THE WORLD BANK W O R L D B A N K O P E R A T I O N S E V A L U A T I O N D E P A R T M E N T The CGIAR at 31 An Independent Meta-Evaluation of the Consultative Group on International Agricultural Research OED PUBLICATIONS Study Series 2002 Annual Review of Development Effectiveness--Achieving Development Outcomes: The Millennium Challenge Agricultural Extension: The Kenya Experience Agricultural Extension and Research: Achievements and Problems in National Systems Bangladesh: Progress Through Partnership Bridging Troubled Waters: Assessing the World Bank Water Resources Strategy 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 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 India: The Challenges of Development India: The Dairy Revolution Information Infrastructure: The World Bank Group's Experience Investing in Health: Development Effectiveness in the Health, Nutrition, and Population Sector Lesotho: Development in a Challenging Environment Mainstreaming Gender in World Bank Lending: An Update The Next Ascent: An Evaluation of the Aga Khan Rural Support Program, Pakistan Nongovernmental Organizations in World Bank­Supported Projects: A Review Paddy Irrigation and Water Management in Southeast Asia Poland Country Assistance Review: Partnership in a Transition Economy Poverty Reduction in the 1990s: An Evaluation of Strategy and Performance Promoting Environmental Sustainability in Development Reforming Agriculture: The World Bank Goes to Market Social Funds: Assessing Effectiveness 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 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 Multilingual Editions Allègement de la dette pour les plus pauvres : Examen OED de l'initiative PPTE Appréciation de l'efficacité du développement : L'évaluation à la Banque mondiale et à la Société financière internationale Determinar la eficacia de las actividades de desarrollo : La evaluación en el Banco Mundial y la Corporación Financiera Internacional Côte d'Ivoire : Revue de l'aide de la Banque mondiale au pays Filipinas: Crisis y oportunidades Reconstruir a Economia de Moçambique : http://www.worldbank.org/oed W O R L D B A N K O P E R A T I O N S E V A L U A T I O N D E P A R T M E N T The CGIAR at 31 An Independent Meta-Evaluation of the Consultative Group on International Agricultural Research 2004 THE WORLD BANK http://www.worldbank.org/oed Washington, D.C. © 2003, 2004 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 Manufactured in the United States of America First edition April 2003 Revised edition February 2004 1 2 3 4 05 04 03 The findings, interpretations, and conclusions expressed here are those of the Operations Evaluation Department and do not necessarily reflect the views of the Board of Executive Directors of the World Bank or the governments they represent. The World Bank cannot guarantee the accuracy of the data included in this work. The boundaries, colors, denominations, and other information shown on any map in the work do not imply on the part of the World Bank any judgment of 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 World Bank encourages dissemination of its work and will normally grant permission 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, www.copyright.com. All other queries on rights and licenses, including subsidiary rights, should be addressed to the Office of the Publisher, World Bank, 1818 H Street NW, Washington, DC 20433, USA, fax 202-522-2422, e-mail pubrights@worldbank.org. This volume is a combined, edited version of the original two-volume evaluation produced by OED. All photographs courtesy of the CGIAR Library. ISBN 0-8213-5645-3 eISBN 0-8213-5646-1 Library of Congress Cataloging-in-Publication Data Lele, Uma J. The CGIAR at 31 : an independent meta-evaluation of the Consultative Group on International Agricultural Research / Uma Lele. p. cm. -- (Operations evaluation study) Contents: v. 1. Overview report. Includes bibliographical references. ISBN 0-8213-5645-3 1. Consultative Group on International Agricultural Research. 2. Agriculture--Research--International cooperation. I. Title: CGIAR at thrity-one. II. Consultative Group on International Agricultural Research. III. Title. IV. World Bank operations evaluation study. S540.8.C66L45 2003 630'7'2--dc22 2003061162 Printed on Recycled Paper Contents vii Acknowledgments ix Foreword ix Prólogo ix Avant-propos xv Preface xv Prefacio xv Préface xxiii Abbreviations and Acronyms PART 1: OVERVIEW 3 1. CGIAR's Changing Mission and Authorizing Environment 7 2. CGIAR Activities and Impacts: What Do We Know? 15 3. New Challenges for the CGIAR: How Is It Responding? 21 4. Restructuring the CGIAR: Lessons of Past Attempts 27 5. The Current CGIAR Reforms: Some Are Creditable But Others Need Revisiting 33 6. The Critical Roles of the World Bank: Convener, Donor, and Development Partner 37 7. Recommendations 37 Focusing the World Bank's Responsibilities 37 Reforming the CGIAR PART 2: TECHNICAL REPORT 41 SECTION I THE CONTEXT 43 8. The Context for the CGIAR Meta-Evaluation i i i T H E C G I A R AT 3 1 43 An Overview of the CGIAR 48 A Conceptual Framework for Assessing CGIAR Effectiveness 59 Previous Evaluations 61 The Change Management and Design Process 65 SECTION II THE DEVELOPMENT EFFECTIVENESS OF THE CGIAR 67 9. High Returns to Germplasm Research 68 Examining the Impacts of Germplasm Research 71 Related Issues Requiring Further Impact Assessments 75 10. Genetic Resources Strategy and Management 76 Genetic Resource Management and Use: Issues for the CGIAR 79 The Challenges of a Rapidly Changing Environment 80 Can the CGIAR Meet this Challenge? 85 Reconfiguring the CGIAR to Meet the Challenge 87 11. Policy Research in the CGIAR 89 IFPRI's Policy Research: 1998 External Program and Management Review 90 Other Reviews of IFPRI Impacts 93 12. Natural Resources Management Research in the CGIAR 94 Thematic Priorities in NRM Research 99 Focus 100 Framework 103 13. National Agricultural Research Systems and the CGIAR 104 The Black Box of High Returns: The Partnership Role of the NARS 105 Pursuing Partnerships with Advanced NARS 107 Ensuring Continued Support to Small-Country NARS 109 14. The CGIAR and Sub-Saharan Africa 110 Activities, Impacts, and Constraints 113 Priority Setting and the Roles of Regions 114 Confronting Shortfalls in Public Agricultural Spending 115 The Role of Donors and the Bank 117 SECTION III THE ORGANIZATIONAL EFFECTIVENESS OF THE CGIAR 119 15. Past Attempts to Restructure the CGIAR 119 The Financial Crisis 119 The Renewal Process 121 The Mixed Impacts of Renewal 121 The Third System Review (1998) 122 The Federation Proposal (2000) 122 The CGIAR's Founding Principles Need Revisiting 124 The Challenge of Reconfiguration 124 Collective Action Problems Are Daunting 127 16. Reform in the CGIAR (2000­Present) 128 The Executive Council 130 The System Office 130 The Science Council 132 Challenge Programs 135 Reforms Do Not Go Far Enough i v C O N T E N T S 137 17. The World Bank and the CGIAR 137 Exercising the Bank's Leadership Role as Founder, Cosponsor, and Convener 139 Funding the CGIAR and Exit Strategy from the DGF? 141 The Bank's Multiple Roles 143 Enhancing the Bank's Role as a Development Partner 147 18. Conclusions 147 Science and Strategy 150 Governance, Organization, Finance, and Management 152 The Role of the World Bank 155 Appendixes 157 1: Current Organizational Chart of the CGIAR 159 2: Meta-Evaluation Terms of Reference 161 3: Meta-Evaluation Team and Advisory Committee: Biographical Information 165 4: Methods and Tools 167 5: List of Working and Background Papers, Authors, and Peer Reviewers 169 6: List of People Consulted 175 7: Management Response 183 8: Chairman's Summary: Committee on Development Effectiveness 187 Endnotes 209 Bibliography Figures 5 1.1 Where the Poor Are 5 1.2 Cereal Yields: Slowdown in Growth 8 2.1 Research with Proven Impacts on Poverty Has Declined Dramatically 9 2.2 Restricted Funding Has Increased 13 2.3 Restricted Funding Varies Greatly among the Top Ten Donors 17 3.1 Private Agricultural Research Expenditures Are Small in Developing Countries, Large in Developed Countries 34 6.1 Bank Lending to Agriculture Declined Dramatically in the 1990s 44 8.1 An Overview of CGIAR Centers 46 8.2 The World Bank Became the Largest Donor in 1994 46 8.3 Top Ten Donors Provide Most of the Funding 48 8.4 Allocations to Commodity and Ecoregional Centers Have Declined 49 8.5 Research with Proven Impacts on Poverty Has Declined Dramatically 49 8.6 Centers' Expenditures on Africa Have Increased Slightly During the Last Ten Years 50 8.7 Global Public Agriculture Research Expenditures, 1995: $21.7 Billion 50 8.8 Global Public Health Expenditures, 1998: $73.5 Billion 51 8.9 Average Annual Growth Rates in Public Agricultural Research Expenditures, by Region and Half Decades, 1976 to 1996 52 8.10 World Bank and USAID Funding of Agricultural Research by Region 53 8.11 Rural and Health Sectors Are Prominent in DGF Allocations to Global Programs (FY02 figures) 53 8.12 Environment, Rural, and Health Sectors Are Prominent in Overall Global Program Expenditures (FY01 figures) v T H E C G I A R AT 3 1 115 14.1 Public Agricultural Research Expenditures, 1976­95 144 17.1 World Bank Lending to Agricultural Research and Extension, 1971­02 Boxes 4 1.1 Portrait of a Global Program 6 1.2 Concepts Relating to Setting and Achieving Priorities 8 2.1 What Is Global and Regional Public Goods Research? 16 3.1 The New World of Intellectual Property Rights 19 3.2 "Golden Rice" Demonstrates the Challenges of Public-Private Partnerships 24 4.1 The CGIAR's Founding Principles Need Revisiting to Maintain Impact on Poverty Reduction 25 4.2 Stakeholders Report Many Concerns about the System 36 6.1 Conflicts of Interest in the Bank's Roles Inhibit Reforms 54 8.1 The Research and Development Continuum 55 8.2 Global versus International Public Goods 56 8.3 Economies of Scale and Scope in Research 58 8.4 The Importance of the Priority Setting Process: The Case of Climate Change 82 10.1 Current CGIAR Biotechnology Research, by Center and Activity 125 15.1 The Unevaluated ILCA/ILRAD Merger Experience 126 15.2 Veterans Cite Multiple Reasons for Consolidation 131 16.1 The Science Council Has Less Power Than Did TAC 132 16.2 Issues in the Transformation of TAC to the Science Council Note: Part 3 of this study, a compilation of extensive supporting material, may be viewed at: www.worldbank.org/oed/cgiar v i Acknowledgments T he work for this review was led by Uma evaluation. Francisco Reifschneider, Executive Lele, Senior Adviser and Task Team Leader Director of the CGIAR and his CGIAR Secretariat for OED's Review of the World Bank's Ap- team (particularly Ernest Corea, Josephine Her- proach to Global Programs; co-authors are Christo- nandez, Sarwat Husain, Manuel Lantin, Gunnar pher Barrett, Carl K. Eicher, Bruce Gardner, Larson, Danielle Lucca, Selcuk Ozgediz, Ravindra Christopher Gerrard, Lauren Kelly, William Lesser, Tadvalkar, Shey Tata, Feroza Vatcha, and Jason Karin Perkins, Saeed Rana, Mandivamba Rukuni, Yauney) deserve special mention for their assis- and David J. Spielman, with valuable contribu- tance, as do Emil Javier, Chairman of the interim tions from Antonio F. D. Avila, Elisio Contini, J. C. Science Council, and his team (especially Amir Katyal, Jamil Macedo, Mruthyunjaya, Cyrus Ndiritu, Kassam, Shellemiah Keya, Timothy Kelly, and Marcio C. M. Porto, and Luis Romano. William Sirkka Immonen), and Hank Fitzhugh and Meryl Hurlbut provided editorial and report production Williams of the Center Directors' Committee. support. Maisha Hyman, Kristina Kavaliunas, Kavita OED also thanks Dana Dalrymple, Robert Mathur, Paramjit Sachdeva, and Christopher Wool- Havener, Robert Herdt, Shantanu Mathur, Alex Mc- ford provided invaluable resource support. Samuel Calla, Rajendra S. Paroda, and Moctar Touré, who Paul served as a consultant on management issues. (along with many other individuals listed in Ap- Caroline McEuen edited the conference volume pendix 6) shared their knowledge of the CGIAR's and the final report. history, served informally as resource persons, The authors and contributors are grateful for and responded to the team's frequent queries the perspectives and inputs from those con- and questions. OED appreciates the comments re- sulted during the preparation of this review (see ceived on earlier drafts from Jock Anderson, Appendix 6), including during the CGIAR Annual Shawki Barghouti, Derek Byerlee, Kevin Cleaver, General Meetings 2001 and 2002 and/or con- Cees de Haan, Dale Hill, Odin Knudsen, Motoo tacted subsequently; from those who provided Kusakabe, Hock-Chye Ong, Eija Pehu, Randall comments on the Working Papers (see Appen- Purcell, John Todd, and Christine Wallich, which dix 5); and from those who responded to the helped to improve these documents. Detailed questionnaire. OED owes special thanks to Ian comments from members of the advisory com- Johnson, the CGIAR Chairman, for his availabil- mittee--Yujiro Hayami, Michael Lipton and Har- ity for consultations, his thoughtful comments ris Mule--helped to clarify questions and on the OED drafts, and his unstinting support concepts. Comments from the Directors-General and respect for the independence of the meta- and the CGIAR membership at the 2002 Annual v i i T H E C G I A R AT 3 1 General Meeting, and a formal response from standards, and unstinting support for this evalu- Adel El-Beltagy, chair of the Center Directors' ation and the global review of which this is a part. Committee, provided their perspectives on key is- The willingness of so many to take time from sues in this report. Last but not the least, the their busy schedules to share views with the team was privileged to work under the leadership meta-evaluation team testifies to their deep com- of Robert Picciotto, Director-General of Operations mitment to the ideals the CGIAR represents. Evaluation until the end of September 2002, who The volume was published in OED's Part- provided a demanding environment, exacting nerships and Knowledge Group. Director-General, Operations Evaluation: Gregory K. Ingram Director, Operations Evaluation Department: Ajay Chhibber Task Manager: Uma Lele v i i i F O R E W O R D Foreword Prólogo Avant-propos ~NOL A ENGLISH ESP FRANÇAIS This meta-evaluation of the Esta metaevaluación del GCIAI La présente méta évaluation du CGIAR is part of OED's review of the integra el examen del DEO sobre la GCRAI a été entreprise par l'OED pour World Bank's involvement in global participación del Banco Mundial en examiner la participation de la Banque programs and the first evaluation of los programas globales y constituye mondiale à divers programmes mon- the CGIAR to be discussed by the la primera evaluación del GCIAI que diaux. Elle représente aussi la première Bank's Committee on Development Ef- será analizada por CODE. évaluation du GCRAI par le CODE. fectiveness (CODE). El GCIAI, creado en 1971, fue el pri- Fondé en 1971, le GCRAI fut le pre- Founded in 1971, the CGIAR was the mer programa global de bienes públi- mier programme mondial de biens pu- first global public goods program to cos que recibió donaciones de los blics à recevoir des dons provenant du receive grants from the Bank's income. ingresos netos del Banco. En la ac- bénéfice net de la Banque mondiale. Today the CGIAR consists of 16 au- tualidad, el GCIAI está compuesto por Le GCRAI se compose aujourd'hui de 16 tonomous research centers, with a 16 centros de investigación autóno- centres de recherche autonomes, il est Secretariat in the World Bank and an mos con una Secretaría en el Banco doté d'un Secrétariat à la Banque mon- interim Science Council (until 2001 the Mundial y un Consejo de Ciencias pro- diale et d'un Conseil scientifique inté- Technical Advisory Committee, or TAC) visional (hasta 2001, denominado Co- rimaire (jusqu'en 2001, Comité housed in the Food and Agricultural mité de Asesoramiento Técnico o CAT) d'assistance technique ou CAT), sis à Organization of the United Nations. ubicado en la Organización de las Na- l'Organisation des Nations Unies pour The CGIAR is the largest global pro- ciones Unidas para la Agricultura y la l'alimentation et l'agriculture (FAO). Le gram supported by the Bank--receiv- Alimentación. El GCIAI es el programa GCRAI est le plus vaste programme ing contributions of $930 million of global más grande que recibe la ayuda mondial soutenu par la Banque--il a completely unrestricted funds since del Banco. Desde que fuera creado, el reçu depuis sa création des contribu- its inception. The CGIAR is exempt GCIAI ha recibido fondos totalmente tions à hauteur de 930 millions de dol- from normal Development Grant Fa- irrestrictos del Banco que alcanzan la lars de fonds entièrement non cility (DGF) requirements that it have an suma de USD 930 millones. El GCIAI conditionnels. Le GCRAI est exempté exit strategy from DGF funding and an está exento de reunir los requisitos des obligations normales du Méca- arm's-length relationship with the habituales para el Fondo de Donacio- nisme d'octroi de dons pour le déve- Bank. Nevertheless, it currently faces nes para el Desarrollo (FDD) en cuanto loppement (MDD), à savoir de disposer increasing competition for its 40 per- a tener una estrategia de salida y una d'une stratégie de sortie des finance- cent share of DGF grants that go to relación independiente con el Banco. ments MDD et d'avoir une relation in- global programs. Sin embargo, actualmente enfrenta dépendante avec la Banque. Il est The meta-evaluation concludes una competencia cada vez mayor por toutefois confronté actuellement à une that the CGIAR's productivity- su participación del 40 por ciento en concurrence accrue pour sa part de 40 enhancing research has had sizable las donaciones del FDD asignadas a % des dons MDD destinés aux pro- impacts on reducing poverty. More- los programas globales. grammes mondiaux. over, further improvements in agri- La metaevaluación concluye que La conclusion de cette méta éva- cultural productivity are critical to las investigaciones destinadas al me- luation est que la recherche effectuée meet the Millennium Development joramiento de la productividad del par le GCRAI pour améliorer la pro- Goal of halving poverty by 2015. But GCIAI, han tenido impactos mensu- ductivité a contribué de façon sensible the CGIAR is less focused on en- rables en la reducción de la pobreza. à réduire la pauvreté. Il est en outre es- hancing agricultural productivity Además, es necesario introducir nue- sentiel de parfaire encore la producti- than it used to be. Its current mix of vas mejoras en la productividad agrí- vité agricole pour pouvoir atteindre i x T H E C G I A R AT 3 1 activities reflects neither its cola para cumplir con las l'Objectif de développement comparative advantage nor Metas de Desarrollo del Mile- pour le millénaire (ODM) qui its core competence. The ~NOL nio, de reducir la pobreza a la est de réduire de moitié la pau- CGIAR's expenditures on A mitad para el año 2015. Sin vretéd'icià2015.MaisleGCRAI productivity-enhancing embargo, el GCIAI está menos est moins axé sur l'améliora- ENGLISH agricultural research--a ESP centrado que antes en mejo- tion de la productivité agricole FRANÇAIS global or regional public rar la productividad agrícola. qu'il ne l'était. Sa combinaison good ideally suited to a pub- La mezcla de actividades que actuelle d'activités ne reflète ni licly funded global network--de- realiza no refleja ni la existencia de su son avantage comparatif ni sa compé- clined by 6.5 percent annually in real ventaja comparativa ni de su compe- tence fondamentale. Les dépenses du terms between 1992 and 2001, and tencia básica. Entre 1992 y 2001, los GCRAI en matière de recherche agri- expenditures on improving policies gastos del GCIAI en investigación cole visant à améliorer la producti- and on protecting the environment para mejorar la productividad agrí- vité-- bien public mondial ou régional increased by 3.1 percent annually in cola --un bien público global o re- parfaitement adapté à un réseau mon- real terms. Overall, CGIAR funding gional idealmente adaptado a una red dial financé par des fonds publics-- declined by 1.8 percent annually global que recibe financiamiento pú- ont diminué de 6,5 % par an en termes during this same time. Meanwhile, blico-- disminuyeron 6,5 por ciento réels entre 1992 et 2001, et les dé- the share of restricted funding in- anual en términos reales, y los gastos penses en matière d'amélioration des creased from 36 percent to 57 per- en el mejoramiento de políticas y la politiques économiques et de protec- cent, with the degree of restriction protección del medio ambiente, se tiondel'environnementontaugmenté accelerating since 1998. acrecentaron en un 3,1 por ciento de 3,1 % par an en termes réels. Les fi- Several factors explain the chang- anual en términos reales. En total, el nancements totaux du GCRAI ont di- ing research mix and the increasing financiamiento del GCIAI disminuyó minuéannuellementde1,8%aucours restrictions, including: (1) the un- en un 1,8 por ciento anual durante el delamêmepériode.Pendantcetemps, popularity of germplasm improve- mismo período. Mientras tanto, el fi- la part des financements conditionnels ment research in the constituencies nanciamiento sujeto a restricciones se estpasséede36%à57%,leniveaud'af- of some key donors due to negative incrementó, del 36 por ciento al 57 fectation allant en s'accélérant depuis perceptions of the Green Revolu- por ciento, y el grado de restricción 1998. tion initially, and of biotechnology se está acelerando desde 1998. Plusieurs facteurs expliquent more recently; (2) the CGIAR's jus- Diversos factores explican los cam- l'évolution dans la recherche et tified response to the second-gen- bios producidos en la combinación de l'augmentation des affectations, eration environmental pressures on investigaciones y las mayores restric- dont : (1) le fait que la recherche en soils and water created by the radi- ciones, entre ellos: (1) la falta de po- matière d'amélioration du matériel cal changes in farming systems dur- pularidad que registra la investigación génétique est relativement mal per- ing the Green Revolution; (3) the sobre el mejoramiento de germo- çue par la base politique de certains rise of environmentalism and grow- plasma entre los integrantes de algu- bailleurs de fonds clés, au départ à ing environmental advocacy in donor nos donantes clave causada por las cause de leur perception défavo- countries; and (4) the weakening of percepciones negativas de la Revolu- rable de la Révolution verte puis many national agricultural research ción Verde inicialmente y de la bio- plus récemment de la biotechnolo- systems (NARS). The last has led tecnología, más recientemente; (2) gie, (2) la réponse justifiée du GCRAI donors to turn to the CGIAR Centers la respuesta justificada del GCIAI a aux pressions écologiques de se- to fill the downstream national and las presiones ambientales de la se- conde génération exercées sur les local public goods gaps closer to the gunda generación sobre suelos y agua sols et l'eau du fait des modifica- farmer, which should ideally be filled creadas por los cambios radicales en tions radicales apportées aux sys- by national systems. Collaterally, the los sistemas de cultivo durante la Re- tèmes d'exploitation au cours de la maintenance and management of volución Verde, (3) el incremento del Révolution verte, (3) la popularité the CGIAR's gene banks of 600,000 ambientalismo y la creciente defensa croissante du mouvement écolo- accessions, a unique global public del medio ambiente en los países do- gique et de défense de l'environne- x F O R E W O R D good, has suffered from lack nantes, y (4) el debilitamiento ment dans les pays bailleurs of support. de muchos sistemas naciona- de fonds, et (4) l'affaiblisse- Two changes in the fund- ~NOL les de investigación agrícola ment de nombreux systèmes ing processes of the CGIAR A (SNIA.) Esto último ha hecho nationaux de recherche agri- since the mid-1990s have also que los donantes se vuelquen cole (SNRA). Ce dernier ENGLISH increased donor influence on ESP hacia los Centros del GCIAI point a amené les bailleurs FRANÇAIS the research of the CGIAR: para reducir las brechas entre de fonds à se tourner vers (1) the Bank's shift in the al- bienes públicos nacionales y les Centres GCRAI pour com- location of its own financial contri- locales, acercándose al agricultor, a bler l'absence de biens publics na- bution from a "donor of last resort" pesar de que lo ideal sería que fueran tionaux et locaux en aval, plus près model to a matching-grant model los sistemas nacionales lo que se ocu- de l'exploitant agricole, tâche qui in response to a funding crisis in paran de esta función. Por otro lado, devrait, dans l'idéal, revenir aux sys- 1993­94, and (2) the redefinition of el mantenimiento y manejo de los de tèmes nationaux. Indirectement, la the CGIAR's "agreed research los bancos de genes del GCIAI, que maintenance et la gestion des agenda" to include both the former cuentan con más de 600.000 acce- banques génétiques de 600 000 ac- "core" agenda (largely the high-re- siones, lo cual constituye un bien pú- quisitions du GCRAI, qui sont un turn global and regional public blico global único en su tipo, sufrió la bien public mondial unique, ont goods research) and the "non-core" falta de apoyo. souffert de ce manque de soutien. agenda (donor-funded, mostly Desde mediados de la década de Deux modifications dans le pro- downstream activities given less pri- 1990, los procesos de financiamiento cessus de financement du GCRAI de- ority by TAC and the interim Sci- del GCIAI sufrieron dos grandes cam- puis le milieu des années 90 ont ence Council). The cumulative result bios, que contribuyeron a aumentar la également accru l'influence des of these changes has been the re- influencia de los donantes en la inves- bailleurs de fonds sur la recherche du duced influence of independent sci- tigación del GCIAI: (1) un cambio de GCRAI : (1) le changement de cap de entific advice provided by TAC, a actitud del Banco en la asignación de laBanquedansl'allocationdesapropre concurrent transformation of the su propio aporte financiero, que pasó contribution financière, passant de « CGIAR's authorizing environment de ser un modelo de "prestador de úl- bailleur de fonds de dernier recours » from being science-driven to being tima instancia" a otro modelo de do- àunmodèlededondecontrepartieen donor-driven, and a shift of the Sys- nación de contrapartida en respuesta réponse à la crise de financement de tem from producing global and a la crisis de financiamiento producida 1993-1994, et (2) la redéfinition du regional public goods toward pro- en el período 1993-94, y (2) la redefi- « programme convenu de recherche » viding national and local services. nición de la "agenda de investigación" du GCRAI pour y inclure à la fois le The CGIAR experience demon- acordada del GCIAI, que debía incluir programme « de base » précédent strates that the sum of the interests tanto la agenda "principal" anterior (dans une grande mesure la recherche of individual stakeholders in a global (compuesta en gran medida, por la in- de bien public mondial et régional à organization need not define a global vestigación de bienes públicos globa- rendement élevé) et le programme public good. les y regionales de alto rendimiento) y « secondaire » (financé par les bailleurs At the same time, the growing la agenda "secundaria" (compuesta de fonds, principalement des activités importance of genetic resource man- por actividades financiadas funda- en aval auxquelles le CAT et le CSI ac- agement, the biotechnology revolu- mentalmente por actividades secun- cordent une moindre priorité). Ces tion, intellectual property rights darias a las que el CAT y el Comité changements ont eu pour résultat cu- (IPRs), and private sector research Científico Provisional asignaron una mulé l'influence réduite des conseils call for System-level responses, prioridad menor.) El resultado final de scientifiques indépendants fournis par strategies, and policies to deal with estoscambiosseplasmóenunamenor le CAT, une transformation simultanée these Systemwide challenges. Un- influencia del asesoramiento científico du contexte d'autorisation du GCRAI, like more recent global programs, independiente provisto por el CAT, en non plus dicté par la science mais par the CGIAR System has no formal or la transformación simultánea del ám- les exigences du bailleur de fonds, et legal persona, written charter, or bito de autorización del GCIAI, que une réorientation du Système passant x i T H E C G I A R AT 3 1 even a memorandum of un- pasó de estar impulsado por la de la production de biens pu- derstanding. Collective action ciencia a estar impulsado por blic mondiaux et régionaux à la problems created by the di- ~ NOL los donantes, y en un cambio prestation de services natio- verse interests of its con- A del Sistema que, de producir naux et locaux. L'expérience stituents have forestalled bienes públicos globales y re- du GCRAI démontre que la ENGLISH fundamental organizational ESP gionales, pasó a prestar servi- somme des intérêts des par- FRANÇAIS reforms. At a minimum, the cios nacionales y locales. La ties prenantes individuelles CGIAR should adopt a written experiencia del GCIAI de- dans une organisation mon- charter that delineates the roles, re- muestra que la suma de los intereses diale ne définit pas nécessairement le sponsibilities, and accountabilities de los participantes individuales de bien public mondial. of the officers and bodies that gov- una organización global no define ne- En même temps, l'importance ern the System, as well as a mecha- cesariamente un bien público global. croissante de gérer les ressources nism to reform the System's cum- Al mismo tiempo, la creciente im- génétiques, la révolution biotech- bersome governance. It should also portancia del manejo de recursos ge- nologique, les droits de propriété analyze the merits of establishing all néticos, la revolución de la intellectuelle (DPI) et la recherche ef- or part of the CGIAR as a separate biotecnología, los derechos de pro- fectuée par le secteur privé nécessi- legal entity attuned to deal with piedad intelectual (DPI) y la investiga- tent des réactions, des stratégies et today's realities in public-private part- ción del sector privado, exigen des politiques à l'échelle du Système nerships. respuestas, estrategias y políticas den- pour relever les défis posés au Sys- The World Bank plays multiple tro del Sistema para poder abordar los tème dans son ensemble. À la diffé- roles in the CGIAR--as convener desafíos que presenta todo el Sistema. rence de programmes mondiaux and donor to the System, and as a A diferencia de los programas globales plus récents, le Système GCRAI ne lender to developing countries for más recientes, el Sistema del GCIAI ca- dispose pas de personne morale ou complementary activities. As a re- rece de estructura formal o de perso- juridique, d'acte constitutif écrit ni sult, the Bank has been the guardian nería jurídica, carta orgánica o de un même de protocole d'accord. Les of the CGIAR and the catalyst that mero memorando de entendimiento. problèmes d'action collective créés makes the System coherent, and Como mínimo, el GCIAI debería adop- par les intérêts divers de la base po- larger than the sum of its parts. tar una carta orgánica que detalle cuá- litique ont devancé les réformes or- Other donors view the Bank's lead- lessonlasfunciones,responsabilidades ganisationnelles fondamentales. Le ership role, its financial contribu- y rendiciones de cuenta, a cargo de GCRAI devrait au minimum adopter tions, and its operational support los funcionarios y organismos que go- un acte constitutif écrit qui établisse as a seal of approval, giving them biernan el Sistema, así como también le rôle, les responsabilités et les obli- confidence to continue to invest in un mecanismo para reformar la engo- gations des dirigeants et des organes the System. But conflicts of interest rrosa estructura de gobierno del Sis- qui régissent le Système, ainsi qu'un among the roles of the ESSD vice tema. También debería analizar los mécanisme d'allègement de la lourde president (who is CGIAR chairman), méritosdeestablecerelGCIAI,entodo bureaucratie de ce dernier. Il devrait the CGIAR director, and other ESSD o en parte, como una entidad jurídica également analyser les mérites d'éri- staff involved with the system, and distinta para poder abordar las reali- ger la totalité ou une partie du GCRAI inadequate Bank oversight (from dades que enfrentan actualmente los en entité légale distincte, habilitée à outside the ESSD vice presidency) partenariados público-privados. gérer les réalités d'aujourd'hui en have compromised the Bank's ca- El Banco Mundial cumple múltiples matière de partenariats entre le sec- pacity to exercise strategic leader- funciones en el GCIAI--como convo- teur public et le secteur privé. ship of the CGIAR and to press for cante y donante al Sistema La Banque mondiale assure plu- reforms at the scale or speed that y como prestador a los países en de- sieurs fonctions à l'intérieur du might be warranted. The Bank sarrollo para las actividades comple- GCRAI-- celle d'organisateur et de do- should address its corporate gover- mentarias. En consecuencia, el Banco nateurduSystèmeetcelled'organisme nance responsibilities in the man- ha sido el guardián del GCIAI y el ca- prêteur pour les activités complé- agement of the CGIAR, separate talizador que hace que el Sistema fun- mentairesdespaysendéveloppement. x i i F O R E W O R D oversight from management, cione en forma coherente y re- C'estainsiquelaBanqueestde- and exercise a degree of sulte en definitiva de una di- venue le tuteur du GCRAI et le oversight consistent with its ~NOL mensión mayor que la que catalyseur qui rend le Système major roles in the CGIAR. A arroja la suma de sus partes. cohérent et plus grand que la Further systemic reforms Otrosdonantespercibenlafun- somme de ses parties. Les ENGLISH are also needed. The strate- ESP ción de liderazgo del Banco, autres bailleurs de fonds voient FRANÇAIS gic priorities of the CGIAR sus contribuciones financieras le rôle de leader de la Banque, should respond more actively y su ayuda operativa como un ses contributions financières et to changes in the global research sellodeaprobación,einspiraconfianza son soutien opérationnel comme un context, giving more prominence to a los inversores para seguir invirtiendo label de qualité qui les rassure pour basic plant breeding and germplasm en el Sistema. Pero los conflictos de in- continuer à investir dans le Système. improvement and reshaping natural terés entre las funciones del Vicepre- Mais des conflits d'intérêts entre les resource management research to sidente de la RDSAyS (que a su vez, es rôles respectifs du vice-président de focus tightly on productivity en- elPresidentedelGCIAI),elDirectordel l'ESSD (qui est le président du Conseil hancement and sustainable use of GCIAI, y otros funcionarios de la RD- du GCRAI), du directeur du GCRAI et natural resources for the benefit of SAyS que participan en el sistema, y la d'autres membres du personnel de developing countries. The Bank inadecuadasupervisióndelBanco(por l'ESSD qui participent au Système, et should lead a concerted effort at the fuera de la vicepresidencia de la RD- une supervision inadéquate de la highest level, much as when the SAyS) han comprometido la capacidad Banque (depuis l'extérieur de la vice- CGIAR was established, to achieve del Banco de ejercer el liderazgo es- présidence de l'ESSD) ont affaibli la fundamental reforms in the organi- tratégico del GCIAI y presionar para capacitédelaBanqueàexerceruneau- zation, governance, finance, and crear reformas a una escala o velocidad torité stratégique sur le GCRAI et à in- management of the CGIAR--partic- que podría ser garantizada. El Banco sister en faveur de réformes qui ularly to encourage donors to re- debe atender sus responsabilidades auraient l'envergure et la rapidité d'ac- verse the trend in restricted funding. empresarialesdegobiernoenlagestión tion nécessaires. La Banque devrait The Bank should also abandon the del GCIAI, separar la supervisión de lo examiner les responsabilités qui lui in- current matching grant model and que es la dirección, y ejercer un grado combent en matière de gouvernance ensure that its resources are allo- de supervisión coherente con las prin- dans la gestion du GCRAI, dissocier cated strategically in support of cipales funciones que desarrolla en el les fonctions de supervision et de ges- global and regional public goods GCIAI. tionetexercerundegrédesupervision that contribute to agricultural pro- También se necesitan otras refor- proportionnel aux principaux rôles ductivity and poverty reduction, mas sistémicas. Las prioridades estra- qu'elle joue au sein du GCRAI. based on long-term priorities, by tégicas del GCIAI deberían responder De plus amples réformes systé- making sure that a strong, qualified, más activamente a los cambios en el miques sont également nécessaires. and independent Science Council contexto de la investigación global, Le GCRAI devrait reformuler ses stra- is established and invested with ap- dar más importancia al cultivo básico tégies prioritaires pour répondre plus propriate responsibilities for trans- de plantas y al mejoramiento de ger- énergiquement aux nouvelles exi- parent advice on System-wide moplasma y reestructurar la investi- gences de la recherche mondiale, priorities, strategies, and resource gación del manejo de recursos ceci en favorisant la recherche de allocation. naturales para hacer hincapié especí- base en matière de sélection végétale If additional reforms are achieved ficamente en el mejoramiento de la et d'amélioration du matériel géné- to address the radically changed ex- productividad y el uso sostenido de los tique, et en redéfinissant cette re- ternal and internal environment fac- recursos naturales para beneficiar a cherche de manière à gérer les ing the CGIAR, there is a strong los países en desarrollo. El Banco de- ressources naturelles pour accroître argument for increased funding for bería conducir un esfuerzo concer- la productivité et permettre une uti- the System, including exploring the tado en los niveles más altos, similar a lisation durable des ressources na- use of grant funds for the provision los realizados cuando se creó el GCIAI, turelles pour le bien des pays en of regional public goods that reduce para lograr reformas fundamentales développement. La Banque devrait x i i i T H E C G I A R AT 3 1 poverty. As a lender to devel- en la organización, gobierno, fi- mener un effort concerté au oping countries, the Bank nanzas y gestión del GCIAI-- plus haut niveau, comme lors also needs to increase lending ~NOL en especial para alentar a los de la création du GCRAI, afin to agricultural research, edu- A donantes a revertir la tenden- de procéder à des réformes cation, extension, and train- cia del financiamiento no su- radicales de la structure or- ENGLISH ing, especially in Sub-Saharan ESP jeto a restricciones. El Banco ganisationnelle, des finances FRANÇAIS Africa, in order to enhance también debería abandonar el et de la gestion du GCRAI -- the performance of NARS. modelo actual de donaciones de façon, en particulier, à en- de contrapartida y garantizar la asig- courager les donateurs à enrayer la nación estratégica de sus recursos en progression du financement condi- apoyo a los bienes públicos globales y tionnel. La Banque devrait également regionales y contribuir a la producti- abandonner le modèle actuel des vidad agrícola y a la reducción de la po- dons de contrepartie et à veiller à ce breza, tomando como base las que les ressources de la Banque prioridades de largo plazo, garanti- soient allouées stratégiquement de zando la creación de un Consejo de façon à favoriser les biens publics Ciencias sólido, calificado e indepen- mondiaux et régionaux qui contri- diente en quien se deleguen las res- buent à la productivité agricole et à ponsabilidades correspondientes que la réduction de la pauvreté, en fonc- permitan brindar, en forma transpa- tion de priorités à long terme. Elle de- rente, asesoramiento sobre priorida- vrait aussi assurer la création d'un des, estrategias y asignación de Conseil scientifique qualifié, fort et in- recursos para todo el Sistema. dépendant chargé d'émettre des avis Si se lograran adoptar reformas transparents sur les priorités, les stra- que permitan atender los cambios tégies et les allocations de ressources radicales que debe enfrentar el GCIAI à l'échelle du Système tout entier. en el ámbito interno y externo, exis- Si de nouvelles réformes sont tiría un argumento sólido que justi- créées pour permettre au GCRAI de ficara la necesidad de aumentar los faire face à des conditions internes et fondos para el Sistema. Esto incluye externes radicalement différentes, un la exploración del uso de los fondos financement accru du Système serait destinados a donación para la provi- alors beaucoup plus justifié et pourrait sión de bienes públicos regionales inclure l'étude de l'allocation de fonds que reduzcan la pobreza. En su cali- de subventions pour des actions ré- dad de prestador de los países en gionales de bien public visant à ré- desarrollo, el Banco también necesita duirelapauvreté.Entantqu'organisme incrementar el servicio financiero prêteur aux pays en développement, destinado a la investigación, educa- la Banque, afin d'améliorer les per- ción, divulgación y capacitación, es- formances des SNRA, doit également pecialmente en la zona del África accroître ses prêts pour la recherche Subsahariana, para poder mejorar el agricole, l'éducation, la vulgarisation et rendimiento de los sistemas nacio- la formation, en particulier en Afrique nales de investigación agrícola (SNIA.) subsaharienne. Gregory K. Ingram Director-General, Operations Evaluation x i v P R E FA C E Preface Prefacio Préface ~NOL A ENGLISH ESP FRANÇAIS This report on the Consultative El presente informe sobre el Le présent rapport sur le Group on International Agricultural Re- Grupo Consultivo sobre Investigacio- Groupe consultatif pour la recherche search (CGIAR) is part of an inde- nes Agrícolas Internacionales (GCIAI) agricole internationale (GCRAI) fait pendent review by the Operations es parte del análisis que realizó el De- partie de l'examen indépendant ef- Evaluation Department (OED) of the partamento de Evaluación de Opera- fectué par le Département d'évalua- World Bank's involvement in global ciones (DEO) sobre la participación tion des opérations (OED) de la programs. The first phase--The World del Banco Mundial en los programas participation de la Banque mondiale à Bank's Approach to Global Programs: globales. La primera fase ya está pu- des programmes mondiaux. La pre- An Independent Evaluation, Phase 1 blicada: The World Bank's Approach to mière phase a été publiée ­ « The Report (OED, 2002b)--has been pub- Global Programs: An Independent Eva- World Bank's Approach to Global Pro- lished. The second phase, due in FY04, luation, Phase 1 Report (El Banco Mun- grams: An Independent Evaluation, involves case studies of 27 programs, dial y su abordaje de los programas Phase 1 Report » (L'approche de la of which the CGIAR is one. globales: Una evaluación indepen- Banque mondiale pour les programmes diente, Primera fase del Informe) (OED, mondiaux : évaluation indépendante, Why evaluate the CGIAR? The inclu- Washington D.C., 2002.) La segunda Rapport phase 1, OED, Washington D.C. sion of the CGIAR evaluation in the fase del informe que estará preparada 2002). La seconde phase, qui sera pu- OED review of the Bank's global para el ejercicio fiscal 2004, incluye bliée au cours de l'exercice 2004, com- programs was requested by the De- estudios de casos de 27 programas, porte des études de cas sur 27 velopment Grant Facility (DGF) and uno de los cuales es el GCIAI. programmes, dont le GCRAI. Bank management in June 2001, and endorsed by OED's global program ¿Por qué evaluar el GCIAI? En junio de Pourquoi évaluer le GCRAI ? L'inclu- advisory committee. The CGIAR was 2001, el Fondo de Donaciones para sion de l'évaluation du GCRAI dans the first program providing global el Desarrollo (FDD) y la Dirección l'examen de l'OED des programmes public goods to receive grants from del Banco solicitaron que se incluy- mondiaux de la Banque a été de- the Bank's net income. It currently era la evaluación del GCIAI en el ex- mandée par le Mécanisme d'octroi faces increasing competition for the amen que realizaría el DEO de los de dons pour le développement 40 percent share that it receives of programas globales del Banco, so- (MDD) et par la direction de la DGF funds for global programs. The licitud que fue avalada por el comité Banque en juin 2001, puis endos- CGIAR has a special status because asesor de programas globales del sée par le Comité consultatif sur les it is exempt from the normal DGF re- DEO. El GCIAI fue el primer pro- programmes mondiaux de l'OED. quirements that it have an exit strat- grama que posibilitó que los bienes Le GCRAI est le premier programme egy from DGF funding and an públicos globales reciban subven- mondial de bien public à recevoir arm's-length relationship with the ciones de los ingresos netos del des dons provenant du bénéfice net Bank. Over the years, the DGF has Banco. Actualmente enfrenta una de la Banque mondiale. Il fait face au- also expressed concerns that the competencia cada vez mayor por la jourd'hui à une concurrence accrue CGIAR is not mobilizing sufficient participación del 40 por ciento que pour la part de 40 % qu'il reçoit des finance from other sources, not recibe de los fondos del FDD desti- dons MDD destinés aux pro- partnering actively with the private nados a los programas globales. La grammes mondiaux. Le GCRAI dis- sector and other key actors in the situación del GCIAI es especial ya pose d'un statut spécial car il est global research system, nor con- que está exento de los requisitos ha- exempté des obligations normales taining the costs of its Secretariat, bituales que exige el FDD: que tenga du MDD, à savoir de disposer d'une x v T H E C G I A R AT 3 1 sharing those costs with una estrategia para salir del fi- stratégie de sortie des finan- other donors, and establish- nanciamiento del FDD y una cements MDD et d'avoir une ing adequate linkages to ~NOL relación independiente con relation indépendante avec World Bank country opera- A el Banco. A través de los años la Banque. Cependant, le tions. el FDD también hizo cono- MDD s'inquiète depuis plu- ENGLISH ESP cer sus inquietudes acerca de sieurs années du fait que le FRANÇAIS Why a meta-evaluation? The que el GCIAI no moviliza fi- GCRAI ne mobilise pas suffi- CGIAR has an impressive tra- nanciamiento suficiente de samment de sources de fi- dition of self-assessments unmatched otras fuentes, no se asocia en forma nancement alternatives, ne forme by other major research systems activa con el sector privado y otros pas de partenariats avec le secteur (even in industrial countries) actores clave en el sistema de inves- privé et autres acteurs clés du sys- through External Program and Man- tigación global, ni tampoco intenta tème mondial de la recherche, qu'il agement Reviews (EPMRs), inter- reducir los gastos de su Secretaría y n'endigue pas les coûts de son Se- Center reviews, and System-wide, compartirlos con otros donantes o crétariat ou partage ces derniers issue-specific reviews. These assess- establecer vínculos con las opera- avec d'autres bailleurs de fonds, et ments, often involving outside re- ciones del país del Banco Mundial. qu'enfin, il n'établisse pas de liai- viewers, have focused on the CGIAR's sons adéquates avec les opérations 16 international research Centers or ¿Por qué una metaevaluación? El GCIAI par pays de la Banque mondiale. on cross-cutting thematic issues. tiene una gran tradición de autoeva- They have been managed by the luaciones que no son comparables con Pourquoi une méta évaluation ? Le CGIAR's Technical Advisory Com- lasqueserealizandeotrossistemasim- GCRAI a une impressionnante tra- mittee (now the interim Science portantes de investigación (aún en los dition d'auto évaluations inégalée Council) and the CGIAR Secretariat. países industrializados) a través de los dans les autres grands systèmes de The CGIAR has also contributed to Exámenes Externos de Programas y recherche (même dans les pays in- a substantial literature focusing on Gestión (EEPG), los exámenes entre dustrialisés). Ces auto évaluations the impacts of its commodity Cen- Centros, y los exámenes realizados en sont effectuées sous forme d'Exa- ters' germplasm research. todo el Sistema, sobre determinados mens de la gestion du programme But evaluations have been few and temasespecíficos.Amenudoparticipan extérieur (EPMR), d'examens inter- far between at the level of the System examinadores externos en estas eva- centres ainsi que d'examens portant as whole. The Third System Review luaciones, que siempre se han con- sur des questions spécifiques à l'in- (TSR) was completed in 1998, 17 centrado en los 16 Centros de térieur du Système tout entier. Ces years after the previous one. System- investigacióninternacionalesdelGCIAI, évaluations, qui font souvent inter- level reviews have been managed by o en cuestiones temáticas interdisci- venir des analystes extérieurs, ont the CGIAR Secretariat and reported plinarias. El Comité de Asesoramiento ciblé à la fois les 16 Centres inter- to the CGIAR chairman. The TSR, Técnico del GCIAI (CAT, actualmente nationaux de recherche du GCRAI et carried out by a distinguished panel el Consejo de Ciencias Provisional) y la des questions thématiques trans- headed by Maurice Strong, drew from Secretaría del GCIAI tienen a su cargo sectorielles. Elles sont gérées par le internal experts intimately knowl- estasevaluaciones.ElGCIAItambiénha Comité d'assistance technique (CAT, edgeable about the CGIAR and from contribuido con importantes trabajos aujourd'hui appelé Conseil scienti- outsiders with fresh scientific and focalizados en los impactos de la in- fique intérimaire, CSI) du GCRAI et strategic perspectives. Yet it engen- vestigación de germoplasma realiza- par le Secrétariat du GCRAI. Le dered little ownership from the mem- dos por sus Centros de productos GCRAI a également contribué d'im- bership and had scant impact on the básicos. portants documents portant sur l'im- System. A few System-level reviews Las evaluaciones han sido esca- pact de la recherche faite dans ses have been undertaken by specific sas, especialmente las que toman el Centres de produits agricoles sur le donors, such as Anderson and Dal- Sistema en su totalidad, y muy se- matériel génétique. rymple (1999) for the World Bank. paradas entre sí. El Tercer Examen Mais trop peu d'évaluations ont Other donors, such as Denmark and del Sistema (TES) finalizó en 1998, 17 été faites à l'échelle du Système dans x v i P R E FA C E the International Fund for años después del anterior son ensemble. Le Troisième Agricultural Development examen realizado en el ám- examen du système (TSR) a (IFAD), have also undertaken ~NOL bito del sistema. Este tipo de été achevé en 1998, soit 17 reviews of their own involve- A exámenes ha estado a cargo ans après l'examen précé- ment in the CGIAR. In view of de la Secretaría del GCIAI, y dent. Les examens à l'échelle ENGLISH the CGIAR's history of limited ESP dependía del Presidente del du Système sont gérés par le FRANÇAIS System-level reviews, the for- GCIAI. El TES estuvo a cargo Secrétariat du GCRAI et les mer Director-General of OED, de un panel muy capacitado, résultats soumis au président Robert Picciotto, proposed an exter- encabezado por Maurice Strong, y du Conseil du GCRAI. Le TSR, ef- nal, independent "blue ribbon" com- contó con la colaboración de exper- fectué par un groupe de spécialistes mission appointed by the World Bank tos internos que conocían íntima- éminents sous la direction de Mau- president to evaluate the CGIAR. The mente el GCIAI y expertos externos rice Strong, a bénéficié à la fois des proposal was not taken up because que aportaron nuevas perspectivas compétences d'experts internes the CGIAR had shown little recep- científicas y estratégicas. Sin embargo, connaissant bien le GCRAI et d'élé- tivity to past System-level external tuvo escasa respuesta de los miem- ments extérieurs apportant de nou- reviews and had just initiated a bros y poco impacto en el Sistema. veaux points de vue scientifiques et Change Design and Management Unos pocos estudios en el ámbito stratégiques. Cet examen n'a ce- Process (CDMP). Therefore, given del Sistema fueron realizados por al- pendant que peu contribué à une the number of previous evaluations gunos donantes específicos, como meilleure prise en charge de la part and its own limited resources, OED por ejemplo el elaborado por An- des membres et n'a eu que peu d'im- determined that a meta-evaluation derson and Dalrymple (1999), para el pact sur le Système. Quelques exa- would most effectively assess CGIAR Banco Mundial. Hubo otros, como mens à l'échelle du Système ont été performance and inform OED's over- Dinamarca y el FIDA, que también entrepris par des bailleurs de fonds all review of the Bank's involvement iniciaron exámenes para evaluar su spécifiques, tels que Anderson et in global programs. propia participación en el GCIAI. En Dalrymple (1999) pour la Banque vista de los escasos antecedentes de mondiale. D'autres bailleurs de Evaluation objectives. This meta-eval- exámenes del GCIAI en el ámbito fonds, tels que le Danemark et le uation is based on a review of previ- del Sistema, el anterior Director Ge- FIDA, ont également entrepris d'exa- ous reviews and impact assessments. neral del DEO, Robert Picciotto, pro- miner leur propre participation au In brief, its objectives are to: puso la creación de una comisión sein du GCRAI. Vu le peu d'évalua- independiente externa altamente tions du GCRAI à l'échelle du Sys- · Evaluate the implementation of cualificada, que sería designada por tème, Robert Picciotto, ancien recommendations made in OED's el presidente del Banco Mundial directeur général de l'OED, avait 1998 internal review of the World para evaluar el GCIAI. La propuesta proposé, pour évaluer le GCRAI, que Bank grant programs relevant to no prosperó porque en el pasado, el le président de la Banque nomme the CGIAR, including an assess- GCIAI mostró escasa receptividad a une commission extérieure indé- ment of subsidiarity, arm's-length los exámenes externos del Sistema pendante composée d'experts « triés relationship with the Bank, and y recientemente, había iniciado un sur le volet » et nommée par le pré- exit strategy. Proceso de Diseño y Gestión del sident de la Banque. La proposition · Review the diagnosis, findings, Cambio (PDG.) Por este motivo, y n'avait pas été retenue car le GCRAI and recommendations made by considerando el número de evalua- s'était montré peu réceptif dans le relevant previous CGIAR evalua- ciones previas y sus limitados re- passé à des évaluations externes à tions relating to the CGIAR's struc- cursos, el DEO decidió que la l'échelle du Système