR/0,,contentMDK: of improvement of desired outcome variables over a 20486236~menuPK:1260741~pagePK:148956~piPK: given period of time up to the point of assessment. 216618~theSitePK:407740,00.html. 4. This assumption relies, most notably, on the find- 4. For complete text of OP/BP 6.0, see http:// ings of influential papers by Burnside and Dollar (1998) wbln0018.worldbank.org/Institutional/Manuals/ 177 E N G A G I N G W I T H F R A G I L E S T AT E S OpManual.nsf/1337075cf0d5ba638525705c0024aa3a/2c (Three others, including Lao PDR, a fiscal 2005 LICUS b575f62255c53a85256e8a0078068c?OpenDocument. country, were found to meet the indebtedness crite- 5. In semistructured interviews, Bank staff were, ria, but have stated that they do not wish to avail however, more positive than indicated by the Stake- themselves of the HIPC Initiative.) In addition to holder Survey about the LICUS Unit’s contribution to these 11 countries, others could be added on a case- harmonization/alignment (for example, through Tran- by-case basis if their data are verified to meet the rel- sitional Results Matrices). evant criteria. For instance, Afghanistan, also a fiscal 6. Only two staff above the GF level, one GE level 2005 LICUS country, would then be included in the list staff, one A–D level staff, three secondees, two junior if, upon reconciliation of its debt, its debt indicators professional associates, and one extended-term tem- are found to be above the relevant thresholds (World porary; staff data provided by LICUS Unit, July 2005. Bank 2004a; World Bank and IMF 2006b). 7. Bank staff may have factored operational use- 4. In the sense that the oil windfall produces sov- fulness into their assessment of quality. ereign resource rents that can generate dysfunctional rent-seeking behavior. Appendix H 5. Lower policy quality (CPIA) implies lower sus- 1. The objectives of the Enhanced HIPC Initiative tainable debt thresholds and, implicitly, a higher grant are, first, “to deal comprehensively with the overall component. debt burden of eligible countries by removing their 6. For instance, the HIPC Initiative has reduced debt debt overhang within a reasonable period of time ratios by half, on average, in 18 countries. But debt sus- and providing a base from which to achieve debt sus- tainability, the primary objective of the initiative, re- tainability and exit the rescheduling cycle,” and, sec- mains elusive. In 11 of 13 countries with available ond, to free up resources for poverty reduction (World data, the key indicator of external debt sustainability Bank and IMF 2006b). has deteriorated since completion point, and in 8 of 2. HIPC decision point is the date at which a heav- these countries the ratios once again exceed HIPC tar- ily indebted poor country with an established track gets (IEG 2006b). record of good performance under adjustment pro- grams supported by the International Monetary Fund Appendix L (IMF) and the World Bank commits to undertake ad- 1. In Angola, for instance, related analytical work in- ditional reforms and to develop and implement a cludes the Public Expenditure Management and Fi- poverty-reduction strategy. A country may start to re- nancial Accountability Review, Oil Diagnostic Study, ceive interim relief at this point. HIPC completion Corporate Social Responsibility Report (with a focus point is the date at which the country successfully com- on the oil sector), and the Oil Revenue Management pletes the key structural reforms agreed at the deci- Study. sion point, including the development and 2. In the Democratic Republic of Congo, mining sec- implementation of its poverty-reduction strategy. The tor management issues are addressed through public country then receives the bulk of debt relief under the financial management studies, including the Public HIPC Initiative without any further policy conditions Expenditure Review, Country Procurement Assess- (Steps of the HIPC Initiative: A Guide, available at ment Report, Public Enterprises Reform Study, Fi- http://web.worldbank.org/WBSITE/EX nancial Sector Assessment Program, Country Financial TERNAL/TOPICS/EXTDEBTDEPT/0„contentMDK: Accountability Assessment, Institutional and Gover- 20655535~menuPK:64166739~pagePK:64166689~pi nance Review, and Country Economic Memorandum. PK:64166646~theSitePK:46904 3,00.html). 