et venait juste ture, governance, financing, and metaevaluación sería una forma efi- d'instaurer un Processus de concep- scientific strategy. caz de evaluar el desempeño del tion et de gestion du changement · Identify issues confronting the GCIAI y informaría al DEO sobre la (CDMP). Par conséquent, compte CGIAR from a forward-looking participación del Banco en los pro- tenu du nombre d'études anté- perspective. gramas globales. rieures et de ses propres ressources x v i i T H E C G I A R AT 3 1 ·Consider the recent recom- Objetivos de la evaluación. Esta limitées, l'OED décida mendations of the CGIAR metaevaluación se basa en qu'une méta évaluation se- Change Design and Manage- ~NOL una evaluación de exámenes rait le moyen le plus efficace ment Team from the stand- A y evaluaciones de impacto d'évaluer les performances point of previous evaluations, previas. En resumen, sus ob- du GCRAI et d'apporter à son ENGLISH including the TSR. ESP jetivos son: évaluation générale les in- FRANÇAIS ·Draw lessons for the Bank's formations sur la participa- overall strategy for global pub- · Evaluar la ejecución de las tion de la Banque à des lic policies and programs, and for recomendaciones formuladas en programmes mondiaux. the CGIAR, for developing and el Proceso de revisión de los pro- disseminating technologies for gramas de subvención del Banco Objectifs d'évaluation. Cette méta éva- agriculture and natural resource Mundial correspondiente al GCIAI, luation est fondée sur l'analyse des management to reduce poverty realizado en el año 1998 por el examens et des études d'impact an- and achieve sustainable devel- DEO, que incluye una evaluación térieurs. En bref, elle a pour objec- opment through productivity de la subsidiariedad, la relación tif de : growth. independiente con el Banco y la estrategia de salida. · Évaluer l'application des recom- The focus of the review is on the · Examinar el diagnóstico, hallaz- mandations pertinentes au GCRAI Bank and on the strategic role it has gos y recomendaciones que sur- faites par l'OED en 1998 lors du played, and might ideally play in the gen de las evaluaciones previas review of the World Bank Grant future, to ensure the CGIAR's de- del GCIAI en relación con la es- Program (Examen du processus velopment effectiveness. Yet it is dif- tructura, gobierno, financiamiento des programmes de dons de la ficult to evaluate the role of one y estrategia científica del GCIAI. Banque mondiale), dont une partner without assessing the per- · Identificar temas que deberá en- étude de subsidiarité, de relation formance and impact of the whole carar el GCIAI desde una pers- indépendante avec la Banque et CGIAR partnership, particularly pectiva de futuro. de stratégie de sortie ; when the partner being evaluated · Tomar en cuenta las recomenda- · Examiner le diagnostic, les plays a pivotal role. Moreover, it is ciones recientes formuladas por el conclusions et les recommanda- difficult to assess the Bank's role Equipo de Diseño y Gestión del tions faites lors des précédentes without assessing the strategic re- Cambio del GCIAI desde el punto évaluations pertinentes du sponses of other partners to the de vista de las evaluaciones pre- GCRAI, concernant sa structure, Bank's decisions and vice versa. Ac- vias, incluido el TES. sa gouvernance, son financement cordingly, the meta-evaluation se- · Tomar las enseñanzas de esta me- et sa stratégie scientifique ; cured the perspectives of key actors taevaluación y aplicarlas a la es- · Identifier les questions auxquelles who hold different views of the part- trategia de políticas y programas sera confronté le GCRAI dans nership and who have responded públicos globales del Banco, y al l'avenir ; differently to changes in the CGIAR. GCIAI, a fin de desarrollar y di- · Examiner les recommandations vulgar tecnologías para la agricul- récentes de la CDMT (Équipe de Evaluation scope and tools. The meta- tura y el manejo de recursos conception et de gestion evaluation is based on a compre- naturales, reducir la pobreza y al- du changement du GCRAI) du hensive desk review, five Working canzar un desarrollo sostenible a point de vue des évaluations pré- Papers prepared by independent través del crecimiento de la pro- cédentes, dont le TSR ; scholars and two by CGIAR clients ductividad. · En tirer des conclusions pour la (Brazil, India), as well as two Back- stratégie globale de la Banque en ground Papers by clients (Kenya, Este examen estará centrado en el matière de politiques et pro- Colombia) (see Bibliography). A Banco y en la función estratégica que grammes publics mondiaux et background note was also commis- ha desempeñado, y en cuál sería su pour le GCRAI ­ afin de déve- sioned on the Change Design and "función ideal" en el futuro, para ga- lopper et de diffuser des techno- x v i i i P R E FA C E Management Process. In ad- rantizar la eficacia del desa- logies pour l'agriculture et la dition, the meta-evaluation rrollo del GCIAI. Sin embargo, gestion des ressources natu- team conducted extensive ~NOL es difícil evaluar la función de relles visant à réduire la pau- consultations with CGIAR, A un asociado sin evaluar el de- vreté et à obtenir un two of the three co-sponsors sempeño y el impacto de développement durable ENGLISH (FAO and IFAD), donors ESP todo el partenariado del grâce à une meilleure pro- FRANÇAIS (USAID, SDC, DFID, Den- GCIAI, en especial cuando la ductivité. mark, Germany, and the Eu- función del asociado que se ropean Union), and private sector evalúa es tan determinante. Además, L'examen est centré sur la Banque representatives and developing es difícil evaluar la función del Banco et sur le rôle stratégique qu'elle a country nationals, and a formal sur- sin evaluar las respuestas estratégicas tenu et qu'elle pourrait idéalement vey of 235 CGIAR insiders and ob- de los otros asociados a las decisio- tenir à l'avenir afin d'assurer un dé- servers. (See Appendixes 4, 5, and 6.) nes tomadas por el Banco y viceversa. veloppement efficace du GCRAI. Il The meta-evaluation report is in Asimismo, la metaevaluación aseguró est néanmoins difficile d'évaluer le three parts. The Overview (Part 1) las perspectivas de actores clave que rôle de l'un des partenaires sans étu- addresses strategic questions re- tenían opiniones distintas del parte- dier les performances et l'impact du garding the organization, financing, nariado y habían respondido de ma- partenariat GCRAI tout entier, tout and management of the CGIAR as nera diferente a los cambios particulièrement lorsque le parte- these have affected research choices, introducidos en el GCIAI. naire évalué joue un rôle essentiel. science quality, and the Bank's rela- En outre, il est difficile d'évaluer le tionship to the CGIAR. The Techni- Alcance y herramientas de la evalua- rôle de la Banque sans évaluer les ré- cal Report (Part 2) explores the ción. La metaevaluación se basa en un ponses stratégiques des autres par- nature, scope, and quality of the Sys- examen integral basado en archivos tenaires aux décisions de la Banque tem's scientific work, assesses the anteriores, cinco Documentos de Tra- et vice versa. C'est pourquoi la méta scope and results of the reviews, bajo preparados por especialistas in- évaluation a pris en compte les dif- and analyzes the governance, fi- dependientes: B. Gardner "Global férentes perspectives des acteurs nance, and management in the Public Goods from the GCIAI: An Im- clé du partenariat, afin d'étudier les CGIAR. The Annexes (Part 3) pro- pact Assessment" (Bienes públicos divers points de vue possibles et les vide supporting materials and are globales desde el GCIAI: Una eva- diverses réactions aux changements available on request, and at luación de impacto), C. B. Barrett, intervenus dans le GCRAI. www.worldbank.org/oed/cgiar. "Natural Resources Management Re- search in the GCIAI: A Meta-Eva- Portée et outils analytiques de l'évalua- Evaluation review process. An external luation"(Investigación sobre el tion. La méta évaluation est basée advisory committee consisting of manejo de recursos naturales en el sur une étude d'ensemble exhaus- Yujiro Hayami, Michael Lipton, and GCIAI: Una metaevaluación", W. Les- tive, cinq documents de travail pré- Harris Mule offered guidance to the ser, "Reviews of Biotechnology, Ge- parés par des universitaires meta-evaluation team. The CGIAR netic Resource and Intellectual indépendants, à savoir B. Gardner chairman and Bank staff commented Property Rights Programs" (Examen « Global Public Goods from the on the first draft report. A Technical de los programas de biotecnología, GCRAI: An Impact Assessment », C. Briefing was given to the Bank's recursos genéticos y derechos de B. Barrett « Natural Resources Ma- Board of Directors on September propiedad intelectual), D. J. Spiel- nagement Research in the GCRAI: A 11, 2002. A second draft was sent man, "International Agricultural Re- Meta-Evaluation », W. Lesser « Re- jointly by the OED director-general search and the Role of the Private views of Biotechnology, Genetic Re- and the CGIAR chairman to the di- Sector" (Investigación agrícola inter- source and Intellectual Property rectors-general of the 16 CGIAR Cen- nacional y la función del sector pri- Rights Programs », D. J. Spielman ters for technical comment. A brief vado) y C. K. Eicher y M. Rukuni, « International Agricultural Re- presentation of key findings was "The GCIAI in Africa: Past, Present, search and the Role of the Private made to, and comments received and Future" (El GCIAI en África: Pa- Sector », et C. K. Eicher et M. Rukuni x i x T H E C G I A R AT 3 1 from, the general member- sado, presente y futuro.) Ade- « The GCRAI in Africa: Past, ship at the CGIAR's Annual más dos clientes del GCIAI Present, and Future ». En General Meeting on Novem- ~NOL (Brasil e India) prepararon outre, deux des clients du ber 1, 2002. Inputs from all A Documentos de Trabajo del GCRAI (Brésil, et Inde) ont consultations are reflected in país: J. Macedo, M. C. M. préparé des documents de ENGLISH this report. ESP Porto, E. Contini, y A. F. D. travail de pays : J. Macedo, FRANÇAIS Avila, "Brazil Country Paper M. C. M. Porto, E. Contini for the GCIAI Meta-Evalua- et A. F. D. Avila « Brazil tion," (Documento de referencia de Country Paper for the GCRAI Meta- Brasil, para la metaevaluación del Evaluation » et J.C. Katyal et GCIAI), y J.C. Katyal y Mruthyunjaya, Mruthyunjaya « GCRAI Effectiveness "GCIAI Effectiveness ­ NARS Pers- ­ SNRA Perspective ». Deux autres pective" (La eficacia del GCIAI, pers- clients (Kenya et Colombie) ont pré- pectiva de los SNIA). Y otros dos paré des documents d'études préli- clientes (Kenya y Colombia) prepa- minaires de pays : L. Romano raron los Documentos de Trabajo del « Colombia Country Paper for the país: L. Romano, "Colombia Country GCRAI Meta-Evaluation », C. Ndi- Paper for the GCIAI Meta-Evalua- ritu « GCRAI-SNRA Partnership: The tion," (Documento de referencia de Case of Kenya ». Un document pré- Colombia para la metaevaluación del liminaire a également été créé sur le GCIAI), C. Ndiritu, "GCIAI-NARS Part- Processus de conception et de ges- nership: The Case of Kenya," (El par- tion du changement (CDMP). tenariado GCIAI-SNIA: El caso Kenia). L'équipe de la méta évaluation a, en También se encargó un trabajo de outre, consulté le GCRAI en détail, antecedentes sobre el Proceso de ainsi que deux des trois co-parrains Cambio de Diseño y Gestión. Ade- (FAO et FIDA), des bailleurs de fonds más, el equipo a cargo de la metae- (USAID, SDC, DFID, Danemark, Al- valuación realizó muchas consultas lemagne et Union européenne) et con dos o tres de los copatrocina- des représentants du secteur privé dores del GCIAI (FAO y FIDA), do- et des ressortissants des pays en dé- nantes (USAID, SDC, DFID, veloppement, et effectué une en- Dinamarca, Alemania y la Unión Eu- quête formelle auprès de 235 ropea), y representantes del sector personnes de l'intérieur et d'obser- privado y nacionales de países en de- vateurs du GCRAI. (Voir Annexes 4, sarrollo, y una encuesta formal a 235 5 et 6). colaboradores y observadores del Le rapport de la méta évaluation GCIAI, y comentarios y revisiones est divisé en trois volumes. Le Rap- entre pares de los documentos de port général (Volume1) traite de referencia. (Ver Apéndices 4, 5 y 6.) questions stratégiques concernant El informe de la metaevaluación l'organisation, le financement et la consta de tres volúmenes. El Informe gestion du GCRAI dans la mesure General (Volumen 1) aborda cues- où ils ont affecté les choix de re- tiones estratégicas relacionadas con cherche, la qualité scientifique et la la organización, las finanzas y la ges- relation entre la Banque et le GCRAI. tión del GCIAI ya que estas cuestio- Le Rapport technique (Volume 2) nes son las que influyen sobre las explore la nature, la portée et la qua- opciones de investigación, la calidad lité des travaux scientifiques du Sys- científica y la relación del Banco con tème, étudie la portée et les résultats x x P R E FA C E el GCIAI. El Informe Técnico des examens et analyse la (Volumen 2) explora la natu- gouvernance, le financement ~NOL raleza, alcance y la calidad del et la gestion au sein du A trabajo científico del Sistema, GCRAI. Les Annexes (Volume evalúa el alcance y los resul- 3) fournissent les documents ESP tados de los exámenes y ana- à l'appui et sont disponibles FRANÇAIS liza el gobierno, finanzas y sur demande. gestión del GCIAI. Los Anexos (Volumen 3) ofrecen materiales de Processus de l'examen d'évaluation. Un respaldo. Están disponibles a pedido. comité consultatif externe compre- nant Yujiro Hayami, Michael Lipton Proceso de examen de la evaluación. El et Harris Mule a prodigué des equipo de metaevaluación tuvo la conseils à l'équipe de la méta éva- orientación de un comité asesor ex- luation. Le président du Conseil du terno integrado por Yujiro Hayami, GCRAI et le personnel de la Banque Michael Lipton y Harris Mule. El pre- ont commenté la première version sidente del GCIAI y el personal del du rapport. Le Conseil d'adminis- Banco hicieron observaciones en la tration de la Banque a assisté à une primera versión del informe. El 11 de séance d'information technique le septiembre de 2002, la Junta de Di- 11 septembre 2002. Une seconde rectores del Banco recibió un com- version du rapport a été adressée pendio de instrucciones técnicas. La conjointement par le directeur gé- segunda versión fue enviada conjun- néral de l'OED et le président du tamente por el Director General del Conseil du GCRAI aux directeurs gé- DEO y el Presidente del GCIAI a los néraux des 16 Centres du GCRAI Directores Generales de los 16 Cen- pour commentaires techniques. Les tros GCIAI, para que efectuaran las ob- conclusions essentielles ont été pré- servaciones técnicas. En la Asamblea sentées à tous les membres lors de General Anual del GCIAI celebrada l'Assemblée générale annuelle du el 1º de noviembre de 2002, se realizó GCRAI le 1er novembre 2002 et ont una breve presentación de los prin- été suivies de commentaires. Les cipales hallazgos y de las observacio- données provenant de toutes ces nes recibidas. Los aportes de todas las consultations figurent dans le pré- consultas realizadas están incluidos en sent rapport. el presente informe. x x i ABBREVIATIONS AND ACRONYMS AGM Annual General Meeting (CGIAR) AIDS Acquired immune deficiency syndrome AKIS Agricultural Knowledge and Information Systems (World Bank) ARD Agriculture and Rural Development Department, formerly RDV (World Bank) ARI Advanced research institution ASARECA Association for Strengthening Agricultural Research in Eastern and Central Africa ASB Alternatives to Slash and Burn (a CGIAR System-wide program) BP Bank Policy BTO Back to Office Report CAPRi System-wide Program on Collective Action and Property Rights (CGIAR) CAS Country assistance strategy (World Bank) CBC Committee of Board Chairs (CGIAR) CBD Convention on Biological Diversity CCER Center-Commissioned External Review (CGIAR) CDC Center Directors' Committee (CGIAR) CDMP Change Design and Management Process (CGIAR) CDMT Change Design and Management Team (CGIAR) CEP CIMMYT Economic Program (CGIAR) CGIAR Consultative Group on International Agricultural Research CIAT Centro Internacional de Agricultura Tropical (CGIAR) CIDA Canadian International Development Agency CIFOR Center for International Forestry Research (CGIAR) CIMMYT Centro Internacional de Mejoramiento de Maïz y Trigo (CGIAR) CIP Centro Internacional de la Papa (CGIAR) CODE Committee on Development Effectiveness (World Bank) CORAF Conférence des Responsables de Recherche Agronomique en Afrique de l'Ouest et du Centre CP Challenge Program (CGIAR) CRM Corporate Resource Management (World Bank) DEC Development Economics Vice Presidency (World Bank) DFID Department for International Development (U.K.) DGF Development Grant Facility (World Bank) Embrapa Brazilian Agricultural Research Corporation EPMR External Program and Management Review (EPMR) ESDAR Environmentally Sustainable Development Agricultural Research and Extension Group (World Bank) ESSD Environmentally & Socially Sustainable Development Network (World Bank) ESW Economic and sector work (World Bank) EU European Union ExCo Executive Council (CGIAR) FAO Food and Agriculture Organization of the United Nations FARA Forum For Agricultural Research in Africa FY Fiscal year GEF Global Environment Facility GFAR Global Forum on Agricultural Research GMOs Genetically modified organisms GPG Global public good GPPPs Global public policies and programs GRPC Genetic Resources Policy Committee (CGIAR) x x i i i T H E C G I A R AT 3 1 HIPC Heavily indebted poor country HRP Special Programme for Research and Development in Human Reproduction IAD Internal Audit Department (World Bank) IBPGR International Board on Plant Genetic Resources (CGIAR) IBRD International Bank for Reconstruction and Development ICAR Indian Council on Agricultural Research ICARDA International Center for Agricultural Research in the Dry Areas (CGIAR) ICLARM International Center for Living Aquatic Resources Management (CGIAR) ICRAF International Center for Research in Agroforestry (CGIAR) ICRISAT International Crops Research Institute for the Semi-Arid Tropics (CGIAR) ICW International Centers Week (CGIAR) ICWG-GR Inter-Center Working Group on Genetic Resources (CGIAR) IDA International Development Association IFAD International Fund for Agricultural Development IFPRI International Food Policy Research Institute (CGIAR) IITA International Institute of Tropical Agriculture (CGIAR) ILCA International Livestock Center for Africa (CGIAR) ILRAD International Laboratory for Research on Animal Diseases (CGIAR) ILRI International Livestock Research Institute (CGIAR) IMF International Monetary Fund INIBAP International Network for the Improvement of Banana and Plantain (CGIAR) INRM Integrated natural resource management IPCC Intergovernmental Panel on Climate Change IPG International public good IPGRI International Plant Genetic Resources Institute (CGIAR) IPR Intellectual property right IRRI International Rice Research Institute (CGIAR) iSC Interim Science Council (CGIAR) ISFM Integrated soil fertility management ISNAR International Service for National Agricultural Research (CGIAR) ITPGR International Treaty on Plant Genetic Resources for Food and Agriculture IWMI International Water Management Institute (CGIAR) LDC Less developed country LIL Learning and Innovation Loan (World Bank) MARP Multi-country Agricultural Research Program for Africa MAS Marker-assisted selection MD Managing Director (World Bank) MDG Millennium Development Goal MTM Mid-Term Meeting (CGIAR) MTP Medium-Term Plan (CGIAR) NARES National agricultural research and extension systems NARS National agricultural research systems NGO Nongovernmental organization NPG National public good NRM Natural resource management ODA Official development assistance OECD Organisation for Economic Co-operation and Development OED Operations Evaluation Department (World Bank) OP Operational Policy (World Bank) x x i v A B B R E V I AT I O N S A N D A C R O N Y M S OPCS Operational Policy and Country Services (World Bank) PREM Poverty Reduction & Economic Management Network (World Bank) PRSP Poverty Reduction Strategy Paper PSI Private Sector Development & Infrastructure Network (World Bank) QTL Quantitative trait loci R&D Research and development RDV Rural Development Department (World Bank) RPG Regional public good SACCAR Southern African Centre for Cooperation in Agricultural and Natural Resources Research and Training SC Science Council (CGIAR) SDC Swiss Agency for Development and Cooperation SGP Special Grants Program (World Bank) SGRP System-wide Genetic Resources Programme SINGER System-wide Information Network for Genetic Resources (CGIAR) SPAAR Special Program for African Agricultural Research SPIA TAC Standing Panel on Impact Assessment (CGIAR) SRGP System-wide Genetic Resource Programme (CGIAR) SRM Strategy and Resource Management Vice Presidency (World Bank) SRO Subregional organization SSP Sector Strategy Paper (World Bank) SWIM System-wide Initiative on Water Management (CGIAR) T&V Training and visit extension system TAC Technical Advisory Committee (CGIAR) TDR Special Programme for Research and Training in Tropical Diseases TF Trust funds administered by the World Bank TFO Trust Funds, Operations Department (World Bank) TRIPS Trade-Related Aspects of Intellectual Property Rights (WTO) TSBF Tropical Soil Biology and Fertility (CGIAR) TSR Third System Review (CGIAR) UN United Nations UNCED United Nations Conference on Environment and Development UNDP United Nations Development Programme UNEP United Nations Environment Programme USAID United States Agency for International Development VP Vice Presidency or Vice President (World Bank) VPU Vice Presidential Unit (World Bank) WARDA West Africa Rice Development Association (CGIAR) WBI World Bank Institute WTO World Trade Organization x x v Part 1 Overview 1 CGIAR's Changing Mission and Authorizing Environment T he Consultative Group on International Agricultural Research (CGIAR) consists of 16 autonomous international research Centers, with a Sec- retariat in the World Bank and a Technical Advisory Committee (renamed the interim Science Council in October 2001) housed in the Food and Agri- culture Organization (FAO) of the United Nations. Membership in the CGIAR has grown from 18 governments and organizations at inception on May 19, 1971, to 62 today (box 1.1). As a convener and donor to the CGIAR and lender to developing countries, the World Bank has played a key role in chairing the System, providing financial support, and housing and funding the CGIAR Secretariat and overhead costs of the System. The CGIAR's total budget in 2001 was $359 million, including $50 million from the Bank.1 The original mission of the CGIAR was a strate- Several forces continue to influence the gic, science-based focus on increasing "the pile CGIAR's mandate. First, the rise of civil society of rice on the plates of food-short consumers," organizations and the empowerment of mar- as characterized by a former chairman.2 It was ginal groups and women have increased donors' to use the best science in advanced countries to attention to social concerns. By restricting their develop technologies for the benefit of food- funding to preferred programs and areas, donors deficit countries and populations. But a rapidly are altering the composition of CGIAR activities. changing external environment has led to an ex- Second, water shortages, soil degradation, cli- panded mission and mandate. The mission state- mate change, and loss of biodiversity have in- ment adopted in 1998 is "food security and creased the prominence of natural resource poverty eradication in developing countries management (NRM), policy, and social science through research, partnerships, capacity build- research. The new research topics (in which ing, and policy support, promoting sustainable proponents argue the CGIAR has developed a agricultural development based on the envi- "dynamic comparative advantage") are down- ronmentally sound management of natural re- stream activities, closer to the farmer, which sources." entail local expertise and solutions, while tra- 3 T H E C G I A R AT 3 1 B o x 1 . 1 P o r t r a i t o f a G l o b a l P r o g r a m The oldest of the global programs, which the Bank helped found tonomous research Centers and 8,500 scientists and staff in and supports, the CGIAR has 62 members, including 24 developing more than 100 countries. From 1972 to 2001 the World Bank con- countries, 22 industrial countries, 12 international/regional or- tributed $930 million of the CGIAR's total support from the inter- ganizations, and 4 foundations. It is now co-sponsored by the national community of $5.6 billion. World Bank, FAO, UNDP, and IFAD. The CGIAR supports 16 au- ditional germplasm3 improvement research In view of the productivity challenge, each of builds on the CGIAR's historical comparative the pressures to expand the CGIAR's mandate or advantage. Third, the growing importance of alter its governance raises questions about its genetic resource management, the biotechnol- comparative advantage and core competencies ogy revolution,4 intellectual property rights (box 1.2) in this new and complex environment. (IPR), and private sector research call for System- Existing evaluations indicate that the CGIAR has level responses, strategies, and policies. This a strong comparative advantage and core com- has led to a vigorous debate about the suitabil- petency in germplasm research and in research ity of the CGIAR's current System-level gover- activities of a transnational scope that draw on its nance, organization, and management to these unique germplasm collection. Many of its new re- new realities. Finally, an effective global agri- search areas--not yet evaluated--involve the cultural research system remains crucial for CGIAR in providing small-scale activities at the na- meeting the Millennium Development Goals tional or sub-national level, areas where national (MDGs), which the CGIAR has undertaken to agricultural research systems (NARS) normally help implement. To halve the number of poor have a comparative advantage and core compe- from 800 million in 1990 to 400 million by 2015 tencies. Poor performance by NARS in some (an MDG goal established by the World Food countries has drawn the CGIAR into developing Summit and adopted by the World Summit on local technologies. To provide System-level Sustainable Development in Johannesburg in responses to the global IPR and biotechnology 2002) will require dramatic increases in total challenges, and to support a more effective net- factor productivity--that is, getting more agri- work-based approach to developing technolo- cultural output from existing cultivated land gies, the Third System Review (TSR) in 1998 had while using fewer natural resources--particularly recommended adopting a corporate model with in Africa and Asia, where 90 percent of the de- a legal entity. This was rejected by the member- veloping world's poor reside (figure 1.1). ship in 1999. But in 2000, in response to the de- The problem is more daunting than it ap- mand of the Bank's Development Grant Facility pears on the surface. According to the CGIAR's (DGF) for increased diversification of funding own analysis, the decline in the number of food- for the System and the Secretariat and increased insecure people in the developing world slowed private sector partnerships, the CGIAR Board considerably in the 1980s and 1990s relative to chairs and Center directors recommended a de- the 1970s, the period of the Green Revolution. centralized Federation of Centers. While the two Indeed, if China is excluded, the number of food proposals differed in the degree of decentral- insecure increased in the rest of the developing ization proposed, both acknowledged the need world in the 1990s, while the annual rate of for a legal entity with a centralized Board to en- growth in cereal yields decelerated from 2.9 per- able System-level responses to IPR issues. cent during 1967­82 to 1.9 percent during Recently announced increases in funding com- 1982­97. The rate of growth in cereal yields is mitments to the CGIAR and revived attention to projected to decline further (figure 1.2) (Pin- agriculture in development5 attest to the Sys- strup-Andersen 2001). tem's reputation as the "single most effective 4 C G I A R ' S C H A N G I N G M I S S I O N A N D A U T H O R I Z I N G E N V I R O N M E N T F i g u r e 1 . 1 W h e r e t h e P o o r A r e Sub-Saharan Africa South Asia 24% 43% Middle East & North Africa East 1% Asia/Pacific 23% Europe & Central Asia Latin America & 2% Caribbean 7% F i g u r e 1 . 2 C e r e a l Y i e l d s : S l o w d o w n i n G r o w t h Percent per year 3.5% 1967­82 3.0% 2.9% 1982­97 1997­20 Baseline 2.5% 2.0% 1.9% 1.7% 1.7% 1.5% 1.2% 1.0% 0.7% 0.5% 0.0% Developing world Developed world Source: Pinstrup-Andersen 2001. 5 T H E C G I A R AT 3 1 C o n c e p t s R e l a t i n g t o S e t t i n g a n d B o x 1 . 2 A c h i e v i n g P r i o r i t i e s Comparative advantage describes the ability of one eco- Core competencerefers to an economic actor's unique set of nomic actor (e.g., the CGIAR System) to produce a good, serv- assets that cannot be replicated easily by other actors. These ice, or knowledge at a lower opportunity cost than another are typically specific production resources, technological and economic actor. Opportunity cost is the cost of forgoing one ac- managerial capabilities, or reputational integrity that help ac- tivity in favor of another, measured in terms of the goods, serv- tors gain an advantage over competitors. ices, or knowledge whose production is forgone. Collectiveactiondilemma refers to the difficulty that rational, Dynamic comparative advantage describes how an eco- self-interested members of a group experience in achieving nomic actor may develop new comparative advantages over their common group interest, unless the number of individuals in time as a result of new investments, learning by doing, or com- a group is quite small, or there is coercion or some other special parable changes experienced by other actors. device to make individuals act in their common interest. use of [official development assistance], bar The CGIAR is at a crossroads. Although fund- none," as it was described in the TSR. Indeed, the ing sources have diversified, the top 10 donors CGIAR is often referred to as a model global part- still contribute two-thirds of the resources to the nership. However, the four most recent chairmen System. Their leaders, along with developing of the CGIAR also share the TSR's conclusion countries, who are the intended beneficiaries of that the CGIAR needed to change fundamentally the System and who make many in-kind contri- to meet the complex challenges of a rapidly butions to the System, must play a major role in changing world. In 2000, Ismail Serageldin (CGIAR addressing and resolving the complex funda- Chairman, 1994­2000) stated in his farewell mental issues of System-level organization, gov- speech to members that the CGIAR faces "the ernance, management, and finance that confront prospect of... gradually fading into obsolescence the CGIAR in much the same way the founders and, ultimately, oblivion, while other actors, more did when the CGIAR was established in 1971. swift, better endowed, and more responsive to Drawing on previous evaluations, extensive the needs of our clients, pass us by" (CGIAR Sec- consultations, and a stakeholder survey, the meta- retariat 2000b). Before he initiated the Change evaluation addresses several strategic questions.7 Design and Management Process (CDMP) in What do we know about CGIAR activities and their 2000, the current CGIAR Chairman, Ian John- impacts on enhancing agricultural productivity son, similarly asserted, "there is strong senti- and reducing poverty? How well is the CGIAR mo- ment for change in the CGIAR. Current realities bilizing the best of global science (biological, phys- require the CGIAR to gear up for change, and ical, and social) to focus on the problems of the such change must be reflected in both form and poor and build capacity in developing countries function" (CGIAR 2000c). The current reform to spread impacts over large areas? How well is the process follows the "Renewal" process of the CGIAR meeting the challenges of biotechnology, mid-1990s, which responded to a financial crisis intellectual property rights, the private sector, and in 1993­94. Yet daunting collective action prob- increased expectations of stakeholders? What are lems have kept the System from making the fun- the lessons from attempts to reform the System damental reforms in its organization, governance, in the 1990s? Will the current reforms enable the management, and finance thought to be needed CGIAR to succeed in a new and more challenging by a wide set of CGIAR stakeholders and analysts.6 environment? To what extent has the Bank been The idea of the need for a legal entity to enable providing the necessary leadership to reposition System-level responses has, for instance, disap- the CGIAR to deliver results in terms of agricultural peared in the recent CDMP. productivity growth and rural poverty reduction? 6 2 CGIAR Activities and Impacts: What Do We Know? T he CGIAR built its reputation on producing improved germplasm and related technologies with sizable impacts on reducing poverty. Germplasm improvement research is a global or regional public good whose production is ideally suited to a publicly funded global or regional net- work (box 2.1), and it is the activity that has been most extensively evaluated by the CGIAR. Because some have doubted the returns and im- torn countries, such as Rwanda and Afghanistan, pacts claimed by its evaluations, the meta-evalu- further testifying to the key global public goods ation assessed the CGIAR's impact studies of its nature of the CGIAR's work in this area. research on germplasm improvement and re- Germplasm improvement research has also lated areas of NRM--such as integrated pest man- brought net positive environmental impacts. agement. Based on that assessment, the The most important of these is the large savings meta-evaluation has determined that the CGIAR in land used in cropping in land-scarce countries, has made important contributions to agricultural many of which have large shares of global productivity and poverty reduction. The studies poverty. Germplasm improvement has helped not only demonstrate extraordinarily high rates poor households to increase production; to sta- of return to germplasm enhancement research-- bilize yields on resource-poor, fragile lands; to 40 to 78 percent1--but also suggest that this re- conserve water, soils, and use of chemical inputs; search has contributed to food security and and thereby to promote biodiversity by enabling helped lift millions out of poverty. The research more diversified farming systems. Even in Africa, benefits the poor directly through increased pro- following an initial failure to appreciate the re- duction of subsistence foods, employment, and gion's particular agro-ecological challenges, the income generation, and indirectly through re- CGIAR has produced some important new tech- duced prices for food-deficit households. CGIAR nologies vital to the poor, several of which have germplasm collections, which are vital for this re- been widely adopted by farmers, while the search, have also been of considerable benefit to process of research has helped the develop- agriculture in industrial countries, such as the ment of the region's NARS. United States and Australia. Recently they have But expenditures on productivity-en- contributed critically to the rehabilitation of war- hancing agricultural research have declined 7 T H E C G I A R AT 3 1 W h a t I s G l o b a l a n d R e g i o n a l P u b l i c B o x 2 . 1 G o o d s R e s e a r c h ? In this report, global and regional public goods research is: · Research in which a global network like the CGIAR, because of its international status and global reach, has a strong com- · Research that produces environmentally friendly technologies, parative advantage and core competencies. knowledge, or information relevant for reducing poverty in de- veloping countries and with potential for large spillovers and See Chapter 8 of Part II of this volume for treatment of the con- economies of scale cepts of public goods, common pool goods, private goods, and merit goods as these apply at the global, regional, national, and · Research that is not easily conducted by national systems of local levels, and the dynamic relationship between these goods developing countries themselves and levels over time and space. dramatically, while those on improving poli- biodiversity grew by 2.7 percent annually and on cies and protecting the environment have in- strengthening NARS declined by 0.8 percent. Train- creased. From 1992 to 2001 (that is, since the ing--which is crucial for accessing global knowl- addition of four new Centers in 1992),2 the edge, which represents about 40 percent of CGIAR's research expenditures on increasing "Strengthening NARS," and which many NARS productivity3 declined by 6.5 percent annually consider the most important contribution of the in real terms, while those on improving policies CGIAR after its germplasm research--has de- and on protecting the environment (largely re- clined by 0.2 percent annually. lated to NRM) increased by 3.1 percent annually At the same time, overall CGIAR funding has (figure 2.1). Research expenditures on saving stagnated in nominal terms, declined in real R e s e a r c h w i t h P r o v e n I m p a c t s o n P o v e r t y F i g u r e 2 . 1 H a s D e c l i n e d D r a m a t i c a l l y Increasing productivity Strengthening NARS Saving biodiversity Protecting environment Improving policies ­8% ­6% ­4% ­2% 0% 2% 4% Average annual change in Centers' expenditures by type of research activity (adjusted for inflation), 1992­01 Source: Calculations based on CGIAR Financial Reports, 1992-01. 8 C G I A R A C T I V I T I E S A N D I M PA C T S : W H AT D O W E K N O W ? terms, and become increasingly restricted over countries should be conducted close to the de- the past decade (figure 2.2). Overall contributions veloping countries whose policy environments grew at an average annual rate of 0.7 percent in are the subject of the research. It should build nominal terms and declined by 1.8 percent per up the policy and analytical capacity of those year in real terms between 1992 and 2001. There- countries to produce what is a national public fore, it appears that the increased expenditures good. Hence, the conduct of research and its ap- on policy and NRM research have come at the plication should actively engage the nationals of cost of germplasm conservation and improve- developing countries. The CGIAR's own external ment. However, it is also possible that, because reviews of the International Food Policy Institute of the principles of donor sovereignty and Cen- (IFPRI) suggest that policy research would ben- ter autonomy under which the CGIAR operates, efit from researchers with first-hand knowledge germplasm research would not have received the of developing countries' policies and processes funding provided to policy and NRM research. (Gardner 2002). The CGIAR is less focused; its current While the quality of the CGIAR's policy re- mix of activities reflect neither its com- search is not in question, the balance of its com- parative advantage nor its core compe- position is. The CGIAR is in a unique position to tence. The variety of research carried out by the explore the incentive effects of Organisation for CGIAR has increased and moved from research Economic Co-operation and Development topics of a global nature to policy, NRM, and so- (OECD) subsidies and trade protection on both cial science research with a national or sub- the rural poor and on investments in agricultural national focus.4 To be relevant, effective, fine- research and development in developing coun- tuned, and have impact, policy research on con- tries (a topic covered in part by IFPRI's work). straints to technology adoption in developing Several CGIAR reviews of commodity Centers F i g u r e 2 . 2 R e s t r i c t e d F u n d i n g H a s I n c r e a s e d Millions of 2000 U.S. dollars 400 350 Non-Agenda -- Restricted 57% 300 Agreed Agenda -- Restricted 250 200 Agreed Agenda -- Unrestricted 150 43% 100 50 0 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 Total (cumulative) contributions to CGIAR research activities (adjusted for inflation) Source: CGIAR Financial Reports, 1988­01. 9 T H E C G I A R AT 3 1 ing countries do not possess, and in conducting research with the potential for wide applicabil- ity. But the global reach of NRM research, beyond developing data sets and understanding natural processes, is limited by the diverse circumstances in which it must be conducted. Moreover, the CGIAR's own reviews have highlighted the small- scale, nonscalable nature of the System's NRM research, its sometimes questionable scientific quality, and its lack of focus on productivity growth or results orientation, and the conse- quent frequent absence of evidence that the re- search will achieve either more efficient use of natural resources or generate relevant knowledge have also concluded that upstream economic of widespread applicability to ensure global or policy analysis is needed to deal with the CGIAR's regional impacts.6 rapidly changing external environment (espe- Evidence on the impacts of NRM, policy, cially on such issues as intellectual property and social science research is lacking. rights and the role of the private sector). The IFPRI's policy research is the only research in the meta-evaluation finds that the CGIAR's policy System that is systematically reviewed by the research will yield higher returns if it addresses System, and hence available for the meta-evalu- commonly observed policy failures in develop- ation. The CGIAR's own reviews suggest that ing countries while helping to build the capac- the quality of both policy and NRM research may ity of developing country NARS to conduct their be higher in the Centers that specialize in such own country-specific policy research. The research, and find that both policy and NRM re- CGIAR's policy research similarly needs to give search boast some notable accomplishments, priority to the external environment facing de- but on the whole their impacts remain largely veloping countries, the improvement of which unevaluated.7 Likewise, although participatory re- is a global public good. search has become popular in the CGIAR, nei- For quite different reasons, most NRM re- ther its methods nor its costs and benefits have search relevant to developing countries must be been evaluated (Gladwin, Peterson, and Mwale conducted on site and be highly sensitive to 2000). A recent CGIAR conference on the Sys- local institutions, given the immense diversity tem's social science research makes a strong and location-specificity of natural resources and case that more upstream social science research the social institutions accompanying them (Bar- on the determinants of household and group rett 2002). Most NRM research is not easily scal- adoption of technologies would avoid wasteful able, and in most cases is a national or a local expenditure on biophysical research. Yet it also public good in the form of knowledge of best concludes that social science research within practices (although the failure of countries to the CGIAR is weak, lacks critical mass, and has provide such national and local public goods been deteriorating over time with a decline in the can have global im- number of social scientists with the necessary Most NRM research pacts).5 The compara- qualifications (Cernea and Kassam 2002). Simi- relevant to developing tive advantage of a lar criticisms have occasionally been leveled at global research system the CGIAR's plant and animal breeding research countries must be in NRM research lies in by scientists in advanced countries, especially conducted on site and bringing to bear ad- given the current highly dynamic state of bio- be highly sensitive to vanced multi-discipli- logical research. But a major difference between local institutions. nary methods and the two types of research is that CGIAR's processes that develop- germplasm research has an unassailable record 1 0 C G I A R A C T I V I T I E S A N D I M PA C T S : W H AT D O W E K N O W ? of positive impact on millions of poor in the de- tem's research priorities Priorities are increasingly veloping world. as articulated by TAC and determined by the A complex combination of factors ex- the financial contribu- preferences of donors and plains the changing research mix. First, tions to those priorities their multiple sources of germplasm improvement and associated bio- by other donors. Under logical research has been unpopular in the con- the matching grant funds. stituencies of some key donors because of model, the Bank's con- negative perceptions of the Green Revolution-- tribution indiscriminately matches funding from that it made the rich richer and the poor poorer other donors, whether in support of System- and caused environmental damage.8 Second, the wide priorities or not.9 Second, to create incen- CGIAR has correctly responded to the genuine tives for Centers to mobilize additional funding second-generation environmental pressures on and to accommodate donors, the CGIAR ex- soils and water created by the radical change in panded the definition of its "agreed research farming systems during the Green Revolution, agenda" to include both the former "core" where research continues to be needed. Third, agenda (largely the high-return global and re- the rise of environmentalism, the 1992 United Na- gional public goods research) and the "non- tions Conference on Environment and Devel- core" agenda (donor-funded, mostly downstream opment (UNCED) in Rio de Janeiro, and growing activities that TAC did not consider as high a environmental advocacy in donor countries led priority). Third, donors have collectively in- to rising demands on the CGIAR to respond to creased their degree of restricted funding from environmental concerns. Fourth, the failure of 36 percent of total funding in 1992 to 57 percent governments of developing countries and their in 2001, with most of the increase in restricted donor supporters alike to make the necessary in- funding occurring since 1998 (figure 2.3).10 vestments in developing countries' own research, There is an overwhelming consensus among education, and development systems limited those whom the meta-evaluation team consulted their ability to adapt CGIAR technologies to their that the growing share of restricted funding is dis- own farming systems. This led CGIAR donors to torting research priorities, increasing transac- turn to the Centers to fill the national and local tions costs, and reducing the efficient use of public goods gaps. Fifth, the biotechnology rev- resources at both the System and Center levels. olution, the emergence of intellectual property Priorities are increasingly determined by the pref- rights, and the associated rapid increase in pri- erences of donors and their multiple sources of vate sector investments have challenged some funds. Centers have less ability and flexibility to CGIAR donor constituencies. Thus, to avoid con- plan, organize, and implement long-term re- troversy among Centers and within its member- search, since they are having to raise funds Cen- ship, which makes decisions by consensus, the ter by Center to conduct research on the most CGIAR leadership has not explored the full im- recent high-profile, usually short-term, issues plications of the biotechnology revolution for with restricted bilateral funding.11 The result is not the System--a response reinforced by the au- only increased time spent by Center scientists, tonomy of the 16 Centers. managers, Center boards, the CGIAR chairman, Three changes in the funding processes of the and Secretariat on fund-raising and negotiating CGIAR since the mid-1990s have also increased with individual donors on research programs,12 the influence of individual donors (and their but also a shift in the balance of power in deci- domestic constituencies) on the research ex- sionmaking from the TAC/iSC to the donors, penditures of the CGIAR. First, in response to a chairman, Secretariat, and Centers. The oppor- funding crisis in 1993­94, the Bank changed the tunity cost of increased fund-raising and negoti- allocation of its own financial contribution from ating is forgone upstream, long-term research a "donor of last resort" model to a matching within the CGIAR's comparative advantage that grant model. Under the former, the Bank's con- is crowded out both by the transactions costs of tribution was used to fill gaps between the Sys- fund-raising and by the tasks the Centers have 1 1 T H E C G I A R AT 3 1 agreed to carry out, but for which NARS have their emerging needs. Both the large and small lower costs and greater comparative advantage. NARS consulted by the meta-evaluation ex- Strong NARS are critical to ensure the pressed interest in the large NARS working with CGIAR's impacts. The CGIAR's impacts result the smaller NARS, since their agro-ecologies and from the joint outputs of CGIAR Centers and development conditions are often similar. Ca- NARS in developing countries. The NARS gen- pacity building efforts of the NARS under the erate location-specific technologies and are es- CGIAR umbrella could be substantially aug- sential for testing, adapting, and disseminating mented, based on clear business-like agreements the products of CGIAR research. Yet the rate of and international financial support for such growth of investments, which would enable south-south cooperation. NARS to undertake such research and which Third, the CGIAR faces structural issues in was quite rapid until the mid-1980s, has slowed Africa. Eighty percent of Africa's agricultural re- for well over 15 years, and has even become searchers are concentrated in 13 large countries. negative in Africa (Pardey and Beintema 2001). The other 30-plus small countries face disec- The meta-evaluation has identified three im- onomies of scale in organizing and managing portant issues with regard to the NARS. First, in their own agricultural research and lack the ca- some regions of the world, most notably Africa, pacity to negotiate with the 16 CGIAR Centers. under-investment in NARS has adversely affected This highlights the importance of relying on sub- their strength and capacity. This weakness has regional research organizations in Africa if the been compounded by economy-wide and sector- African agricultural productivity challenge is to be level policy and institutional failures that inhibit addressed. Many argue convincingly that Africa's the development of rural infrastructure, input de- excessive donor dependence has reduced the livery systems, and output markets. Therefore, incentive for countries to invest in their own in- technology adoption has not been accompanied stitutions, create domestic constituencies for re- by effective delivery of services and increases in search, and ensure long-term stability and national productivity and incomes. This failure has led the priorities. They believe that recent proposals (by CGIAR to conduct, adapt, and disseminate re- Canada, the EU, the World Bank, and the Chal- search, extension, and information at the na- lenge Program on Sub-Saharan Africa) to increase tional level. donor support to the region may waste resources Second, the CGIAR has not kept pace with the without providing high-priority long-term re- changing and highly divergent needs of NARS. search of regional significance and without cre- Large and small NARS alike acknowledge that the ating long-term domestic capacity and a political CGIAR has made major contributions to their or professional constituency for research. The growth. But the capacity of NARS in large and overall approach to improving agricultural pro- middle-income developing countries now sub- ductivity and reducing poverty in Africa, includ- stantially exceeds that of the CGIAR. NARS argue ing the role of the CGIAR, requires fundamental that the CGIAR has not kept pace with their in- collective rethinking. creasingly complex needs and does not con- Reforms in the CGIAR System alone, sider them equal partners. They lament the while essential, will not solve these prob- decline in research collaboration and the limited lems. Clearly, more unrestricted donor fund- efforts of the CGIAR to draw ing and a strong Science Council are needed to Capacity building on their expertise and expe- ensure that funds are directed toward strategic, efforts of the NARS rience to build capacity in long-term research priorities that are clearly fo- under the CGIAR smaller, less-advanced NARS. cused on the CGIAR's mission of sustainable umbrella could be Smaller and weaker NARS are poverty reduction and agricultural productivity concerned that the CGIAR's improvement. Well-defined strategic priorities substantially training and collaborative re- should drive funding rather than funding driving augmented. search have not kept up with priorities.13 Concurrently, substantial comple- 1 2 C G I A R A C T I V I T I E S A N D I M PA C T S : W H AT D O W E K N O W ? mentary investments need to be made in the thirds of the recorded funding, and the top three agricultural sectors of developing countries to en- (the World Bank, United States, and Japan) sup- sure that they become more effective partners ply about 60 percent of the critical unrestricted of the CGIAR. funding that finances Center overheads, gene Diversification, which is symptomatic of what bank operations and maintenance, and longer- Center directors and board chairs have described term research programs (figure 2.3).14 In addi- as a "marketplace" for public goods research tion, developing countries make contributions (CGIAR Committee of Board Chairs and CGIAR of genetic materials, land and buildings for the Center Directors' Committee 2000), has broad- Centers and their regional offices, and devote ened political support and stabilized financing for funds to collaborative research. NARS scientists the System. But it has also increased the trans- spend time on CGIAR matters, none of which are actions costs of managing small grants and keep- recorded in their contributions. A concerted ef- ing all on board, while changing the long-term fort to improve the CGIAR's effectiveness must nature of the CGIAR. Although funding has di- therefore involve both the major donors and versified, the top 10 donors still provide two- developing country members. R e s t r i c t e d F u n d i n g V a r i e s G r e a t l y a m o n g F i g u r e 2 . 3 t h e T o p T e n D o n o r s EC 96% United Kingdom 95% Germany 82% Switzerland 65% Japan 50% United States 43% Denmark 40% Netherlands 31% Canada 31% World Bank 0% 0% 20% 40% 60% 80% 100% Restricted funding as percent of total, 2001 Source: CGIAR Financial Report, 2001. 1 3 3 New Challenges for the CGIAR: How Is It Responding? T he radically changed external environment offers challenges and op- portunities. These include: the growing scarcity and degradation of soil and other natural resources; the reduced costs of information and communications; the growth of private sector investments and food and bio- safety regulations; and the lack of progress on agricultural trade liberalization. Related challenges include the growth of mod- changes in the research methods in genomics. ern biotechnology and bioinformatics; concerns The reviews stress that the critical investment in about genetically modified organisms; the grow- equipment needed for long-term research is ing importance of intellectual property rights; hampered by the short-term project-by-project and an increasingly diversified set of NARS in de- financing the CGIAR receives from donors and veloping countries. The substantially improved by the fragmentation of research across several food situation in most of the developing world-- Centers. They have urgently recommended con- with the notable exception of Sub-Saharan solidation, long-term financing, and investment Africa--has limited the incentives to invest in in human capital to upgrade the CGIAR's ge- agriculture in developing countries, especially netic improvement work, and establishment of in view of the agricultural trade restrictions, strategic partnerships with advanced-country subsidies, and resulting surpluses of OECD institutions that possess the huge capital in- countries. vestments needed in biotechnology research. Investing in biotechnology and estab- Instead, funding shortages have led the CGIAR lishing partnerships among CGIAR Centers commodity Centers to rapidly downsize. and between the CGIAR Centers and oth- Effective genetic resource conservation ers. Informed experts consider the CGIAR's es- and management. CGIAR possesses 600,000 timated investment in biotechnology research of accessions of genetic material collected mainly $25 million, spread over several Centers, to be from developing countries and held in trust by too small, leading it to obsolescence (Lesser the Centers with the Food and Agriculture Or- 2002). They consider the Center-by-Center and ganization (FAO) as the trustee.1 Although this is commodity-by-commodity approach that was a true global public good, donors have been un- pertinent when the CGIAR was established to be willing to fund the overhead costs of managing inappropriate in light of the revolutionary the gene banks, and they have even been in- 1 5 T H E C G I A R AT 3 1 creasingly unwilling to fund germplasm im- plant genetic material requires vigilance, in much provement. Yet biodiversity conservation was the same way as the TSR and the CGIAR board high on the agenda of UNCED in Rio de Janeiro chairs/Center directors' Federation proposal in 1991 and at the recent Johannesburg Sum- (discussed in the next section) acknowledges mit. The overheads of gene banks have been the need for the formation of a System-level supported by the ever-diminishing unrestricted legal entity to deal with the growing challenges funding provided by a handful of donors together of intellectual property rights and to position the with the World Bank. Hence, the gene banks are CGIAR for the new environment. under-funded and an unknown amount of ma- The growth of private agricultural re- terial has deteriorated over time. search. The private sector is an increasingly The growing importance of intellectual important actor in agricultural research. It funded property rights in agricultural research. 35 percent of the $33 billion (in 1993 interna- A combination of the convention on biodiversity, tional dollars) spent on agricultural R&D world- the World Trade Organization (WTO) rules, farm- wide in 1995/96 and 70 percent of the research ers' and indigenous people's rights, and the on genomics, while possessing 80 percent of 2001 International Treaty on Plant Genetic Re- the intellectual property emanating from it. Yet sources for Food and Agriculture (ITPGR)--all in 1995/96, private sector expenditure on agri- relevant to the CGIAR's large germplasm col- cultural R&D in developing countries was less lection--have dramatically changed the intel- than 6 percent of all private sector expenditures lectual property environment for the CGIAR. A in agricultural R&D worldwide (figure 3.1), which recent British commission on intellectual prop- is also less than 6 percent of the total expendi- erty rights (IPRs) reviewed these and other tures (both public and private) on agricultural changes (box 3.1) and advised global programs R&D in developing countries. The public sector such as the CGIAR to confront the full implica- and the CGIAR continue to have a strong com- tions of IPRs.2 It made recommendations to deal parative advantage vis-à-vis the private sector in with "the fundamental asym-metry in relations developing technologies for poor and marginal between developed and developing countries."3 farmers in developing countries, because of the The Commission also warned that the compro- limited markets for such technologies, the lim- mise wording of the ITPGR and its scope for ited purchasing power of poor and marginal patents on genes to be isolated from the CGIAR's farmers, and long-term returns to such research T h e N e w W o r l d o f I n t e l l e c t u a l B o x 3 . 1 P r o p e r t y R i g h t s "Over the last twenty years or so there has been an unprece- · The extension of protection into new areas such as soft- dented increase in the level, scope, territorial extent and role ware and business methods, and the adoption in some coun- of IPR protection. Manifestations of this include: tries of new sui generis regimes for semiconductors and databases · The patenting of living things and materials found in nature, · A new emphasis on the protection of new knowledge and tech- as opposed to man-made products and processes more read- nologies produced in the public sector ily recognizable to the layman as inventions · The focus on the relationship between IP protection through · The modification of protection regimes to accommodate new the TRIPS agreement, and of higher standards through bilat- technologies (particularly biotechnology and information eral and regional trade and investment agreements technology), such as the EU Biotechnology Directive or the · The widening of exclusive rights, extension of the duration Digital Millennium Copyright Act (DMCA) in the United States of protection, and strengthening of enforcement mechanisms." Source: Commission on Intellectual Property Rights 2002, p. 2. 1 6 N E W C H A L L E N G E S F O R T H E C G I A R : H O W I S I T R E S P O N D I N G ? P r i v a t e A g r i c u l t u r a l R e s e a r c h F i g u r e 3 . 1 E x p e n d i t u r e s A r e S m a l l i n D e v e l o p i n g C o u n t r i e s , L a r g e i n D e v e l o p e d C o u n t r i e s $10.2 billion Public $11.5 billion Developed Public Developing Private Private $10.8 billion $0.7 billion Agricultural R&D expenditures, 1995 Total: $33.2 billion Source: Pardey and Beintema 2001. (see Lesser 2002; Pardey and Beintema 2001; and on the distinct accountabilities and obliga- Spielman 2002). tions of the partners. Collaborative successes Responding to these challenges requires will depend on the political support for public- a System-level approach. These challenges re- private collaboration from domestic con- quire the CGIAR to develop a System-level pol- stituencies of donors and developing countries, icy and strategy on intellectual property and strong research programs and financial resources public-private partnerships so it can speak with in both the public and private sector, well- one voice and become a powerful force in in- thought-out benefit-sharing arrangements, IPR ternational negotiations, and to forge active part- regimes that provide for commercial incentives nerships with the private sector, universities of within public-private research initiatives, and advanced countries, and the national systems market segmentation where it is needed for the of developing countries. To do so the CGIAR benefit of the poor.4 needs to confront five major questions: (1) In The CGIAR is responding--but slowly. which areas should the System partner with the To maintain its genetic material, the CGIAR has private sector to conduct research? (2) In which worked with the Swiss Agency for Development areas do costs make it more efficient to access and Cooperation (SDC) and FAO to establish the private sector technologies for the benefit of Global Conservation Trust and helped launch a the poor? (3) How can benefit-sharing arrange- fund-raising effort. The prospects for raising the ments be defined to ensure such access? (4) In conservatively estimated $260 million endow- which areas should the CGIAR's own research be ment--the interest would support the gene actively commercialized by the private sector? banks--are unclear at this stage.5 By November (5) How would having CGIAR members with 2002, commitments of $60 million had been ob- specific interests in the choices made by the tained. These steps are appropriate, but are un- CGIAR affect its decisions? likely to be sufficient. The mandate of the Public-private partnerships have the potential System-wide Genetic Resources Program (SGRP) to increase the efficiency of the CGIAR's work in applies only to the Centers fulfilling their obli- biotechnology, but the results of the partner- gations under the FAO agreement, and does not ships are highly dependent on their objectives extend to the program on the conservation, man- 1 7 T H E C G I A R AT 3 1 ficult to achieve, and the lack of consensus within the CGIAR threatens the System.6 Even if achieved, the consensus will need to be fol- lowed by concrete actions by the CGIAR Sys- tem as a whole to achieve poverty-reducing outcomes in a situation where national laws and regulations may conflict, and international trade rules will be invoked as in the current African food crisis.7 Given the importance of partnerships and consensus building, the commercial private sec- tor and the nongovernmental organization (NGO) community acquired seats at the CGIAR table in 1995, and on the Executive Council when it was established in 2001. A debate continues among agement, and enhancement of genetic resources, the CGIAR membership as to whose interests which is the responsibility of individual Centers. NGOs represent, since developing countries and The System Office being established under the farmer groups have asserted that they prefer CDMP is a move in the right direction, but it does their own representation.8 In addition, a new not yet go far enough in its coverage of services private foundation, Syngenta, joined the CGIAR that could be effectively centralized and are cur- membership in 2002. It differs in character from rently duplicated. Some recent developments the older foundation members since it has a di- compound the problem. The CGIAR's System- rect interest in agricultural biotechnology and wide office on IPRs, located at and paid for by the in public-private partnerships. International Service for National Agricultural The CGIAR lacks a System-level strategy Research (ISNAR), is understaffed and inade- for public-private partnerships. Several quate to meet the System's need for IPR capac- CGIAR Centers are engaged in public-private ity. Its funding has dwindled as other CGIAR partnerships, but CGIAR reports are uninfor- Centers have developed their own capacity, and mative regarding the System-wide policies and ISNAR is being restructured with uncertainty as strategies on private sector issues. Neither a uni- to the future location and role of the System-wide fied strategy nor a System-level reporting mech- IPR office. The example demonstrates that the anism for IPR issues is currently in place for a Center-by-Center and issue-by-issue approach system that holds a large collection of global to reforms being pursued by the CGIAR under germplasm. The value of private sector part- the CDMP is insufficient for the times. nership derives from its expertise and know- The World Bank launched a new global ini- how in taking technologies to the market. It is tiative at the Johannesburg Summit to assess unclear that the private sector would bring new the potential of agricultural science and tech- funding for the CGIAR Centers even in the best nology in boosting agricultural productivity in de- of circumstances. veloping countries, with a view to generating a CGIAR needs to mobilize the best practical political and scientific consensus similar to that System-level expertise in global public policy, developed for climate change. Its relationship to law, and ethics, and develop, monitor, and report the CGIAR reform regularly on effective partnership arrangements Consensus on the most process is unclear. to the System as a whole, while actively devel- controversial issue, Consensus on the oping strategies and policies that advance its genetically modified most controversial poverty alleviation mission. It needs to docu- issue, genetically ment and learn concrete lessons from its own ex- organisms, may be modified organisms perience with public-private partnerships and difficult to achieve. (GMOs), may be dif- management of IPRs at the System level. But 1 8 N E W C H A L L E N G E S F O R T H E C G I A R : H O W I S I T R E S P O N D I N G ? while there have been many meetings, commit- System-level attention, and that its stakeholders tee reports, and declarations, there is no evi- undertake bold reforms not only in the CGIAR dence that the CGIAR has learned enough System, but also in shaping the environment in System-level lessons from its own experience which the CGIAR operates. With a System-level and reviews in dealing with modern technology. policy framework that provides authoritative, It has been slow to respond to these growing transparent, and accountable System-level re- challenges in a concrete way or in allowing Sys- sponses to the changing technological and in- tem-level outcomes and impacts to be meas- stitutional environment, the CGIAR would be a ured on the ground. A partnership between powerful force in global negotiations backing IRRI, the Rockefeller Foundation, Syngenta, and the interests of developing countries. In its re- others, which has developed "golden rice," high- sponse to the ITPGR, the CGIAR showed that it lights the issues in taking a new technology from can develop an appropriate System-level re- the laboratory to the consumer (box 3.2). sponse. But Center autonomy and System-level Regaining focus requires that the System de- governance weaknesses have prevented the termine key issues of high priority that require CGIAR from addressing other challenges. " G o l d e n R i c e " D e m o n s t r a t e s t h e B o x 3 . 2 C h a l l e n g e s o f P u b l i c - P r i v a t e P a r t n e r s h i p s Golden rice is a well-known result of a public-private partnership are the intellectual property of 32 companies and universities. in which private companies have granted royalty-free licenses to To navigate these legal complexities, the scientists negoti- public sector research institutions to work with their patented ated a deal in which Syngenta acquired the rights to golden rice, technologies on behalf of poor farmers in developing countries. allowing the company to exploit the commercial potential of In August 1999, Swiss researchers collaborating on a Rocke- the technology, in exchange for allowing royalty-free distri- feller Foundation­funded project succeeded in genetically mod- bution of the technology to poor farmers in developing coun- ifying the rice genome to express beta-carotene, a precursor of tries. They are negotiating similar deals with other firms, such vitamin A. With its "golden" tint, the rice is a potential solution to as Monsanto and Bayer, who hold rights over key technologies vitamin A deficiency (a condition causing blindness) for over 100 used in golden rice. million people in the developing world, most of whom are children. The CGIAR needs to be empowered and equipped to under- However, some 70 process and product patents are asso- take such partnerships with full understanding of their impli- ciated with the technology, and the genes and methods used cations for its mission. Sources: Interview with Robert Herdt; R. Herdt 2000; Commission on Intellectual Property Rights 2002, p. 129. 1 9 4 Restructuring the CGIAR: Lessons of Past Attempts T he CGIAR is a unique instrument of international cooperation with a solid record of achievements. In the research and development continuum spanning from basic, strategic, applied, and adaptive research to tech- nology transfer, the CGIAR has made its mark and demonstrated the genius of its framers. It has succeeded mostly because of its emphasis on strategic research of a global or regional public goods nature, with benefits that spill across na- tional boundaries and cannot easily be obtained through private, national, or regional research, and its practical, problem-solving focus on bringing the best of known science to address the problem of food security. However, the governance, management, and fi- the time, responded by merging the two live- nancing of the CGIAR have become increasingly stock Centers (ILCA and ILRAD) into one Center cumbersome in recent years, and fundamental (ILRI), both to reduce costs and to strengthen links changes are called for. Successfully restructuring between upstream research and the downstream the CGIAR in the light of current challenges and needs of clients experiencing rapidly growing assessing how well the current reform process livestock demand. He also commissioned a report is addressing issues raised in previous evaluations from the Technical Advisory Committee (TAC) require an understanding of past attempts at re- for a System-wide consolidation. When Ismail form and of the evolution of the CGIAR gover- Serageldin became CGIAR chairman in 1994, he nance, management, and financial systems over felt that further mergers were inappropriate in a the past decade. period of financial weakness and embarked on a The mid-1990s "Renewal" demonstrates program of political and financial mobilization, how changing the rules can have unintended which he called "Renewal."1 Renewal solicited consequences. Soon after the CGIAR acquired developing country membership and increased four existing Centers and established a new one their ownership of the System.2 It increased par- in the early 1990s, unexpected funding cuts from ticipation by the private sector and the NGO com- key donors in 1993­94 produced a serious fi- munity, but also increased the number of nancial crisis. V. Rajagopalan, CGIAR Chairman at committees to reflect diverse viewpoints. Ser- 2 1 T H E C G I A R AT 3 1 ageldin successfully persuaded the Bank and oth- From the perspective of the System, it: ers to sustain and even increase their funding, but also introduced important policy changes, in- · Reduced funding for the five commodity-ori- cluding redefining the "agreed research agenda" ented research Centers by an average of 3.3 per- to create incentives for Centers to mobilize ad- cent annually (in real terms) from 1992 ditional funding and to accommodate donors.3 · Diminished the voice of TAC in priority setting Renewal stabilized the System's finances, did not and quality control result in any fundamental reforms, and expanded · Reduced the strategic use of World Bank fi- the System's management superstructure. Chang- nancing in ensuring the provision of global ing the allocation of the World Bank's financial con- and regional public goods research tribution from a "donor of last resort" model to a · Increased the fragmentation of the CGIAR re- matching grant formula unwittingly reinforced a search portfolio tendency for funding to drive programs, rather · Increased dependence on three donors (World than for strategic priorities to drive funding, and Bank, United States, and Japan) to about 60 per- gradually de-linked System-wide priorities from re- cent of the unrestricted funds that finance Sys- source allocation. The results of the matching tem overheads. grant model vary among the stakeholders. From the perspective of the CGIAR Chairman, From the perspective of the Centers, the match- it: ing grant formula: · Required presenting to donors a menu of op- · Increased incentives for Centers and donors to tions for them to finance, in order to mobilize negotiate directly for funding, leading to a the additional resources that the stakehold- balkanization of fund-raising ers expect the chairman to deliver, a conse- · Increased financing and accounting require- quence of the current political realities of ments that donors imposed on Centers agenda-based, small-scale, decentralized fund- · Pulled Centers downstream and shifted re- ing search into short-term or nonstrategic areas · Increased the number of committees needed · Increased micro-management of research port- to achieve consensus on the System issues folios by donors without a decisionmaking mechanism to follow · Led Centers to conduct some research inap- through on committee suggestions. propriate for a global or regional research sys- tem, despite its worthwhile value to specific The "corporate model" recommended donors or civil society by the Third System Review (TSR) to for- · Caused Centers to spend more time and re- malize decisionmaking, transparency, and sources preparing and reporting on specific accountability was rejected for being "top- projects and meeting individual donor re- down" and contrary to CGIAR's founding quirements, thus reducing time devoted to principles. The TSR addressed two broad areas-- research. science and strategy for the new millennium, and improving the organization, governance, From the perspective of the donors, it: and management of the CGIAR. It recognized many of the problems outlined in this meta- · Maintained the CGIAR's attraction in their evaluation and recommended radical reforms countries, relative to other programs to address them. These included establishing a · Increased a sense of ownership among indi- centralized body with the necessary authority and viduals in those agencies who worked hard to responsibility to deal with IPR issues and public- secure CGIAR funds private partnerships, and divesting the CGIAR's · Created a vested interest in donor agencies current "downstream" activities to NARS of de- that supported the arrangement. veloping countries in proactive partnerships 2 2 R E S T R U C T U R I N G T H E C G I A R : L E S S O N S O F PA S T AT T E M P T S with them to increase the efficiency and rele- ating IPRs and partnerships with the Federation, vance of the CGIAR. which would be a legal body, move some strate- Unfortunately, the TSR had little ownership gic planning functions from TAC, increase the among the membership. CGIAR stakeholders voice of NARS and developing regions in prior- criticized the review process for being costly, ity-setting, and unify the CGIAR Secretariat with lacking in analysis, and not involving them ade- the Federation on the grounds that the CGIAR quately--even though the review had conducted Secretariat had not served the Centers well. substantial consultations with the CGIAR Center Given its profound implications for the or- directors, boards, donors, and members.4 The ganization and governance of the System, the recommendations on governance and manage- proposal needed thorough discussion. A few key ment clashed with vested interests and with the donors saw the bottom-up Federation as taking CGIAR culture of consensus decisionmaking. away donor prerogatives on System-level ac- Ultimately, the recommended reforms made countability and on strategic planning through the slow progress because there was no systematic existing units such as the CGIAR Secretariat and mechanism in the major donor agencies to con- TAC. Reportedly, one developing country ob- sider and follow-up on them. Nevertheless, in ret- jected. But the Federation proposal too was rospect, the TSR had a significant effect on the dropped. Those who designed and advocated it CGIAR by exposing the System to high-level sci- contend that the concerns that prompted the ence managers and private sector actors in in- proposal (sharing of the CGIAR Secretariat costs, dustrial countries, by reaffirming donor support gearing up organizationally for private sector part- for the System, by formally broadening the nerships, and mobilizing additional resources) CGIAR mission to include integrated natural re- and the momentum around them were altogether source management research, and by offering les- lost in the subsequent CDMP. Some of these issues sons for the current CDMP. were never taken up by the CDMP. In retrospect, A subsequent "Federation proposal" by this can be seen as a loss of two years in the re- the CGIAR board chairs and Center direc- form process. tors was opposed because it was "bottom- The CGIAR's founding principles are un- up" and would have increased Center suited to ensuring poverty impacts in a control at the expense of donors and the changed environment. The founding princi- CGIAR Secretariat. Following the CGIAR mid- ples that underlie the CGIAR were adopted when term meeting in May 2000, an electronic TAC con- the System consisted of fewer Centers and less ference to collect ideas for reforms resulted in diverse constituents, and setting priorities to a proposal to unify management of some func- achieve poverty impact (through its governance, tions related to germplasm, intellectual prop- management, and financing processes) was rel- erty, and public awareness. Then in response to atively simple. But in today's more politically the DGF demand that the CGIAR Secretariat driven authorizing environment, and with a wider costs and overall funding become more diversi- research agenda and expanding membership, fied, the CGIAR board chairs and Center direc- the CGIAR's ability to address its mission is now tors came up with a more ambitious proposal for undermined by the six founding principles, a "Federation of Centers" comprising the 16 which exacerbate the System's collective action Centers, a Federation office, and a centralized problem (box 4.1). In particular, both the TSR board.5 The proposed Federation would per- and the CGIAR board chairs/Center directors' form six functions: (1) strategic planning and Federation proposal ac- science quality enhancement, (2) resource mo- knowledged the need CGIAR's ability to bilization, (3) public awareness, (4) science ad- for a legal entity with a address its mission is vocacy, (5) Federation undertakings, and (6) centralized board to en- now undermined by providing common services to Centers, donors, able System-level re- and the CGIAR chairman. The Centers would sponses (although they some of its six founding give up certain powers, for example, in negoti- differed in other re- principles. 2 3 T H E C G I A R AT 3 1 T h e C G I A R ' s F o u n d i n g P r i n c i p l e s N e e d B o x 4 . 1 R e v i s i t i n g t o M a i n t a i n I m p a c t o n P o v e r t y R e d u c t i o n Donor sovereignty--Various interests in donor domestic and the lack of consensus has become an excuse for failing to constituencies have encouraged members to tie their contribu- reach resolution on important issues. tions to specific regions of the world, Centers, and programs, and to their own national personnel or institutions. While broaden- Independent technical advice--Undermining this de- ing the CGIAR's political support, this has created a chaotic sirable principle by increasing restricted funding has dimin- marketplace for global public goods research and shifted the ished the authority of the independent technical advice of the composition of the overall program from strategic research to TAC (now the interim Science Council) in priority setting and re- development and dissemination activities tied to short-term source allocation. donor agendas, in which the CGIAR does not have a compara- tive advantage or core competencies. Informal status of the System--The lack of memo- randa of understanding, constitution, legal status, or explicit by- Center autonomy--Having 16 research Centers as the only laws at the System level has constrained the ability of the independent legal entities governed by self-nominating boards CGIAR to speak with a single voice and to develop System-wide has increased fragmentation, inter-Center rivalry, and board policies and long-term strategies. Even with the newly estab- membership (to 220 members); diluted board accountability and lished ExCo, its informal status is ill-suited to rapid changes in responsibility for quality; and contributed to collective action science, to the increasing role of the private sector and intel- problems. lectual property, and for determining accountabilities and re- sponsibilities. Consensus decisionmaking--Expanding and diversifying membership has broadened ownership and provided seats at the Nonpolitical (nonpartisan, nonideological) nature-- table for the private sector and NGOs. But this has made it dif- The need to raise resources for a wider mission has revealed ficult to reach consensus among members on the governance, the different priorities of the constituencies within each of the organization, management, and financing decisions needed to industrial and developing countries, and has undermined the achieve the CGIAR's mission. Every member effectively has a veto, CGIAR's non-political nature. spects). Yet neither proposal for a legal entity was committees of TAC, and the CGIAR's own internal fully considered by the membership. Unlike reviews have proposed many types of consolida- more recent global programs such as the Global tions. A large majority (79 percent) of CGIAR stake- Environment Facility or the Global Fund for holders surveyed by OED agree that consolidating AIDS, TB, and Malaria, the CGIAR System has no the number and functions of Centers is advisable formal or legal persona, written charter, or even (box 4.2). African countries as a group have called a memorandum of understanding. The only legal for consolidation to reduce the transactions costs entities in the CGIAR are the 16 Centers. Re- of dealing with 16 Centers. The System is being sponsibilities and accountabilities remain ill-de- pulled in two opposite directions. On the one fined as the considerable superstructure of hand, the CGIAR Centers are not conducting suf- committees has evolved over the years to deal ficiently coordinated research on the highly de- with a complex set of stakeholders. As a result, centralized nature of NRM research, which calls greater responsibility for managing the overall for effective partnerships with NARS to produce System has accrued by default to the World Bank regional and national public goods in NRM. On the and the CGIAR Secretariat. other hand, the System is not sufficiently cen- Resistance to consolidation demonstrates tralized to deal with advances in the biological sci- the Olsonian, collective action dilemma ences and IPRs, which call for a more unified (Olson 1965). Since 1994, stakeholders, scientific approach to research strategies and policies. 2 4 R E S T R U C T U R I N G T H E C G I A R : L E S S O N S O F PA S T AT T E M P T S S t a k e h o l d e r s R e p o r t M a n y C o n c e r n s B o x 4 . 2 a b o u t t h e S y s t e m In support of the meta-evaluation, OED administered a ques- · 51 percent indicated that TAC's scientific quality has de- tionnaire to solicit input from 235 CGIAR stakeholders and clined this past decade; 23 percent disagreed; 26 percent did outside observers in December 2001. Thirty-three percent not know. (mostly insiders) responded to the questionnaire. Among the · 54 percent said that the Science Council should have the responses: lead in System-level priority setting; 30 percent disagreed; and 16 percent did not know. · 68 percent believed that "knowledge of germplasm and · 79 percent believed that a consolidation in the number and germplasm research appropriately sensitive to agro-ecolog- functions of Centers is advisable. ical conditions" is the core competency of the CGIAR. 38 per- · 33 percent believed that the Challenge Programs are the best cent said that the CGIAR should focus primarily on this area approach to achieve consolidation. of research; 55 percent disagreed. · 33 percent agreed that the Challenge Programs are sufficient · 99 percent agreed that the Centers should pursue meaning- to open up the CGIAR to obtain/produce the best science. ful collaborative partnerships with strong developing coun- · 77 percent indicated that the World Bank and some other try NARS in strategic research and help build capacity of the donors' financing of overhead costs of Centers has ensured weaker developing country NARS. 70 percent said that the stability of the System and enabled the CGIAR to focus on a Centers are not doing enough in this area. longer-term research agenda. · 79 percent agreed that there should be a System-wide policy · 22 percent said that the recent changes emerging from the on intellectual property right (IPR) matters. Change Design and Management Process go far enough. 53 · 67 percent believed that TAC's role in priority setting has de- percent said they do not go far enough; and 25 percent did not clined in the past decade. know. Source: See Annex O in the Annexes (Part 3 of this study), www.worldbank.org/oed/cgiar. The collective action problems are daunting. tralized and others decentralized. Functions to be Individual countries that host Centers oppose centralized include System-level priority setting consolidation, as do donors, Center chairs and and quality assurance on science; policies and boards, scientists, and Center directors of "their" strategies concerning IPRs and public-private Centers. Yet the long-run benefits of consolida- partnerships; and monitoring implementation tion will undoubtedly be significant. The CGIAR of these policies and strategies. Functions that has some successful examples of consolidation, should remain decentralized include Center-level which in the livestock sector have positioned the priorities, planning, implementation, and quality CGIAR to address global issues.7 Getting key assurance of science, and collaborations, training stakeholders to agree that the time has come for and other activities. In addition, some functions both real reform and organizational change re- could be decentralized from the Centers to NARS, mains a major challenge. the private sector, and other suppliers of services The skills, expertise, and scale economies in- in areas where they can conduct them more cost- volved have led previous reviews to recommend, effectively, increase efficiency of resource use, and OED concurs, that some functions be cen- and assure greater development impact. 2 5 5 The Current CGIAR Reforms: Some Are Creditable But Others Need Revisiting T he CGIAR initiated its current reforms at International Centers Week in October 2000. Under the leadership of the incoming Chairman, Ian Johnson, the Group established a Change Design and Management Team (CDMT) to recommend concrete proposals for change. Consisting of both CGIAR stakeholders and professionals from outside the System, and chaired by Margaret Catley-Carlson, former president of the Canadian International Development Agency (CIDA), the CDMT issued its report in April 2001 for con- sideration at the Mid-Term Meeting in May 2001. Learning from the experience of the Third Sys- stakeholders surveyed by OED believe they do tem Review, the CDMT consulted widely to en- not go far enough (box 4.2). hance greater ownership of the reform process, At the Mid-Term Meeting in May 2001, the but it carried out little objective analysis of the Group adapted the CDMT's proposals into four System and its functioning.1 Seeking "internal tol- actionable areas, each of which has become one erance" and "quick wins," it avoided some of pillar of the current reform effort. First, the the most contentious issues in its terms of ref- CGIAR established an Executive Council to erence, including a restructuring action plan for improve the efficiency of decisionmaking and im- the entire System based on a clear rationale for plementation. Second, it initiated a System Of- integrating programs and/or consolidating Cen- fice to bring together the CGIAR's previously ters. Dropping the idea of merging Centers or uncoordinated and independent administrative creating a "Federation of Centers," it opted for and management units. Third, it proposed to an "evolutionary approach" in which restruc- transform TAC into a Science Council in order turing would emerge from other reforms. Given to improve the quality of science. Fourth, it has the past difficulty of changing the System, the re- established Challenge Programs to, in the forms are significant. Yet many of the CGIAR words of the chairman, "elevate the game to ad- 2 7 T H E C G I A R AT 3 1 The fundamental dress issues of global accountability that the CGIAR's own reviews challenge is to and regional signifi- have detected in their self-nominating Center appropriately balance cance." The CGIAR also boards. Given the importance of ExCo, an effort eliminated its Mid-Term should also be made to improve the quality of legitimacy and Meeting, dissolved developing country and other stakeholder par- efficiency. many of its committees ticipation, and to monitor its accountability to and then reconstituted stakeholders. Without these steps, the author- some of them,2and up- izing environment and legitimacy of ExCo will be graded the position of undermined, and the CGIAR's collective action CGIAR director from executive secretary, giving problem will not be reduced.5 There is currently it more executive powers. (See Appendix 1 for a lack of clarity as to whether ExCo is a decision- the current organizational chart of the CGIAR, making body or a body that proposes decisions subsequent to these changes.) to be considered by the membership at the An- The chairman gets high marks for es- nual General Meeting. tablishing the long-overdue Executive The System Office has promise, although Council (ExCo). ExCo members and others it is a work in progress. The System Office interviewed by the meta-evaluation team at the links 10 independent units in order to increase Annual General Meetings in 2001 and 2002 are their coordination and ability to serve the Cen- optimistic that this is a step in the right direction, ters and membership.6 Issues of authority, re- but they are reserving judgment until they see sponsibility, and accountability between the how ExCo operates. The fundamental challenge CGIAR Secretariat (which is the coordinating is to appropriately balance legitimacy and effi- unit), the Centers, and other units are still being ciency. That ExCo is a stakeholder committee sorted out. Monitoring is required of the System rather than a shareholder committee enhances Office's functions, resources, responsibilities, its legitimacy. All stakeholders are represented-- accountabilities, and performance through rou- both developed and developing countries, the tine evaluations to ensure its effectiveness in three cosponsors, foundations, the Center chairs serving the members and clients. Having lost and directors, TAC/SC, and GFAR, as well as civil the momentum on divesting certain Center re- society and the commercial private sector.3 While sponsibilities to the System level--as proposed ExCo members are selected through caucuses of in the CGIAR board chairs/Center directors' Fed- stakeholder groups and can only make decisions eration proposal--there is a concern that the Sys- on matters delegated to them by the member- tem Office may be duplicating services already ship, they are not formally accountable to those performed at the Center level. groups or obliged to solicit the views of their con- The transformation of the Technical Ad- stituent groups before decisions are made. In ad- visory Committee (TAC) into a Science dition, most developed country groups are more Council (SC) raises many questions. His- organized than developing country groups, and torically, TAC played a powerful role in the only members "in good standing" (whose annual CGIAR's governance and organizational struc- membership dues of $500,000 are paid in full) ture by setting System-level priorities; recom- are eligible for ExCo membership. These fac- mending allocations of resources among Centers, tors limit the effectiveness and extent of devel- programs, and activities; monitoring budgets; oping country membership.4 conducting Center-level and System-level re- OED concurs with the proposal that the TSR views; and, more recently, assessing impacts.7 had made for an executive committee with for- However, the TAC's influence declined during the mally elected members accountable to the par- 1990s (box 4.2). In the view of the meta-evalua- ticular groups they represent. This is necessary tion team, its advisory committee, and certain key to increase the legitimacy and effectiveness of donors, the most significant reasons for this de- ExCo. Representation will avoid some of the cline have been the rise in restricted funding problems concerning quality, responsibility, and and the change to a matching grant formula for 2 8 T H E C U R R E N T C G I A R R E F O R M S : S O M E A R E C R E D I TA B L E B U T O T H E R S N E E D R E V I S I T I N G the allocation of the Bank's resources.8 Together bership when they are transmitted to ExCo. Yet these have allowed donor preferences to drive the Working Group on the Science Council has programs and decoupled resource allocation not determined the time and financial resources from TAC's medium- and longer-term priority set- needed, or the remuneration arrangements for ting. Some donors have also acknowledged to the Science Council members and its chairman. the meta-evaluation team that their priorities The Challenge Programs (CPs) are not are influenced by domestic constituency inter- addressing issues of System-level funding, ests, which tend be short-term, while research priority setting, science quality, and gov- is a long-term endeavor requiring steady, long- ernance. The CDMT recommended the cre- term commitments that are not necessarily po- ation of CPs as high-visibility, time-bound litically popular. partnerships to address complex issues of global Hence, the TSR recommended strengthening or regional significance. These would increase the TAC, and the CDMT recommended transforming scope for inter-Center collaboration, facilitate a it into a Science Council in order to "ensure that wider range of partnerships, tap new sources of the science practiced in the System meets world funding from current and new donors, and im- class standards." OED strongly agrees with this prove output accountability. The CDMT and the objective. The CGIAR needs a strong, qualified, chairman also viewed CPs as a way of address- and independent Science Council to set overall ing consolidation indirectly by building new pro- System priorities, to support ExCo, and to ensure grams and partnerships while, in the words of the quality and impact of all System-level pro- some interviewees, letting the weak Centers and grams, including the Challenge Programs.9 The programs "die on the vine" rather than face po- decline in independent scientific advice in the litically unpopular consolidation. Their design CGIAR went hand-in-hand with the decline in the was also intended to address three important strategic nature of the CGIAR's research during shortcomings that had plagued previous the- the 1990s. But OED is not convinced that the matic and eco-regional System-wide programs: transformation of TAC into a Science Council insufficient funding, poorly defined timelines, will achieve the desired objective for a number and a lack of adequate governance and man- of reasons. agement. In October 2002, after a year-long re- First, this transformation is taking place with- view process involving both the interim Science out the TAC having been evaluated while major Council and ExCo, the Annual General Meeting decisions on resource allocations were being approved the implementation of the first two made. Second, relative to TAC, the Science Coun- pilot CPs: Water and Food and Biofortified Crops cil's role is greatly diminished. It is expected to for Improved Human Nutrition.12 While they focus mainly on science quality; to have a more passed scientific review, these two pilot pro- limited (if any) role in priority setting, medium- grams do not address the three shortcomings of term planning, and monitoring resource alloca- thematic and ecoregional programs (funding, tion; and to play no role in the annual financial timelines, and governance) noted above. planning process.10 Third, the FAO has indicated OED believes that, properly developed, the to the meta-evaluation team that it is not being CPs could enhance the CGIAR's effectiveness adequately consulted about the role of the Sci- and impact. One of their strong and positive ence Council. Fourth, the roles and responsibil- features is that they are helping to open up the ities of the Science Council vis-à-vis the Challenge System by promising to allocate substantial re- Programs and the new and emerging Finance sources to advanced research institutions and the and Program Committees are unclear, particularly NARS of developing countries. Yet, only one- as the permanent Science Council is not in place third of stakeholders surveyed by OED believe while major decisions on Challenge Programs the CPs are sufficient Challenge Programs are are being made.11 Fifth, Science Council members to open up the CGIAR, helping to open up the need to be independent and objective, and their to produce the best findings should be shared with the full mem- science, or to achieve System. 2 9 T H E C G I A R AT 3 1 allocation of Bank funds to CPs is a pivotal con- cern, given that Bank funds are unrestricted and have traditionally provided the System with a seal of approval for science quality and man- agement. Instead of allocating Bank funds to areas other donors are most willing to finance, the Bank should use its funds to achieve strate- gic reforms in the System as a whole. Strategic Priorities. The tension between topics that can generate funds and those that are needed to ensure impacts on the largest num- ber of the poor has been evident from the out- set of the CPs. The CP process is becoming time-consuming, and their proposed sequential approval is distracting from System-level prior- ities and strategies.15 Past reallocations, based partly on the revealed preferences of donors, have led to reductions in productivity-enhancing consolidation. OED's interviews with CGIAR strategic activities of a global or regional public members reveal much ambiguity and confusion goods nature and in research on commodities of about the CPs and a range of concerns about their importance to the poor.16 The Interim Science development--concerns articulated by mem- Council (iSC) therefore recommended to ExCo bers at Annual General Meetings (AGMs) in 2001 in October 2002 that the introduction of new CPs and 2002--although some supported moving into the CGIAR research agenda be slowed down full speed ahead with CPs.13 to enable the CGIAR to (1) review the evaluation Funding. The CGIAR leadership has sought criteria designed by the Task force on CPs and far higher funding for CPs than was originally un- endorsed by the Group at AGM 2001, (2) learn derstood when the idea of CPs was approved-- lessons from the pilot process, and (3) review the $82 million for Water and Food, $42 million for implications of the CPs for System-level resource Biofortification, $69 million for Unlocking Ge- allocations. OED considers this an important netic Diversity, and $100 million for Sub-Saharan signal that should be acted upon before approval Africa. Center directors have stressed to OED that of any additional CPs. Researchers are trying to each fully funded CP is equivalent to an estab- convert almost every important research theme lishment of an additional new Center. Moreover, into a CP as a way of raising its profile and mo- the CPs were to be financed with additional bilizing funds. funds, but initial pledges for the first two are far Science Quality. The approved CPs do not below their costs.14 If additional funds are not sufficiently spell out their methodology and lack forthcoming, then the CPs will either be under- well-defined end points.17 The announcement of funded or funded from existing programs. Re- large sums by donors for individual programs in- ducing core funding to the Centers poses the volving individual Centers in advance of the review danger of process compromises the Science Council (SC) The tension between topics that further un- review process for scientific quality. The uncertain can generate funds and those that dermining role of independent scientific advice in relation are needed to ensure impacts on the CGIAR's to the CPs is exemplified by the limited resources p r o v e n allocated to the interim Science Council for eval- the largest number of the poor germplasm uating concepts, pre-proposals, and proposals has been evident from the outset i m p r o v e - for the pilot CPs.18 Related concerns are the cur- of the Challenge Programs. ment pro- rent and future roles of the SC, and the roles of gram. The ExCo, the Finance Committee, and the Program 3 0 T H E C U R R E N T C G I A R R E F O R M S : S O M E A R E C R E D I TA B L E B U T O T H E R S N E E D R E V I S I T I N G Committee relative to the SC, and of the CPs issued. The ad hoc nature of the governance and themselves in the context of the System (Dal- management arrangements in the pilot programs rymple 2002). Without the Science Council's are nevertheless establishing precedents in stan- strong and independent analytical input, ExCo, the dards and procedures. CGIAR Secretariat, and the Finance and Program Global Public Policy Issues. It is unclear Committees are not equipped to deal with the ap- how the CPs will address the key global public pol- propriateness or science quality of proposals or icy issues associated with public-private partner- with the role of CPs in System-wide priorities. ships and intellectual property rights identified in Nor is the membership likely to be able to make this report. The Biofortification CP has only par- informed decisions without timely, widely shared, tially addressed this. Avoiding undesirable out- high-quality input from a strong SC with well-de- comes will require System-wide priorities, fined, transparent procedures. monitoring of individual CPs, and assessing the op- Governance and Management. The gov- portunity cost to developing countries of under- ernance and management structures proposed taking one program relative to another. The for individual CPs vary, seem ad hoc, and pose CDMT's recommendation that one-half of CGIAR concerns about transparency, accountability, likely research be moved to CPs by 2006, while making science quality, and ultimately about science lead- those programs responsible for raising their funds, ership. CPs and their evolving reporting arrange- seems premature. It appears prudent not to ap- ments (as reported at AGM 2002) seem to be prove any more CPs until a new Science Council autonomous. For example, the Biofortification CP is in place, and its roles, responsibilities, ac- proposal states that its independent Project Ad- countabilities, independence, reporting arrange- visory Committee will make an annual progress ments, and supporting resources in the context report to the SC, and that the two sponsoring of the rest of the System are fully established. Center directors will report to ExCo. Yet guide- The Bank should act as a leader in helping to set lines on reporting arrangements have yet to be high standards for the conduct of the CPs. 3 1 6 The Critical Roles of the World Bank: Convener, Donor, and Development Partner T he Bank has played multiple roles in the CGIAR--as convener (founder and cosponsor), as donor to the System, and as a lender to develop- ing countries for complementary activities. Convener. Historically, leadership by individuals As one of four cosponsoring agencies, the shaped the CGIAR. But the CGIAR's increased Bank has fully financed and housed the CGIAR and diverse membership, its need to mobilize Secretariat, and contributed to TAC (now the and maintain political and financial support interim Science Council) and other System-level from a diverse constituency, its more complex support units. However, the Bank's financing of authorizing environment, and its competing the CGIAR Secretariat does not yet accord with priorities have made leadership by individuals new guidelines adopted by the DGF in June more difficult. Former Bank president Robert 2000. These guidelines state that the Bank should McNamara was a key player in founding the Sys- not fund more than 50 percent of in-house sec- tem, and every Bank president since then has retariat costs in order "to avoid a program's over- been strongly committed to the CGIAR. All but reliance on the Bank."1 OED would go even the first CGIAR chairman have been World Bank further. All major donors should share the full vice presidents responsible for agriculture and costs of running the CGIAR Secretariat. rural development. Each has been a strong, While donors have been interested in the committed champion, advocating the CGIAR Secretariat's efficacy, they indicated to OED that cause and mobilizing resources for it. The Bank's they have been less concerned with its efficiency chairmanship has given the CGIAR access to because they do not contribute financially to its governments at the highest levels, ensuring operation, although some indicated that they continued political and financial support. The ac- would be willing to share CGIAR Secretariat tive leadership of the chairmen has expanded costs. The advantages of cost-sharing are wider membership, which donors and member coun- ownership and greater accountability of the tries value. CGIAR Secretariat to its shareholders; rein- 3 3 T H E C G I A R AT 3 1 forcement of the CGIAR partnership (the DGF funds from its contribution year-to-year to relieve rationale for its guidelines); enhancement of the unexpected shortfalls in other donors' contri- arm's-length relationship between the Bank and butions. But shifting to a matching grant for- CGIAR management; and reduced conflicts of in- mula in 1994 and allocating a significant portion terest between resource mobilization and re- of its resources to Challenge Programs in 2002 source allocation. The current CGIAR chairman have reduced the strategic use of Bank funds. has noted drawbacks to cost sharing, including Development partner. The Bank is the possible instability or unpredictability in Secre- largest lender to agricultural research and ex- tariat financing, accrual of savings to the DGF tension in developing countries, having com- rather than to the CGIAR, and likely diminished mitted $6 billion to 173 projects in 91 countries Bank influence. since 1971. In addition, as the largest lender to Donor. The Bank has been the largest fi- agricultural development--$85.6 billion to 1,770 nancial contributor to the CGIAR during the projects since 1971--the Bank has helped de- past 10 years, contributing $500 million (15 per- veloping countries build the policy, institutional, cent of total CGIAR resources) between 1992 and infrastructure foundation necessary to adapt, and 2001. The Bank's funds are the more valu- adopt, and generate technologies needed for able because they have been unrestricted, and rapid and broad-based productivity growth therefore indispensable in supporting both Sys- among small farmers. But new commitments to tem- and Center-level overheads and long-term agricultural projects and to research and exten- research agendas. The Bank also helped to sta- sion projects have declined dramatically from bilize the System by providing extra funds dur- their peaks in 1985 and 1992, respectively (fig- ing the 1993­94 financial crisis, and by advancing ure 6.1). Linkages between the CGIAR and the B a n k L e n d i n g t o A g r i c u l t u r e D e c l i n e d F i g u r e 6 . 