3. In September 2004, the Boards of IDA and the Appendix M IMF decided to extend the “sunset clause” of the En- 1. The sample is not restricted to LICUS and in- hanced HIPC Initiative to end-2006, as well as to iden- cludes a total of 83 operations. tify countries that at the end of 2004 had estimated 2. There is a potential upward bias in the ratings debt burden indicators above the enhanced HIPC Ini- of adjustment operations because of the nature of pol- tiative thresholds. As of April 2006, staff identified 11 icy actions against which performance is monitored. countries as potentially eligible for the HIPC Initiative. For instance, the ratings could incorporate the suc- 178 ENDNOTES cessful passage of necessary legislation. Whether this diverse, or unprecedented. Category B: The project’s legislation is being applied in practice is more difficult potential adverse environmental impacts on human to observe unambiguously, which could result in the populations or environmentally important areas are overestimation of the result. less adverse than those of Category A. Category C: The 3. The sample included projects approved in fiscal project is likely to have minimal or no adverse envi- 2000 or thereafter, for which IEG ratings already exist: ronmental impacts. Category FI: The project involves 31 operations in total, of which 9 are adjustment and investment of Bank funds through a financial inter- 22 are investment. mediary in subprojects that may result in adverse en- 4. Including marginally unsatisfactory, unsatisfac- vironmental impacts. tory, and highly unsatisfactory. 2. The sample was stratified by two periods: proj- 5. It is important to note here that SWAps should ects approved over fiscal 2000–02, and projects ap- not be viewed as a direct alternative to DPL in the sense proved over fiscal 2003–05. that DPL is a lending instrument whereas SWAps are 3. An IBRD/IDA commitment level of $13 million an approach to development program planning and was used to divide “large” from “small” projects. implementation that can rely on various lending in- struments (including budget support). Appendix BB 1. Other Good Practice Notes, as noted in IEG’s re- Appendix O view, include World Bank 2005f and 2005k; and World 1. The Institutional and Governance Review has also Bank and UNDP 2005. been included in the list of LICUS core diagnostic re- 2. IEG notes that these results are not necessarily ports because the 2002 LICUS Task Force Report iden- fully attributable to the LICUS Initiative, as only one tified the IGR as an essential piece of ESW for LICUS. project covered by these ratings was approved in the period following its inception. Appendix P 3. IEG notes that the report is referring to the 1. QAG has carried out six assessments of quality CASCR Review ratings for those CASCRs that covered at entry and five of quality of supervision. Three as- at least part of the period since the inception of the sessments for both were conducted prior to fiscal LICUS Initiative—of the four such CASCRs reviewed by 2000. Because QAG conducts these reviews for a very IEG, three were rated moderately unsatisfactory or un- small sample of projects, results for pre-fiscal 2000 are satisfactory, and one was rated moderately satisfactory. compared with results of fiscal 2000 onward. 