1 D r a m a t i c a l l y i n t h e 1 9 9 0 s To all agricultural projects To research and extension projects 8,000 1,600 7,000 To all agricultural 1,400 projects 6,000 1,200 5,000 1,000 4,000 800 To research and 3,000 extension projects 600 2,000 400 1,000 200 0 0 1971 1973 1975 1977 1979 1981 1983 1985 1987 1989 1991 1993 1995 1997 1999 2001 Volume of new commitments in 2000 U.S. dollars Source: OED data. 3 4 T H E C R I T I C A L R O L E S O F T H E W O R L D B A N K : C O N V E N E R , D O N O R , A N D D E V E L O P M E N T PA R T N E R Bank's agricultural development strategies were burdened by other Conflicting political strong when the CGIAR was established. But in heavy managerial pressures and the need to an increasingly decentralized and demand-led responsibilities; maintain political support Bank, linkages between the CGIAR and Bank and (3) reporting for the System can come at operations have weakened. Bank staff serve on arrangements both Center boards, participate in External Program for the CGIAR Sec- the cost of reforms. and Management Review teams, and engage the retariat and the CGIAR in other matters, but usually on an ad hoc Bank that are fraught with real or potential con- basis. An effort to systematize this relationship flicts of interest. These features limit the capac- is underway, as is an inventory of CGIAR­World ity of the Bank to provide the objectivity and Bank linkages. Looking ahead, the role of the leadership needed for far-reaching reforms of the CGIAR in implementing the MDGs has major CGIAR and to allocate the Bank's financial con- implications for policy dialogue and lending to tribution in a strategic way. As a result, the Bank agricultural research and development. has been more successful in using its convening The Bank's role is pivotal. As a result of its power to raise additional resources for the Sys- multiple roles, the Bank has been described to tem than to provide strategic leadership to the OED as the indispensable guardian of the CGIAR, CGIAR. and as the "glue" that makes the System coher- Conflicts of interest and inadequate over- ent and larger than the sum of 16 research Cen- sight constrain the Bank's strategic lead- ters. The Bank's leadership role, its financial ership. Many donors have indicated that they contributions, and its operational support are want the Bank to continue to chair the CGIAR, viewed by other donors as a seal of approval, in part because they believe Bank and other giving them confidence to continue to invest in donor funding will decrease if the chairmanship the System. But some argue that the Bank has not is moved outside the Bank.3 However, it is prob- exercised the influence others say it possesses lematic for the chairman to be both judge and through its multiple roles as convener, donor, and advocate--to acknowledge the need for and fund-raiser for the System as a whole. It has not press for major reforms while also making the displayed the leadership needed to stimulate case for continued funding to the Bank and long-term reforms of the System, which is fraught donors. Conflicting political pressures and the with collective action problems in a highly com- need to maintain political support for the System plex external environment. Donors and Centers can come at the cost of reforms in the System. have stated that there has been too close a rela- Having a chairman who is not at arm's length tionship between the World Bank and the CGIAR from the Bank can compromise the Bank's abil- Secretariat without sufficient accountability to ity to press for reforms on a scale or speed that the members and stakeholders at large. In the vac- might be warranted, and can increase its expo- uum created by the absence of an empowered sure and risks (box 6.1). TAC/SC and the uncertain and evolving role of There is currently no effective independent ExCo, over-reliance on the CGIAR Secretariat oversight within the Bank of its involvement in poses the risk of creating conflicts of interest in the CGIAR (as is also the case for a number of resource mobilization, policy and strategy for- other global programs), nor a mechanism to mulation, and resource allocation.2 conduct, assess, or follow up on System-level The multiplicity of roles the Bank has as- evaluations.4 The ESSD vice president has, among sumed has led to (1) excessive Bank involve- other things, wide latitude to determine how ment in the day-to-day management of the the Bank's contribution will be spent, given a lack System and dependence of the System on the of guidance from managing directors or the DGF Bank, yet little use by the CGIAR of the Bank's on how the DGF grant should be allocated. It is country assistance capacity; (2) a dispropor- understandably difficult for the rural research ad- tionate share of CGIAR management responsi- viser to exert oversight over his own manager. bility allocated to a Bank senior manager, already Beyond sponsoring the CGIAR's annual appli- 3 5 T H E C G I A R AT 3 1 C o n f l i c t s o f I n t e r e s t i n t h e B a n k ' s R o l e s B o x 6 . 1 I n h i b i t R e f o r m s The ESSD vice president: · Represents the ESSD vice presidency on the DGF Council · Chairs the CGIAR · Reports to the ESSD vice president. · For fund-raising, represents the CGIAR to the Bank · To the CGIAR donors, represents the Bank's imprimatur of The Bank's Rural Research adviser: good housekeeping for the System. · Is responsible for Bank oversight of the CGIAR · Reports to the ARD Director and ESSD Vice President. The Agriculture and Rural Development (ARD) director: · Serves as the Bank's cosponsor representative to the CGIAR The CGIAR director: · Officially represents the Bank in the CGIAR and ExCo · Reports to the ESSD vice president. · Chairs the ExCo Finance Committee cation to the DGF, the Rural Sector Board has · Bring about fundamental reforms in the System played little role in monitoring the use of the · Mobilize more unrestricted funding funds from a strategic perspective, and is unlikely · Ensure that Bank resources are being allocated to have the will to displease its chairman. Further, to supplying global and regional public goods while the Bank used to conduct periodic finan- for the benefit of the largest number of poor cial audits of the CGIAR, the insightful 1995 in a catalytic way audit was the last of its kind (although the In- · Ensure that DGF funds are generating the ternal Audit Department is currently helping in- greatest impact? dividual CGIAR Centers to build up their auditing capacity). Because many donors have indicated The Bank should fully exercise its leadership that they prefer the Bank to continue to chair the role, if necessary by convening the best exter- System, independent oversight may need to be nal expertise to bring about additional reforms exercised at the managing director level. Alter- in the System. If additional reforms are natively, an eminent person could chair the achieved, there is a strong argument for in- CGIAR, with oversight exercised by the ESSD creased funding for the CGIAR, including ex- vice president. In either case, the Bank should ploring the use of grant funds for the provision exercise a degree of oversight consistent with the of regional public goods, and eventually global major roles that it plays in the CGIAR. public goods, that reduce poverty.5 As a lender The long-standing allocation to the CGIAR of to developing countries, the Bank also needs to large amounts of DGF funding and the lack of increase lending to agricultural research, edu- an exit strategy for the program have raised cation, extension, and training, especially in concerns within the Bank about the different Sub-Saharan Africa, in order to enhance the per- standards that apply to the CGIAR compared formance of NARS. with other programs. The relationship between OED also concludes, based on the evidence the DGF and the CGIAR and the extent of ob- the meta-evaluation team has assembled, that the jective oversight of such large sums need to be rapidly growing but under-evaluated areas of clarified. policy and NRM research and the CGIAR's ca- The $50 Million Question. How, then, can pacity building activities would benefit from an the Bank, as one independent impact evaluation. Yet it would be If additional reforms are partner (albeit a more appropriate to conduct such an evalua- achieved, there is a strong key partner), use tion after the more fundamental, longer-term its financing and issues related to System-level strategy, gover- argument for increased its convening nance, organization, management, and finance funding for the CGIAR. power to: identified in this report are fully vetted. 3 6 7 Recommendations Focusing the World Bank's ence Council. This will require the Bank to as- Responsibilities sure itself that a strong, qualified, and inde- pendent Science Council is established and 1. As a convener and donor to the CGIAR and lender vested with the role and resources to establish to developing countries, the Bank should address its System-wide priorities, policies, and strategies, corporate governance responsibilities in the man- and to monitor and report to the member- agement of the CGIAR and exercise a degree of over- ship on the uses and allocations of CGIAR re- sight consistent with the major roles that it plays in the sources toward fulfilling these priorities, CGIAR. This will require the Bank to: policies, and strategies. · Carry out independent triennial appraisals of · Lead a concerted effort at the highest level, the CGIAR, with Board approval as the basis of much as when the CGIAR was established, to continuing Bank support. achieve fundamental reforms in the organiza- tional structure, finance, and management of Reforming the CGIAR the CGIAR--particularly to encourage donors to reverse the trend in restricted funding and 2. The strategic priorities of the CGIAR should respond to establish clear targets for an increased share more actively to changes in the global research con- of unrestricted funding. text, giving more prominence to basic plant breeding · Separate oversight and management functions and germplasm improvement and reshaping natural re- within the Bank to address the conflicts of in- source management research to focus tightly on pro- terest among the roles of the ESSD vice pres- ductivity enhancement and sustainable use of natural ident, the ARD director, the research advisor, resources for the benefit of developing countries. This and the CGIAR director. will require CGIAR to: · Abandon the current matching grant model; re- port to the Board on a regular basis on the im- · Postpone the approval of new Challenge Pro- pact of the allocation of the Bank's resources grams (beyond the first two already approved) on the incentive structure of the System; and pending the installation of the new Science ensure that the Bank's resources are allocated Council, an assessment of System-level prior- strategically in support of global and regional ities, and a thorough review of the design and public goods that contribute to agricultural approval process of the first two programs to productivity and poverty reduction, based on learn lessons for the selection, design, se- long-term priorities as articulated by the Sci- quencing, and phasing of future Challenge 3 7 T H E C G I A R AT 3 1 Programs in the context of System-level pri- 3. The governance of the CGIAR should be reconfigured orities and strategies. to promote greater efficiency, tougher priority setting, · Increase funding for conventional germplasm and scientific excellence without sacrificing legiti- enhancement and plant and animal breeding macy and ownership. This will require the CGIAR to: research, in which the CGIAR possesses a com- parative advantage; conduct an independent re- · Adopt a written charter that clearly delineates the view of NRM, policy, and social science research roles, responsibilities, and accountabilities of the from a global and regional public goods per- various officers and bodies that govern the Sys- spective to help it address country- and re- tem,aswellasamechanismtofurtherreformSys- gional-level issues constraining productivity tem governance. Analyze the advantages and enhancement and the sustainable use of nat- disadvantages of establishing all or part of the ural resources; and devolve that portion of CGIAR as a separate legal entity geared to deal the CGIAR's applied and adaptive NRM re- with today's partnership realities. search program that does not constitute global · Make ExCo members more fully representative or regional public goods research to national and accountable to the CGIAR membership and regional agencies, supported by substan- and have donors share in the costs of the tially larger funding for national and regional CGIAR Secretariat, the Science Council and its agricultural research and development from Secretariat, and other central bodies in the both developing country governments and CGIAR System. donors. · Increase the efficiency of the System, from the · Develop effective System-wide strategies and viewpoint of generating global and regional policies that facilitate businesslike partner- public goods, through appropriate consolida- ships with NARS, agricultural research institu- tion, decentralization, streamlining, and ab- tions, NGOs, and the private sector; strengthen sorption of marginally effective Centers, based the management and use of intellectual prop- on a management review of the organization erty and genetic resources; and use new sci- of Centers, programs, and science quality. entific areas such as biotechnology and · Ensure that there is a body that reports to bio-informatics to complement its conven- ExCo with responsibility for (a) annual Sys- tional research. tem-level audits and System-level external re- · Enhance collaborative research as a means views of the CGIAR every three to five years in of capacity building and training and en- consultation with the Science Council, and (b) gage qualified developing country NARS to enhancing transparency in the reporting of provide similar services to smaller and the System's expenditures to ensure the strate- weaker NARS. gic public goods nature of CGIAR research. 3 8 Part 2 Technical Report Section I The Context 4 1 8 The Context for the CGIAR Meta-Evaluation T his chapter defines the context for the meta-evaluation. It includes an overview of the CGIAR, a conceptual framework for assessing the CGIAR's effectiveness, a brief description of recent evaluations, and a brief overview of the current Change Design and Management Process. An Overview of the CGIAR national Board on Plant Genetic Resources Founded on May 19, 1971, the CGIAR originally (IBPGR). The new consolidated Center was re- consisted of four agricultural research Centers named the International Plant Genetic Resources established by the Rockefeller and Ford Foun- Institute (IPGRI). Today, there are 16 CGIAR dations: the Centro Internacional de Mejo- Centers conducting research on a variety of is- ramiento de Maíz y Trigo (CIMMYT) in Mexico, sues worldwide (figure 8.1).2 the International Rice Research Institute (IRRI) Membership has also expanded substantially in the Philippines, the International Institute of since the CGIAR's inception. The first develop- Tropical Agriculture (IITA) in Nigeria, and the ing countries to become members were two Centro Internacional de Agricultura Tropical OPEC countries--Nigeria and Saudi Arabia, in (CIAT) in Colombia. It had 20 members in 1971: 1975--and the first non-OPEC developing coun- 11 governments of industrial countries, 6 inter- try members were the Philippines and Mexico, national organizations, and 3 foundations. Its in 1980. As of January 2003, the CGIAR had 62 initial annual budget was about $20 million in members: 24 developing and transition countries, nominal dollars. 22 industrial countries, 12 international and re- By 1993, the CGIAR had 18 Centers, 39 mem- gional organizations, and 4 foundations. Mo- bers, and annual funding of $328 million.1 A rocco, Malaysia, Israel and the Syngenta funding crisis in 1993­94, along with other con- Foundation, the latest additions, joined the siderations, led to the merger of the Interna- CGIAR in 2002. tional Livestock Center for Africa (ILCA) and the Organizationally, the CGIAR System has four International Laboratory for Research on Ani- components: the Consultative Group (the mem- mal Diseases (ILRAD) into the International Live- bers); the 16 Centers; CGIAR committees (in- stock Research Institute (ILRI), and the merger cluding standing, advisory, partnership, and of International Network for the Improvement Center committees); and the CGIAR Secretariat. of Banana and Plantain (INIBAP) into the Inter- The Consultative Group is led by a chairman, his- 4 3 T H E C G I A R AT 3 1 F i g u r e 8 . 1 A n O v e r v i e w o f C G I A R C e n t e r s ISNAR The Hague, Netherlands ICARDA Aleppo, IFPRI IPGRI Syrian Arab Republic Washington, D.C. Rome, ICLARM ICRISAT United States Italy Penang, ITA Patancheru, Malaysia CIMMYT Ibadan, India Mexico City, IRRI, Nigeria ICRAF Mexico Los Baños, CIAT Nairobi, Philippines Cali, Kenya IWMI Colombia WARDA Colombo, Bouaké, CIP ILRI Sri Lanka Côte d'Ivoire Lima, Nairobi, CIFOR Peru Kenya Bogor, Indonesia torically a vice president of the World Bank (ex- informal status, and a nonpolitical (nonpar- cept for the CGIAR's first few years), nominated tisan, non-ideological) nature (see box 4.1). by the World Bank's president in consultation Some of these principles have come under con- with the group's cosponsoring agencies. The siderable stress and some are no longer rele- CGIAR has four cosponsors: the World Bank, vant in the current highly changed circumstances, the Food and Agriculture Organization of the as demonstrated in this report, yet they remain United Nations (FAO), the United Nations De- operative even in the context of the recent velopment Programme (UNDP), and the Inter- Change Design and Management Process national Fund for Agricultural Development (CDMP) introduced by the current CGIAR Chair- (IFAD).3 The principal advisory committee has man Ian Johnson, and discussed in Chapter 16. historically been the Technical Advisory Com- The CGIAR is a two-level financial system. The mittee (TAC), which has been housed in FAO, System level comprises the financial contribu- supported by a secretariat. It is now being trans- tors (including cosponsors, members, and non- formed into a Science Council (the existing TAC members), the Consultative Group itself, the has been operating as the interim Science Coun- CGIAR Secretariat, and the TAC/iSC.5 Tradition- cil since October 2001). In 2001, the group es- ally, TAC recommended System-level priorities tablished an Executive Council (ExCo) to facilitate and strategies, while the Centers developed their decisionmaking and implementation.4 own Center-level priorities and strategies within The CGIAR is an informal organization. It is the overall System context. TAC and the CGIAR remarkable that such a large program has oper- Secretariat have jointly carried out periodic Ex- ated for so long without a written agreement, ternal Program and Management Reviews charter, or formal definition of roles, responsi- (EPMRs) of the Centers and inter-Center the- bilities or accountabilities. Six main principles matic reviews, while TAC's Standing Panel on have governed the organization and evolution of Impact Assessment (SPIA) has had responsibility the CGIAR from the beginning: donor sover- since the second half of the 1990s for System-level eignty, Center autonomy, consensus decision- monitoring, evaluation, and impact assessment. making, independent technical advice, The Centers themselves have conducted similar 4 4 T H E C O N T E X T F O R T H E C G I A R M E TA - E VA L U AT I O N reviews of their own research activities. The in 1993 and 1994 The proportion of restricted CGIAR chairman, Center directors, and Center when the United funding increased from 36 boards have shared responsibility for fund- States and Canada re- percent in 1992 to 57 raising. The CGIAR Secretariat has provided sup- duced their funding percent in 2001. port services, including allocation of the Bank's by $24 million and $6 unrestricted funds (until 1994), in accordance million, respectively, with priorities established by TAC/iSC, channel- over the two-year pe- ing of Bank-managed trust funds to Centers, riod. The changes in monitoring, and overall reporting of the System- the financial arrangements adopted in 1994 sus- level resources to the membership.6 Over time, tained the aggregate level of support for the these roles and responsibilities have evolved con- System, although they fundamentally changed siderably, however. The full range of reviews of the way the System was financed.9 Among the top the CGIAR and its Centers is discussed later in this 10 donors, the World Bank made up $12.4 mil- chapter, and a detailed analysis of the evolution lion of this shortfall over the two-year period; and implications of the CGIAR's organization, Japan another $10 million; Denmark, $3.3 mil- governance, management, and financing are con- lion; and the Netherlands, $2.8 million (figure tained in Chapters 15 and 16. 8.2). Some diversification of funding has oc- The System as a whole received financial con- curred over the past 10 years: the share of con- tributions of $337.3 million toward its research tributors other than the top 10 increased from agenda in 2001, 95 percent of the total came 25 percent in 1992 to 34 percent in 2001. The top from members, and 5 percent from non-mem- three donors still contributed 35 percent of the bers.7 Overall contributions have been relatively total in 2001 compared with 41 percent in 1992, stable since the addition of four new Centers in and the top 10 contributed 66 percent (figure 1992--growing at an average annual rate of 0.7 8.3) compared with 75 percent in 1992. In 2001 percent in nominal terms, and declining by 1.8 the top three donors also supplied about 60 percent in real terms between 1992 and 2001 (fig- percent of the unrestricted funding that finances ure 2.2). However, the proportion of restricted Center overheads, gene bank operations and funding increased from 36 percent in 1992 to 57 maintenance, and longer-term research pro- percent in 2001, with most of the increase oc- grams, compared with 45 percent in 1992. curring since 1998. While Centers may allocate The World Bank provides three kinds of fi- unrestricted funds to any program or cost ac- nancial support to the CGIAR System--as a cording to a Center's institutional needs or pri- donor contributing more than $800 million to the orities, donors may restrict funds either by CGIAR research activities since 1971, as a cospon- attribution (to a particular research program or sor contributing $5 million a year since 1991 to region) or by contract (to a project, subproject, the operating costs of the CGIAR Secretariat and or activity). But the reporting arrangement on re- the TAC/iSC, and as a lender to the client coun- stricted funding is confusing at best. For exam- tries of the CGIAR.10 The first two combined ple, an allocation to one Center as opposed to make the World Bank the largest grantor to the another is not classified as restricted, since, in ac- CGIAR System during the past 10 years (con- cordance with the principle of donor sover- tributing $500 million between 1992 and 2001, eignty, donors allocate funds to the Centers of compared with $450 million for the United their choice (either directly or through a World States). The Bank's contribution toward CGIAR Bank trust fund), not to the System as a whole. research activities is more valuable than most It is presumed that the allocations to the new donors' contributions, since it is entirely unre- Challenge Programs will now be classified in the stricted. The Bank provided almost one-third of same way--as unrestricted, unless restricted by all unrestricted funding for CGIAR research ac- attribution or contract.8 tivities in 2001. The overall picture in figure 2.2 hides a fund- The overall picture in figure 2.2 masks sub- ing crisis (discussed in Chapter 15) that occurred stantial differences in research funding received 4 5 T H E C G I A R AT 3 1 T h e W o r l d B a n k B e c a m e t h e L a r g e s t D o n o r F i g u r e 8 . 2 i n 1 9 9 4 Millions of 2000 U.S. Dollars 140 Next 7 120 100 Others 80 United States 60 World Bank 40 Japan 20 0 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 Contributions to CGIAR research activities (adjusted for inflation) Source: CGIAR Financial Reports, 1988-2001. T o p T e n D o n o r s P r o v i d e M o s t o f t h e F i g u r e 8 . 3 F u n d i n g United States 13% Others 34% World Bank 13% Denmark Japan 3% 9% Canada EC 3% 6% The Netherlands United Kingdom 4% 6% Germany Switzerland 4% 5% Share of total contributions to CGIAR research activities, 2001 Source: CGIAR Financial Report, 2001. 4 6 T H E C O N T E X T F O R T H E C G I A R M E TA - E VA L U AT I O N by each of the 16 Centers. Allocations to the percent of total ex- The share of expenditures on four natural resource management Centers com- penditures in 1992 increasing productivity and bined grew by 4.2 percent annually over the pe- to 44 percent in germplasm enhancement is riod, allocations to the three policy and 2001, while those declining. capacity-building Centers grew by 2.5 percent, for West Asia and and allocations to the five commodity-oriented North Africa de- Centers and four ecoregional Centers, which clined from 12 percent to 9 percent. The shares largely conduct research on germplasm im- of Asia and Latin American and the Caribbean provement, declined by 3.3 percent and 4.2 per- have been relatively constant at 32 percent and cent, respectively (figure 8.4). 17 percent respectively. The CGIAR classifies the research expendi- In the early years, sustained investment in tures of the 16 Centers into five categories. the CGIAR stimulated investments in research at Among them, the CGIAR's research expendi- the national level in developing countries. Con- tures on increasing productivity11--widely versely, and by comparison, the limited invest- viewed as the traditional strength and compar- ment in health research at the global level is ative advantage of the CGIAR System--declined reflected in the limited investment in health re- by 6.5 percent annually in real terms between search in many developing countries (figures 1992 and 2001, while those on improving poli- 8.7 and 8.8). But this pattern has begun to cies and on protecting the environment (largely change. related to NRM) both increased by 3.1 percent The growth of complementary public sector annually (figure 8.5). Research expenditures on investments in agricultural research by devel- saving biodiversity grew by 2.7 percent annually, oping countries has slowed down from an aver- and on strengthening NARS declined by 0.8 per- age of 7.0 percent in 1976­81 to 3.6 percent in cent. Training, crucial for accessing global knowl- 1991­96. Across regions, public sector research edge, represents about 40 percent of expenditures appear to have rebounded in China "Strengthening NARS," and is considered by and Latin America during 1991­96, but contin- many NARS to be the most important contribu- ued their decline in the other three developing tion of the CGIAR after its germplasm research, regions (figure 8.9). Excluding the three large but it has declined by 0.2 percent annually.12 countries--China, India, and Brazil--that have Among the subcategories that make up "in- increased their agricultural research expendi- creasing productivity," research expenditures tures in recent years, the overall situation is even on fisheries and forestry grew by 3.4 percent worse (Pardey and Beintema 2001). During the annually in real terms between 1992 and 2001, same period, there has been a decrease in fund- and that on enhancing germplasm, livestock, ing from donor countries for agricultural re- and crop production systems declined by 5.6 search (figure 8.10). percent, 8.9 percent, and 9.1 percent respec- These trends are important reasons for the tively. This confirms the continuation, indeed, the CGIAR's limited impact in some regions, such as acceleration of trends that became apparent in Africa, and perhaps overall in more recent years. the 1980s toward a declining share of expendi- Without these complementary investments at tures on increasing productivity and germplasm the national level (and increasingly at the re- enhancement (Lipton 1999, p. 18). The Chal- gional and subregional levels, especially where lenge Programs that are being approved by the countries are too small to have a critical minimum CGIAR (and discussed later in this report) are mass of scientists), the CGIAR's role will remain projected by the commodity Centers to further particularly important, but its impact would likely reduce their budgets for research. remain limited, because the donors are expect- The regional allocation of research expendi- ing the CGIAR to fill the gap by moving down- tures has been fairly stable since 1992, with one stream into areas that are not its comparative major difference (figure 8.6). Expenditures on advantage or core competencies, as discussed Sub-Saharan Africa increased slightly, from 39 below. This report returns frequently to the issue 4 7 T H E C G I A R AT 3 1 A l l o c a t i o n s t o C o m m o d i t y a n d E c o r e g i o n a l F i g u r e 8 . 4 C e n t e r s H a v e D e c l i n e d 8% 6% Commodity Ecoregional Centers Centers 4% ­3.1% ­4.1% 2% 0% ­2% ­4% NRM Policy Centers Centers ­6% 3.3% 3.1% ­8% ­10% CIP ILRI IRRI IITA CIAT CIFOR ICRAF IWMI IPGRI FPRI ISNAR CIMMYT WARDA ICARDA ICRISAT ICLARM Average annual increase in Center funding (adjusted for inflation), 1992 to 2001 Source: OED calculations based on CGIAR Financial Reports, 1992­01. of core and complementary investments in de- While overall official development assistance veloping countries, particularly in Chapter 17 on (ODA) has remained stagnant at about $45 mil- the role of the World Bank in the CGIAR, as the lion annually (Ferroni and Mody 2002, p. 19), the critical determinant of impact, but it is an area in share of ODA being spent on global programs which both developing countries and donors has increased from 5.0 percent in 1980­82 to 6.8 have failed miserably in recent years, with the no- percent in 1990­92 and 8.8 percent in 1996­98 table exception of large developing countries (Hewitt, Morrisey, and Veldt 2001). This increase such as China, India, and Brazil. has largely been justified on the grounds that such programs supply global and regional public A Conceptual Framework for Assessing goods. But there is much confusion among the CGIAR Effectiveness funders of global programs as to which global This section explores issues relating to the or- programs are actually supplying global and re- ganization, financing, and objectives of agricultural gional public goods and, indeed, whether the research at the global, regional, and local levels provision of global public goods should be a pri- to provide a framework for assessing the CGIAR's mary criterion for the support of global pro- effectiveness in achieving its mission. There is par- grams.13 This increase also raises issues of criteria ticular focus on the relevance and applicability of for allocating grant funds among a competing set the concept of global and regional public goods of programs (figures 8.11 and 8.12). to the CGIAR in the same way that the OED The CGIAR was the first program providing global review is doing in its other case studies of global public goods to receive funds from the global programs. Bank's net income. It currently faces increasing 4 8 T H E C O N T E X T F O R T H E C G I A R M E TA - E VA L U AT I O N R e s e a r c h w i t h P r o v e n I m p a c t s o n P o v e r t y F i g u r e 8 . 5 H a s D e c l i n e d D r a m a t i c a l l y Enhancing germplasm Increasing Fisheries productivity Systems Forestry ­6.5% Production Livestock Crops Protecting environment 3.08% Saving biodiversity 2.7% Improving policies 3.09% Networks NARS Strengthening Training NARS Strengthening Org / Mgt ­0.8% Doc/Pub/Info ­12% ­10% ­8% ­6% ­4% ­2% 0% 2% 4% 6% 8% Average annual increase in Centers' expenditures by type of research activity (adjusted for inflation), 1992 to 2001 Source: OED calculations based on CGIAR Financial Reports, 1992­01. C e n t e r s ' E x p e n d i t u r e s o n A f r i c a H a v e F i g u r e 8 . 6 I n c r e a s e d S l i g h t l y D u r i n g t h e L a s t T e n Ye a r s 50% Sub-Saharan Africa 45% 40% 35% Asia 30% 25% Latin America and Caribbean 20% 15% 10% 5% West Asia and North Africa 0% 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 Source: CGIAR Financial Reports, 1988­01. 4 9 T H E C G I A R AT 3 1 G l o b a l P u b l i c A g r i c u l t u r e R e s e a r c h F i g u r e 8 . 7 E x p e n d i t u r e s , 1 9 9 5 : $ 2 1 . 7 B i l l i o n Developed countries 47% Developing countries 53% Source: Pardey and Beintema 2001. G l o b a l P u b l i c H e a l t h E x p e n d i t u r e s , 1 9 9 8 : F i g u r e 8 . 8 $ 7 3 . 5 B i l l i o n OECD investments 97% Developing country expenditures 3% Source: Global Forum for Health Research and Development 2001. competition for the 40 percent share that it re- production of global, regional, and national pub- ceives of DGF funds for global programs, and is lic goods; (3) core and complementary activities; one of 12 global programs that the Bank's ex- and (4) the respective roles of public and private ecutive directors have excluded from the normal actors in producing public goods through re- DGF requirement for an exit strategy from DGF search. funding.14 Faced with this increasing competition from global programs in other sectors, as well as The Agricultural Research and Development challenges to its perceived sense of entitlement, Continuum the CGIAR is increasingly being evaluated ac- The concept of an agricultural research and de- cording to several concepts related to global velopment continuum provides a framework for public goods. For the purpose of understanding considering the role of different actors in research the role of the CGIAR in the global agricultural and the kinds of research they conduct. Research research system, four sets of issues are impor- administrators distinguish four types of research tant: (1) the different types of research along the along this continuum--basic, strategic, applied, research and development continuum; (2) the and adaptive (box 8.1)--all of which represent 5 0 T H E C O N T E X T F O R T H E C G I A R M E TA - E VA L U AT I O N A v e r a g e A n n u a l G r o w t h R a t e s i n P u b l i c F i g u r e 8 . 9 A g r i c u l t u r a l R e s e a r c h E x p e n d i t u r e s , b y R e g i o n a n d H a l f D e c a d e s , 1 9 7 6 t o 1 9 9 6 12% 10% 5.2% 6.5% 8% 4.8% 6% 2.5% 4% 1.9% 1.5% 2% 0% ­2% Asia and Pacific China Middle East Latin America Sub-Saharan Developed excluding China and North and the Africa countries Africa Caribbean 1976­81 1981­86 1986­91 1991­96 Source: Pardey and Beintema 2001. steps in the process of generating new knowledge developing countries lacked their own capacity associated with improved agricultural technolo- to do so. Today, Brazil, China, India, and some gies. Knowledge transfer to the intended clien- other large and middle-income countries conduct tele--in the CGIAR's case, to farm households in a considerable amount of upstream strategic re- developing countries--requires institutions ca- search in addition to applied and adaptive re- pable of discovering, testing, applying, and adapt- search, and smaller countries, particularly in ing technologies developed elsewhere and Africa, still look to the CGIAR to fill downstream ensuring their prompt dissemination. Achieving gaps in their research capacity. effectiveness and impact requires strong link- ages and information flows across networks, re- Global Public Goods and Scale Economies gions, and researchers in industrial countries, The literature highlights the following rationales the CGIAR, and developing countries, combined for the CGIAR: (1) providing a global public good with a supportive economy-wide and sectoral that would not otherwise be produced because policy and investment framework. of lack of incentives for private sector invest- Conventionally, the CGIAR has been viewed ment (that is, goods with limited markets and as conducting strategic research in the research long-term payoffs), and (2) tapping economies and development continuum. Representing less of scale and scope in agricultural research that than 4 percent of the global total of public and cannot be accessed through country-based re- private sector agricultural research expenditures, search. In addition, support for the CGIAR is it was designed to use the best science in ad- often justified on the grounds of generating vanced countries to develop technologies for spillovers beneficial to both public and private ac- the benefit of food-deficit countries and popu- tors and contributing to the improvement of in- lations. That said, the CGIAR has also had to centives to international and national-level private conduct applied and adaptive research when investment in agricultural R&D. 5 1 T H E C G I A R AT 3 1 W o r l d B a n k a n d U S A I D F u n d i n g o f F i g u r e 8 . 1 0 A g r i c u l t u r a l R e s e a r c h b y R e g i o n World Bank (1974­00) Million 1993 US dollars 500 450 400 350 300 250 200 150 100 50 0 1974 1976 1978 1980 1982 1984 1986 1988 1990 1992 1994 1996 1998 2000 USAID (1952­96) Million 1993 US dollars 250 200 150 100 50 0 1952 1955 1958 1961 1964 1967 1970 1973 1976 1979 1982 1985 1988 1991 1994 Sub-Saharan Africa Asia Latin America & Caribbean Other Source: OED data, and Philip Pardey and Nienke Beintema 2001. Bank data includes research and extension; USAID data includes research only. Although the CGIAR is the oldest global pro- prompted by the emergence of this theme in the gram supported by the Bank, a long-time ob- operational strategies of U.N. agencies and the server of the CGIAR has recently noted that the multilateral development banks for the reasons use of the global public goods concept, per se, discussed in OED's Phase 1 Report (World Bank is relatively new in the CGIAR (Dalrymple 2002b; 2002a). However, the literature related to the or- Winkelman 1994). The recent usage has been ganization of research, including the CGIAR's 5 2 T H E C O N T E X T F O R T H E C G I A R M E TA - E VA L U AT I O N R u r a l a n d H e a l t h S e c t o r s A r e P r o m i n e n t F i g u r e 8 . 1 1 i n D G F A l l o c a t i o n s t o G l o b a l P r o g r a m s ( F Y 0 2 f i g u r e s ) Environment 5.7% ICT PSI 4.2% 12.1% PREM 5.5% Rural 42.2% Social protection 0.0% Health 17.4% Social Education development 3.8% FSE 8.7% 0.4% Source: DGF Secretariat. E n v i r o n m e n t , R u r a l , a n d H e a l t h S e c t o r s F i g u r e 8 . 1 2 A r e P r o m i n e n t i n O v e r a l l G l o b a l P r o g r a m E x p e n d i t u r e s ( F Y 0 1 f i g u r e s ) PSI ICT 5.8% 1.9% PREM 3.6% Environment Social 35.3% protection 0.1% Health 19.6% Education 2.8% FSE 0.5% Social Rural development 29.2% 1.2% Source: OED data. 5 3 T H E C G I A R AT 3 1 B o x 8 . 1 T h e R e s e a r c h a n d D e v e l o p m e n t C o n t i n u u m Basic research involves the search for knowledge and un- Adaptive researchis applied research to produce local pub- derstanding of basic organism functions or physical phenomena. lic goods and evaluation of technological innovations to assess The unraveling of the structure of DNA by Watson and Crick is their performance in a particular agricultural system (including an example of basic research. its socioeconomic factors) and to then adjust the technology to fit the specific needs identified. Screening and testing are typi- Strategic research is mission-oriented application of basic cally conducted both on-farm in farmer participatory trials and scientific knowledge to produce global and regional public on-station. For instance, when several varieties are available, goods. It is aimed at solving specific problems that may be of in- adaptive research would seek to determine those most suited to terest to developing countries and has broad application over local conditions and to identify additional research problems that several areas using known methodologies and techniques to pro- need to be overcome for adoption by farmers. duce new knowledge, principles, and understanding of critical needs in developing countries. Enhanced plant parent lines that Technology transferto farm households utilizes institutions can be used by numerous countries in applied crossbreeding pro- such as national extension and outreach services, the private grams are a good example. sector, NGOs, cooperatives, and rural development programs to inform the clientele in the use of the new knowledge or tech- Applied research develops tangible inventions by adapting nology. Techniques include on-farm demonstrations, incubator strategic and basic research to solve or meet field problems or labs, adoption networks, and farmer-to-farmer exchanges. The needs. An improved parent line, for instance, may gain resist- importance of extension increases in situations of poor com- ance to major local pests through genetic engineering tech- munications infrastructure and low levels of education among niques. farmers. Source: Adapted from Lele and Coffman 1996. role, has long considered factors such as benefits associated with the CGIAR research ac- spillovers, economies of scale and scope, and the crue to both developing and industrial coun- size of targeted markets integral to the economic tries, although some of those benefits were theory of public goods. Frequent reference to in- inadvertent and unplanned.15 Thus, global pub- ternational or global public goods can also be lic goods are those public goods with benefits found in documents of the CGIAR's TAC and its that spill across national boundaries and can be current successor, the interim Science Council consumed by people in more than one country (iSC). at the same time. National and local public goods Public goods are distinguished from private only benefit those who live in a single country goods by two characteristics: nonrivalry (one or locality. person's use does not reduce the ability of an- From such a perspective, conservation of bio- other person to use or enjoy the good at the diversity through gene banks and plant breed- same time) and nonexcludability (a person who ing has been widely viewed as a global public does not pay for or otherwise contribute to the good. The CGIAR's collection of genetic re- supply of the good cannot easily be excluded or sources--with more than 600,000 accessions-- prevented from using the good). Global public and its research in plant breeding have benefited goods are distinguished from international (be- both developing and industrial country agricul- tween two or more developing regions), re- ture, although the research was undertaken for gional, national, and local public goods by their the benefit of developing countries. These ben- reach (box 8.2). Research on the CGIAR's impact efits illustrate the global public goods nature of has now clearly demonstrated that the spillover the CGIAR's work, and the impact assessments 5 4 T H E C O N T E X T F O R T H E C G I A R M E TA - E VA L U AT I O N B o x 8 . 2 G l o b a l v e r s u s I n t e r n a t i o n a l P u b l i c G o o d s The recent spate of literature on global public goods is a re- The use of the term "global" is also broader than the CGIAR's sponse to globalization in the sense of the deterritorialization traditional use of the word "international" (between two or of space, which increasingly affects both developed and de- more developing regions) in another sense. "Global" incorpo- veloping countries. The emerging global public goods literature rates emerging challenges to organizing international research stresses the mutuality of benefits (or costs associated with pub- that the CGIAR did not previously have to contend with, such as lic "bads" such as climate change, conflict, the spread of com- WTO trade rules, intellectual property rights, and international municable diseases, and the like), which affect both developing biosafety regulations. and advanced industrial countries. The absence of a global For the World Bank's grant programs, the issue of global government to deal with these issues requires voluntary col- and regional public goods has been important since the incep- lective action and investments at the global level (beyond con- tion of the grant programs, long before the DGF was established sideration of aid as charity) to realize mutual benefits for both in 1997. Grants from net profits are to be used only when other industrial and developing countries. Much of the Bank's grow- instruments, such as loans and credits, which are available to ing activity at the global level has been justified in this new con- individual countries, would not be used by individual countries text of cross-border spillovers and cross-border benefits. The because of significant cross-border spillovers of benefits. A rapid rise of supra-territoriality, however, coexists with territorial global program may also be able to mobilize resources and issues, and the two interrelate in a complex fashion. (See Kaul, know-how that a single country may not be able to manage, even Grunberg, and Stern 1999; Gerrard, Ferroni, and Mody 2001; Lele if it is able to borrow. and Gerrard 2003; and Dalrymple 2002b; among others). of this work demonstrate how it has generated oping countries for some time to come, requir- more benefits than the intended international ing continued public investments, especially public goods (Pardey and others 1996; Nelson given absent or incomplete markets and limited and Maredia 2001). But there are few truly "pure" purchasing power of large segments of the pop- global public goods. Most are local or national, ulation to pay for technological products (Pardey or regional at best. and Beintema 2001, among others). Plant breeding, for example, also illustrates the Economies of scale and scope are realized limits of nonrivalry and nonexcludability. While in the production of improved agricultural tech- knowledge pertaining to plant breeding can in nologies when knowledge, methods, processes, principle be nonrival, seeds are not. And knowl- and information are mobilized in ways that in- edge embodied in seeds is excludable where dividual countries are unable to do because of there are intellectual property rights (IPRs), a lack of funds or a lack of scientists or institu- where hybrid lineages are known only by breed- tional infrastructure (box 8.3). Plant breeding ers, or with technologies such as "terminator" research and research on certain animal dis- genes. (Issues of IPRs as they affect public and eases conducted by the CGIAR Centers enjoys private investment in agricultural research are dis- substantial economies of scale, and relatively cussed in Chapter 10.) large markets for these technologies, with strate- Clearly, what constitutes a public or a private gic research eventually leading to production good is also highly contextual and varies across of improved breeding material.16 regions and stages of development, as shown by Research on agricultural production systems the increased role of private sector research in and natural resource management tends to have advanced countries in recent years. This requires relatively low economies of scale and small "mar- a highly nuanced application of the concept of kets" because such research often produces global public goods. But agricultural research location-specific knowledge and information will remain an important public good in devel- about processes and practices. It is less often de- 5 5 T H E C G I A R AT 3 1 B o x 8 . 3 E c o n o m i e s o f S c a l e a n d S c o p e i n R e s e a r c h Economies of scale refer to declining per unit costs as the num- Spillovers or positive externalities occur where the actions of ber of units of output produced increases. They are often the re- firms or individuals generate beneficial effects for other agents, sult of greater labor and managerial specialization and the such that total social cost exceeds private costs. Spillovers affordability of more efficient capital as the quantity of goods pro- may also be negative. duced increases. Spillins occur where firms or individuals benefit from positive Economies of scope occur when the cost of producing two prod- externalities or spillovers generated from the actions of others. ucts in combination is less than the total cost of producing each They are, in effect, spillovers viewed from the perspective of the separately. Scope economies often arise because production agents benefiting from the spillover. processes require much the same overhead, or because fixed investments are large and result in some unused capacity. signed for applications across multiple agro-eco- where it should be conducted--that is, at the in- logical, demographic, and policy/institutional ternational as opposed to the regional or the conditions, and therefore does not generate country level--depends on (1) whether it re- knowledge, information, and technologies rel- quires the advanced scientific methods that only evant to a broad set of clients. Hence, to gener- developed countries, or increasingly the private ate products (in this case knowledge rather than sector laboratories, can provide; (2) the extent goods such as seeds and pesticides) that have a to which it requires national research systems of large market, broad applicability, and spillovers developing countries to bring to bear the criti- across several regions requires highly coordi- cal national and local knowledge on policy, in- nated research networks with a complement of stitutional, and resource issues; and (3) the many different disciplines. capacity and sustainability of the national systems. By these criteria, the CGIAR should focus on The CGIAR is ideally suited for "global" or re- strategic research related to the production of gional public goods research, such as plant breed- public goods that may be global, international, or ing and associated natural resource management regional in character, such that they are charac- research, provided it focuses on research with po- terized by economies of scale and scope, and tential for wide impacts (particularly when com- that cannot be supplied by the private sector or bined with the rapidly advancing science in other sources, such as national systems or al- industrial countries) and enables developing ternative sources of supply in general, even if the countries to participate actively in the research latter capacity is built.17 Where public goods can process to help build their capacity. be supplied if national or regional capacity is es- What the research objectives should be, at tablished, for instance, all effort should be geared what level the research should be conducted, and to generating capacity at the appropriate level how its benefits should be conceived and gen- rather than filling such gaps through interna- erated may be quite different depending on the tional research, except on a short-term, tempo- nature of the research problem, the type of so- rary basis. That the CGIAR is increasingly filling lution needed, and how the outcome of the re- such gaps at the national and regional levels is an search would impact the problem. The need for area of concern for the System that is discussed advanced molecular biology techniques and throughout this report. methodologies available only in the laborato- Whether, how, and how much truly "global" ries of industrial countries to solve disease or pest public goods research should be financed and problems, for example, would primarily require 5 6 T H E C O N T E X T F O R T H E C G I A R M E TA - E VA L U AT I O N a voice for the scientists in those laboratories. But themselves. Hence, How much of its effort to control or eliminate infectious plant and an- critically evaluating should be devoted directly imal diseases in tropical countries will call for col- the role of core and to the provision of "merit laboration of national research systems of complementary ac- goods" with high social developing countries where the problem exists tivities is important. (but currently cannot be controlled due to the value. limited physical and human capital or financial Core and capacity), given that it is the "weakest link" that Complementary Activities will determine the final outcome. The former re- A World Bank report, Global Development Fi- quires the best of advanced science, while the lat- nance, 2001, distinguishes between core and ter requires much more local capacity complementary global activities. It defines core development, as well as input of local knowledge activities as global and regional programs that in research. produce public goods (for example, vaccine re- search) undertaken with transnational consid- Merit Goods erations, as well as activities in one country that An important question for a global research net- generate beneficial spillovers for others. The work is how much of its effort should be devoted report defines complementary activities (for ex- directly to the provision of "merit goods" with ample, seed and fertilizer trials, or animal vac- high social value, such as reducing poverty or cination programs) as designed to assist malnutrition, or the increased participation of developing countries in consuming the public women in the development process.18 This ques- goods that core activities of global programs tion has become pervasive in the debate about make available, while at the same time creating the current and future directions of the CGIAR. valuable national public goods.19 The downward drift in the CGIAR's resource al- The range of the activities that the CGIAR (as location and research thrusts documented in this well as other global programs OED is reviewing) report has resulted in part from the frustration is undertaking suggests that a neat division of of donors. They perceived that the national sys- labor between the production of global public tems of developing countries were too weak to goods at the global level and consumption at the provide the legitimate public goods function country level is not always possible. In some because of the neglect by both their own gov- cases, developing countries "produce" the global ernments and donors of crucial national in- public good (such as conservation of biodiver- vestments and capacities. As a result, global sity or the containment of communicable dis- programs such as the CGIAR have been ex- eases), which both they and the global pected to fill the gap. But the direct provision community enjoy. Hence, investment is needed of merit goods may not always be the compar- in developing countries to research and pro- ative advantage of a global program and may di- duce some global public goods, and assessment vert it from generating global or regional outputs criteria of core and complementary activities with potential for large spillovers, provided com- need to account for spillovers from the local to plementary investments are promoted at the the global level. national level. Besides, global public goods are The lack of good water management or pest often only one input into achieving merit good and disease control at the local level may have outcomes. Poverty alleviation or women's par- global implications such as the size and direction ticipation require both appropriate public goods of food trade and food aid flows, as indicated by such as technologies, which cannot be pro- IFPRI's modeling work. But water should be a duced by the countries themselves, and other topic for a global public goods research agenda complementary conditions--policies; socio- only when the research is helping to contribute cultural, legal, and economic institutions; in- to new scientific knowledge to develop a better frastructure; agricultural inputs; and markets-- understanding of the processes, policies, insti- that can only be provided by the countries tutions, or organizational innovations across re- 5 7 T H E C G I A R AT 3 1 gions so as to transfer technologies or knowl- dersupply it, while oversupplying "public bads" edge. The knowledge generated through such such as air pollution or biodiversity loss. Hence, research is global if it is perceived by CGIAR public resources are generally required to in- clients (developing countries) to be of relevance crease the supply of public goods. But incentives to them.20 By these criteria, much of the re- can also be put in place for the private sector to search on gender issues would be local, or at best increase the supply of public goods or reduce the national, unless it is designed to learn lessons supply of "public bads" through taxes, subsi- across the immense diversity of local social sys- dies, and creating new markets. This helps to ex- tems that influence women's participation--not plain both, in general, why global organizations an easy task. Climate change also illustrates the such as the United Nations and the World Bank complex research choices facing the CGIAR (box have become increasingly involved in global pro- 8.4). grams for the provision of global public goods and for the development of global markets such The Funding of Public and Private Sector as the Prototype Carbon Fund, and, in particu- Agricultural Research lar, why the CGIAR was established not just as an The essential problem with strategic agricultural international mechanism to finance the pro- research is that market mechanisms tend to un- duction of agricultural technologies of impor- T h e I m p o r t a n c e o f t h e P r i o r i t y S e t t i n g B o x 8 . 4 P r o c e s s : T h e C a s e o f C l i m a t e C h a n g e What research to conduct with respect to climate change rep- quences of climate change, but also research into the causes resents an example of the complex choices facing the CGIAR, of climate change. Important as this is, it would seem to be bet- illustrating the importance not only of science quality but also ter done elsewhere. of relevance to client needs and the CGIAR's comparative ad- Of course, one should always look for win-win opportunities vantage, although in each case the potential research pertains where both developed and developing countries benefit. But the to global issues or global concerns. It further illustrates both the primary goal of the CGIAR must always be to address the urgent importance and the difficulty of priority setting among compet- needs of the poor in developing countries, while considering the ing demands and the tendency for mission creep in response to magnitude of the spillovers and the likely speed of the impacts global concerns in the donor circles. on the poor. Even where there are win-win opportunities at the Evidence amassed by the Intergovernmental Panel on Climate local and global levels, however, it is necessary to assess the Change suggests that climate change has substantial adverse opportunity cost of resources used in downstream activities in impacts on the poor, lowland tropical countries. Therefore, re- particular research (such as the cost of field testing in developing search could be aimed at mitigating global climate change, for countries compared with the likely impacts and time profile of example, by enhancing carbon sequestration in small farmer rice alternatives to achieve impacts on the poor), as well as their or- production, or by increasing agroforestry coverage in develop- ganizational implications for the CGIAR. These issues are raised ing countries. Alternatively, research could be aimed at helping by technologies such as golden rice (discussed in Chapter 10). poor households in developing countries to cope with the local They emphasize the fundamental importance of a strong scien- consequences of global climate change or to benefit from the tific review process for priority setting in the overall allocation emerging carbon markets. The priorities within CGIAR's re- of resources, not simply the science quality of the individual pro- search on climate change are confusing, however, and illustrate grams and projects proposed, and highlight the risks of under- the temptation to adapt to "fashions" or donor interests, a sub- mining or taking away altogether the important System-wide ject explored in more detail later. Some activities mention not priority setting and advisory function from the proposed Science just strategies (to help poor farmers) to adapt to the conse- Council. 5 8 T H E C O N T E X T F O R T H E C G I A R M E TA - E VA L U AT I O N tance to the food-short and mostly poor devel- ing or IDA credits). But private research will oping countries. This is consistent with not replace public research However, the biotechnology revolution and the notion that Bank- of importance to the growing importance of intellectual property supported global pro- developing countries any rights (IPRs) for improved agricultural tech- grams should not be nologies are greatly changing the nature of re- producing purely na- time soon. search and technology from public goods to tional public goods, proprietary goods (box 3.1). More and more agri- since the Bank's grant cultural research has acquired private goods char- resources are limited and have many competing acteristics and has become attractive to private demands on them (box 8.2), nor should they be financing. This can help reduce the need for substituting for national-level or private efforts fo- public finance to a certain extent. But private re- cused on activities that are not of high national search will not replace public research of im- priority, since the opportunity cost of developing portance to developing countries any time soon. countries' own resources used in such pursuits The private sector funded 35 percent of the $33 tends to be high. billion (in 1993 international dollars) spent on A related question for the CGIAR, then, is agricultural R&D worldwide in 1995/96 and 70 how much of the Bank's grant funds are lever- percent of the research on genomics, while pos- aging other donor funds for activities that should sessing 80 percent of the intellectual property em- more legitimately be funded out of the much anating from it. But private sector expenditure on larger pool of Bank funds for country assistance agricultural R&D in developing countries was for lending and loan/credit administration--a less than 6 percent of all private sector expendi- question explored in greater detail in Chapter 13. tures in agricultural R&D worldwide (figure 3.1), This chapter argues that the failure of govern- which is also less than 6 percent of the total ex- ments and donors to invest in national agricul- penditures (both public and private) on agricul- tural systems of developing countries has led tural R&D in developing countries. Nonetheless, the CGIAR to move downstream to invest in de- private sector research is already raising impor- velopment and dissemination activities that tant policy issues pertaining to research priorities should ideally be conducted at the national or for the public sector (including the CGIAR), pri- regional levels to fill the gap. The possibility of orities for public-private partnerships, the control using IDA grants rather than credits for national of research outputs, and access to technologies and regional activities now offers an opportunity resulting from them. An improved policy envi- to increase investments at the national and re- ronment in developing countries, including more gional level, which developing countries, espe- secure IPRs, will help to increase private invest- cially in Africa, have not been willing to ments in developing countries. undertake.21 Thinking about what financial interventions are needed, at what level, and through what kind of Previous Evaluations institution, in order to achieve the ultimate ob- OED reviewed a wide variety of evaluations of jective of poverty reduction in the most cost-ef- the CGIAR. While many of these reports are fective, equitable, and sustainable way is critical, treated in detail in the subsequent chapters, and is increasingly challenging in the context of annexes, Working Papers and Background Papers the growth of the private sector. Nonetheless, the accompanying this report, a typology of CGIAR principle of financial subsidiarity is a particularly evaluations is presented below. The meta-eval- important and useful concept for the World Bank uation team paid particular attention to three in this regard, since DGF guidelines call for grants evaluations: the OED internal review of the to follow this principle. DGF grants should fund World Bank grant programs (1998), the OED only those activities that would not qualify for review of the CGIAR (1999), and the Third Sys- other financial instruments (such as IBRD lend- tem Review of the CGIAR (1998). 5 9 T H E C G I A R AT 3 1 Typology of Evaluations Internal reviews commissioned by the CGIAR CGIAR evaluations can be classified into five cat- or the Centers include impact assessments of egories: external System and Center reviews CGIAR programs that are conducted either in- commissioned by the CGIAR; external Center re- house or by independent CGIAR panels such as views commissioned by the Centers; internal re- the Impact Assessment and Evaluation Group views commissioned by the CGIAR or its Centers; (IAEG) and its successor, the Standing Panel on external reviews and audits commissioned by Impact Assessment (SPIA). CGIAR members; and external audits of the External reviews and audits commissioned CGIAR or its Centers.22 by CGIAR members include studies conducted External reviews commissioned by the CGIAR by CGIAR donors and other partners, and span include several types of evaluations conducted a variety of topics including science and research at different levels within the System. External Pro- strategies, governance and management, fi- gram and Management Reviews (EPMRs) of in- nances, and impact assessment. dividual Centers (more recently referred to as External audits of the CGIAR or its Centers in- External Reviews, or ERs) were organized ex- clude financial and management audits con- clusively by TAC until 1983, at which time the ducted by internationally recognized firms and CGIAR Secretariat was given shared responsi- by the donors. The World Bank conducts its au- bility with the incorporation of Center manage- dits of the CGIAR through the Internal Auditing ment reviews into the evaluation process. At the Department. System-wide level, external reviews also form Since the CGIAR's inception in 1971, the an important evaluation tool for the CGIAR. A focus of reviews has evolved from evaluations of total of three such evaluations have been con- the quantity and quality of research to research ducted: the First System-wide Review (1976), results, to management efficiency, to strategic di- the Second (1981), and the Third (1998). System- rections, to impact and, currently, to science level reviews of crosscutting, thematic, or pro- quality (see Anderson and Dalrymple 1999, p. 39; grammatic topics are also commissioned by the CGIAR Secretariat 1995d; Ozgediz 1995, p. 3). CGIAR or TAC, and include various management Indeed, the meta-evaluation team has found reviews, "stripe" reviews, and impact assess- that the CGIAR possesses a rich history in the ments undertaken with increasing frequency area of monitoring, evaluation, and impact as- since the mid-1980s. Finally, the CGIAR (or TAC) sessment.23 However, as will be discussed commissions reviews of central CGIAR compo- throughout this report, the CGIAR has also nents, such as the CGIAR Secretariat (1988) and come up short on ensuring the quality and in- TAC Secretariat (1989) (CGIAR Secretariat 2000a, dependence of some of these reviews, as well pp. 3­8). as providing systematic information on follow- External reviews commissioned by the Cen- up processes arising from this massive invest- ters include Center-commissioned external re- ment in reviews and evaluations. views (CCERs), designed to evaluate program and management topics at individual Centers OED's 1998 Review and to support and complement CGIAR- or OED's 1998 review of grant programs (OED 2002) TAC-commissioned reviews such as the EPMRs, was done to inform the Bank's Executive Board dis- ERs, and System-wide reviews. External reviews cussion of funding for grant programs in fiscal of Centers are conducted by individual Cen- 1999 and beyond, under the auspices of the Spe- ters about once every five years, with the ob- cial Grants Program (SGP) and its successor, the jective of monitoring the institutional strength Development Grants Facility (DGF). The review and contributions of a Center in both a retro- focused on three issues: the relevance of the spective and a prospective manner. Centers Bank's grant-making programs to its overall strat- also commission other types of external re- egy and developmental role; the quality of grant views, including impact assessments, program programs management; and the efficacy of the reviews, and audits. development results of grant programs. As the 6 0 T H E C O N T E X T F O R T H E C G I A R M E TA - E VA L U AT I O N largest and oldest of the Bank's grant programs, need for production of global and regional pub- the CGIAR figured significantly in the review. lic goods by the CGIAR and recommended: The review recommended that the World Bank's grant program be governed by three · Continued Bank sponsorship of the CGIAR guiding principles, as follows: · Improvement of Bank-CGIAR linkages in agri- cultural and natural resource programs, espe- 1. Subsidiarity. The review observed that grants cially loans for agricultural research. should be given in situations where lending is inappropriate and there is no other source The Third System Review (1998) of funding to ensure that grants do not com- The TSR, completed in 1998 (CGIAR Secretariat pete with the Bank's other instruments, such 1998e), was the first comprehensive System-wide as its own lending programs. This principle is external evaluation of the CGIAR since 1981. The a defense against having grant money drive TSR was commissioned by the CGIAR and con- out the need for lending--a form of moral ducted by a distinguished panel chaired by Mau- hazard. rice Strong. The panel focused on issues relevant 2. Maintain an arm's-length relationship. to developing a strategy for the new millennium, The review observed the potential for real or given an increasingly complex external environ- perceived conflicts of interest when the grantor ment, and on ensuring that resources were made is too closely related to the grantee. A de facto available for the CGIAR to pursue this strategy. dependency arrangement arises, especially The TSR addressed two broad areas: (a) science where the Bank is called upon to handle a com- and strategy and (b) organization, governance, bination of fund-raising, fiduciary, and admin- and management. Among its many recommen- istrative responsibilities within a collaborative dations for the CGIAR, the TSR emphasized that arrangement between the Bank and its grantee. the CGIAR should: 3. Vigorously follow an exit strategy policy. The review observed a risk of dependency if · Pursue a dual strategy of integrated gene man- grants continue over a long period, potentially agement and natural resource management undercutting the independence of the grantee, · Pursue more meaningful partnerships with impeding sustainability of program benefits, NARS, advanced research institutions, NGOs, and inhibiting proactive management of the and the private sector grant portfolio and new programmatic priorities. · Undertake major reforms in governance, in- cluding a corporate governance structure with OED's 1999 Review of the CGIAR legal status, and streamlining of governance In 1999, OED prepared a study on the CGIAR systems (Anderson and Dalrymple 1999) to provide back- · Undertake a more effective public awareness ground for the larger evaluation of the Bank's and fund-raising campaign. grant programs discussed above. The report was prepared by OED staff in collaboration with The specific recommendations set forth by the USAID staff, and its primary objective was to ad- TSR are examined in detail throughout this re- dress broad questions posed by the 1998 OED port through a similar division of topics--the im- internal review of the grant program with re- pact of CGIAR science and strategy in Chapters spect to the CGIAR.24 Specifically, it sought to ad- 9 to 14 and an evaluation of CGIAR governance dress the subjects of (1) alignment with Bank and management issues in Chapters 15 to 17. goals and programs; (2) implementation of su- pervision, leverage, and evaluation; and (3) ac- The Change Management and Design complishments in grant efficacy and program Process effects. Anderson and Dalrymple awarded the The CGIAR embarked on an internal exercise to CGIAR high marks in all three areas. Further- determine how to position itself in the future be- more, they recognized the continuing and future ginning at International Centers Week 1999. The 6 1 T H E C G I A R AT 3 1 exercise consisted of several initiatives: (1) a vi- gional organizations, and civil society institu- sion and strategy document from TAC, (2) an tions/NGOs analysis of the System's structure by the Center · Managing the NARS-CGIAR relationship dif- Directors' Committee (CDC), (3) an examination ferently to reflect the changes in NARS of governance issues by the Committee of Board · Expanding interaction with the private sector Chairs (CBC), and (4) an outline of options for · Outlining new arrangements to deal with action based on these documents by a Synthe- patents and intellectual property rights sis Group established by the Oversight Com- · Clarifying the CGIAR's role in institution build- mittee. To create a concrete proposal for change, ing and strengthening the Group established a Change Design and · Addressing internal inefficiencies, including Management Team (CDMT) at International Cen- decisionmaking processes ters Week 2000, consisting of both CGIAR stake- · Achieving congruence among strategy, struc- holders and professionals from outside the ture, financing, management systems, organi- System.25 The CDMT was guided by an ad hoc zational structure, and the like.27 steering group of CGIAR stakeholders, led by the CGIAR chairman.26 The team's work, as The CDMT's report, issued in April 2001, con- agreed by the Group, was to focus on: tained seven specific proposals for reform: · A restructuring action plan for the entire Sys- · Create Global Challenge Programs, focused tem, with a clear rationale for program inte- on specific outputs, based on an inclusive ap- gration and/or consolidation of Centers proach to priority setting, drawing on research (including analysis of options) competencies of the Centers and other part- · A governance plan to streamline CGIAR decision- ners, and funded largely by additional resources making and to clarify the roles of all components · Enhance NARS through full "mobilization" of in order to achieve net efficiency gains their capacities in design and implementation · A business plan to increase efficiency in the pro- of the Global Challenge Programs, and through vision of common services, coordinate Sys- an initiative to promote financial support to tem-wide programmatic activities, and reduce NARS overheads in order to transfer more resources · Enhance science output through the trans- to research. formation of TAC into a Science Council · Engage in specific efforts to ensure longer- Within these broader terms of reference, the term financing of the System Steering Group also asked the CDMT to respond · Reduce annual meetings of the Group to to several specific issues: one, with an Executive Council appointed by the Group to carry out delegated functions · An urgent need for the CGIAR to "elevate the in the interval between the annual general game"--to demonstrate the salience of its meetings work in relation to key interests and concerns · Create a CGIAR System Office to enhance ef- of the international community ficiencies in System-level management and to · Finding new ways to appeal to traditional serve the CGIAR chairman, members, com- donors mittees, and the System at large, as well as to · Improving System-wide synergies offer some services to the Centers · Placing greater emphasis on strategic oppor- · Adopt an "evolutionary restructuring ap- tunities (for example, climate change and sus- proach," which is to "flow from the imple- tainable agricultural development in Sub- mentation of the change proposals."28 Saharan Africa and South Asia) · Increasing inclusiveness in agenda setting, in- At the Mid-term Meeting 2001, the Group cluding a need to interact effectively with the adapted these recommendations into four ac- Global Forum on Agricultural Research, re- tionable areas, which have become the four pil- 6 2 T H E C O N T E X T F O R T H E C G I A R M E TA - E VA L U AT I O N lars of the current reform effort: (a) establish- as treatment of the CDMT proposals, are dis- ment of a representative Executive Council, (b) cussed in Chapters 15 and 16, and in Annex J. It transformation of TAC into a Science Council, (c) is important to note that this meta-evaluation is a programmatic approach to research through not intended to evaluate the reform process Challenge Programs, and (d) creation of a virtual currently under way, but only to assess whether, System Office. how, and how well the Change Design and Man- The quality, content, and follow-up of the key agement Process is addressing issues raised in the internal and the three external reviews, as well previous evaluations. 6 3 Section II The Development Effectiveness of the CGIAR T he major issues identified by the review of the CGIAR literature were examined in a detailed series of papers produced for the meta-evalua- tion. This section presents the findings of those papers. For the detailed evidence that forms the basis for these findings, the reader is directed to the Working and Background Papers themselves (see Bibliography). The main top- ics presented here are the returns to germplasm research; genetic resource management, biotechnology, and intellectual property rights; policy research; natural resource management research; NARS capacity building and the im- pacts of NARS research; and the challenges of Sub-Saharan Africa. In addition to discussing research conducted by the CGIAR and other experts, this sec- tion also incorporates NARS views, a perspective that the meta-evaluation team has found to be invaluable in its review of CGIAR impacts and development effectiveness. 6 5 9 High Returns to Germplasm Research C GIAR's germplasm research constitutes a long-term investment program with a record of high rates of return, and an activity with strong posi- tive impacts on agricultural growth, poverty reduction, and the envi- ronment. The CGIAR's research in germplasm improvement is an important function that has the continuing potential to increase agricultural productiv- ity, generate positive spillover effects, and exploit economies of scale.1 The CGIAR's most lauded achievement, the in- policy to be essential, and complementary to the troduction of improved cereal varieties and com- CGIAR's objectives. The issues raised by the team plementary resource management techniques in are the insufficient evaluation of research con- Asia during the Green Revolution, exemplifies the ducted in these other areas, particularly given the importance of germplasm improvement. The rapidly changing allocation of existing and new increase in cereal production--a doubling of resources in the System and the CGIAR's com- rice and wheat yields in Asia and Latin America, parative advantage in relation to other suppliers. a doubling of rice and wheat production in Asia, The bulk of the CGIAR's impact literature and similar gains in Latin America--brought comprises cost-benefit and rate-of-return stud- about by the Green Revolution demonstrates ies of its germplasm research. The studies have the contribution of enhanced germplasm not been criticized on methodological and data only to agricultural productivity improvement, grounds, and their claims of high returns have but also to poverty reduction and land savings sometimes been questioned. Doubts have also (see IFPRI 2002). However, germplasm research been raised about their utility as guidance to has suffered more than other research areas in CGIAR research priorities and resource allocation recent funding allocations. The decline in fund- decisions.2 For these reasons, the OED meta- ing for this research seems counterintuitive given evaluation team thoroughly reviewed these stud- its well-documented impacts. ies (Gardner 2002; Eicher and Rukuni 2002; The meta-evaluation team is not suggesting, Katyal and Mruthyunjaya 2002; Romano 2002; however, that the CGIAR revert to exclusive re- Macedo and others 2002). search on plant breeding and germplasm im- Methodological difficulties notwithstanding, the provement. Rather, the team considers the meta-evaluation team concludes that the impact research on natural resource management and studiesprovideconvincingevidencethatthe(mostly 6 7 T H E C G I A R AT 3 1 Crop breeding research of conventional) crop tain over half the world's poor living on $1 per the CGIAR Centers, together breeding research day by the year 2035.5 Furthermore, only 11 per- with the follow-up work of the CGIAR Cen- cent of the agricultural land in the developing ters, together with world is "favored" land--area having no or mod- with developing country the follow-up work erate limitations to sustained application of in- NARS, continues to generate with developing puts under a given use--and only 35 percent of extraordinarily high country NARS, con- the rural population occupies this land tinues to generate (CGIAR/TAC 2000d). The rural poor occupying returns to investment. extraordinarily high both favorable and unfavorable land are often returns to invest- vulnerable, powerless, and voiceless with ment. Those returns, ranging from 40 to 78 percent, respect to assets and income; food, water, and soil; are well above the returns attainable from many al- markets for agricultural inputs, credit, and outputs; ternative uses of public funds.3 accessing entitlements and institutions; and Even for crops such as sorghum, millet, and applying science, technology, and knowledge. cassava, which are not readily amenable to pro- There is no agreement, nor is there likely to ductivity enhancement because of the marginal be, on the best way to reduce poverty--for ex- agro-climatic environments in which they are ample, by intensifying areas with favorable re- grown, the CGIAR has made important progress. source endowments and promoting healthy After some initial missteps and learning, the migration to these areas; by creating rural em- CGIAR Centers concerned with these crops have ployment opportunities in agriculture and value- generated new technologies, and their adop- added enterprises; by reducing urban food prices tion by poor people is said to be considerable, and promoting migration to urban areas; or by even in Sub-Saharan Africa, where successes improving agricultural productivity in resource- have been more limited. There are also cases of poor areas. Many agricultural economists work- high estimated rates of return to research on ing on development widely share the view that maize, wheat, and other crops in Africa.4 "the power of technological solutions to solve The meta-evaluation team further observes poverty problems is extremely limited in the ab- that the returns to germplasm research would be sence of economy-wide growth and rural-urban even higher without the OECD agricultural sub- migration" (Binswanger 1994, pp. 624­28). It sidies of nearly $1 billion per day--a massive can thus be argued that complementing CGIAR constraint on the effectiveness of the CGIAR's in- research with research on high-value, nonfood vestment of $340 million per year. Reduction or agricultural activities typically carried out by elimination of these subsidies and trade pro- non-CGIAR actors, increasing market opportu- tections would raise prices, increase market ac- nities by adding value to agricultural output, cess, and induce developing countries to invest and improving policies and investments through- more in their own agricultural research systems. out the economy simultaneously is critical for re- ducing rural poverty and creating economic Examining the Impacts of Germplasm opportunity. In short, poverty reduction entails Research going beyond what the CGIAR can offer. The strategic allocation of scientific efforts in But economy-wide growth and structural germplasm research is of considerable impor- transformation from predominantly agricultural tance for the CGIAR and public research pro- to industrial economies depend on agricultural grams throughout the developing world, given growth, and in turn on innovations in agricultural the high incidence of rural poverty and the con- technology (Hyami and Rutlan 1971). Increases tinuing contribution of agricultural productivity in agricultural productivity as the result of tech- improvement to poverty reduction. Of the 1.2 bil- nological innovations stimulate nonfarm eco- lion people subsisting on less than $1 per day, nomic activities through various linkages, 75 percent of those live and work in rural areas. including (a) reallocation of labor and entre- Projections suggest that rural areas will still con- preneurship to non-farm production within farm 6 8 H I G H R E T U R N S T O G E R M P L A S M R E S E A R C H households as the result of the reduced labor But given that re- In resource-poor areas, input for farm production to produce the same search is a long-term research takes much longer amount of food for home use and market sale, endeavor and given to produce benefits. (b) increased real incomes (including home- the complex dynamic produced food) of the farm population that interactions among stimulate local manufacturing through expanded population growth, poverty incidence, and the consumption, and (c) increased demands for end of the extensive margin, what can the evi- farm inputs associated with technological inno- dence of the CGIAR's impact to date tell us about vations having the same effects on the expansion the areas that will offer the CGIAR the best of demand for nonfarm goods. chance of future impacts on poverty reduction? More specifically, IFPRI's Hazell and Haddad Studies suggest that relatively better-endowed have argued that agricultural research that leads areas, crops, and farmers have benefited more to improved technology can benefit the poor in from research than resource-poor areas, crops, a number of ways: and farmers. These benefits accrue primarily from increased productivity and lower food · Increasing farm production and marketed out- prices, but accrue unevenly when poorer house- put, and providing more food and nutrients for holds are unable to adopt new technologies.6 In consumption resource-poor areas, however, research takes · Increasing employment for landless laborers much longer to produce benefits. And even and small farmers where new technologies are widely adopted and · Improving opportunities for economically ben- hold promise for resource-poor areas, economy- eficial migration wide and sector policy and investment failures · Inducing growth in rural and urban nonfarm can hamper productivity increases. The experi- income ences of research in crops such as sorghum, · Reducing food prices millet, and cassava in Africa provide evidence of · Increasing economic and physical access of these issues, although increased food supply poor women to better foods and greater stability in production have clearly · Empowering the poor (Hazell and Haddad helped poor subsistence households.7 2001, p. 9). Reduced commodity prices are usually taken to be an obstacle to technology adoption and There is considerable evidence that crop im- technology investment progress.8 But others provement research has contributed to increased argue that incentives to invest have weakened be- food availability, reduced prices, employment cause, absent large new scientific thrusts, high generation, and higher wages, often to the ben- returns are increasingly provided by success in efit of landless, near landless, or migrant rural "defending" Green Revolution gains against suc- workers. These gains have led to growth in urban cessive pest biotypes, and hence do not show up and rural nonfarm sectors and have increased as sensational yield increases in the way that inter-sectoral linkages, further improving em- earlier Green Revolution advances did. This ployment opportunities for the poor (Kerr and makes investments in research crucial. But sev- Kolavalli 1999; Lipton with Longhurst 1989; Hazell eral experts also argue that changing donor pref- and Haddad 2001; IFAD 2001; Gardner 2002; erences have reduced the funding for research Delgado, Hopkins, and Kelly 1998). Similarly, re- and hence the incentives, as has the nonappro- search focused on resource-poor environments priable nature of public sector research. has contributed to improving crop tolerance to An additional complicating factor is the in- drought, salinity, pests, disease, weeds, tem- teraction between research and farm commod- perature, and soil nutrient deficiencies, thereby ity taxation or support programs. In OECD addressing a variety of economic and environ- countries, there has been criticism of spending mental concerns critical to poor rural producers on research to boost production, at the same in semi-arid and arid areas (IFAD 2001). time that commodity programs are trying to re- 6 9 T H E C G I A R AT 3 1 duce production (for example, by the acreage today? The meta-evaluation team thinks not. Re- set-asides implemented by the United States search takes a long time to deliver results, and and the European Union). Several studies by by the time research results are available, through agricultural economists have indicated that in improved policies and investments, markets may this situation research spending that would oth- well have been liberalized or established. erwise have a high rate of return becomes a Studies suggest that productivity growth aris- waste of the taxpayers' dollars (Alston, Edwards, ing from germplasm improvement has also had and Freebairn 1987). Still, several experts con- important environmental impacts. With the end clude that even the largest commodity program of the extensive margin, land savings represent distortions in developing countries have not the most important positive environmental (and changed the story of high social returns to pub- poverty-reducing) impact of productivity-en- licly supported agricultural research (Alston and hancing germplasm research. There is consider- Pardey 1996). able evidence of the reduced area needed to Some studies suggest that falling food prices cultivate food for rapidly growing populations contribute to the persistence of rural poverty, al- in developing countries. Estimates of area saved though lower prices help the urban and nonag- vary widely, however, ranging from approximately ricultural poor. However, reduced food prices 2.5 to 3.6 billion hectares of crop and pasture land resulting from productivity increases also create saved since the 1960s. A CGIAR study estimates the opportunity to shift resources from food land savings in Asia, Africa, and Latin America at- production to non-food production activities in- tributable to all Green Revolution research at side and outside the farm sector, leading to 426 million hectares, while savings attributable to economy-wide development (Hayami and Rut- research in seven of the CGIAR-mandated com- tan 1971). In order to achieve success in this modities are estimated at 100 to 300 million process without imposing undue sacrifice on hectares in the developing world.9 food producers, public assistance must be pro- But studies also suggest that not all germplasm vided for smoothing the intersectoral resource improvement research results in land savings. transfer--for example, by means of education, In land-surplus, labor-short economies, (a) the in- training, communication, and transportation in- troduction of new agricultural technology will frastructure. One form of such assistance that the likely lead to increased land clearing, and (b) lib- CGIAR may be able to provide is germplasm re- eralization of trade and investment is likely to pro- search for nonfood crops where there is no pri- vide additional incentives to clear land for vate sector research. So far, the CGIAR's agriculture, as evidenced in the Amazon in Brazil germplasm research has been targeted almost ex- and the outer islands of Indonesia.10 clusively at subsistence food crops in support of The evidence on returns to recent research smallholders. However, with the very success on germplasm improvement is limited but omi- of the research on food crops, the improvement nous. Several of the most recent studies pro- of nonfood, high-value crops suited to small- vide disconcerting evidence of a slowdown in holders may have to be included in the major yield growth, even in favorable areas such as agenda of the CGIAR's germplasm research. For rice and wheat.11 Slowing yield growth has been this research to be effective, it would require attributed to decaying irrigation infrastructure close coordination with natural resource man- and resource degradation--for example, falling agement and cropping system research, as well groundwater tables, micronutrient depletion, as private sector research. and low-level pest buildup. Recent reports, such Should re- as IFAD 2001, take this slowdown seriously and Productivity growth arising search avoid stress the importance of renewed public sector from germplasm improvement ger mplasm research at the international and national levels improvement to maintain the gains made in yield improve- has also had important if markets are ment and to ensure continued increases in pro- environmental impacts. not liberalized ductivity. 7 0 H I G H R E T U R N S T O G E R M P L A S M R E S E A R C H Resource degradation, often associated with and donors, are not Linkage between germplasm productivity growth, can be alleviated through already doing.12 improvement and NRM research. Resource degradation is commonly at- Linkage be- research is thus essential to tributed to (a) continuous cropping made pos- tween germplasm achieve sustainable, high- sible by new technologies and (b) changes in improvement and farming systems prompted by differential pro- NRM research is productivity production ductivity growth. There are isolated NRM tri- thus essential to systems. umphs in the CGIAR, notably reducing the use achieve sustain- of harmful agrochemicals and increasing sus- able, high-produc- tainability by integrated pest management. But tivity production systems. The TSR, in its most successes are breeding-related: breeding for discussion of the need for a "twin pillars" strat- robust resistance or tolerance to pests and for egy emphasizing productivity enhancement sustainable yields in the face of soil-nutrient de- through integrated gene management and in- ficiencies. One of the CGIAR's more successful tegrated natural resources management, stressed efforts to address the issue of resource degra- the two legs on which the CGIAR must walk. dation and slowdown in yield increases has been The CGIAR has recently adopted an integrated in rice-wheat cultivation. The initiative, under- NRM approach, yet these two areas have fre- taken by CIMMYT, IRRI, and the NARS of quently remained distinct in the System's re- Bangladesh, India, Nepal, and Pakistan evolved search portfolio and generated considerable and into a System-wide initiative under NARS lead- continuing debate, as discussed further in Chap- ership in 1994 to promote research on issues crit- ter 12. ical to maintaining and enhancing productivity Overall, the impacts of conventional and sustaining rice-wheat systems in South Asia germplasm have been strongly positive. However, (Ladha, Fischer, Hossain, and others 2000). But certain issues where CGIAR impact research has the lack of stable, predictable funding for this ef- been weak, or where CGIAR strategy needs fur- fort has been a problem. The record on similar ther examination, are beyond research disin- System-wide initiatives has been mixed. centives and resource degradation. Yet while improved NRM is crucial (and is dis- cussed in Chapter 12 in detail), there are no Related Issues Requiring Further Impact major scientific breakthroughs in the manage- Assessments ment of tropical soils or water. To justify divert- Externalities such as negative human and animal ing large sums away from germplasm research, health impacts or genetic diversity losses warrant Centers need to make the necessary scientific further examination by the CGIAR. Health im- breakthroughs relevant to the problems of de- pacts are a continuing area of concern in the dis- veloping countries. At the same time, germplasm cussion of productivity growth arising from improvement (for example, for drought-prone research in germplasm improvement, although areas or areas with limited water availability) can there is little agreement among experts on the ease the natural resource management prob- subject, particularly the role of global public lems that farmers face. Researching NRM as such, goods research. Human pesticide poisonings with little new contribution by the CGIAR Cen- associated with the use of improved seed and ters to the basic science of productivity growth planting materials are clearly a significant cost to or increasing the efficiency of resource use, does society, and there is empirical evidence of these not seem justified. There is currently insufficient costs. Evidence suggests that similar arguments independent evaluation of the CGIAR's NRM re- can be made about chemical fertilizers. How- search that convincingly documents the negative ever, assessments of the negative impacts of impacts of productivity growth on natural re- chemical inputs on human and animal health and sources and the possible contributions the CGIAR on productivity are limited in terms of data and can make to this issue in ways that others, in- coverage. This problem is also present in the few, cluding national systems of developing countries but often contradictory, studies on the impacts 7 1 T H E C G I A R AT 3 1 eficial technologies and processes to research sys- tems of developing countries and farmers world- wide and in an appropriate, effective, and timely manner? Again, studies on the determinants of technological adoption would need to be con- ducted to help accomplish this. Because the CGIAR has an extensive collection of genetic materials, germplasm research and returns are often attributed to the CGIAR alone. But if the germplasm were more readily available to NARS and other research institutions, then many other actors in development research could work to improve germplasm and increase agricultural productivity. of genetic diversity losses on crop productivity Research on the productivity of fish and live- and yield stability (Maredia, Pingali, and Nelson stock is also a topic that warrants increased em- 2002). In short, the CGIAR has not studied these phasis by the CGIAR. As income levels rise in issues. If they are of high priority to NARS, they developing countries, per capita consumption of should receive attention. meat, milk, and fish also increases. These trends Technological adoption and variability in adop- have important implications for the global sup- tion rates also need further research, notwith- ply of and demand for fish, livestock products, standing the extensive literature available. and feed grains, and for productivity-enhancing Despite all the impact studies carried out by the technologies (see, among others, Delgado, Cour- CGIAR on productivity growth and germplasm bois, and Rosegrant 1998). While key Centers improvement, empirical knowledge on how and such as ILRI have developed strategic responses why certain technologies spread while others to the challenges posed by these trends, they re- do not remains limited. The obvious explana- main underfunded due to inadequate prioriti- tions--profitability, risk, and access to informa- zation of research that addresses these issues tion--are not sufficient determinants of strategically and effectively.13 The meta-evaluation adoption. This points to a need for studies by so- team investigations suggest that ILRI lacks the cial scientists working jointly with their coun- critical mass of scientists to be a world leader. terparts in the biological sciences to understand ILRI's director general attributes this to funding the constraints to accelerated technology adop- constraints. tion and to meet yield and natural resource man- Finally, the CGIAR's research and develop- agement gaps through the use of existing ment strategy needs vigorous examination. In re- knowledge. But again, NARS need to lead the cent years the focus on alleviating poverty, rather change. than on conducting global and regional public Technology dissemination and the role of the goods research that is likely to have a direct im- CGIAR is yet another area in need of strategic ex- pact on large numbers of the poor, has led to a amination. Numerous technologies developed by lack of focus on research in which the System non-CGIAR sources need to be disseminated. should have a strong comparative advantage, Some large NARS are now world leaders: China with significant potential for reducing poverty. In- in hybrid rice, Brazil in minimum tillage, and stead, the CGIAR has addressed issues that may India in information systems. Can the CGIAR be be important at the local or the national level, but a catalyst in dis- marginal from a global public goods perspective. Can the CGIAR be a catalyst seminating their Several examples of this are discussed below. i n n o v a t i o n s The meta-evaluation team concurs with Hazell in disseminating globally, moving and Haddad that CGIAR research must focus on innovations globally? potentially ben- (a) maintaining efforts to reduce food prices 7 2 H I G H R E T U R N S T O G E R M P L A S M R E S E A R C H through productivity growth, (b) intensifying Mwale 2002). More- Breeding to improve less-favored lands, (c) helping smallholders di- over, participatory nutrient content of staple versify into high-value crops, and (d) increasing research requires crops is an area with both employment and incomes for landless and small multidisciplinary past successes and future farmers. Equally challenging, but no less im- work, a task that is portant, is research on increasing the access of costly and difficult potential for the CGIAR. the poor, especially poor women, to foods rich to organize and in crucial micronutrients. Critics argue that in- replicate for developing countries on any sig- creasing nutrient content at the cost of yields nificant scale. Farmer participation increased raises issues of profitability and likely adoption costs by 66 percent and accounted for 80 per- of the new technologies, while running the risk cent of the researchers' time in Ghana, although of drawing resources away from other objec- increased costs can be compensated by in- tives, such as strengthening crop tolerance to var- creased adoption later (Asenso-Okyere, Agble, ious stresses. Others argue that focusing on yield Attah-Krah, and others 1998). Besides, em- increases alone reduces the constraints placed powerment can only be achieved through in- on researchers, while nutrient access can be im- creasing the access of the poor to education proved with more cost-effective efforts such as and increasing community action, areas in which the promotion of home gardening and infor- the CGIAR does not have a demonstrated com- mation and educational interventions by na- parative advantage relative to other actors.14 In tional governments. Yet despite the poor addition, there are concerns about the quality outcomes of early attempts in breeding to en- of the CGIAR's social science research, as dis- hance nutrients in staple crops (such as quality- cussed in Chapter 11. protein maize), breeding to improve nutrient An important question is the comparative ad- content of staple crops is an area with both past vantage of the NARS (as well as universities and successes and future potential for the CGIAR NGOs in developing countries) relative to CGIAR (Hazell and Haddad 2001). Moreover, it has the Centers in participatory research or research on potential to work affordably for the poor, and is empowerment issues (see Barrett 2002). Relative directly in the CGIAR's area of research special- to the NARS, the CGIAR is able to generate high ization, wide country application, and compar- payoffs to research in areas that NARS cannot ative advantage, although the CGIAR does not easily undertake, such as research on plant breed- make a compelling case that plant breeding from ing (Byerlee and Eicher 1997; Maredia and Byerlee micronutrients will benefit the poor. 1999). But whether the CGIAR can do this for par- Equally uncertain are the CGIAR's actual or ticipatory research or research on empowerment potential impacts on what Hazell and Haddad issues is unclear. An assessment needs to be con- term the "newer pathways" by influencing the ducted of the CGIAR's relative capacity in areas access of the poor to the components of power beyond germplasm research. The need for such such as decisionmaking processes, information, an assessment has been voiced by the Brazilian and authority. Many other research institutions-- NARS on previous occasions, requesting the representing an important alternative source of World Bank to undertake comparative cost-ben- supply--have conducted more extensive re- efit analysis of research at the CGIAR Centers search in this area than the CGIAR, and it is not relative to research at the NARS of advanced and clear the CGIAR has a comparative advantage in developing countries (see Macedo, Porto, Con- this field (see Barrett 2002). Agricultural re- tini, and Avila 2002). And with respect to issues search can contribute to the empowerment of of empowerment, even panel members of the the poor if conducted in a participatory way, TSR indicated to the meta-evaluation team that but much of the CGIAR's participatory research they did not expect the CGIAR Centers to un- is in testing products of research, not in setting dertake such research, but instead to stimulate research priorities (see Gladwin, Peterson, and NARS in leading such research.15 7 3 10 Genetic Resources Strategy and Management G enetic resources lie at the heart of the CGIAR System. The System's collection of genetic resources is a unique, critical element in the agri- cultural research that has generated spectacular rates of return on in- vestment and positive impacts on poverty reduction. The genetic resources are an important input to complementary research undertaken by NARS throughout the developing world. They have played an important but largely unrecognized role in agricultural reconstruction in post-conflict nations such as Afghanistan, Burundi, and Rwanda. They represent a significant contribution toward lic-private interaction over proprietary knowl- preserving the world's biodiversity for use by edge. However, the scientific advice set forth present and future generations.1 And they have by these reviews either failed to provide focused the attributes of a pure public good with almost priorities and financial options, or was under- unlimited potential for producing large eco- played for complex political, organizational, and nomic benefits of global significance. social policy reasons, thus limiting their overall Yet the continued maintenance and use of impact on the way the CGIAR manages its genetic these genetic resources by the CGIAR faces un- resources. precedented challenges. TAC-commissioned The CGIAR's priorities, governance system, studies such as the First External Review of the and financial prospects limit the System's ability System-wide Genetic Resources Programme to effectively manage and use its vast collection (1999) and the System-wide Review of Plant of genetic resources. Concurrently, the rapidly Breeding Methodologies in the CGIAR (2001) changing market, institutional, and technologi- highlighted many of the challenges, and the TSR cal conditions in which the CGIAR operates ne- brought these challenges into a broader per- cessitate more dynamic responses and spective for the CGIAR. Specifically, the TSR