4. See, for example, World Bank, 2005e, pp. 3–6, and World Bank, 2005h, pp. 10–13. See also the LICUS Appendix U Web site for knowledge work on political economy and 1. Category A: The project is likely to have signif- state building, including the joint PREM-OPCS work- icant adverse environmental impacts that are sensitive, shop on state building. 179 BIBLIOGRAPHY Beynon, Jonathan. 2003. “Poverty-Efficient Aid Al- cation and Poverty Reduction.” European locations: Collier/Dollar Revisited.” Economic Economic Review 45: 1–26. and Statistics Analysis Unit Working Paper No. ———. 2001. “Can the World Cut Poverty in 2, Overseas Development Institute, London. Half? How Policy Reform and Effective Aid Birdsall, Nancy, and Devesh Kapur. 2005. The Can Meet International Development Goals.” Hardest Job in the World: Five Crucial Tasks World Development 29 (11). for the New President of the World Bank. Collier, Paul, and Anke Hoeffler. 2002. “Aid, Pol- Washington, DC: Center for Global icy and Growth in Post-Conflict Countries.” Development. Policy Research Working Paper No. 2902, Boesen, Nils, and Ole Therkildsen. 2003. Be- World Bank, Washington, DC. tween Naivety and Cynicism. A Pragmatic Dalgaard, Carl-Johan, and Henrik Hansen. 2001. Approach to Donor Support for Public Sector “On Aid, Growth, and Good Policies.” CREDIT Capacity Development. Copenhagen: Danida. Research Paper No. 00/17, Centre for Research Burnside, Craig, and David Dollar. 1998. “Aid, Poli- in Economic Development and International cies and Growth.” Policy Research Working Trade, University of Nottingham, U.K. Paper No. 1777, World Bank, Washington, DC. DFID (Department for International Develop- Carnegie Endowment for International Peace. ment). 2005. “Why We Need to Work More Ef- 2004. States at Risk—Stabilization and State- fectively in Fragile States.” DFID Policy Paper. Building by External Intervention. Wash- London. ington, DC. Dollar, David, and Victoria Levin. 2004. “The In- Chauvet, Lisa, and Paul Collier. 2005. Policy Turn- creasing Selectivity of Foreign Aid, 1984–2002.” arounds in Failing States. Oxford, U.K.: Ox- Policy Research Working Paper No. 3299, ford University Press. World Bank, Washington, DC. ———. 2004. Development Effectiveness in Frag- Dollar, David, and Lant Pritchett. 1998. Assessing ile States: Spillovers and Turnarounds. Ox- Aid—What Works, What Doesn’t, and Why. ford, U.K.: Oxford University Press. Washington, DC: World Bank. Cliffe, Sarah, Scott Guggenheim, and Markus Easterly, William, Ross Levine, and David Rood- Kostner. 2003. “Community-Driven Recon- man. 2003. “New Data, New Doubts: A Com- struction as an Instrument in War-to-Peace ment on Burnside and Dollar’s Aid, Policies, Transitions.” CPR Working Paper No. 7, World and Growth` (2000).” NBER Working Paper Bank, Washington, DC. No. 9846, National Bureau of Economic Re- Collier, Paul. 2005. Is Aid Oil? An Analysis of search, Cambridge, MA. Whether Africa Can Absorb More Aid. Ox- Foreign Policy and the Fund for Peace. 2005. ford, U.K.: Centre for the Study of African “The Failed States Index.” Foreign Policy Economics, Oxford University. (July/August). ———. 2003. Breaking the Conflict Trap: Civil François, Monika, and Inder Sud. 2006. “Pro- War and Development Policy. Washington, moting Stability and Development in Fragile DC: World Bank. and Failed States.” Development Policy Review Collier, Paul, and David Dollar. 2002. “Aid Allo- 24 (2): 141–60. 181 E N G A G I N G W I T H F R A G I L E S T AT E S Grindle, Merilee S. 2004. “Good Enough Gov- port No. 34351. IEG Study Series. Washington, ernance: Poverty Reduction and Reform in DC: World Bank. Developing Countries.” Governance: An In- ———. 2005b. The Effectiveness of World Bank ternational Journal of Policy, Administra- Support for Community-Based and -Driven tion, and Institutions 17 (4): 525–48. Development. IEG Study Series. Washington, Guillamont, Patrick, and Lisa Chauvet. 