em- interactions on the part of the System. Com- phasized the need for the CGIAR to pursue bined, these factors pose a serious challenge to efforts in genetic resource conservation; the ongoing, productive use of the CGIAR's ge- biosafety, bioethics, and public information; a netic resources for international agricultural re- legal entity to hold patents; and rules for pub- search. This chapter demonstrates that the 7 5 T H E C G I A R AT 3 1 The genius of the CGIAR CGIAR's genetic re- management, on the other.5 Third, donor in- was to crossbreed sources still represent terest in genetic resource management is limited, indigenous genetic an important input notwithstanding the importance placed on ge- into increasing agri- netic resources in the United Nations Conference material resistant to pests cultural productivity on Environment and Development in 1992 and and diseases with higher- and reducing poverty, the World Summit on Sustainable Development yielding varieties. but that new ap- in 2002. proaches are required While donors have been willing to fund phys- to (a) effectively man- ical structures for gene banks, they have expected age and use the existing collection of genetic re- Centers to fund the recurrent costs through un- sources, (b) incorporate modern biotechnology restricted contributions. But the declining share tools and methods into CGIAR research, (c) en- of unrestricted funds in recent years has made this gage the private sector and other actors in the difficult for the Centers. And as the share of re- global agricultural research community, and (d) stricted funds has increased, donors have given establish an effective policy that addresses the higher priority to policy and NRM research at related issues of proprietary knowledge and in- the cost of gene bank operations. Donors have tellectual property. also shown a preference for funding final-prod- uct technologies such as improved varieties and Genetic Resource Management and Use: cropping systems that offer visible, measurable Issues for the CGIAR outputs and that are easier to justify to their con- The genius of the CGIAR was to crossbreed in- stituencies, rather than intermediate products digenous genetic material resistant to pests and such as genetic resource management. More- diseases with higher-yielding varieties to de- over, donors are often unwilling to fund projects velop more productive plants and, more re- such as gene banks whose benefits extend over cently, animals. Successes led to a collection of several countries or regions when the donor is valuable material from a wide range of develop- mandated to assist development of a certain re- ing countries to support further breeding ef- gion or a country. This dilemma underlying per- forts. Today, the CGIAR holds the single largest vasive under-investment in the provision of public collection of plant genetic material, comprising goods necessitates greater allocations of unre- 600,000 accessions, or about 10 percent of the stricted multilateral funding to the maintenance world's collection, in 11 CGIAR Centers.2 The pre- and management of genetic resources.6 vious chapter demonstrated that the uses and The allocation of responsibility for genetic benefits of this collection are well documented resource management is an additional issue for with respect to developing countries and to cer- the CGIAR. Currently, the management of the ge- tain industrial countries.3 Thus, conservation of netic resources collections is governed by a stan- agro-biodiversity through establishment of ex dard agreement between the FAO and each situ gene banks has already proven its global Center that holds the genetic material. The Sys- public goods nature and value. tem-wide Genetic Resource Programme (SGRP) However, the CGIAR faces serious internal is charged with assisting the Centers in fulfilling challenges to effective management and use of their obligations under the FAO agreement. The its collection of genetic resources. First, the gene SGRP is administered through the International banks are severely underfunded, and an un- Plant Genetic Resources Institute (IPGRI) as the known amount of the material may have dete- convening Center. Policy advice is provided by riorated and been lost over the years.4 Second, the broadly representative high-level Genetic there are competing demands from Centers for Resources Policy Committee (GRPC), established financing genetic resource management for ex in 1994 to monitor and analyze developments in situ conservation of crops, on the one hand, genetic resources policy both internationally and conservation of aquatic, livestock, forest, and within the System, and to recommend ap- and microbial biodiversity, or in situ ecosystem propriate CGIAR action as necessary. This has in- 7 6 G E N E T I C R E S O U R C E S S T R AT E G Y A N D M A N A G E M E N T cluded advice regarding the engagement of the CGIAR in the negotiations of the FAO Commis- sion on Genetic Resources for Food and Agri- culture that led to the adoption of the International Treaty on Plant Genetic Resources for Food and Agriculture in 2001.7 The SGRP has been an effective System-wide program, in part, because of the clear authority vested in it to ensure compliance with stan- dardized procedures under FAO auspices. Yet the SGRP, as a System-wide program, has no inde- pendent authority. This problem is exacerbated by two System-level issues: (a) the lack of agree- ment in the System over genetic resource man- agement and the System's inability to determine their "Federation proposal" for reorganization, what should be included or excluded from the which was also rejected (see Chapter 15). SGRP, and (b) the role or position the SGRP Key international agreements governing the should assume when organizational needs control, management, and use of genetic re- emerge outside the scope of the FAO agree- sources include the Convention on Biological Di- ment, but are not supported by all the Centers versity (1992), the International Treaty on Plant with gene banks (Lesser 2002). These persistent Genetic Resources for Food and Agriculture issues have significant implications for program (2001), the International Convention for the cohesion and for the strategic objectives of the Protection of New Varieties of Plants (1961 and CGIAR (see CGIAR Secretariat 1999c). subsequent revisions), and the World Trade Or- The CGIAR also faces significant external con- ganization's Trade-Related Aspects of Intellec- straints that affect the management and use of tual Property Rights (TRIPS), all of which raise its genetic resources. Among them are the in- important issues both for CGIAR access to ge- creasing complexity of international agreements netic resources and for access to CGIAR's own on genetic resources and the CGIAR's role genetic resources by other parties. The CGIAR's therein (box 3.1), the difficulty in mobilizing effective engagement in the preparation of the international support and resources for long- International Treaty (discussed below) is an in- term genetic resources conservation, and the dication of the System's important role in shap- need to ensure access to its collections to the ing international opinion on key topics in genetic NARS. The Third System Review concluded that resource management. the CGIAR's current governing principles--in The CGIAR has also made significant efforts to particular its nonpolitical nature and informal mobilize international support for genetic re- status--constrain its ability to fully set policy on sources conservation. At International Centers key issues in this area, which are of vital im- Week 2000, a feasibility study of an endowment portance to the System's ability to address its campaign for genetic resource conservation was mission and to participate in international fo- initiated and, with endorsement from the CGIAR, rums concerning those issues. The need to be the FAO, and the World Bank, a high-profile cam- able to influence international debate and pol- paign was proposed for the creation of the Global icy setting contributed to the TSR's recom- C o n s e r v a t i o n The CGIAR has also made mendation that the CGIAR become a more Trust. The trust's formal body with clear legal standing, which feasibility study, significant efforts to mobilize could develop and support unified policies that completed in April international support for touch on genetic resources and related mat- 2001, set an initial genetic resources ters. The CGIAR board chairs and Center di- target of $260 mil- rectors recommended a similar approach in lion to be raised conservation. 7 7 T H E C G I A R AT 3 1 from private foundations, corporations, and gov- ing countries, both because of the relatively higher ernments to develop and sustain a global system cost of refrigeration in the high-temperature, of gene banks, including those of the CGIAR high-humidity tropics, and the threat posed by Centers. Switzerland confirmed its support to conflicts.9 the Trust, while the United States, the U.N. Foun- Of additional concern are CGIAR's policies dation, and Egypt announced their support at the and strategies for meeting the needs of devel- World Summit on Sustainable Development in Jo- oping country NARS with respect to genetic re- hannesburg in August 2002. So far, commitments sources. Studies commissioned by this of $60 million have been obtained. meta-evaluation cite concerns or tensions be- But the long-term prospects for establishing a tween Centers and NARS over the ownership gene bank, as envisioned by the Global Conser- and flow of genetic materials for research and ex- vation Trust, are unclear. These have been con- perimental use in several countries.10 To the ex- strained by the magnitude of resources required tent that these concerns are valid, impediments to operate the gene bank in perpetuity, and by the to the free exchange of genetic resources pose donors' unwillingness to pledge funding until the a serious threat to the System's underlying prin- negotiations of the Treaty were concluded and the ciples and objectives. Another important issue support of the FAO Commission was assured. is whether developing country NARS should The adoption of the Treaty and the FAO Com- have their own germplasm collections included mission's statement in October 2002 of universal in financing schemes for gene bank facilities, appreciation and support to the Trust are ex- operations, maintenance, and training, and how pected to make a difference. A detailed IFPRI these collections, and the resources necessary to study for the SGRP of the costs of conserving fund them, would be managed to complement and distributing the current holdings of the the CGIAR collection. CGIAR's 11 gene banks in perpetuity estimates a The World Bank and various bilateral donors required endowment of $149 million, although have supported the operations of some national- plausible variations in interest rates and regen- level gene banks, but others have argued that the eration cycles of genetic materials could cause this maintenance of international collections inad- estimate to range from $100 to $325 million (Koo, vertently reduces the incentive for developing Pardey, and Wright 2002). Prospects for raising the countries to save their own biodiversity, and entire initial target of $260 million are unclear at even undertake their own breeding programs, this stage, as are the priority activities for opera- leaving such work to the CGIAR (Shands 2001). tions and maintenance of genetic resource con- The proposed campaign to fund the Global Con- servation, including, for example, support to the servation Trust addresses some of these issues funding of gene banks in developing countries. In- by providing for sustainability grants on a long- formed sources argue that $400 million to $500 term basis to holders of national and interna- million may be required to maintain the gene tional ex situ plant genetic resource collections banks.8 The initial target of $260 million for an en- meeting international standards, and a capac- dowment for genetic resources conservation was ity-building fund.11 set on the basis of donor prospects at the time of What remains to be determined is how the the feasibility study, in 2000, when the negotiations subsidiarity principle that drives DGF funding will of the Treaty were not yet concluded. Moreover, be applied to this program, as well as a deter- some founders mination of what should be funded at the global Impediments to the free and long-time ob- level, what should be funded through country as- servers of the sistance programs, and how global-level funding exchange of genetic resources CGIAR have raised can complement investments made by devel- pose a serious threat to concerns about oping countries at the national level, for exam- the System's underlying the risks of main- ple, through building their capacity for taining these gene maintenance, operations, public awareness, pol- principles and objectives. banks in develop- icy formulation, and implementation.12 7 8 G E N E T I C R E S O U R C E S S T R AT E G Y A N D M A N A G E M E N T The Challenges of a Rapidly Changing allowed private Possibly the most significant Environment firms to enter do- trend in the rapidly The need to ensure wide access, sustainable mestic seed mar- changing environment is the use, and continued protection of the CGIAR's ge- kets, often starting growth of the private sector netic resources for developing countries is chal- with hybrid maize lenged by the rapidly changing market, seed, and helped to in agricultural input institutional, and technological context in which increase the intro- markets and agricultural the CGIAR operates. However, these changes-- duction of im- research. including breakthroughs in genetics and genetic proved seed and engineering that have led to a new round of bi- complementary ological innovation in agriculture, the increasing technologies. Despite the limited number of importance of proprietary knowledge and in- studies on private sector seed production and tellectual property rights (IPRs) in agricultural re- marketing, some evidence suggests rapid growth search, and the rapid growth of the private sector in this sector. For instance, maize seed supplied in agricultural input markets and agricultural by the private sector constituted 89 percent of R&D--also represent a critical opportunity for the total in seven Asian countries as of 1997, the CGIAR. compared to only 30 percent a decade ago.14 Possibly the most significant trend in the rap- Growth in applied and adaptive research con- idly changing environment is the growth of the ducted by private firms also has the potential to private sector in agricultural input markets and contribute to agricultural productivity improve- agricultural research. Growth in market infra- ment by complementing upstream public sector structure and private firm participation in mar- research at both the national and international kets can potentially reduce costly state levels. Worldwide, the private sector currently ac- interventions in the provision and pricing of counts for approximately 70 percent of total in- agricultural inputs and commodities, interven- vestment in research for genetically modified tions that have had serious effects on agricultural organisms (GMOs). As a result, nearly 80 percent growth throughout the developing world. of agricultural biotechnology­related patents Consider, for instance, the process by which are owned by the private sector (see Lessor new seed technologies were disseminated dur- 2002; Spielman 2002). ing the Green Revolution in India. The Rockefeller What remains to be detailed is the CGIAR's po- Foundation and CGIAR scientists recommended tentially complementary role with private sector that USAID and the World Bank support public growth in agricultural development. Private sec- sector seed companies in India during the early tor investment in agricultural R&D is often de- 1970s because of the limited farmer demand for termined by a very specific set of factors that has improved seed and a lack of institutional infra- potential ramifications for dissemination and structure for delivering productivity-enhancing adoption of new technologies, including overall technologies to farmers. At the time, there was market size, crop value, farmers' purchasing little understanding of new technologies among power, farmers' repayment capacity on input farmers, limited regulatory capacity in the gov- loans, farmer knowledge of new technologies, re- ernment to ensure seed quality, and a lack of in- sponsiveness of the crop to inputs, relative pro- terest and incentives for private sector investment portions of purchased versus saved seed used by in improved seed. The public sector interventions farmers, crop processing and storage qualities, in seed production and distribution played a cru- and consumer preferences. Moreover, access to cial role in exposing farmers to the new tech- technologies emerging from the private sector nology and creating a market. depends critically on the ways in which propri- Since that time, India and many other Asian etary rights are allocated to innovators and farm- countries have developed a thriving seed mar- ers; the ways in which benefits are distributed ket.13 Privatization of state-owned seed compa- among private firms, rural producers, consumers, nies and expansion of market infrastructure has and government; and the nature of institutions 7 9 T H E C G I A R AT 3 1 The role of CGIAR research formed at the na- CGIAR priorities and strategies, and require de- is more important than tional and interna- tailed and integrated consideration by the CGIAR ever, especially in regions tional levels to and its stakeholders. address these issues. Also worth mentioning is the role of the such as Sub-Saharan Africa. The complexity CGIAR's genetic resources in the noisy debate of these factors over GMOs and transgenic crops. Controversy sheds light on the relatively unexplored impli- over GMOs has led to concerns about food safety cations of the investment choices made by the and trade, famine relief supplied to southern private sector over specific technologies, crops, African countries, and NGO and civil society move- and regions; the types of farmers that are likely ments in industrial and developing countries. to benefit from private sector technologies; the The World Bank announced a new global initia- regulatory regimes that will make available such tive at the Johannesburg Summit to assess the po- technologies; the intellectual property that they tential of agricultural science and technology in embody; and the forums that exist to allow the boosting agricultural productivity in developing voices of stakeholders to be heard. countries with a view to generating the same kind These issues are critical to the CGIAR insofar of consensus as was developed on climate change. as its collection of genetic resources represents It will be difficult to achieve such consensus with- the raw materials that enable biotechnology to out conclusive scientific evidence either for or produce new processes and products to im- against GMOs, and even if achieved, the consen- prove agricultural productivity and reduce sus will need to be followed by concrete actions poverty. The control, management, and use of by the CGIAR System as a whole to achieve genetic resources are major issues both within poverty-reducing outcomes on the ground. the System and in the international arena. Both Thus, the CGIAR faces a number of complex internal systems and international agreements issues raised by the intricate relationships among raise important questions relating to the social, genetic resources, biotechnology, and IPRs. The economic, and ecological impacts of genetic re- CGIAR is in an enviable position to help devel- source management, modern biotechnology, in- oping countries impartially evaluate risks and tellectual property, biosafety and field testing advise them in selecting biotechnologies, espe- regulations, and the CGIAR's role in addressing cially given that some governments in industri- these questions. alized countries are themselves embroiled in In this context, the role of CGIAR research is controversies over biotechnology and unable to more important than ever, especially in regions provide objective leadership or direction.15 Some such as Sub-Saharan Africa, where the CGIAR, the developing countries are proceeding with field World Bank, and the donor community can as- tests of transgenic crops, while others get left be- sist in developing programs that support re- hind. The TSR recommendations on IPRs provide search and concurrently promote markets and evidence of the urgent need for an effective IPR commercialization of new technologies. The strategy in the CGIAR, and the CGIAR's en- changing environment and the rapid pace of in- dorsement of the use of an integrated gene man- novation in the private sector requires that the agement approach at the Centers further CGIAR (a) develop a more sophisticated and illustrates the support for such an approach. timely response to issues such as IPRs, (b) ex- pand its engagement with private firms, both Can the CGIAR Meet this Challenge? multinational and domestic, and with the NGO Given the challenges posed by a rapidly chang- community on issues of technology choices and ing environment, is the CGIAR prepared to (a) institutional arrangements, and (c) remain a key ensure sound management of use of its genetic actor with respect to international agreements resources, (b) invest in new biological tech- on genetic resources, intellectual property, nologies, (c) collaborate with the private sector biosafety and field testing, and other key issues. and other actors in research and development, These issues have important implications for and (d) address critical issues pertaining to pro- 8 0 G E N E T I C R E S O U R C E S S T R AT E G Y A N D M A N A G E M E N T prietary knowledge and intellectual property? the SGRP Review The primary concern for the To be sure, the CGIAR has proved its ability in Panel argued that CGIAR is developing options using conventional plant breeding tools to de- the totality of in- that address the issue of velop new varieties of many types of crops. The ternational agree- managing its collection of System-wide Review of Plant Breeding Method- ments relating to ologies in the CGIAR (2001) recommends, and genetic resources genetic resources. the meta-evaluation team concurs, that con- provides for a pol- ventional breeding be maintained at or above its icy environment that requires quick and appro- present level in the CGIAR, even if biotechnol- priate responses by individual Centers and the ogy advances are also to be used to advance CGIAR as a whole, responses that could not be plant breeding. But beyond this, the meta-eval- achieved in the current structure.17 But the CGIAR uation concludes that the CGIAR must undertake still lacks a person with the authority and re- urgent discussion and analysis of these issues to sponsibility to address issues beyond the Inter- develop an appropriate and effective set of poli- national Treaty on Plant Genetic Resources for cies and programs for action. Food and Agriculture. In other words, substan- tial reconfiguration of the System is necessary to Managing Genetic Resources ensure sound management and use of the The primary concern for the CGIAR is develop- CGIAR's genetic resources. ing options that address the issue of managing its collection of genetic resources. The CGIAR has Investing in Biotechnology posed a number of solutions: the Global Con- How much the CGIAR should invest in new servation Trust discussed earlier, and the con- biotechnologies is unclear, even in the CGIAR's tributions of IPGRI and the GRPC to maintaining own reviews. Currently there is great variation open access to genetic resources for developing in the degree to which Centers use new biotech- country research in the FAO's International Treaty nology tools and methods and the degree to on Plant Genetic Resources for Food and Agri- which these tools are integrated into conven- culture. Moreover, the CGIAR has made infor- tional plant breeding (box 10.1). Several Centers mation on its genetic resources available to have used genetic transformation to create trans- researchers through its System-wide Informa- genic cultivars and introduce pest and disease tion Network for Genetic Resources (SINGER), control where no satisfactory genetic resistance an important contribution to improved man- can be found.18 But Centers responsible for sev- agement of genetic resources. eral "orphan" crops such as cassava and yams The TSR attempted to address the internal have enjoyed fewer opportunities to apply challenges to integrated genetic resource man- biotechnology, despite the highly advanced state agement, in part, by recommending the estab- of their work with tissue cultures from some of lishment of a centralized governance mechanism these crops. The allocation and distribution of for the CGIAR. However, the idea was rejected by these new technologies within the CGIAR sug- the CGIAR membership. The board chairman gest that there is scope for further study and im- and director general of IPGRI, the focal Center of provement. Moreover, the CGIAR has annually the CGIAR's plant genetic resource management invested only an estimated $25 million, or about efforts, have had similar reactions to the recom- 25 percent, of its plant breeding budget in mendations. In arguing that the recommenda- biotechnology research, a fraction of the amount tions were inconsistent with the principle of invested by the private sector globally and a fig- Center autonomy and would not have Center ure widely considered ownership, the chairman and director suggested inadequate to meet Currently there is great instead that the Intercenter Working Group on the challenges of the variation in the degree to Genetic Resources (ICWG-GR) be granted greater future (Lesser 2002). authority to make decisions relating to the Sys- Yet the costs associ- which Centers use new tem's genetic resources.16 Moreover, the chair of ated with biotechnol- biotechnology tools. 8 1 T H E C G I A R AT 3 1 C u r r e n t C G I A R B i o t e c h n o l o g y R e s e a r c h , b y B o x 1 0 . 1 C e n t e r a n d A c t i v i t y The interpretation of the scope of activities that can be classi- Molecular-level analysis (for example, mapping, DNA se- fied as biotechnology research can be quite broad. For the pur- quencing, fingerprinting): IRRI, rice; ILRI, livestock and crop poses of this meta-evaluation, biotechnology research is stover, including sorghum; ICARDA, lentil and barley. classified into the categories shown below, based on generally accepted terminology and applied to current CGIAR research. Transformation(forexample,Bacillusthuringiensis,orBt,her- Note that tissue culture is not included under biotechnology bicide resistance, drought tolerance): CIMMYT, IRRI, rice; CIAT, procedures. The information on current CGIAR biotechnology re- cassava; IITA and ICARDA, cowpea; ILRI, livestock vaccines, search is drawn from individual Center Web site lists of current ICRISAT. projects and work plans. Due to the difficulty of determining what is and is not biotechnology-related based on a brief program de- Functional foods (for example, starch and nutritional quality): scription, as well as interim changes in projects and delays in CIAT, cassava; IITA, cowpea; ICRISAT, stover nutrition. posting updates, the list is not exhaustive. Research methods (for example, new marker types, trans- Germplasm enhancement, including genetic diversity formation techniques): IITA, cowpea; CIMMYT. (for example, Marker Assisted Selection, gene identification and characterization): CIMMYT, maize, wheat; IRRI, rice; CIAT, cas- sava, rice, beans; CIP, potato and sweet potato viruses, ICRISAT. ogy research--modern equipment, genome cause of budget constraints. The short-term and data, and bio-infomatics; skills training for sci- unpredictable nature of donor funding makes it entists; biosafety and field testing compliance; difficult to develop long-term strategies for re- expertise on regulatory regimes that vary from tooling the Centers, even if they engage in part- country to country; and so on--are substantial nerships and contract out most research to (Petit and others 1996). The System-wide Review agencies in industrial countries. of Plant Breeding Methodologies in the CGIAR In short, the CGIAR may be threatened with (2001) observed that (a) breeding programs are technological obsolescence should it fail to in- understaffed because of recurrent budget cuts, vest significantly in developing its capacity to (b) vital breeding programs in some Centers engage in partnerships promoting biotechnology have been de-emphasized in favor of new and research and its use. Conversely, according to the different (non-plant-breeding) programs, often CGIAR's own System-wide reviews, the addi- due to donor preferences, (c) Centers are hand- tional research costs associated with biotech- icapped in the use of new biotechnology tools nology are unlikely to lead to any savings in the by the lack of a global base for information and short and medium term. expertise to draw on, particularly for the or- phan crops, (d) field breeders and biotechnol- Engaging the Private Sector ogy experts have insufficient communication Public-private partnerships represent a poten- linkages, and (e) utilization and competence in tially cost-effective means of transferring tech- the new field of biotechnology are less than in nology and capturing technological spillovers the older conventional plant breeding field (TAC from private sector R&D that may benefit agri- Secretariat 2001a, p. 5). culture in developing countries. According to Adding to the challenges facing the CGIAR is private sector representatives in the CGIAR, the uncertainty in the future of both conventional more public-private partnerships in the CGIAR plant breeding and modern biotechnology, and would facilitate the dissemination of CGIAR tech- many commodity Centers are downsizing be- nologies at a much faster rate than at present. For 8 2 G E N E T I C R E S O U R C E S S T R AT E G Y A N D M A N A G E M E N T such arrangements to be forged, the CGIAR will from the Bank's Agri- There is little up-to-date have to develop flexible and responsive methods culture and Rural De- information on private of engagement with private sector firms. It may velopment Department sector investment in also have to consider reallocating some of its own (ARD) provide direc- agricultural R&D, or on research efforts where private sector investment tion for future consul- is large enough to warrant CGIAR withdrawal tations and research the potential or actual and where access by developing countries to through activities such benefits and costs of the fruits of such research can be ensured.19 as the Agricultural public-private Surprisingly, there is little concrete, docu- Knowledge and Infor- mented, Center-by-Center, up-to-date informa- mation Systems (AKIS) partnerships. tion on private sector investment in agricultural Thematic Team or the R&D, or on the potential or actual benefits and roundtable discussions held with CEOs of major costs of public-private partnerships. At the System agribusiness companies and with leaders of civil level, very limited information is available from the society in 2000­01 (see Byerlee and Fischer 2000). CGIAR on its relationship with, or the role of, the The global initiative on agricultural science and private sector in agricultural research and devel- technology launched by the World Bank in Jo- opment on activities in which the Centers are en- hannesburg is expected to address many of the gaged, except in the reports of the Private Sector issues relevant to the private sector, but its im- Committee (PSC) (see, among others, CGIAR Sec- plications for the CGIAR are unclear. retariat 2000d, 2001c, d). Minutes of the PSC re- Public-private partnerships will contribute to flect high-level, exploratory dialogues on issues the CGIAR's work in biotechnology. But they need including, but not limited to, potential areas for to be crafted with care since their results are partnership, IPRs and regulatory issues in devel- highly dependent on the objectives of the part- oping countries, CGIAR governance and man- nership and the distinct accountabilities and ob- agement, and the need to improve public ligations of partners. The most significant awareness on biotechnology. However, the PSC it- partnerships to date are concentrated in Brazil, self provides little evidence of tangible, System-level China, and India, and fall into one of five cate- progress toward supporting collaboration or part- gories: basic and applied research initiatives led nership between the CGIAR, the private sector, by the public sector; outsourcing of private sec- NARS, and other stakeholders, even though the tor research to public institutions; joint public-pri- CGIAR has been financing the activities of the PSC vate ventures in applied research; public in much the same way it has financed the activi- partnerships with research foundations estab- ties of the NGO Committee. lished by the private sector; and technology trans- At the Center level, there is similarly limited in- fer systems between the public and private sectors. formation that is systematic or transparent, and that Collaborative successes depend critically on (a) considers accountability expectations on public-pri- strong research programs and financial resources vate partnerships. The CGIAR's recent conference in both the public and private sector, (b) politi- on impact assessment, held in Costa Rica, did not cal support for public-private collaboration with highlight this issue: few panels or papers directly domestic and foreign firms, and (c) an IPR regime addressed private sector R&D or its impacts.20 Sev- that provides for commercial incentives within eral Centers have informed the meta-evaluation public-private research initiatives (see Spielman team that they have forged extensive linkages with 2002; Pray 2001). the private sector. Yet there is insufficient infor- More analysis is needed of the alternative mation regarding the larger, System-wide implica- arrangements for sharing the benefits of research tions of these linkages for the meta-evaluation between and among parties to public-private part- team to make a useful assessment of their contri- nerships. The few analyses of public-private part- bution to the CGIAR or its objectives. nerships that do exist make little distinction The World Bank itself is hoping to contribute between research for development purposes--a to the issue of public-private partnerships. Papers strategy consistent with the CGIAR's mandate to 8 3 T H E C G I A R AT 3 1 The question is whether the serve the poor, and need to cope with the growing complexity of CGIAR System has an overall potentiallysupported IPRissues,ensuretransparencyandaccountability vision, strategy, and by some types of co- in the use of public funds, facilitate the intro- financing, joint ven- duction of new scientific information into the management system for ture, nonprofit public domain for eventual or downstream use addressing the biotechnology partnerships and by private firms, and generate income. These agenda through public- technology transfer needs parallel those faced by the CGIAR. arrangements--and A recent British commission on IPRs headed private partnerships. research for com- by John Barton and staffed by some of the most mercial purposes--a knowledgeable experts on IPRs globally22 argues, strategy more consistent with contracting arrange- "concerns about the operation of the intellectual ments. Indeed, R&D for social and economic de- property system and the extension of IPRs are not velopment is increasingly giving way to a more confined to their application to developing coun- complex and fashionable terminology. Today, de- tries" (Commission on Intellectual Property Rights velopment R&D is more appropriately referred to 2002, pp. 2, 4). The commission advises global pro- in terms of (a) research for development, (b) de- grams such as the CGIAR, international organi- velopment of research for commercial purposes, zations, and developing countries to confront (c) dissemination of technology and information, the full implications of IPRs with the knowledge and (d) development occurring as a result of and sophistication called for. The CGIAR needs to technological dissemination and adoption. confront five major issues: (a) areas in which the In sum, the overarching question is whether System should partner with the private sector to the CGIAR System has an overall vision, strategy, conduct research, (b) areas in which costs make and management system through which it is ad- it most efficient to simply access private sector dressing the biotechnology agenda through pub- technologies for the benefit of the poor, (c) de- lic-private partnerships as a way of implementing velopment of benefit-sharing arrangements to its mission. The meta-evaluation's review of pre- ensure such access, (d) areas in which the CGIAR's vious studies on this issue finds the CGIAR se- own research should be actively commercialized verely wanting. by the private sector, and (e) the implications of membership of actors with specific interests in in- Addressing Intellectual Property Rights Issues fluencing the choices the CGIAR will make in The CGIAR is similarly inadequately equipped to these new, complex areas. The British commission address IPR issues to benefit the System and its on IPRs makes a variety of recommendations to developing-country stakeholders. Progress to- deal with what they term "the fundamental asym- ward remedying this problem has been slow. In metry in relations between developed and de- its review of the TSR recommendations on IPRs, veloping countries," including: the CGIAR decided against a single entity to hold patents, reportedly based on legal advice.21 · International negotiations of the IPR policies How then will the CGIAR pursue a compre- and their implementation to ensure global IP hensive IPR strategy given the articulated need systems "evolve so that they may contribute to and support for one? Consider the CGIAR in a the development of developing countries by context similar to that of research institutions in stimulating innovation and technology trans- industrial countries, such as agricultural univer- fer relevant to them while also making prod- sities or public research agencies. Many such in- ucts of technology available to them at stitutions have annual budgets for germplasm competitive prices" (Commission on Intellec- improvement work similar to that of the CGIAR, tual Property Rights 2002, pp. 2, 4) and complement this research with consistent, · Learning from international experience, in- centralized (for example, university-wide) policies cluding understanding the rapidly evolving na- and offices for managing intellectual property ture of international public-private partnerships assets. Such an approach is necessitated by the (box 3.2) 8 4 G E N E T I C R E S O U R C E S S T R AT E G Y A N D M A N A G E M E N T · Assessing the implications of IPRs for and of the ing agricultural pro- Only recently has the Convention on Biological Diversity ductivity, reducing CGIAR recognized the · Strengthening the capacity of international poverty, and incorpo- importance of IPRs. and national institutions involved in IPRs rating the demands of · Supporting the international architecture. smallholders and other types of agricultural pro- The Commission points out, for example, ducers. Third, the CGIAR must determine what that implementation of the International Treaty structure of governance and management is most on Plant Genetic Resources for Food and Agri- conducive to supporting a common IPR policy, and culture (2001) requires vigilance because of the whether it is prepared to accept the greater degree compromise wording and the scope for patents of centralization necessitated by such a policy. on the genes isolated from the plant genetic In addition to its own internal concerns over material under the aegis of the CGIAR and its im- IPRs, the CGIAR faces significant demand from de- plications for developing countries--hence the veloping country NARS for technical input on broad-based need for education and information how to address their own IPR issues, a demand on IPRs in global programs, international agen- that the CGIAR has not effectively met. Papers cies, and developing countries. commissioned by the meta-evaluation to capture Only recently has the CGIAR recognized the perspectives from developing countries indicate importance of IPRs and the need for the Centers an urgent need for assistance in the interrelated to accommodate IPRs as part of their overall ac- areas of biotechnology, IPRs, and management of tivities. The CGIAR now has System-wide IPR genetic resources. They consider such help es- guidelines, and several Centers have IPR policies. sential not simply to meet the needs of their poor, A Central Advisory Service was established at but as a way of maintaining their competitive- ISNAR in 1999 to assist Centers in understand- ness in the world markets even while being aware ing the effects of IPRs, to develop training for sci- of the urgent need to manage the biosafety aspects entists, and to establish policies and procedures. of the new technology (Romano 2002; Eicher Subsequently, several Centers have hired their and Ndiritu 2002; Macedo, Porto, Contini, and own IPR specialists and the CGIAR has also con- Avila 2002). ducted an audit of IPRs and IPR issues in each Thus, while the CGIAR has taken steps nec- Center. The unit has suffered from funding short- essary to keep the System's options open with ages, and ISNAR itself is being restructured, leav- respect to IPRs and biotechnology, the meta- ing the unit's future uncertain. This response is evaluation team does not consider this response inadequate to deal with the current challenges sufficient and cohesive at the System level, nor facing the CGIAR, a concern shared by many re- adequate in terms of the CGIAR's commitment spondents to the survey conducted by the meta- to developing country NARS. evaluation team. Results from the survey show that 79 percent of the respondents recognized Reconfiguring the CGIAR to Meet the need for a System-wide policy on IPRs, par- the Challenge ticularly in light of the growing importance of The present principles of governance in the IPRs in agricultural research.23 CGIAR have prevented the System from making But if the CGIAR is to develop a common IPR a variety of key decisions on the organization, policy for the System, three substantive issues management, and financing of a long-term ge- must be addressed immediately. First, the CGIAR netic resources strategy. The CGIAR does not must analyze the commonalities and differences have a structure that is The CGIAR's present in existing IPR agreements signed by various Cen- designed to address ters, and determine whether such an analysis, complex and contro- principles of governance along with input from other sources, can con- versial issues of global have prevented the System tribute to forming an appropriate IPR policy. Sec- public policy and sci- from making key ond, the CGIAR must provide a definition of entific considerations, "appropriate" that strikes a balance between rais- an issue raised by a re- decisions. 8 5 T H E C G I A R AT 3 1 structuring proposal from the CGIAR Commit- Another implication is the need to improve the tee of Board Chairs and the Center Directors' System's approach to long-term priority-setting Committee in 2000 (discussed further in Chap- and financing. At the System level, the CGIAR has ter 15). The challenging combination of free- yet to determine what types of technologies it will dom and collective action that characterizes the invest in over the long term, what types of col- System is described by one reviewer as "the laborations it is willing to enter into with the pri- tragedy of the commons" of the CGIAR. Although vate sector or other actors in agricultural research, TAC has commissioned System-wide reviews to and how it will distribute its resources between examine many of these contemporary issues, integrated genetic resources and natural resource the committee has been toothless in that it has management. Moreover, the CGIAR must con- no authority or ability to mobilize and promise tend with limited, restricted, or year-to-year donor support to the CGIAR on the key issues of sci- funding that is ultimately detrimental to planning ence policy. This paralysis has critical implications and executing long-term research.24 for how the CGIAR addresses related issues of In sum, the CGIAR's strategy, policy, organi- plant breeding research, gene banks, biotech- zational structure, and financing mechanisms nology, and IPRs. for genetic resources are currently inconsistent One possible way of examining this issue is to with the challenges it faces. The CGIAR urgently discuss the need for reconfiguration within the needs an integrated strategy that incorporates CGIAR System. While many reviews of the CGIAR plant breeding, genetic resource conservation, discuss System reconfiguration, the System-wide modern biotechnology, and IPR policy, backed up Review of Plant Breeding Methodologies (2001) by long-term investment and funding and a sup- put the necessity of such a process in stark per- portive organizational structure that brings the spective with respect to the CGIAR's research in most appropriate science to bear on the provi- plant breeding, genetic resource management, sion of global public goods to meet the needs of and biotechnology. The review observed the its poorest clients. It must also have a credible need for a greater interaction among and within process for priority-setting with a results orien- Centers or among researchers--within and be- tation that is independent of specific interests yond the System--working on similar crop types within the System and is held accountable to or similar research tasks, to generate beneficial the CGIAR membership at large on a regular synergies. It also makes a case for centralization basis, and so is able to explain the different op- of key operations or even consolidation of Cen- tions open to the System and the paths recom- ters to improve current research methodologies, mended and adopted. Both the strategy and the realize economies of scale and scope, and im- organizational structure must have the legiti- prove access to key inputs for research on orphan macy provided by the authorizing environment crops that would benefit most from improved ac- and sufficient authority vested in it to best serve cess to bioinformatics services. the interest of its ultimate clients. 8 6 11 Policy Research in the CGIAR P olicy research has grown more rapidly than any other research area in the CGIAR in recent years. At the same time, the lack of a conducive agricultural policy environment at both the global level and in devel- oping countries and a shortage of needed social science research and research capacity are frequently identified as major constraints to the adoption of new technologies and rapid, broad-based agricultural growth. The meta-evalua- tion team paid special attention to the scope, quality, and impact of this re- search.1 Policy and social science research are critical to systematic assessments. Thus, while the meta- ensuring the development, dissemination, and evaluation has been directed at the System level, adoption of new technologies that are beneficial in the arena of policy research it has had to to poor households. The CGIAR is not obligated focus on a specific Center, IFPRI, to a degree not to undertake such research simply because of its found with respect to other issues.2 international status and mandate. Rather, global The role of policy research in a science-driven research on policy and social science issues CGIAR was a matter of debate when IFPRI was should be carried out where it (a) is relevant to formed in 1973 and brought into the CGIAR in agricultural productivity and poverty reduction, 1977. The primary objective of establishing IFPRI (b) is high quality and cost-effective, and (c) was to improve the policy framework in devel- contributes to building long-term, high-quality oping countries, a national public good, which can research capacity in developing countries. constrain broad-based agricultural development.3 Policy research has been carried out at many Some feared that IFPRI would duplicate analyses CGIAR Centers, but nearly half of the CGIAR of trends and international food trade carried policy work is carried out in IFPRI, and the rest out by the FAO. TAC stressed that IFPRI should is spread throughout the other 15 CGIAR Cen- emphasize the problems of developing coun- ters. The predominant body of work, in both tries and the TAC chairman stressed the impor- scope and depth, is that of IFPRI. Moreover, tance of engaging and developing capacity of only IFPRI's policy research has been the subject nationals of developing countries to undertake of previous evaluation through EPMRs and other policy work so that they would influence policies 8 7 T H E C G I A R AT 3 1 work of the kind IFPRI has conducted--for ex- ample, on the reforms of the food price subsidy regimes. Even more important, they said, is as- sistance with improving their long-term capability to develop policy analysis and advice on their own. Both priorities follow from the original TAC guidance that IFPRI's research should em- phasize the problems of developing countries. An issue in the current CGIAR policy research program is the extent to which the program is giving sufficient priority to the areas of policy re- search where its comparative advantage is great- est relative to the FAO, the World Bank, and universities in advanced countries.5 This includes both the generation of research products and and their fine tuning in developing countries. building capacity for policy analysis in develop- TAC also suggested that IFPRI's in-house research ing countries. function should not compete in the provision of Policy research in the CGIAR is distinct from policy analysis services to donors, particularly social science research. Policy research is largely the World Bank, and to other CGIAR Centers. Fi- perceived as the province of economists and ad- nally, there was concern that locating IFPRI in dresses mostly sector, macroeconomic, and in- Washington, D.C., would give it a privileged po- ternational issues. Social science research sition and expose it to undue donor influence.4 (research by sociologists, anthropologists, and so Policy research can use one of two main ap- forth) is considered to focus on household and proaches: methodological and applied. IFPRI's community-level issues and is to help plant breed- methodological research, for example, has fo- ers and others to improve the relevance of their cused on ways to adapt and extend standard work. The distinction is important, but not with- welfare economics to agricultural and rural de- out problems. For example, in the past, IRRI and velopment policy work. Its applied research has ICRISAT economists carried out some of the best assessed existing and proposed policy alterna- CGIAR household research. Increasingly, social tives to estimate their benefits and costs. The ap- science research, broadly defined, has become plications fall into two main categories: policies more multidisciplinary and quantitative in the of national governments and international pol- universities of advanced countries. But there has icy issues that, since the formation of IFPRI, been relatively little evaluation of the CGIAR so- have emerged as important constraints to in- cial science research. Before turning to discussion centives for investment in agriculture in devel- of the impact of the CGIAR policy research, it is oping countries. IFPRI's publications in both worth reporting the findings of the social re- areas are widely read and respected. search conference organized at CIAT in mid-Sep- Closely related to the quality and relevance of tember 2002 involving participants (sociologists, the research products is building policy research anthropologists, and economists) from 13 CGIAR capacity in developing countries. Officials in de- Centers, as well as representatives of donors and veloping country governments and national agri- reputable social scientists from the universities culture research of developed and developing countries and sev- Developing countries seek systems told the eral national programs on all continents. assistance with improving evaluators that The meeting endorsed the role of social science their long-term capability to one of their pri- research, particularly upstream strategic research, mary needs is for in the CGIAR, and stressed the strong contribu- develop policy analysis and more country- tion such research makes when it is of the high- advice on their own. specific empirical est quality. But the meeting also concluded that 8 8 P O L I C Y R E S E A R C H I N T H E C G I A R social science research has declined in several funded by the inter- There is critical need for an CGIAR Centers over the past three-to-five years, national donor com- independent external review including in Centers that had a strong tradition of munity provided of social science and policy social science research. It particularly noted the they are relevant to research in the CGIAR. loss of critical mass and skill mix "impoverish- the long-term vision ment" among noneconomists, a 24 percent drop and strategy for the compared with a 2.2 percent increase in all sci- CGIAR. Ideally, unrestricted funds mobilized under entists since 1995/96. The report on the meeting an international umbrella organization such as the concluded that the research has been "pushed too CGIAR could be allocated to such proposals. Cen- much downstream at the expense of upstream ters with competitive capacity in policy and social knowledge generation for inclusion in and guid- science research could engage in this process, ance of integrated research programs [and that] while those without would ultimately reduce their service research often takes precedence over burden on the System and permit resources to be strategic research." It also noted that, while re- allocated to those with greater capacity--for ex- search on participation has grown, "genuine par- ample, other Centers, NARS, advanced research in- ticipatory research is often replaced with stitutions, universities, and the like. As designed, rhetorical/broad brush `participatory painting' at the first two Challenge Programs did not entail the surface of the research programs, in the ab- such competition. The CGIAR Centers are in the sence of specialized socio-cultural skills for sub- lead both in managing and in running the com- stantive research" (Cernea and Kassam 2002). petitive grants. While this concept comes close to One notable exception to these observations is the that envisioned by the CDMT for the Challenge Pro- research on Common Property Resources grams (discussed in Chapter 16), there are early in- (CAPRi), recognized for its outstanding quality dications that the selection of proposals is not both in its EPMR and at the Annual General Meet- consistent with this spirit. ing 2002. Even then the meta-evaluation team has concluded that the best, most advanced re- IFPRI's Policy Research: 1998 External search on common property resources in devel- Program and Management Review oping countries is not being carried out in the IFPRI's third EPMR reviewed the programs of CGIAR Centers, but rather in several U.S. univer- IFPRI's one outreach and four research divisions sities. CAPRi is also much better funded relative separately, with generally positive assessments of to other social science research in the CGIAR the quality and quantity of work done. The EPMR Centers. made four main recommendations. First, it en- The recommendations set forth in the report couraged continuing effort to document im- of the social science conference are thus a step pacts. Second, it recommended adjusting the in the right direction. They emphasize strength- research agenda to reemphasize water issues ening capacity, institutional support, and part- and take into account developing countries' in- nerships. Still, there is critical need for an teractions with the world economy. Third, it rec- independent external review of social science and ommended that relevance be increased by taking policy research in the CGIAR, not only to examine developing country concerns more fully into ac- the System's comparative advantage, but also count and by diversifying the staff to include the allocation of resources among various more staff with policy as well as research expe- System-wide programs. rience. Fourth, it recommended improving out- The meta-evaluation team has concluded that reach by better integration and mutual one potential means of improving social science re- reinforcement of research and outreach activi- search (and policy research, discussed below) on ties. The latter two recommendations are sup- a global basis is the use of genuinely competitive ported by TAC, which suggested that there approaches to research--unlike that being used should be more country-specific, focused re- currently in the Challenge Programs and discussed search (as opposed to large-scale, multi-country in Chapter 16, in which the best proposals are efforts), and that IFPRI should become more 8 9 T H E C G I A R AT 3 1 actively involved in "emerging policy debates" veloping country nationals with experience in pol- (TAC Secretariat 1998a, p. vi). The EPMR also icy analysis remain few in number on IFPRI staff. recommended higher priority, even at the ex- Comments received by the meta-evaluation pense of work in other areas, on "a more in- team suggest that outreach in developing coun- depth involvement with policy research and tries (and genuine participation of developing advice on low-income transitional economies country institutions) could be enhanced in ways about how to institute a functioning market that would make IFPRI's research more pivotal economy." While this may seem counter-intu- in strengthening the capacity for policy analysis itive for a system oriented toward global and re- in those countries. Internally, this would require gional public goods, so much of the donor better integration between the outreach and the conditionality in small and least-developed coun- research divisions. Externally, it implies stronger tries, such as those in Africa, has lacked an em- linkages with organizations dedicated to research pirical base. Informed debate on policy capacity building in developing countries. improvements would be enhanced, for example, by a better understanding of the functioning of Other Reviews of IFPRI Impacts the markets and institutions involved, or of the In evaluating the impact of policy research and impacts of policy reforms. Such research could outreach in developing countries, case studies best be conducted with the active participation of IFPRI impacts on policy decisions are helpful. of nationals across several countries. Ryan (1999a) investigated the returns to IFPRI's These issues temper the reception of IFPRI's rice market and policy research in Vietnam. The generally well-regarded 2020 initiative as well as work was well integrated with Vietnam's Ministry other research that has focused on professional of Agriculture and Rural Development, thus min- audiences. The IFPRI response to the EPMR imizing the outreach problem of getting an au- agrees with the need to become more proactive dience for the findings. IFPRI made a large on outreach. But the response suggests a struc- number of generic and specific policy recom- tural problem. It states: "While most of IFPRI's mendations, among them recommendations to outreach activities have been and will continue liberalize rice trade both internally and externally. to be undertaken within the research divisions Vietnam did relax rice export quotas and inter- as integral parts of research projects, we visual- nal restrictions on rice trade. Increased rice ex- ize that the Outreach Division will play the dual ports under these policy changes were estimated role of supporting such outreach activities while by Ryan to benefit Vietnam by about $60 million undertaking other outreach activities that are per year during 1996­01. A "conservative" view more appropriately done outside the research di- of IFPRI's causal role in the policy change-- visions, such as capacity strengthening and in- which attributes to IFPRI an acceleration of pol- formation dissemination that cut across divisions" icy change that resulted in $45 million in gains (TAC Secretariat 1998a, p. xiii). The structural during 1996/97--obtains a benefit-cost ratio of issue is whether the kind of outreach that would at least 45 based on policy research that cost just best serve CGIAR purposes can be done with this under $1 million. division of labor. A quite different picture emerges from a re- IFPRI's response to the recommendation on rel- view of IFPRI's 10-year program of work in evance says that the institute's priorities already are Malawi.6 The review cites "the adverse percep- "heavily influenced by developing countries' tion that IFPRI is an expensive organization, needs," and IFPRI already has senior researchers which is too close to donors, too Washington- with significant policy experience. The response centric, and too possessive of the databases it goes on to agree to take steps in these directions, generates." It concludes by quoting with ap- Outreach could be but evidently IFPRI proval an interviewee, "it seems IFPRI may have does not see a prob- concentrated too much on data collection and enhanced to strengthen lem where the EPMR too little on building solid linkages with the pol- country capacity. does, although de- icy environment in Malawi. Links with the donor 9 0 P O L I C Y R E S E A R C H I N T H E C G I A R community are not a substitute for this, as staff market-distorting poli- CGIAR's role in turnover and changing priorities are not con- cies in developing coun- stimulating capacity ducive to sustainability" (Ryan 1999b, pp. 35­36). tries; and analysis of needs more attention Similarly, the 2020 Vision Project's efforts in the relationships be- through independent regions of Africa have received both praise as in- tweeneconomicgrowth, novative initiatives and blame as activities pe- environmental improve- evaluation. ripheral to and largely divorced from IFPRI's ment, institutional de- core research program "as an afterthought or to sign, and poverty reduction, including issues in placate the critics amongst its donors," as one nutrition and health. As an example of a positive as- person put it. Experts consulted for this review sessment along these lines, Farrar (2000), in his disagreed about some aspects of IFPRI's sup- broad review of IFPRI's food subsidy work, does not port for capacity building in policy analysis and estimate quantitative impacts but makes the case social science research to support such analysis. well that IFPRI's long series of work on food sub- Some give high marks to what IFPRI has done in sidies made it the recognized world source of ex- improving the capacity of government analysts pertise on the subject. and others involved in policy in Africa, espe- But in other areas the priorities are question- cially in the period up to the early 1990s. able. It is arguable that IFPRI gives too high a pri- Others point to what has been described as ority to world supply-demand projections and a Washingtoncentric or stop-and-go characteristic technical pursuits such as developments in com- of IFPRI's involvement in African policy issues and putable general equilibrium modeling and simu- education as evidence of a lack of commitment lations, relative to applied policy analysis of by IFPRI. At the same time, proponents point to developing countries (carried out in those coun- IFPRI's ongoing research on the economics of tries with IFPRI participation in collaborative re- African agriculture, reasonable advice, and train- search and guidance), analysis of political economy ing of personnel from African governments, and issues, country-by-country analysis of the damage note that no other outside government, inter- being done to developing countries by industrial national agency, or NGO can claim even that countries' agricultural subsidy policies, and trans- much influence or effect. Yet without question, parent assessment of the impact of developed capacity for policy research in developing coun- country policies on both the rural poor and in- tries and the CGIAR's role in stimulating such ca- centives to investment in agricultural research pacity through collaborative research with and development in developing countries. IFPRI nationals of developing countries needs more at- has carried out several valuable studies on these tention through independent evaluation of the subjects. CGIAR research thus provides a wel- CGIAR's policy research. Several CGIAR Centers come strengthening of the World Bank and oth- have indicated to the meta-evaluation team that ers on OECD subsidies to agriculture and barriers the Centers located in developing countries can to agricultural exports from developing countries be more effective in this regard than those such in forums such as the 2002 World Summit on as IFPRI and ISNAR, located in the capitals of in- Sustainable Development in Johannesburg. The dustrial countries--a hypothesis worth investi- issue is one of further work and dissemination ef- gating in a System-wide review of policy research. fort to deliver possibly unwelcome messages to Overall, the most important issues concern- developed countries. A need has also been ex- ing IFPRI, and policy research in the CGIAR more pressed for research on the short-term impacts broadly, are not the quality or quantity of work, of liberalization, if and when it occurs, in devel- but priorities in research topics and genuine out- oping countries; specifically, for research that rec- reach to developing countries through closer in- ognizes the complex interactions and responses volvement of local institutions and capacity of private entrepreneurs, farmers, government building partnerships. IFPRI has generated good agencies, and consumers to liberalization, and research in many areas, such as developing meth- provides advice on minimizing the adjustment ods of impact assessment; policy analysis of costs to more vulnerable sectors of society. 9 1 T H E C G I A R AT 3 1 IFPRI's main objective A research agenda side, country-specific advice is by definition not should be the simpler one of of special interest to a global public good unless it influences poverty adding to the world's pile of the CGIAR as well as reduction on a global scale, and IFPRI's efforts donors is IFPRI's ef- could become too diffuse if it attempted to pro- good policy choices. forts to develop and vide, in effect, host-country advice on a quasi- carry out impact as- global basis without a strategic approach to sessments of individual Centers and of the whole ensuring spillovers from its research. CGIAR System. Especially notable is the assess- The questions raised by the meta-evaluation re- ment of impacts on poverty and the spillover garding policy research parallel those about the gains developing countries have received as a re- role of the CGIAR that occurred at its inception. sult of germplasm research in Centers. IFPRI According to one of the participants interviewed has also carried out helpful meta-reviews of Cen- by the meta-evaluation team, one proposed ob- ters' impact assessments.7 In addition, IFPRI has jective of the CGIAR was to make the Centers done some notable analytical work on how to the focus of world research on crop science; the evaluate its own products.8 Based on its self- alternative was the more homely objective of hav- evaluations, the areas of policy research that ap- ing the Centers add as much as possible to the de- pear most promising in generating net social veloping world's "pile of grain." The two are gains are those that influence policy decisions by related, of course, but the claim of the inter- national governments or international institu- viewee is that had the crop-science objective pre- tions. Increases in incomes or other social indi- vailed, the Centers might have become great cators that result from appropriate policy sources of articles in scientific journals, but would decisions can then be credited in part to the not have produced the series of yield-increasing findings and influence of that research. varieties that cemented the reputation of the A more concentrated focus on policy advice CGIAR as a solver of the world food problem. for developing countries has both pluses and mi- The point for IFPRI is that its main objective nuses. On the plus side, this is where the likely should be the simpler one of adding to the world's net social benefits of policy research have been pile of good policy choices, which in turn can re- and are expected to be highest. Some members sult in journal articles of missions accomplished of the meta-evaluation's advisory committee and and reasons why. More generally, the CGIAR faces others believe the opportunities here are so a growing challenge between the scale of impact promising that IFPRI should be largely devolved on the ground, on one hand, and international vis- to developing countries so as to be in a better ibility to maintain donor support on the other. And position to conduct research in collaboration in the view of the meta-evaluation team, the Sys- with developing country NARS and to formu- tem balance has lately tilted too much in favor of late and deliver such advice in active partnership meeting donor and industrialized country ex- with the countries concerned. On the minus pectations rather than achieving results. 9 2 12 Natural Resources Management Research in the CGIAR R esearch on natural resource management and its links to productivity growth have been a matter of considerable and continuing debate in the CGIAR, raising issues of strategic focus.1 The meta-evaluation team identified natural resource management (NRM) research as a topic for focused treatment because its position in the CGIAR research portfolio has become increasingly prominent over the past decade or so. This chapter assesses the quality, coverage, and impact of the CGIAR's NRM research and extracts im- plications for the System's science strategy, organizational structure, and fi- nancing mechanisms. The meta-evaluation team also sought to estab- or scope for beneficial spillovers that are closely lish the global public goods nature of research tied to the CGIAR mission. on NRM--a difficult task since resource degra- Other evaluation difficulties arise because pri- dation has local, national, regional, and global mary source material for an evaluation of NRM re- consequences. search is thin, despite the rapid growth of this The difficulty arises because, in contrast to research area. The CGIAR has produced few im- germplasm research, NRM research is not per- pact assessments on NRM research, so it is diffi- formed in laboratories, it addresses less homo- cult to say whether NRM research has been geneous issues, and its benefits are spread over effective within the CGIAR. Of course, the absence several timeframes and across different scales and of such evidence does not imply an absence of can have broad spillover effects. Scalability, time impact, rather that there is no established set of horizons, externalities, and measurement--is- methodologies for NRM impact assessment. Yet sues common to NRM research--are far more there is enough information about NRM activities complex than those encountered in other kinds worldwide and within the CGIAR to draw some of research. Furthermore, the scope of NRM re- plausible conclusions. search is potentially so broad that establishing First, NRM research is central to sustainable priorities requires the use of clear criteria, based productivity increases in agriculture and to on comparative advantage, core competencies, improvements to rural livelihoods worldwide. 9 3 T H E C G I A R AT 3 1 CGIAR is correct to The CGIAR is cor- capacity building is a serious challenge that de- emphasize integrated rect to emphasize mandsSystem-wideattentionanddonorfinance.3 natural resource integrated natural Fifth, the resources-oriented Centers may be resource manage- doing more and better work in integrated NRM management and integrated ment and integrated than are the more established, commodity-ori- genetic resource genetic resource ented Centers, with the ecoregional Centers management as its twin management as its falling somewhere in between. Although a few twin pillars support- System-wide programs are making significant pillars supporting ing agricultural pro- advances toward addressing global problems, agricultural productivity ductivity enhance- their objectives generally far outreach their re- enhancement. ment. sources or authority, thereby limiting their ef- Second, NRM re- fectiveness.4 search has the potential to generate global pub- Sixth, legitimate concerns have been raised lic goods in the form of new knowledge, especially about the NRM research portfolio. Some of these on core processes and on methods of analysis and concerns appear in Center or System-wide pro- measurement, as well as in meta-datasets with gram reviews, others are apparent more at the global coverage. NRM research has made im- System level than at the Center or program level. portant intellectual contributions to water man- These concerns can be usefully sorted into two agement, tropical deforestation, characterization sets of issues--focus and framework--and have of agro-ecosystems, and sustainable resource been identified by TAC (see Barrett 2002). management on marginal lands. But these ac- Seventh, and most important, NRM research complishments notwithstanding, the CGIAR cur- has attracted increasing interest and resources rently falls short of realizing its considerable over the past decade. But these may not have fo- potential to generate global public goods, pri- cused on the topics and functions where the marily because of System-level issues of focus CGIAR can make tangible, high-return contri- and framework (discussed below). Satisfactory butions to global public goods, that is, in con- resolution of these issues would do much to tributing to sustainable agricultural productivity push the CGIAR to its possibilities frontier. increases and improving the livelihoods and re- Third, and related to the preceding point, ducing the vulnerability of the rural poor. the CGIAR's NRM research programs sometimes In sum, the CGIAR's NRM research can be venture beyond the System's core competen- justified by the System's impressive, well-estab- cies without providing a compelling case for the lished agricultural impacts, but only as long as strategic importance of the research. The ex- the NRM research portfolio stays true to the Sys- pansion of the System in the early 1990s added tem's core agricultural productivity agenda. scope without commensurate growth in funding, It is understandable that donors, bearing a fi- which increased the pressure to leverage donor duciary responsibility for wise use of their re- resources and led to a drift in the research pro- sources, demand impact assessment of the NRM gram. This has undermined the traditional ex- portfolio. cellence of CGIAR science, including its NRM research. Thematic Priorities in NRM Research Fourth, the CGIAR has made significant pro- Since pushing more heavily into NRM about 10 ductive investments in training individual sci- years ago, the CGIAR has been experimenting entists from developing-country NARS and, in a with a variety of approaches to NRM research, few cases, in helping develop NARS institutional using different, evolving terms and modalities-- capacity, regional networks, and subregional or- such as sustainability research, ecoregional ap- ganizations relating to NRM.2 This capacity build- proach, integrated natural resources management ing has declined in recent years, although the (INRM)--as it has thought through a coherent need remains acute. Given the funding and per- NRM research strategy. There is now essentially sonnel challenges many NARS face, NRM-related universal recognition in the CGIAR of the need 9 4 N AT U R A L R E S O U R C E S M A N A G E M E N T R E S E A R C H I N T H E C G I A R for NRM research to address not only current mealy bug through- The CGIAR's successes in poverty and food insecurity but, at least as im- out Africa. ICRISAT research on terrestrial portant, to prevent future poverty and food in- has similarly enjoyed resources underscore the security by protecting the natural resource base great success in de- complementarity between on which future productivity improvements de- veloping and pro- pend. Yet, as a recent TAC report notes, "In the moting effective IPM NRM and germplasm past, research on natural resources has been too for pigeon pea in improvement research. often conducted in a disjointed, fragmented fash- South Asia.5 ion" and "[n]otably absent...is a coherent System- What IPM means in concrete terms has been wide strategy for INRM priority setting and for a matter of debate.6 A recent World Bank Work- operationalizing a more effective set of strategic ing Paper observed that the CGIAR's System- INRM activities within the CGIAR" (TAC Secretariat wide program on IPM is still struggling to deliver 2001d, pp. 1, 4). In hindsight, it is clear that the on its promises to increase the impact of IPM re- CGIAR joined the early-1990s, "sustainable de- search on pest management practices. A recent velopment" bandwagon and is now recognizing review has suggested elevating the SP-IPM to a that the resulting NRM research portfolio did Global Challenge Program with improved links not necessarily match either its core competen- to other players in IPM research and outreach ac- cies or its strategic objectives. Nor did it neces- tivities (Guitierrez and Waibel 2002). "Major con- sarily fill gaps left by the rest of the scientific straints to a more significant role for SP-IPM are community with related research interests--is- the lack of incentives for participatory multidis- sues that are beginning to receive greater atten- ciplinary research, the gap between scientific tion in the CGIAR. IPM information and user-friendly outreach, and The thematic priorities within the CGIAR's limited impact on national policies" (Sorby, Fleis- NRM research can be roughly summarized as (1) cher, and Pehu 2003, p. 28). The World Bank management of terrestrial resources (soils, flora, Working Paper also emphasizes "the lack of doc- fauna) to enhance sustainable agricultural pro- umented experiences from scaling up pilot proj- ductivity; (2) integrated water management for ect approaches . . . [and] the lack of standards both quality and quantity as an input to agricul- of impact assessment, especially regarding the so- ture and as a habitat for living aquatic resources; cial and environmental benefits of IPM" (Sorby, (3) management of forests for enhancing rural Fleisher, and Pehu 2003, p. ix). livelihoods and providing sustainable sources of Similar yield-enhancing achievements have fuelwood and non-timber forest products; and (4) been made through research on land manage- incentives and policies for improved NRM man- ment practices. The conservation technique, or agement (TAC Secretariat 2001d, pp. 1, 4). minimum tillage, that originated in the NARES The CGIAR's successes in research on terres- and spontaneously among farmers in South trial resources cover several areas and under- America, has been effectively analyzed and is score the complementarity between NRM and being promoted by the CGIAR Centers. But the germplasm improvement research. Among the extent of adoption and spread are unknown, as most well-known examples are the achievements is the extent to which the impacts can be at- in integrated pest management (IPM), especially tributed to the CGIAR's effort since there are a by IRRI and the FAO in Southeast Asian rice. The variety of other actors promoting similar prac- CGIAR played a pioneering role in the emer- tices. Improved crop varieties have in many cases gence of IPM, now a global staple for effective encouraged monoculture that can leave farmers management of pests while minimizing the nec- vulnerable to pests and disease on a catastrophic essary use of potentially dangerous agro-chemi- scale. Work on farm-level biodiversity and ef- cals. The CGIAR's IPM work has had an impact on fective mixed systems and crop rotations, for many different systems using the basic princi- example, by CIAT and ICRISAT in rain-fed small- ples it helped pioneer. IITA, for example, devel- holder agriculture, has been central to combat- oped successful biological control for cassava ing these prospective problems and enabling 9 5 T H E C G I A R AT 3 1 some considerable yield gains. Various Centers work in process research. ICRAF, CIP, IFPRI, and have likewise been at the forefront of research ILRI have been actively pushing the frontiers of to improve soil conservation methods, and to es- process modeling of complex agro-ecosystems, tablish their relative benefits. Methods include especially in capturing the interaction of naturally tied ridges, vegetative and stone bunds, and ter- occurring biophysical processes (nutrient cy- racing--techniques that have been shown to in- cling, soil erosion, biomass regeneration) and crease yields and profitability for small farmers those managed by farmers. These efforts often (Sanders, Nagy, and Ramaswamy 1990; Barrett, heavily leveraged ARI resources to address Cen- Place, and Aboud 2002). ter objectives. Examples include CIP's work with The CGIAR significantly revised its research Montana State University in modeling tradeoffs program with respect to terrestrial resources after between pasture-potato systems productivity, TAC's 1996 study, Priorities for Soil and Water As- human health, and soils sustainability in the pects of Natural Resources Management Research Andes, and IITA's work with Texas A&M Univer- in the CGIAR (TAC Secretariat 1996c). This ad- sity in developing useful Spatial Characteriza- vanced a research vision, also anticipated in the tion Tools. Unfortunately, such examples remain INRM System-wide program, focused on broad- more the exception than the rule, and their im- based management of land and biological re- pacts or relevance to developing-country needs sources, including genetic material, to meet are not known. productivity, poverty, and sustainability goals. In re- Water is central to agriculture. Water is also, sponse, several Centers, notably CIAT, through according to the Bank's World Development Re- Tropical Soil Biology and Fertility (TSBF), and port 1992: Development and the Environment ICRAF have advanced a paradigm of integrated soil (and subsequent editions of the WDR), the sin- fertility management (ISFM) that is currently at- gle biggest environmental challenge facing the tracting considerable attention. Research on soil poor. Many of the original CGIAR Centers helped fertility management previously focused on the use foster great improvements in water control across of mineral inputs to sustain crop production, and Asia and, to a lesser degree, Latin America dur- showed substantial crop yield and profitability in- ing the Green Revolution. More recently, IFPRI, creases from fertilizer use. ISFM addresses small IRRI, and CIMMYT research has reconfirmed farmers' low mineral fertilizer application rates in what has been recognized by agencies such as Africa and environmental degradation problems as- the World Bank for some time: that poor water sociated with high application rates in Asia and pricing policy in the wake of rapid expansion of Latin America since the Green Revolution. The irrigation has contributed significantly to emer- ISFM paradigm has further broadened the scope gent, widespread water shortages, salinization, for potential yield-increasing interventions in a and waterlogging of lands. The World Bank's number of ways, helping direct and use germplasm WDRs and other reports have stressed that ur- improvements. ISFM has established the impor- banization and rising incomes and population are tance of and promoted research on soil biology as also rapidly increasing competition for water central to making nutrients available to plants and between agricultural and nonagricultural uses. forages and to creating and maintaining produc- The water research challenge has thus shifted in tive soil structure. ISFM also highlights the need parts of the world to require far more policy re- for improved germplasm. Improved crop search than previously seemed necessary. Al- germplasm has a major role to play not only in im- though the global problem remains acute, IFPRI proving nutrient acquisition but also in providing and IWMI have been doing useful modeling work more organic inputs, prompting efforts by sev- on these problems. These various efforts by eral Centers to develop dual-purpose or multi- many agencies have heightened awareness of Water is also the single purpose grain the water issue among donors, as emphasized at legume varieties. the World Summit on Sustainable Development biggest environmental The CGIAR has in Johannesburg. Pressing problems develop- challenge facing the poor. done pioneering ing countries face are increasing water use effi- 9 6 N AT U R A L R E S O U R C E S M A N A G E M E N T R E S E A R C H I N T H E C G I A R ciency (that is, the proportion of surface and groundwater that reaches crops' root zones) and land reclamation in response to salinization and waterlogging. An important strategic question for the CGIAR in its policy research is whether its comparative advantage lies in addressing the science related to the agricultural uses of water or in down- stream institutional issues such as conflict reso- lution. To date, the CGIAR has not demonstrated a significant comparative advantage in these areas, and has no demonstrated record. Elsewhere, particularly in Africa, water con- servation remains a serious challenge, as does in- creased water use efficiency. Less than 11 percent of agricultural lands in Africa were irrigated in lacking coordination and cooperation among 1999, the latest year for which data are avail- potential partners. Surface waters were treated able. Germplasm improvements will likely prove separately from ground waters; water quality, difficult absent better water management in drier independently from water quantity and each areas that lack good water control and efficient sector of users (i.e., agriculture) was ignorant of use. The CGIAR does not seem to have tackled all the others. This approach to research often this challenge adequately, and it is likely to re- led, not surprisingly, to inadequate policies that quire new science to attend to basic issues of were not well suited to solve problems ad- water production and distribution, not just im- dressed" (TAC Secretariat 2001d, p. 6). Research proved management of existing water resources. on integrated water management appears to Improved water science and technology appear have underperformed its potential within the to be vital if Africa is to enjoy significant sus- CGIAR, largely for organizational reasons related tainable gains in agricultural productivity. to insufficient multidisciplinarity, especially weak IWMI has been a leader in advancing the ho- incorporation of social science research, inade- listic approach of whole catchment/basin analy- quate incorporation of water productivity re- sis in water management. It has produced key, search into crop productivity research, degraded global-scale strategic analyses of water resources international hydrological data collection infra- and valuable new methodologies for water ac- structure, and perhaps excessive concentration counting, and the most authoritative data ref- of CGIAR water research capacity in a single erence on world water (the World Water and Center, without the involvement of key com- Climate Atlas); developed methodologies for modity Centers. Given the massive amount of re- the measurement of water productivity; and search on water management done within ARIs done quality research on organizational design and with the advanced NARS in developing coun- for irrigation systems. But the impact of this re- tries, there are also crucial questions as to the search is not known. ICLARM has developed CGIAR's niche in integrated water management key global databases on fish and reef systems that research. are an important global public good, and its One major question surrounds the future of tilapia programs have brought sharp productiv- the System-wide Initiative on Water Management ity gains in inland aquatic systems. (SWIM). The objectives of SWIM and IWMI, These successes notwithstanding, TAC has SWIM's convening Center, are essentially indis- expressed worries that "[f]or too long, research tinguishable. SWIM lacks the focus of most of the on water issues has been disjointed, based on tra- CGIAR's other System-wide programs and has ditional disciplinary sciences without crossing become largely a vehicle for IWMI to obtain ad- boundaries, focused on short-term issues, and ditional funding to extend its partnerships with 9 7 T H E C G I A R AT 3 1 Core design questions other Centers. Moreover, on sustainable exploitation of forest resources remain, including the most of these partner- and forest recovery after fire have generated relationship between ships are bilateral, limit- high-quality publications and filled voids in the ing effective System-wide global scientific community. ICRAF has evolved and among SWIM, the collaboration on issues rapidly from an institution focused narrowly on Challenge Program on germane to multiple Cen- agroforestry into a leading Center for integrated water and food, and ters. For these reasons, nutrient management research, with an em- SWIM's external review phasis on improved fallows, biomass transfer IWMI's work program. panel questioned whether systems, green manure cover crops, and mixed SWIM really functions ef- sylvicultural systems. ASB has been applauded in fectively as a System-wide program and recom- its most recent external review for innovative field mended that TAC consider phasing it out. TAC research, strong science, and for going furthest rejected this suggestion, but the core design within the CGIAR toward effectively imple- questions clearly remain, including such issues as menting a holistic, ecoregional approach the relationship between and among SWIM, the founded on in-depth local research linked recently approved Challenge Program on water methodologically across long-term benchmark and food, and IWMI's work program.7 sites around the world to permit effective scal- The two lead Centers on forest resources re- ing up to the global level. It forms the backbone search, CIFOR and ICRAF, and the System-wide of the Rainforest Challenge Program now be- Alternatives to Slash and Burn (ASB) Program fore the iSC. The intellectual value of this work have all received highly laudatory external re- has derived from the synthesis afforded by care- views, and justly so. Given the global impor- ful methodological coordination across sites on tance of the tropical deforestation problems and different continents, and close working rela- the previous dearth of high-quality research link- tionships with ARIs and NARS (Angelsen and ing agricultural technology development, sus- Kaimowitz 2001; Lee and Barrett 2001). ASB has tainable intensification, and tropical forests, this also contributed to methodological research area of work seems a wise investment by the into indicators of above-ground biodiversity and CGIAR. Nonetheless, there are tendencies, per- carbon stocks and spatially explicit land use haps most evident at CIFOR, for NRM research modeling, as well as to policy research on quan- to drift from research squarely focused on im- tifying tradeoffs among agronomic, conserva- proving agricultural productivity and rural liveli- tion, and socioeconomic objectives and on the hoods in the low- and middle-income countries opportunities conservation credits potentially toward topics that are of more interest to envi- afford for small farmers in the tropics. It would ronmental groups in the high-income countries, be useful to assess the impact of this research on such as the tropical moist forests. For example, developing countries' own capacity to address is- although some CGIAR research on carbon se- sues of poverty alleviation and sustainable en- questration and climate change mitigation has vironmental management. been at the forefront of scientific efforts to de- In sum, some of the CGIAR's NRM research is velop good estimates of carbon stocks--some of well-regarded science, as reflected by peer- the ASB work has been used as an input into re- reviewed publications. More important, there is cent IPCC guidelines--it is not clear whether evidence of impact through significant im- the CGIAR's work on climate change mitigation, provements in the portfolio of NRM practices regardless of its high quality, fills a significant void and technologies available to small farmers, as in in the broader scientific community. the cases of IPM and the development and pro- CIFOR's work on the relationship between motion of improved nitrogen-fixing fallows. Doc- agricultural technologies and deforestation has umenting the adoption of these practices and the established the intellectual frontier in this area global returns to NRM research remains elusive, (Angelsen and Kaimowitz 2001), and its work in however, in large measure due to serious developing biodiversity assessment tools and methodological challenges and the limitations 9 8 N AT U R A L R E S O U R C E S M A N A G E M E N T R E S E A R C H I N T H E C G I A R posed by location-specific circumstances (Barrett 2002). The agricultural sciences community ap- preciates that researchers cannot breed endlessly for soil nutrient deficiencies or drought toler- ance. At some point, soils become exhausted and water scarcity sharply limits productivity im- provements. Improved resources management is, in most cases and over the longer term, a com- plement to development of improved germplasm. But since these two research foci compete for scarce resources within the CGIAR, significant ef- fort needs to be made to establish more pre- cisely how NRM research contributes to agricultural productivity improvements and poverty reduction, where these investments are most likely to generate acceptable rates of return The justification for a system of international in the future, and how the research contributes agricultural research Centers lies in the CGIAR's to the supply of global or regional public goods role as a producer of global public goods. This that cannot be, or are not already being, pro- poses a challenge for much NRM research be- vided at the national level. cause most of the current empirical and applied or adaptive research is inherently site-specific, Focus and therefore does not generally produce global The focus of the CGIAR derives from its core public goods (TAC Sec- competencies and the justification for the Sys- retariat 1999c). Several The CGIAR's core tem. The CGIAR's core competencies lie in agri- external reviews indi- competencies lie in cultural issues, not in environmental and natural cate that there is some resource issues, so the System should not be ex- apprehension and mis- agricultural issues, not in pected or assigned to make fundamental scien- understanding within environmental and tific contributions in NRM that do not directly Centers as to how the natural resource issues, so support agriculture. NRM research, like local-global research germplasm research, is indispensable to the link needs to be made. the System should not be CGIAR's agricultural research program, but it This is most clearly cap- expected or assigned to must not proceed independently of that focus. tured in the 1998 re- make fundamental The issue is not whether NRM research comes port of the CIFOR second to work on varietal improvement: there review panel.9 Although scientific contributions in are places where improved NRM likely would not explicitly high- NRM that do not directly have a greater impact on agricultural productivity lighted in other re- support agriculture. than would germplasm research, and vice versa.8 views, the CGIAR Rather, the System needs to "walk on both legs," leadership faces signif- without one drifting too far from the other. icant challenges in financing, priority setting, There is some potentially valuable NRM research and scientific criteria for approval when apply- being conducted that is nonetheless tangential ing the global public goods criterion, while at the to the CGIAR's core mission, perhaps because same time seeking in-depth research in specific, donor funding available to skilled scientists some- carefully chosen sites, which emphasizes the times diverts the research agenda in Centers importance of strong scientific review capacity suffering budgetary stress. This is occurring at within the System, albeit in a smallholder de- multiple levels, as manifested by EPMR concerns velopment context. about focus in multiple Centers (CIFOR, CIP, The CGIAR foci that satisfy both the above ICARDA, ICRISAT, and others). points are process and methodological research 9 9 T H E C G I A R AT 3 1 and the compilation and distribution of meta- This may reflect insufficient recognition of the data. The CGIAR is generally doing well in the first global public good value of producing and dis- and second of these, contributing important, seminating high-quality, well-documented data. fundamental research on systems modeling, ap- When only publications resulting from such data propriate sustainability indicators, methods of es- count in personnel management, scientists and tablishing and quantifying tradeoffs between Centers face strong disincentives to the timely, biophysical and socioeconomic objectives, nu- public release of data they collect. Access to trient cycling and soil microbiological processes, such data by developing countries' NARS is an and impact assessment and evaluation meth- even more serious issue. There are good exam- ods. The global public good value added from the ples within the CGIAR, such as the System-wide knowledge produced could nonetheless be en- Information Network for Genetic Resources hanced in many Centers and System-wide pro- (SINGER), that have already increased access grams. This could be accomplished by an and offer useful examples of how to remedy this increased commitment to peer-reviewed publi- deficiency, although several reviewers of this re- cation and to ensuring the accessibility of sci- port have noted data sharing and data loss prob- entists' research, particularly for the NARS of lems within SINGER and many other studies the developing countries, through journals, books, Centers have conducted. and working papers readily available on the In- Institutional capacity building at the national ternet. Some Centers have outstanding publi- and subregional levels is fundamental to doing ef- cations records (such as CIMMYT, ICRAF, IFPRI), fective NRM research. In the CGIAR, however, al- but many important CGIAR NRM research find- though considerable individual-level training is ings are not distributed broadly or quickly being done, there is scant evidence of systematic enough to have full impact, particularly for po- capacity building or maintenance at the institu- tential users in developing countries. tional level within NARS that can be directly linked Meta-datasets offering truly global coverage, to CGIAR NRM research. ISNAR has thus far such as the recent IFPRI-WRI agro-ecosystems proved unable to fulfill this crucial mission. If the mapping and IWMI's World Water and Climate CGIAR is to increase its emphasis on strategic Atlas, offer opportunities to improve the tar- partnerships, then institutional capacity devel- geting of research and technology development, opment--not just individual professional devel- as well as crucial baseline information on which opment for the best NARS scientists--seems to found ongoing agro-ecological monitoring essential.10 This is especially challenging in low- activities (Wood, Sebastian, and Scherr 2001). income countries with weak NARS. Some nas- When well documented and readily accessible to cent efforts at building subregional organizations prospective collaborators worldwide, such (SROs) among NARS and other stakeholders to datasets can generate important knowledge achieve critical mass across countries facing sim- spillovers globally. But such meta-datasets, in- ilar challenges show promise, such as CIMMYT's cluding those that interface with poverty inci- Soil Fertility Network in southern Africa. The in- dence data, are under-produced, even though terim Science Council has proposed a System-wide this was an explicit objective of the CGIAR's training review, and although this is a high-prior- ecoregional initiative. ity issue for developing countries, donor funding Like too many publicly funded ARIs, some has been difficult to mobilize. Centers and System-wide programs seem to treat data as proprietary. Framework Some Centers and System- Even NARS and ARI The framework for NRM research in the CGIAR wide programs seem to treat partners can have System encompasses a variety of organizational data as proprietary. difficulty accessing and financial issues. The CGIAR has special skills relevant data nec- and a unique mandate, but as noted in the case essary to do their own analyses, although some of social science research (discussed earlier), of these data originate in developing countries. these do not endow it with comparative advan- 1 0 0 N AT U R A L R E S O U R C E S M A N A G E M E N T R E S E A R C H I N T H E C G I A R tage for all areas of NRM research. In a few areas such as ASB the ex- The collaborative model it is best suited to be the direct producer of ception rather than offers an opportunity for the NRM research, such as NRM that helps improve the rule. Hence, CGIAR to overcome its production practices in connection with how the Challenge capacity limitations. germplasm improvement and in areas such as Programs are being tropical deforestation, where ARI and NARS re- developed is an search has been deficient. Yet the research model issue of particular interest, which is discussed in pioneered within Centers with global commod- Chapter 16. ity mandates may not be well suited to NRM re- The number and organization of Centers was search. One indicator is that the Centers widely also addressed by the CDMT (2001). In NRM re- evaluated as doing the best work in NRM have search, just as in agricultural research more either had ecoregional mandates (CIAT and broadly, there is a nontrivial issue of the critical ICRISAT in the past) or have an explicit NRM mass of skilled scientists and research infrastruc- emphasis that enables them to attain the scale ture needed to undertake world-class research. and scope needed to make significant advances The critical mass needed for multi-disciplinary (CIFOR and ICRAF). But this raises issues of the NRM research is even larger. Henzell and others relationship and integration of their research (TAC Secretariat 1999) recommended that "fre- with cropping and livestock research. quently observed imbalances between biophysi- Given the importance of both hierarchical cal and social science research must be redressed" and disciplinary connectivity to NRM research, as must capacity development in the NARS, where there is a strong argument for organizing more "social science capacity in NRM is one of the of the CGIAR's NRM research following the col- weakest disciplines." The review noted that in- laborative or facilitative models. The new Chal- creased social science research has the unique po- lenge Programs are clearly oriented toward the tential to shed light on what might be done about collaborative model. But given the site-speci- the apparently large stock of unused and under- ficity of much NRM research and the lack of crit- used CGIAR research results in NRM and that it ical mass to do good NRM work in many NARS, needs to be expanded in almost all Centers un- more attention needs to be given to the facili- dertaking significant NRM research. Moreover, tative model, in which the CGIAR serves more ensuring quality in social science research is as im- as a broker between NARS and ARIs and among portant as ensuring the same in the biological NARS--in line with the subsidiarity principle and physical sciences. In addition, as the number promulgated by the DGF, and develops method- of donors increases, and each requires its own in- ologies, tools, data, information, and results that dependent reporting, the critical mass needed to have broader spillovers. conduct the research increases. The fixed costs The collaborative model offers an opportunity of high-quality science for global public goods for the CGIAR to overcome its capacity limitations production must not be underestimated. These in the social sciences and to raise the quality of fixed costs argue for concentrating restricted proj- the science in the Centers that lag in this area. ect funding in multifunctional Centers where the The CGIAR can leverage its resources better by average fixed cost becomes relatively low, and partnering with ARIs and leading NARS doing re- probably also for moving toward clusters of the lated work in which they have independent in- sort suggested by the CDMT. terest. While there have been some good A decision to reduce the number of Centers examples of such partnerships on an informal through consolidation or closure, which seems ad- basis, what has been lacking to date are strate- visable given that System resources are spread ex- gic collaborative and facilitative partnerships cessively thinly today, should probably include with the more secure, longer-term funding to serious consideration of the option to divest to pursue innovative, longer-term strategic research. NARS social science and water research of an ap- Even inter-Center strategic cooperation has been plied and adaptive nature--two of the less effec- scant, with successful System-wide programs tive threads within the CGIAR in the NRM domain. 1 0 1 T H E C G I A R AT 3 1 The System has so far concentrated most of its so- tinuation of System-wide programs where they cial science research capacity in IFPRI and ISNAR cannot be more tightly focused. Are the System- and most of its water research capacity in IWMI. wide programs to continue with the emergence Most other Centers lack critical mass in either of of Challenge Programs? Might or ought they be these areas, and several EPMRs suggest that this placed directly under the control of the new Sci- limits the efficacy of the broader NRM research ence Council as a manifestation of the programs' portfolio in the Centers. Key commodity Centers roles in addressing System-wide scientific prior- do not seem to be participants in the new Water ities? Or will they have independent governance Challenge Program. System-wide programs have structures, similar to those currently proposed for thus far proved ineffective in remedying this prob- Challenge Programs, thereby further fragmenting lem, not least because they are generally too re- the oversight for science quality of the System? liant on short-term project funding and have Finally, the CDMT made a number of rea- insufficient autonomy and core resources to em- sonably conservative suggestions regarding the bark on significant collaborative research ven- provision of Center support services. Unlike tures within the System. The quality of the science the reviews related to genetic resources and and relevance to problem solving on the ground plant breeding, the NRM-related reviews do have also been issues. not really speak to the System-wide question of At least in the NRM domain, it appears worth centralization of certain functions, including exploring whether the System-wide programs impact assessment; much external communi- have sufficient focus (a weakness of SWIM), suc- cations; and legal, financial, and management cessfully avoid duplication of ARI research (an ap- services. Clearly, some efforts are duplicated parent weakness of CAPRI), and should be across Centers and, as TAC's 1994 restructuring separated as a System-wide program rather than report notes, this drains staff time and resources mainstreamed within all the Centers (an issue at the Center level, so there seems to be merit with INRM). The independence of the programs, in the CDMT suggestion. But this would re- in terms of priority setting and resource alloca- quire more detailed scrutiny than can be un- tion proposed under the CDMP though Chal- dertaken in this meta-evaluation in order to lenge Programs, also needs to be reviewed. ensure that the sustainable operating costs sav- System-wide programs play an important role ings would be sufficient to justify the transac- in coordinating among Centers and between the tion and transition costs of the changes. CGIAR and external collaborators and stake- Similarly, it would seem prudent to explore holders. Yet they are universally hosted by a Cen- greater outsourcing of support services at the ter, have relatively meager budgets and support Center level. Many such functions continue to staffs, and therefore have little authority or flex- be organic to the institutions, but in many host ibility to deviate from the prevailing wishes of the countries, recent rapid changes in private serv- host Centers, a weakness continued in the two ices availability, quality, and cost may make it recently approved Challenge Programs. Henzell worthwhile to look into a switch to increased and colleagues (1999) recommend the discon- outsourcing. 