2001. “Aid DC: World Bank. and Performance: A Reassessment.” Journal ———. 2003a. Debt Relief for the Poorest: An of Development Studies 37 (6): 66–87. OED Review of the HIPC Initiative. IEG Study Hansen, Henrik, and Finn Tarp. 2001. “Aid and Series. Washington, DC: World Bank. Growth Regressions.” Journal of Develop- ———. 2003b. Extractive Industries and Sus- ment Economics 64: 547–70. tainable Development: An Evaluation of Hassan, Fareed. 2004. “Lessons Learned from World Bank Group Experience. A Joint IEG- World Bank Experience in Post-Conflict Re- World Bank, IEG-IFC, and IEG-MIGA evaluation. construction.” IEG Conference Note, World IEG Study Series. Washington, DC: World Bank. Bank, Washington, DC. ———. 2002. Social Funds: Assessing Effective- IDA (International Development Association). ness. IEG Study Series. Washington, DC: World 2005. Additions to IDA Resources (14th Re- Bank. plenishment). Working Together to Achieve ———. 2000. “Utilization of Project Implemen- the Millennium Development Goals. Report tation Units (PIUs). Lessons and Practices.” No. 31693. Washington, DC: World Bank. Washington, DC. ———. 2004. IDA’s Performance-Based Alloca- ———. 1998. The World Bank’s Experience with tion System: Update on Outstanding Issues. Post-Conflict Reconstruction. IEG Study Se- Report No. 27837. Washington, DC: World ries. Washington, DC: World Bank. Bank. Kanbur, Ravi. 2005. “Reforming the Formula: A ———. 2002. Additions to IDA Resources (13th Modest Proposal for Introducing Develop- Replenishment). Supporting Poverty Reduc- ment Outcomes in IDA Allocation Procedures.” tion Strategies. Report No. 34298. Washington, Discussion Paper No. 4971, Center for Eco- DC: World Bank. nomic Policy Research, Washington, DC. IEG (Independent Evaluation Group–World Koeberle, Stefan, and Zoran Stavreski. 2005. “Bud- Bank). 2006a. Assessing the Effectiveness of the get Support: Concept and Issues.” Paper pre- Bank’s Fiduciary Work: An Evaluation of sented at the Practitioners’ Forum on Budget Country Financial Accountability Assess- Support, Cape Town, South Africa, May 5–6. ments and Country Procurement Assessment Lensink, Robert, and Howard White. 2001. “Are Reports, July 1999-December 2004. Washing- There Negative Returns to Aid?” The Journal ton, DC: World Bank. of Development Studies 37 (6): 42–65. ———. 2006b. Debt Relief for the Poorest: An Levin, Victoria, and David Dollar. 2005. “The For- Evaluation Update of the HIPC Initiative. gotten States: Aid Volumes and Volatility in Dif- IEG Study Series. Washington, DC: World ficult Partnership Countries (1992–2002).” Bank. Summary Paper for DAC Learning and Advi- ———. 2006c. “Timor-Leste: Community and sory Process on Difficult Partnerships, OECD- Local Governance Project” (P069762), “Second DAC, Geneva. Community Empowerment Project” Levy, Brian, and Sahr Kpundeh, eds. 2004. Build- (P072356), “Third Community Empowerment ing State Capacity in Africa: New Approaches, and Local Governance Project” (P075342), Emerging Lessons. Washington, DC: World “Agricultural Rehabilitation Project” (P070533), Bank Institute. and “Second Agriculture Rehabilitation Pro- Mitra, Saumya. 2004. “Aid and Development.” ject” (P073911). World Bank, Washington, DC. Comments at the conference on the Future of ———. 2005a. Capacity Building in Africa: An Intervention: Intervention, Legitimacy and the OED Evaluation of World Bank Support. Re- Reconstruction of Statehood. Berlin, July 13. 182 BIBLIOGRAPHY OED (Operations Evaluation Department, The and Reconstruction Working Paper No. 2, Re- World Bank). See IEG. port No. 26361. World Bank, Washington, DC. OECD (Organisation for Economic Co-operation Roodman, David. 2004. “The Anarchy of Num- and Development. 2005a. “Chair’s Summary: bers: Aid, Development, and Cross-country Senior Level Forum on Development Effec- Empirics.” Working Paper No. 32, Center for tiveness in Fragile States.” Development Co- Global Development, Washington, DC. operation Directorate and Development Schiavo-Campo, Salvatore. 2003. “Financing and Assistance Committee. Paris: OECD. Aid Management Arrangements in Post- ———. 2005b. “Piloting the Principles for Good Conflict Situations.” CPR Working Paper No. International Engagement in Fragile States.” 6, World Bank, Washington, DC. Draft Concept Note, Development Coopera- UNCTAD (United Nations Conference on Trade tion Directorate and Development Assistance and Development). 2000. The Least Devel- Committee. DCD (2005)11/REV1, Paris. oped Countries Report, 2000: Aid, Private ———. 2005c. “Principles for Good Interna- Capital Flows, and External Debt. New York. tional Engagement in Fragile States.” OECD- UNDG (UN Development Group) and The World DAC Draft Paper, Development Cooperation Bank. 2005. “Operational Note on Transitional Directorate and Development Assistance Results Matrices, Using Results-Based Frame- Committee, DCD (2005)8/Rev2, Paris. works in Fragile States.” Washington, DC. ———. 2005d. The DAC Fragile States Agenda: UNDG (United Nations Development Group), Progress and Issues Development. Paris. UNDP (United Nations Development Pro- ———. 2005e. “Work Plan for the Fragile States gram), and The World Bank. 2004. Practical Group (FSG) 2005–2006.” Development Coop- Guide to Multilateral Needs Assessments in eration Directorate, DCD(2005)12/REV2, Paris. Post-Conflict Situations. Report No. 29822. ———. 2003. “Draft Mandate and Terms of Ref- Washington, DC: UNDG, UNDP, World Bank. erence for the Joint Learning and Advisory UNDP (United Nations Development Program). Process on Difficult Partnerships.” Paris. 2005. Human Development Report: Interna- Ottaway, Marina. 2003. Democracy Challenged. tional Cooperation at a Crossroads. New York. The Rise of Semi-Authoritarianism. Wash- Unsworth, S. 2003. “Better Government for Poverty ington DC: Carnegie Endowment for Inter- Reduction: More Effective Partnerships for national Peace. Change.” Consultation document, Department ———. 2002. “Rebuilding State Institutions in for International Development, London. Collapsed States.” Development and Change Van de Walle, Nicolas. 2005. Overcoming Stag- 33 (5): 1001–24. nation in Aid-dependent Countries. Wash- Perlez, Jane. 2006. “From Bad to Worse in East ington, DC: Center for Global Development. Timor.” International Herald Tribune, May 31. World Bank. Forthcoming a. Aid That Works: Rajan, Raghuram, and Arvind Subramanian. 2005. Successful Development in Fragile States. “Aid and Growth: What Does the Cross- Washington, DC. Country Evidence Really Show?” NBER Work- ———. Forthcoming b. Strengthening the Or- ing Paper No. 11513, National Bureau of ganizational Response to Fragile States. Wash- Economic Research, Cambridge, MA. ington, DC. Reuters. 2005. “Lots of Fanfare, Billions of Dol- ———. 2006a. “Are We Really Getting Rid of lars and Not Much to Show.” April 3. PIUs?” World Bank, Washington, DC. Rogier, Emeric. 2005. Designing an Integrated http://web.worldbank.org/WBSITE/EXTERNAL/ Strategy for Peace, Security and Development TOPIC S/EXTCDRC/0,,contentMDK: in Post-Agreement Sudan. The Hague: Nether- 20845819~pagePK:64169192~piPK:64169180 lands Institute of International Relations. ~theSitePK:489952,00.html Rohland, Klaus, and Sarah Cliffe. 2002. “The East ———. 2006b. Annual Report on Portfolio Per- Timor Reconstruction Program: Successes, formance: Fiscal Year 2005. Quality Assur- Problems and Tradeoffs.” Conflict Prevention ance Group. Washington, DC. 183 E N G A G I N G W I T H F R A G I L E S T AT E S ———. 2006c. “Cambodia: World Bank Releases ———. 2005l. Papua New Guinea Interim Strat- Latest Statement and Update.” News Release egy Note. Washington, DC. No. 2006/493/EAP. June 26. ———. 2005m. “World Bank Operational Man- ———. 2006d. Global Monitoring Report 2006: ual.” Washington, DC. Strengthening Mutual Accountability, Aid, ———. 2004a. Enhanced HIPC Initiative: Pos- Trade, and Governance. Washington, DC. sible Options Regarding the Sunset Clause. ———. 2006e. IDA’s Implementation of the Mul- Washington, DC. tilateral Debt Relief Initiative. Vol. 1. Wash- ———. 2004b. “The LICUS Challenge.” Corpo- ington, DC. rate Day Presentation. Washington, DC. ———. 2005a. Building Effective States—Forg- ———. 2004c. “The Role of the World Bank in ing Engaged Societies. Report of the World Conflict and Development: An Evolving Bank Task Force on Capacity Development in Agenda.” CPR Seminar Series. Washington, DC. Africa. Washington, DC. ———. 2004d. “Workshops for Leaders, and ———. 2005b. Country Assistance Strategy for Leadership Workshops: Support for Meeting the Kingdom of Cambodia. Report No. 32118- the Challenges of Transition.” Synthesis Note KH. Washington, DC. on Leadership Workshops 2001–2004. Wash- ———. 2005c. Enabling Country Capacity to ington, DC. Achieve Results—2005 CDF Progress Report. ———. 2003a. Low-Income Countries Under OPCS Report No. 34102. Washington, DC. Stress: Implementation Overview. OPCS Re- ———. 2005d. Engaging Civil Society Organi- port No. 27536. Washington, DC. zations in Conflict-Affected and Fragile States: ———. 2003b. Low-Income Countries Under Three African Case Studies. Washington, DC. Stress: Implementation Trust Fund. OPCS Re- ———. 2005e. Fragile States: Good Practices port No. 27535. Washington, DC. in Country Assistance Strategies. Report No. ———. 2002. World Bank Group Work in Low- 34790. Washington, DC. Income Countries Under Stress: A Task Force ———. 2005f. Good Practice Note for Devel- Report. Report No. 26903. Washington, DC. opment Policy Lending: Development Policy ———. 1998. Post-Conflict Reconstruction: The Operations in Fragile States. Report No. Role of the World Bank. Report No. 17752. 32636. Washington, DC. Washington, DC. ———. 2005g. Low-Income Countries Under ———. 1997. World Development Report 1997: Stress (LICUS) Implementation Trust Fund: The State in a Changing World. Washington, Request for Replenishment and Special DC. Support for Sudan. Report No. 34831. Wash- World Bank and International Monetary Fund. ington, DC. 2006a. Heavily Indebted Poor Countries ———. 2005h. Low-Income Countries Under (HIPC) Initiative — Listing of Ring-Fenced Stress: Update. Report No. 34789. Washington, Countries That Meet the Income and In- DC. debtedness Criteria at End 2004. Washing- ———. 2005i. Meeting the Challenge of Africa’s ton, DC: World Bank and IMF. Development: A World Bank Group Action ———. 2006b. HIPC—Status of Implementa- Plan. Washington, DC. tion. Washington, DC: World Bank and IMF. ———. 2005j. World Development Indicators. World Bank and United Nations Development Washington, DC. Programme (UNDP). 2005. An Operational ———. 2005k. Good Practice Note for Devel- Note on Transitional Results Matrices: Using opment Policy Lending: Development Policy Results-Based Frameworks in Fragile States. Operations and Program Conditionality in Washington, DC, and New York: World Bank and Fragile States. Washington, DC. UNDP . 184 IEG PUBLICATIONS Study Series 2004 Annual Review of Development Effectiveness: The Bank’s Contributions to Poverty Reduction Addressing the Challenges of Globalization: An Independent Evaluation of the World Bank’s Approach to Global Programs Agricultural Extension: The Kenya Experience Assisting Russia’s Transition: An Unprecedented Challenge Bangladesh: Progress Through Partnership Brazil: Forging a Strategic Partnership for Results—An OED Evaluation of World Bank Assistance Bridging Troubled Waters: Assessing the World Bank Water Resources Strategy Capacity Building in Africa: An OED Evaluation of World Bank Support The CIGAR at 31: An Independent Meta-Evaluation of the Consultative Group on International Agricultural Research Country Assistance Evaluation Retrospective: OED Self-Evaluation Debt Relief for the Poorest: An OED Review of the HIPC Initiative Developing Towns and Cities: Lessons from Brazil and the Philippines The Drive to Partnership: Aid Coordination and the World Bank Economies in Transition: An OED Evaluation of World Bank Assistance The Effectiveness of World Bank Support for Community-Based and –Driven Development: An OED Evaluation Evaluating a Decade of World Bank Gender Policy: 1990–99 Evaluation of World Bank Assistance to Pacific Member Countries, 1992–2002 Financial Sector Reform: A Review of World Bank Assistance Financing the Global Benefits of Forests: The Bank’s GEF Portfolio and the 1991 Forest Strategy and Its Implementation Fiscal Management in Adjustment Lending IDA’s Partnership for Poverty Reduction Improving the Lives of the Poor Through Investment in Cities India: The Dairy Revolution Information Infrastructure: The World Bank Group’s Experience Investing in Health: Development Effectiveness in the Health, Nutrition, and Population Sector Jordan: Supporting Stable Development in a Challenging Region Lesotho: Development in a Challenging Environment Mainstreaming Gender in World Bank Lending: An Update Maintaining Momentum to 2015? An Impact Evaluation of Interventions to Improve Maternal and Child Health and Nutrition Outcomes in Bangladesh The Next Ascent: An Evaluation of the Aga Khan Rural Support Program, Pakistan Nongovernmental Organizations in World Bank–Supported Projects: A Review Poland Country Assistance Review: Partnership in a Transition Economy Poverty Reduction in the 1990s: An Evaluation of Strategy and Performance The Poverty Reduction Strategy Initiative: An Independent Evaluation of the World Bank’s Support Through 2003 Power for Development: A Review of the World Bank Group’s Experience with Private Participation in the Electricity Sector Promoting Environmental Sustainability in Development Putting Social Development to Work for the Poor: An OED Review of World Bank Activities Reforming Agriculture: The World Bank Goes to Market Sharing Knowledge: Innovations and Remaining Challenges Social Funds: Assessing Effectiveness Tunisia: Understanding Successful Socioeconomic Development Uganda: Policy, Participation, People The World Bank’s Experience with Post-Conflict Reconstruction The World Bank’s Forest Strategy: Striking the Right Balance Zambia Country Assistance Review: Turning an Economy Around Evaluation Country Case Series Bosnia and Herzegovina: Post-Conflict Reconstruction Brazil: Forests in the Balance: Challenges of Conservation with Development Cameroon: Forest Sector Development in a Difficult Political Economy China: From Afforestation to Poverty Alleviation and Natural Forest Management Costa Rica: Forest Strategy and the Evolution of Land Use El Salvador: Post-Conflict Reconstruction India: Alleviating Poverty through Forest Development Indonesia: The Challenges of World Bank Involvement in Forests The Poverty Reduction Strategy Initiative: Findings from 10 Country Case Studies of World Bank and IMF Support Uganda: Post-Conflict Reconstruction Proceedings Global Public Policies and Programs: Implications for Financing and Evaluation Lessons of Fiscal Adjustment Lesson from Urban Transport Evaluating the Gender Impact of World Bank Assistance Evaluation and Development: The Institutional Dimension (Transaction Publishers) Evaluation and Poverty Reduction Monitoring & Evaluation Capacity Development in Africa Public Sector Performance—The Critical Role of Evaluation All IEG evaluations are available, in whole or in part, in languages other than English. For our multilingual selection, please visit http://www.worldbank.org/ieg THE WORLD BANK ISBN 0-8213-6847-8