1 0 2 13 National Agricultural Research Systems and the CGIAR C GIAR Centers have worked closely with the national research systems of developing countries in several strategic areas, and some of the CGIAR's most spectacular successes are a result of these relationships. The Third System Review and other reviews of the CGIAR acknowledge that CGIAR-NARS partnerships are one of the System's great strengths (CGIAR Sec- retariat 1998a; Baum 1986, pp. 315­16). But as the CGIAR adopts an explicit dual focus on research in integrated germplasm and natural resource man- agement, the increased complexity of research questions places increasing demands on the CGIAR-NARS relationship. The CGIAR's contributions to NARS capacity some observers of the CGIAR have said should building, much of which has occurred through be commissioned systematically to identify more research collaborations or networks, have rarely clearly client demand for CGIAR services (IFAD been systematically evaluated. Capacity build- 2001). These studies are also an important con- ing through training--the activity many NARS tribution to the meta-evaluation, as they cap- consider the most important contribution of the ture the issues facing the NARS at different stages CGIAR after its germplasm research--has suf- of development as CGIAR partners and clients, fered. Expenditures on training declined by 0.2 and as research systems in their own right. More percent annually in real terms between 1992 important, the studies contribute to the meta- and 2001. Yet training is a particularly impor- evaluation's efforts to understand the needs of tant component of a network-based organization, the NARS at the System level. which the CGIAR is evolving to become, both in These studies, along with input from other accessing and contributing to global knowledge. sources, highlight several issues. First, despite the To begin to understand the needs of the NARS major contributions of the CGIAR-NARS part- and their capacities in agricultural research, the nerships to the CGIAR's success, the limited meta-evaluation commissioned a series of four credit awarded to the NARS weakens the mutual papers by developing-country authors in Brazil, trust needed to sustain that outcome and de- Colombia, India, and Kenya.1 This exercise is an tracts from identification and prioritization of illustration of the type of "demand survey" that NARS needs. Second, the CGIAR is not keeping 1 0 3 T H E C G I A R AT 3 1 The CGIAR is not keeping pace with the global with what many NARS can now do, and whether pace with the global challenges facing alternative sources of supply have been suffi- challenges facing NARS. NARS, such as mod- ciently explored. Unfortunately, it appears that ern biotechnology, TAC/iSC has been unable to examine the actual intellectual property rights, private sector growth, research agenda with respect to NARS because trade liberalization, and the impact of nontariff of the restricted nature of donor funding that is barriers and agricultural subsidies in industrial increasingly tied to downstream activities (Baum economies. Third, the donor community has and others 2001). become increasingly unwilling to fund activities The CGIAR could do more to meet the needs that address these global challenges, instead al- of the NARS, not only through Center-level col- locating its resources to more regional, national, laborations but also through greater partner- or local challenges that appeal to their domes- ship at the System level. Clearly the CGIAR cannot tic constituencies, and causing the CGIAR System undertake the task of developing capacity to pursue research and activities that may be through greater collaboration and partnership in more appropriately carried out by the NARS. the NARS of 180 countries across the globe. Fourth, the CGIAR Centers report that they have Donors and developing country governments also become a conduit for donor funding to must also provide resources for such capacity NARS as a means to minimize the uncertainty, building. What the CGIAR can do, however, is risk, and accountability issues that plague many pursue its comparative advantage relative to NARS, further complicating the bureaucratic and donors, advanced research institutes (ARIs), and financial processes facing NARS and exacerbat- others in conducting mission-oriented compar- ing the historic teacher-student relationship be- ative collaborative research across many coun- tween the CGIAR and NARS, which many NARS tries in issues of interest to developing countries. argue they have outgrown. In this context, the meta-evaluation team ex- These four issues represent a fundamental amined the following issues with respect to NARS challenge to the international agricultural re- from the viewpoint of their lessons and impli- search system: without a strategic approach to the cations for System-level reforms: (a) the impor- NARS, it is likely that the CGIAR, the donor com- tance of recognizing NARS contributions to munity, and governments of both developing agricultural research outputs, (b) the need for and industrial countries will prevent advanced new and meaningful partnerships between the NARS from realizing their full potential, limit the CGIAR and advanced NARS, and (c) the need for maturation of smaller NARS, and render the continued CGIAR support for small-country CGIAR's contributions to these processes moot. NARS. The chapter concludes with a discussion Yet the approach to the NARS pursued by the of the global challenges facing the NARS and CGIAR must be balanced against the research recommendations for the role the CGIAR can criteria identified by the CGIAR's scientific advi- play in assisting NARS to address these chal- sory body (formerly TAC, currently the interim lenges. Science Council), specifically that research focus on the production of global and regional public The Black Box of High Returns: goods, offer good probabilities of success, be The Partnership Role of the NARS cost-effective, and consider alternative sources of The contribution of the NARS to CGIAR research supply and comparative advantage. is of great importance to international agricultural Both the meta-evaluation and input from the research. Although many NARS are weak in adapt- NARS suggest that these criteria are not applied ing CGIAR research outputs to the national level, vigorously. Both the meta-evaluation team and thus constraining the CGIAR's potential impact, NARS have concerns about the extent to which their contributions to CGIAR research are some- global and regional public goods are being pro- times overlooked. Specifically, NARS scientists be- duced, whether the research that is actually lieve that their role is not sufficiently recognized being conducted is cost-effective in comparison in rate-of-return studies or in successes attributed 1 0 4 N AT I O N A L A G R I C U LT U R A L R E S E A R C H S Y S T E M S A N D T H E C G I A R to the Centers. Indeed, many System-wide stud- of supportive and The challenge of building ies make little distinction between NARS and complementary re- genuine partnerships with Center research contributions in basic science, lationships be- developing-country NARS will hybrid development, farming methods innova- tween advanced make or break the future tion, training of scientists, or institutional de- and small-country velopment, all of which contribute to yield NARS. Third, the CGIAR. increases.2 Typically, these studies take the view failure to ade- that research systems are a "black box" that quately recognize the NARS' contribution im- transforms research funds into new crop varieties pedes the development of a cohesive global in farmers' fields. strategy by the CGIAR, its donors, and its de- Although disaggregating the contribution of veloping-country members that plays up to the the NARS would not change the high returns to catalytic role the CGIAR must play in agricul- the collaboration, it would help show that (a) tural research, given that the CGIAR constitutes strategic research must be conducted at the in- only 4 percent of the global research expendi- ternational level owing to scale economies, the tures. This leads donors to expect the CGIAR to need for state-of-the-art methodologies and sci- fill a gap that cannot be filled by the NARS, ence, and the desirability of likely spillovers, and thereby diverting attention and reducing the that (b) other research, including applied and System's effectiveness in providing global and re- adaptive research, must be primarily conducted gional public goods that contribute to increas- by national systems to address issues such as local ing agricultural productivity and reducing natural resources and agro-ecology, institutions, poverty. and policies. The challenge of building genuine partner- Moreover, disaggregating NARS contributions ships with developing-country NARS will make would have important policy and financing im- or break the future CGIAR, according to former plications beyond the CGIAR itself that have not CGIAR Chairman Ismail Serageldin. In the same been adequately addressed by the CGIAR. Re- spirit, the TSR recommended that the expanded forms in the CGIAR alone cannot ensure its im- mission of the CGIAR be achieved through part- pact. Investment in agricultural R&D by nerships, capacity building, and policy dialogue developing-country governments and donors that includes the NARS. NARS scientists con- must be addressed simultaneously. First, the sulted by the meta-evaluation provide similar CGIAR is not effectively using its "bully pulpit" comments (see Katyal and Mruthyanjaya 2002; to persuade developing-country policymakers Macedo and others 2002). Unless incentives to and donors of the necessity for investment in na- investment in agricultural R&D in developing tional research systems as an essential comple- countries are fundamentally changed, partner- ment to international research. This is particularly ships with NARS will not improve. important given that investments in agricultural and rural development have declined substan- Pursuing Partnerships with Advanced tially, as shown in Chapter 8. Second, an aggre- NARS gated approach to estimating returns and China has approximately 50,000 agricultural re- assessing impact makes it difficult for the CGIAR search scientists, while India boasts 26,000, and to divest itself of activities that some large- Brazil 7,500. In the United States, about 25,000 country NARS indicate they are now perfectly ca- scientists work on agricultural research topics, pable of performing more cost-effectively than 45,000 scientists if related disciplines such as the CGIAR, and that some smaller, weaker NARS biomedical research are included. The CGIAR are keen to undertake as a way of strengthening Centers, on the other hand, have just 8,500 sci- their own capacity. And the continued partici- entists and staff working in more than 100 coun- pation of the CGIAR in national or local-level tries. The issue of CGIAR-NARS partnership research diverts the CGIAR from the provision therefore has special meaning with respect to the of global public goods and impedes the growth advanced NARS in countries such as Brazil, China, 1 0 5 T H E C G I A R AT 3 1 and India. It raises the questions of (a) whether agricultural research. Yet despite this emerging the CGIAR is advancing rapidly enough to keep relationship between the CGIAR and EMBRAPA, up with the scientific needs of advanced NARS, joint activities in research, capacity building, and (b) whether the CGIAR is facilitating meaning- institutional strengthening have actually de- ful and constructive partnerships between ad- clined. Scientific cooperation and training of sci- vanced NARS and other, smaller-country NARS, entists decreased considerably in the 1990s, as and (c) whether the CGIAR and advanced NARS did the contribution of genetic materials from the are pursuing activities in which they have a com- CGIAR to the Brazilian seed market during the parative advantage. The more advanced NARS same period (Macedo and others 2002, table 1, stress that the CGIAR has not kept pace with p. 14, and figure 2, p. 19). rapid changes in the external environment--in An equally relevant example is India's national areas such as biotechnology, intellectual property research system. The CGIAR has played a signif- rights, near-frontier science, information man- icant role in supporting the growth and evolution agement, public-private partnerships, and so- of this NARS, and today, India's NARS is a highly cial science and policy research. This has limited complex system led by the Indian Council on the Centers' ability to contribute to the chang- Agricultural Research (ICAR). With persistent ing needs of advanced NARS. Moreover, the challenges in food and nutrition security, poverty, CGIAR has contributed minimally to the inclu- equity, and deterioration of natural resources, sion of the NARS of advanced developing coun- as well as the myriad problems associated with tries in the global research system, although managing an intricate and diversified research sys- these NARS now possess the human talent and tem, India argues that its NARS requires research experience necessary to constructively contribute in the areas of germplasm improvement, genetic to the research and capacity building in other, resource management, and modern biotechnol- small-country and weaker NARS. ogy from the CGIAR. This would allow the NARS A useful example of these concerns is the on- to focus on national and local issues such as en- going collaboration between the CGIAR and EM- vironment and equity, or on internationalizing its BRAPA, the Brazilian Agricultural Research perceived comparative advantages in manage- Corporation. Given that Brazil is a country with ment training and capacity strengthening. Efforts continental dimensions, and hosts a broad di- in these directions will require that ICAR and versity of climate, soil, agricultural production sys- India's national research system expand the tems, and cultural values, its NARS argues that breadth and depth of their linkages nationally and it should be considered a unique partner by the internationally, and raises questions about CGIAR, requiring a flexible and multifaceted ap- whether the CGIAR can constructively contribute proach to cooperation. CGIAR collaboration with to meeting these needs. The task ahead is for the EMBRAPA and other Brazilian institutions rep- CGIAR to collaborate with India's NARS as a val- resents an opportunity to generate synergies ued partner not only for research purposes, but with and facilitate technology transfers to Brazil also to support other developing country NARS, and other developing countries, ultimately gen- particularly in semi-arid regions (Katyal and erating numerous direct and spillover benefits. Mruthyanjaya 2002). It is likely that the same ar- Collaboration with the CGIAR entered a new gument would be made by China were it to be phase when EMBRAPA recently became an active consulted. stakeholder in the CGIAR System, and this new However, the role of advanced NARS in col- status reflects both EMBRAPA's progress over laborating with smaller, weaker NARS remains a the past three complex issue. Donors argue that small NARS The role of advanced NARS decades and the from developing countries typically are reluc- in collaborating with CGIAR's recogni- tant to work with larger NARS from developing tion of EMBRAPA's countries and to involve them in research. smaller, weaker NARS potential to be an Donors also indicated to the meta-evaluation remains a complex issue. equal partner in team in response to this offer by the NARS of ad- 1 0 6 N AT I O N A L A G R I C U LT U R A L R E S E A R C H S Y S T E M S A N D T H E C G I A R vanced developing countries that advanced NARS out that the There is potential to overcome tend to be reluctant to involve smaller NARS CGIAR has com- the persistent teacher/student when such collaborations open access to the pensated for the relationship between the valuable technologies and intellectual property general decline CGIAR and small-country of the advanced NARS. Yet there remains ample of Colombia's scope for partnership and collaboration between NARS through NARS. NARS of developing countries, and an impor- CIAT research in tant role for the CGIAR and the donors to con- cassava, bean, grasses, soils, fruit trees, and tribute to such partnerships both in areas of maize in areas that are not being addressed by mutual research interest and on more contro- Colombian institutions. It raises the larger issue versial issues such as intellectual property rights. of whether the CGIAR has a role in taking over In sum, evidence suggests that future coop- functions of a national research system in times eration between the CGIAR and advanced NARS of conflict or upheaval. The recent examples of should (a) focus on the generation of new tech- WARDA and Côte d'Ivoire and CIAT in Colom- nologies and on the applicability of such tech- bia demonstrate the difficulty in operating in nologies to agricultural development; (b) include countries experiencing internal conflict. the participation of diverse research institutions Second, as during the "meeting of the minds" that comprise or complement the systems, in- consultations organized by ILRI in Kenya that cluding universities, state research organizations, brought together NARS and other CGIAR Cen- and the private sector; (c) use CGIAR resources ters, there is potential to overcome the persist- to leverage NARS activities in both research and ent teacher/student relationship between the scientific training in countries that need help to CGIAR and small-country NARS (Ndiritu 2002). reach the same stage as those with advanced The lack of trust and cooperation between in- NARS by actively planning for large spillovers; dividual Centers and NARS, including contro- and (d) evolve toward collaboration with the versies over NARS appropriating plant breeding NARS of other countries as equal partners. research conducted in collaboration with the CGIAR, challenges these valuable partnerships.3 Ensuring Continued Support Third, small-country NARS are particularly sub- to Small-Country NARS ject to inadequate financing and restrictive bu- The CGIAR plays a different but crucial role with reaucratic rules and regulations on employment respect to small-country NARS. Centers facili- and procurement (especially of imported equip- tate linkages between the international scientific ment and chemicals). Donors conveyed an over- community and small-country NARS, provide whelming concern over fiduciary management access to germplasm for crop improvement pro- of funding and delivery of quality output, an grams, offer invaluable research inputs and ex- issue that they and the CGIAR Centers indicated pertise, and provide a means of circumventing leads donors to turn to the Centers even when the arcane national bureaucratic rules and reg- the tasks are mostly of a national nature.4 ulations that otherwise slow down research The search for solutions to these issues is activities. partly constrained by the limited forums through However, small-country NARS face signifi- which small-country NARS are able to voice their cant issues and impediments to growth, a prob- concerns to the larger agricultural research com- lem that involves the CGIAR. First, many weaker munity. Recent attempts to establish global fo- small-country NARS believe that the CGIAR is rums, regional forums, and subregional being used to compensate for the failures of organizations in the developing world are a step their national systems, forcing the CGIAR to in the right direction, but discussion alone is bring its resources to bear on national-level not enough. Given the critical importance of agricultural research issues. This diverts the NARS to determining impacts on the ground, the CGIAR's efforts to build small-country NARS. In Bank, the CGIAR, and donors must consider Colombia, for example, Romano (2002) points concrete actions to strengthen such organiza- 1 0 7 T H E C G I A R AT 3 1 tions to address concrete issues that seem to stitutes (NARIs) to private institutions (NGOs, uni- come in the way of enhancing partnerships, par- versities, and commercial enterprises) (see TAC ticipation, and interaction between NARS, re- Secretariat 1996b, especially p. 52, 1996c). Ac- gional and subregional organizations, the CGIAR, cordingly, ISNAR described its role as a focal and other actors in agricultural research. This in point for both research and provision of services turn must be balanced with concerns among in the organization and management of "new in- the NARS that these emerging global and re- novation systems" for NARS and other public gional forums will simply contribute to the du- and private entities. plication of research administration tasks without However, the Fourth External Programme adding value, or that they will be overwhelmed and Management Review concluded that ISNAR by large sums of donor support without clear and has been only a modest contributor to research, shared objectives and performance indicators and its performance has been below required to assess progress in achieving them. standards. The EPMR concluded that ISNAR did The more essential solution--building capa- not follow recommendations of the Third EPMR ble and effective national agricultural research to improve its performance by revitalizing its systems--is an undertaking that may extend far staff competencies and skill mix, or to develop beyond the CGIAR's mandate or capacity, de- strategic partnerships needed to maximize the spite the existence of ISNAR, discussed below.5 global and regional public good dimensions of The need for developing countries to place a high its research and service work. The EPMR panel priority on fortifying their NARS partners with fi- made three alternative recommendations for nancial and institutional support is a point raised ISNAR's future: (1) to rejuvenate ISNAR as a re- by former Chairman Serageldin and again em- search and service-oriented Center by retool- phasized by the TSR (Serageldin 1996, pp. 7­9). ing and restaffing; (2) to reconfigure ISNAR as a If the Centers are expected to contribute to decentralized, service oriented Center; or (3) NARS capacity building and strengthening as a to phase ISNAR out and pass its relevant activi- means of improving the potential for appropri- ties on to other Centers. Following discussion at ation and local adoption of international re- AGM 2002, efforts are under way to pursue the search, then national governments must be second option under the guidance of a restruc- responsible for improving the operation and the turing team. organization of the NARS (see Romano 2002 for In sum, though the CGIAR believes that ca- a discussion of this issue in the context of Colom- pacity building is an important objective and bia). At the same time, donors and the CGIAR claims substantial achievements in this area, must help devise concrete, workable solutions and though the small-country NARS attach great that are routinely monitored and evaluated in- importance to such inputs, relatively little effort dependently for effectiveness. is devoted to assessing the value of these ac- System-wide coordination and provision of tivities by the CGIAR or the donors, and the as- technical assistance and training have become sessment that has been carried out is not principally the purview of ISNAR. As discussed in encouraging.6 The need to address the CGIAR's TAC's review of policy and management research role and comparative advantage is overdue, in the CGIAR, ISNAR's dual role--as a provider and greater effort to measure the impacts of of management and institutional development training and its comparative advantage in de- services and the Center for research on improved veloping the essential methodologies in this management methods--needs strengthening on field is urgently needed. Hence, the evaluation the research side, and on the services side, needs of training and capacity building getting under expanding from services to public research in- way is welcome. 1 0 8 14 The CGIAR and Sub-Saharan Africa A fter 40 years of development programs and billions of dollars of foreign aid, Africa still faces rising poverty rates and remains the only region of the world where per capita food production has stagnated over the past 40 years.1 The World Bank, the CGIAR, Canada, and the EU have once again ex- pressed interest in considerably increasing their support to African agriculture, but it is not clear how well coordinated these efforts will be. Of the 29 countries facing food emergencies in 2001/2002, 16 were in Africa (Sanchez 2002; FAO 2002). Food security in the region is compromised by in development assistance. These factors are droughts, internal conflicts, and the grim im- more acute in Africa than in South Asia. In the pact of HIV/AIDS, which to date has taken the latter, the poor distribution of massive surpluses lives of 7 million agricultural workers and threat- and the lack of purchasing power in the hands ens to take another 20 million before 2020. Agri- of the poor lie at the heart of the problem, al- cultural labor shortages, rather than labor though population growth and land pressure will surpluses, are a problem in many HIV/AIDS-rav- call for continued emphasis on productivity en- aged countries, and have tremendous implica- hancement such that the poor directly partici- tions for the kind of agricultural technologies that pate in the growth process and many move out are promoted to alleviate hunger. of agriculture. But in Africa, increasing the sup- Because agriculture is the primary source of ply of food and agricultural products through livelihood for two-thirds of its people, Africa's productivity growth is a far more urgent chal- quest for poverty alleviation, food security, and lenge. It is key to increasing incomes and em- balanced regional growth depends critically on ployment. Low productivity has eroded the broad-based agricultural growth. Africa's food in- competitiveness of African agriculture on world security is directly related to insufficient food and markets. Export crop production has shifted to agricultural production, appropriate science meet subsistence food needs. Despite devalu- and technology gaps, extreme urban biases in ations, which have increased competitiveness, national resource allocations, OECD agricul- Africa's share in world agricultural trade fell tural policies that militate against Africa's agri- from 8 percent in 1965 to 3 percent in 1996, cultural trade, and poor policy and coordination mainly due to low volumes resulting from low 1 0 9 T H E C G I A R AT 3 1 productivity and supply (World Bank 2002d). unlike in Asia, rates of return studies are more There is enormous potential for improving agri- limited in Africa.4 culture and rural livelihoods for the poorest Many of these studies highlight successes farmers in Africa. Doing so requires that African such as the high-yielding TMX cassava varieties, governments assign top priority to agriculture improved hybrids and open-pollinated varieties and rural development, including agricultural re- of maize in eastern and southern Africa, higher- search, and the CGIAR to productivity growth. yielding wheat in eastern and southern Africa, hy- While the CGIAR has made many useful con- brid sorghum in Sudan, semi-dwarf rice for tributions to African agriculture, there is far irrigated regions in West Africa, early maturing greater potential in the region yet to be realized. cowpeas in West Africa, and disease-resistant po- Every African consulted for this review agrees tatoes in the eastern and central African highlands that the pipeline of new technologies produced (Maredia, Byerlee, and Pee 2000, p. 554; Oehmke by the CGIAR has been important and must con- and Crawford 1996). CGIAR Centers, including tinue. But from Africa's perspective, the current IITA, CIAT, WARDA, and CIMMYT, have played a configuration of CGIAR Centers in Africa com- significant role in leading the research and col- pound Africa's technology problems. African laborating with NARS to disseminate new tech- countries incur high transaction costs in dealing nologies. with multiple CGIAR Centers, costs that could be Other studies cite concerns such as low yield reduced under a service provision arrangement increases in other crops, limited adoption, and set up regionally to exploit economies of scale and institutional constraints. Centers face a complex scope.2 The Centers are currently underfunded set of problems: the large number of mandated and overstretched. To have an impact in the re- commodities, problems of technological trans- gion, the CGIAR needs to play a greater role in ferability, lack of research priorities appropriate addressing science and technology gaps, and to Africa's needs, inter-Center competition, donor needs to structure its engagement in the region conflicts over research priorities, and complex in a way that is more conducive to forming an ef- relationships with NARS and SROs.5 Also, Cen- fective strategy to this end. ters have been buffeted by unexpected political In an effort to examine the CGIAR's chal- disasters including civil wars and civil strife. Yet lenges in Sub-Saharan Africa, the meta-evaluation perhaps the most inhibiting factors the Centers carefully considered the TSR. The TSR has been have had to overcome have been the limited criticized for not adding value to the CGIAR's ac- scientific knowledge base about African agricul- tivities in Africa since it devoted only one of its ture and an insufficient agricultural science base 29 recommendations to the region. in terms of human and institutional capacity. While the Centers and other development prac- Activities, Impacts, and Constraints titioners in Africa have built a body of knowledge Today, all 16 of the CGIAR Centers have pro- over the past half century, inappropriate donor grams in Africa, although the continent physically and national expectations regarding the time headquarters only four Centers: WARDA, ILRI, needed to join the ranks of industrial nations, the ICRAF, and IITA. Per- emphasis on agricultural extension and inte- Perhaps the most inhibiting formance of the grated rural development in boosting food out- factors have been the limited CGIAR Centers oper- put, and wholesale importation of rural ating in Africa is of institutions and technologies were all part of scientific knowledge base central importance to the problem. about African agriculture this meta-evaluation, And even where Centers have produced adopt- and an insufficient because Africa re- able technologies, the technologies are not trans- agricultural science base in ceives 43 percent of lating into increased productivity because yields the CGIAR's $350 mil- on research tend to be lower due to extraordi- terms of human and lion annual research narily complex patterns of pests, diseases, and institutional capacity. expenditures.3 Yet, soils. Low farm yields are primarily the result of 1 1 0 T H E C G I A R A N D S U B - S A H A R A N A F R I C A massive constraints in delivery of inputs and lack culture. Although Commodity and input of value-adding infrastructure and marketing there has been a markets, and the policies services, including prices and information. A re- substantial intro- needed to facilitate them, cent study shows that while adoption of im- duction of new have been a critical issue in proved crop varieties have been similar in Asia, sorghum and mil- Latin America, the Middle East, and Sub-Saha- let cultivars in semi- Sub-Saharan Africa. ran Africa during the past 38 years, such vari- arid Sub-Saharan eties are responsible for 66 to 88 percent of the Africa, there has been minimum impact on yields crop yield increases in the first three regions, because of the lack of fertilizer, improved seeds, but only 28 percent in Africa. The region shows irrigation, or improved water retention (Ahmed, not only a crop-mix less oriented to the main high- Sanders, and Nell 2000). In Nigeria, for exam- yield variety (HYV) crops, but also a lower share ple, fertilizer use fell from 450,000 tons in 1993/94 of crops in HYVs for each specific main staple, and to 100,000 tons in 1999/2000, a decline of 80 per- lower yields for HYVs and non-HYVs for each cent (IFDC 2001, p. 5). In some countries, the pre- main staple separately (Lipton 1994, pp. 131­56). cipitous drop in fertilizer use stems from high The reasons include a flagging government and prices for inputs--the result of lower input sub- donor commitment to agricultural development, sidies and market liberalization--combined with poor policy planning, lack of rural infrastructure, lower output prices--the result, many will argue, poorly functioning commodity and input markets, of agricultural policy liberalization in Africa com- human capital constraints, and poor strategic ad- bined with continued OECD subsidies to their do- vice and support from all sides.6 mestic agricultural sectors. But in other countries, Commodity and input markets, and the poli- the problem stems simply from incomplete or cies needed to facilitate them, have been a crit- weak markets and supporting institutions. ical issue in Sub-Saharan Africa. In the 1970s and Human capital constraints contribute further 1980s, assistance from the donor community to the limited productivity gains in Sub-Saharan greatly increased government employment in African agriculture. African research systems are agricultural services--especially extension--at at an unnecessarily premature stage of scien- the cost of investment in agricultural infrastruc- tific and institutional development given the ture, input supply, or in the development of com- strengths of the colonial research and extension modity markets (to say nothing of research). But systems (Eicher and Rukuni 2002). This situation doubling the number of extension agents failed might not have developed had the subsequent to increase the agricultural growth rate. Extension institutional development been built on both raises the speed of research diffusion, yield in- the positive and negative lessons offered by the creases, and agricultural growth, but only where colonial experience, rather than allowing the in- there are new, profitable technologies to extend stitutions to be replaced completely by new in- to the farm level.7 In Africa, the lack of useful tech- stitutions, driven more by political imperatives nologies has severely limited the success of the than by scientific and development imperatives.8 "extension first" model in which the Bank alone Eicher and Rukuni (2002) argue that the colo- invested about $4 billion to promote the training nial experience in African agricultural develop- and visit (T&V) system, which, like the earlier in- ment offers some useful lessons for research tegrated rural development bandwagon, it later strategies in the region, despite the many short- abandoned (see Gautam 2000). In the mean- comings of colonial policy (Eicher and Rukuni time, complementary initiatives such as research 2002). Colonial decisionmakers introduced a were eroded by a combination of political strife, number of innovations in research--largely using lack of recurrent public finances, and unpre- internal African financing through export taxes-- dictable and fragmented donor assistance. to develop lean and self-financed regional re- Poorly functioning input markets are a major search institutes to serve a large number of small constraint on the contribution that CGIAR Cen- countries. Since the colonies were expected to ters can make to the performance of African agri- rely fully on their own financing, at relatively 1 1 1 T H E C G I A R AT 3 1 low cost, the colonial governments generated have requested Brazil's EMBRAPA to help them substantial increases in agricultural productivity build scientific capacity to plan and execute agri- by investing in regional research stations placed cultural research programs, although resources strategically at locations selected for specific for such initiatives remain scarce. Turning this sit- commodities to maximize spillovers of research uation around should be the primary responsi- findings. Research also made effective use of bility of African governments. Thus, the lack of the basic science and technological research African scientific capacity is impeding the tran- produced by institutions within the imperial sition from the current, what Eicher and Rukuni countries. The colonial governments built in- (2002) term, the "CGIAR-first" model to African- frastructure and assured input supply and export led research partnerships, networks, and al- markets for products. But the fatal flaw in this sys- liances with public and private research tem was the failure to (1) encompass food com- organizations. Unfortunately, the stagnation or modities and natural resource management decline in domestic public spending on African concerns, (2) build African institutions, and (3) agricultural research systems suggests that there train and retain a competent cadre of African is very limited hope for this model. scientists. Thus, the shortage of human capital An additional constraint is the small size of in the post-colonial period slowed the develop- many NARS with a lack of a critical minimum ment of African scientific leadership and con- number of scientists. Currently, 80 percent tinues to arrest the performance and impact of (4,800) of Africa's agricultural researchers are the CGIAR in the region.9 concentrated in 13 large countries, while the re- The problem has been compounded by a de- maining 20 percent (1,200) of the scientists are cline over the past 15 years in external aid for agri- dispersed across 35 smaller countries, a distri- culture in general, and training in particular. In bution not necessarily reflecting the relative size 1998, only 20 USAID scholarships were available of population or agricultural GDP (Mrema 2001). to Africans for studying agriculture in the United The 80/20 ratio highlights the diseconomies of States, down from 250 in l985 (Eicher 1999). scale of small NARS and the fundamental im- This poses two questions: who will train the portance of building subregional and regional re- next generation of African agricultural scientists, search organizations to take advantage of scale and will they be competitive with scientists in the economies in research as well as to ensure ef- traditional and emerging industrial world, par- fective intermediation between the CGIAR Cen- ticularly in fields such as biotechnology? One ters and the national systems. In contrast, there solution is to develop more relevant and cost- are currently 16 Centers operating in 17 coun- effective training programs in the region itself, tries in eastern and southern Africa, with 13 Cen- focusing on agricultural science and technology, ters operating in Kenya alone. This concentration data gathering and statistical analysis, and other of CGIAR personnel and resources in eastern relevant topics. The 2020 Vision Initiative Net- and southern Africa raises issues of strategic lo- work on East Africa is a useful example of the cation of research in Africa from an agroecolog- CGIAR engaging NARS in the region to conduct ical perspective.10 policy research and build capacity. Several African scholars have stressed to the Another solution is to use expertise in ad- meta-evaluation team that managing CGIAR net- vanced NARS to train African scientists. Cur- works has high opportunity and transaction rently, only about 100 African agricultural costs. The networks risk diverting research away scientists (mainly from Angola and Mozambique) from country needs unless international insti- are enrolled in tutions keep their research programs closely Who will train the next training programs aligned with those needs (through their subre- generation of African in Brazil (Macedo gional collaborating mechanism). They use scarce and others 2002). talent that might otherwise be used to manage agricultural scientists, and Moreover, several the NARS.11 Additionally, a network is only ef- will they be competitive? African countries fective if its members are strong. There is al- 1 1 2 T H E C G I A R A N D S U B - S A H A R A N A F R I C A ways a risk that strong members of a network dertaken jointly After nearly two decades in benefit disproportionately, especially if it entails with universities of which the donors and the access to resources, financial or technical, from industrial and de- CGIAR dominated the outside. The CGIAR will not achieve its full po- veloping countries. dialogue, the voice of tential until the African NARS, like their coun- The proposal of the terparts in Brazil and India, are productive and current CGIAR Africans is increasingly their regional and subregional organizations are management to reg- being heard. effective and financially sustainable. An effec- ularize and institu- tive, sustainable NARS at the national level, com- tionalize such training is worthy of urgent bined with constructive networking at the support by donors, provided African institutions regional level, can only be achieved through in- of higher education are also concurrently sup- vestment in capacity building for individual NARS. ported.13 The evaluation of the training activities The Kenyan and South African NARS are among of the CGIAR being undertaken by SPIA should the strongest systems in Africa, with the potential pay particular attention to these training issues. to significantly contribute to emerging networks In this context, the CGIAR and the NARS need of agricultural research systems in the region. effective collaborations with the African academic However, experts observe that these NARS also system, the private sector, and civil society. One face increasing pressures to privatize agricultural of the key shortcomings of both the region's research to the benefit of middle- or large-size agricultural research systems is their inability to farmers. This has the effect of leaving agricultural build strong partnerships with indigenous polit- research for small farmers as an increasingly under- ical, social, and economic forces in the region. To resourced residual, increasingly unable to retain date, the region's research systems seem to be good scientists or conduct strategic research rel- more accountable to donors and select admin- evant not only to small farmers in the domestic istrators in the ministries of agriculture than to agricultural sectors, but across the region. Public finance ministries, the market, and civil society sectors must play an important role in addressing actors in their own countries. The challenges problems of small, poor farmers. facing the CGIAR and NARS in Sub-Saharan Africa, Nevertheless, after nearly two decades of ten- as well as the donors, require greater levels of sion between the CGIAR Centers and the African partnership with development actors within the NARS, in which the donors and the CGIAR dom- region to help create strong long-term con- inated the dialogue, the voice of Africans is in- stituencies for agricultural R&D. creasingly being heard.12 The contribution of the Forum for Agricultural Research in Africa Priority Setting and the Roles of Regions (FARA) to the CDMT exercise flagged the lack of A regional approach to research is needed be- congruence of priorities between CGIAR Centers, cause Sub-Saharan Africa consists of many small NARS, and the SROs. The FARA report has been countries with diverse agro-climatic conditions, well received within the CGIAR and the donor thereby limiting opportunities for continent- community. This may be a barometer of the wide economies of scale in research, and be- growing countervailing power of the African sci- cause most of the region's NARS lack the entific community. But further dialogue is needed resources and capacity to contribute to research at the subregional level to determine the serv- effectively. The Special Program for African Agri- ices needed by the SROs, gaps in CGIAR pro- cultural Research (SPAAR), initiated at the Mid- grams, coordination issues relating to Center Term Meeting in 1985, was designed to facilitate mandates, the appropriateness of CGIAR struc- agricultural research in the region and has helped tures, and the key issue of retention of CGIAR to establish three SROs and FARA as an apex or- and NARS scientists in the region. ganization.14 The CGIAR has a sizable comparative advan- Subregional organizations such as the Asso- tage in training through collaborative research ciation for Strengthening Agricultural Research with M.Sc.-level and Ph.D.-level scientists if un- in Eastern and Central Africa (ASARECA), the 1 1 3 T H E C G I A R AT 3 1 Without greater priority to Southern African fore, without strengthening the NARS, the prob- research and investment by Centre for Cooper- ability of success is low. More important, they Africans in their NARS, ation in Agricultural argue that the research challenges, the lesson and Natural Re- learning, and the economies of scale and scope partnerships in the region sources Research do not neatly coincide with the political and ad- will not improve. and Training (SAC- ministrative regional borders in Africa. Thus, a re- CAR), and the Con- gional concentration might detract from the férence des Responsables de Recherche need for research links and networks on specific Agronomique en Afrique de l'Ouest et du Cen- commodities or themes that would otherwise be tre (CORAF) are playing an increasingly impor- lost in a more regional approach to research in tant role in the region.15 SROs have the potential Africa. Finally, experts point out that greater re- to provide leadership on critical agricultural re- gional and subregional integration runs the risk search policy issues facing the NARS in Africa of simply generating costs associated with cre- today, and can offer an "African voice" in such key ating new institutions and building systems for organizations as GFAR, the World Bank, and the governance, management, and finance, without CGIAR System. SROs also have the potential to addressing the key scientific, technological, and help some countries reduce their research out- policy issues in African agriculture. lays on certain crops and increase their capacity Ultimately, there is a question of what the pri- to access technology through SRO networks, orities should be for CGIAR work in Africa and and from the CGIAR System. who should set them. There is increasing senti- Moreover, there is a strong, and as yet unreal- ment that the subregional organizations should ized, potential for advanced NARS in countries guide subregional priorities, but this is a chal- such as Brazil, India, and China to play a more ac- lenging problem since subregional capacities for tive role in building African research capacity articulating priorities are both diverse and limited. through SROs, a topic the TSR and other reviews of the CGIAR have commented on extensively, but Confronting Shortfalls in Public that remains unattended because of the nature of Agricultural Spending CGIAR funding. Strong, African-led national and What can the CGIAR do, given that Africans and regional research organizations are needed to donors, including the World Bank, have failed to set appropriate CGIAR priorities in the region. Al- strengthen research and research capacity build- though the move toward regional and subre- ing efforts in Africa, thereby pressuring the CGIAR gional collaboration and integration has been to move downstream? Public spending for agri- strong in Africa, funding of regional research ac- cultural research in Africa stagnated in the tivities still amounts to less than 2 percent of total 1980s and 1990s at about $1,200 million per year, spending on agricultural research. Without greater only slightly higher than 1976 levels (Figure 14.1). priority to research and investment by Africans in After 40 years of independence, many NARS in their NARS, along with universities, policy think Africa are scientifically weak, oversized, finan- tanks, farmers' organizations, and the private sec- cially unstable, and heavily dependent on un- tor, partnerships in the region will not improve. predictable and fragmented donor assistance. Yet some experts, including members of the Donors fund an average of 40 percent of all NARS meta-evaluation's advisory committee, express expenditures on research in Africa and around 60 concern over the greater focus by donors on percent in some countries (Pardey, Roseboom, the regional and subregional organizations in and Beintema 1997). Only five African coun- Africa, which they fear will come at the cost of tries--Botswana, Ethiopia, Mauritius, Nigeria, building the strong foundation of the national re- and South Africa --are paying the recurrent search systems. They argue that as long as many budget of their NARS from national sources. This, African NARS remain weak, these constitute an compounded by waning donor interest in agri- unstable foundation upon which to build re- culture, has led to severe financing shortfalls and gional and subregional collaborations. There- disincentives to continued investment in agri- 1 1 4 T H E C G I A R A N D S U B - S A H A R A N A F R I C A P u b l i c A g r i c u l t u r a l R e s e a r c h F i g u r e 1 4 . 1 E x p e n d i t u r e s , 1 9 7 6 ­ 9 5 Millions of 1993 international dollars 12,000 Developed countries 10,000 8,000 Asia & the Pacific 6,000 4,000 Latin America & the Caribbean 2,000 Sub-Saharan Africa 0 1976 1985 1995 Source: Pardey and Beintema 2001. cultural research by African governments. The massive institution-building programs over the meta-evaluation has found that most of the African next 30 years. Donor coordination has the po- NARS attempted to adjust to the shortfall in an tential to facilitate improved collaboration be- uncoordinated fashion--for instance, by inde- tween the African NARS and CGIAR Centers. pendently approaching donors with individual The Bank's revised strategy for Sub-Saharan funding requests (Ndiritu 2002). Ways must be Africa effectively argues that increased public fi- found for Africa's leaders to provide leadership, nancing of agricultural research is important coordination, ownership, and responsibility for given its public goods nature and its crucial role public financing of its NARS and SROs. in poverty reduction, and because of the mod- est role of private sector research in Africa. The The Role of Donors and the Bank Bank recognizes the need for building regional The CGIAR must assist in generating African po- and subregional cooperation and for NARS spe- litical will and commitment for science and tech- cialization, but how can the Bank and other nology and sustained funding of agricultural donors collectively and strategically support this research. There is an urgent need for African po- initiative with a combination of support for re- litical leaders to elevate agricultural research to gional and national agricultural research sys- a national priority and move aggressively to pay tems when donor aid to agricultural research is the recurrent budget of research on a timely highly fragmented? The implications for how basis for decades to come. In order for the CGIAR Bank grant financing can support regional and to fulfill such a role, the Bank and donors must subregional organizations and how the Bank's adhere to a proactive, well-coordinated strategy lending activities could support research have not for agricultural research and development. Much been fully articulated.16 of the CGIAR's future in Africa will be determined The Bank's Sub-Saharan Africa proposal by the ability of donors to mount and sustain adopts a realistic 20-to-25-year timeframe and it 1 1 5 T H E C G I A R AT 3 1 calls for the Bank to lead the preparation of a should be used as research foci, and a strategic Multi-country Agricultural Research Program for realignment based on regional agreements Africa (MARP). But the country focus of the should be encouraged by grant funding from Bank's organization is hindering the develop- external sources. This approach would help pro- ment of a long-term agricultural research lend- mote the kind of research that will have spillover ing strategy for Africa. Strategic decisionmaking effects for smaller countries with similar agro-eco- is required to target which NARS should be sup- logical conditions. ported now, and it would be especially useful to The New Partnership for Africa's Develop- explore the use of IDA grants for regional pro- ment (NEPAD) is crucial in this regard. The grams for agricultural research, education, and Bank's cooperation with NEPAD includes a pro- training. The Bank would be better off using gram for agriculture, emphasizing accelerated part of its resources to develop regional research growth in productivity, improved food security, organizations jointly with other donors. In turn, better management of natural resources, im- this would help develop research investments in proved market access, and restructuring of agri- strategically located countries whose develop- cultural research and development. Though ment should be supported on a long-term basis. NEPAD does not hold all the solutions to re- This will take time and perseverance, but in the ducing poverty and increasing growth in Africa, long run, it will be more rewarding than dotting it offers a mechanism to draw attention to the un- Africa with unsustainable research projects on a derlying problems identified in this report and country-by-country and multi-country basis using provides African governments and other insti- grant funds. It is neither CGIAR's mandate nor tutions a leadership role in seeking solutions. an appropriate use of DGF grant funds to help In conclusion, three questions must be ad- the Bank develop its lending program. This dressed in order to tackle the CGIAR's dilemma should take place using the Bank's own budg- in Sub-Saharan Africa. First, what can be done to etary resources. persuade African governments to elevate agri- Furthermore, while the Bank proposal is cultural research to a national and continental po- timely and makes repeated reference to the need litical priority? Second, can the Bank and other to coordinate the proposal for MARP with the on- major donors collectively and strategically sup- going reform of the CGIAR System, the multi- port regional and national agricultural research country approach the Bank has been using with systems at a time when donor aid to agricultural respect to HIV/AIDS, and proposes to use in the research is highly fragmented? Third, can the case of agricultural research, is not suited to the CGIAR develop a well-coordinated and inte- formation of a strategic approach to the devel- grated strategy for agricultural research in Africa opment of research programs without consid- that increases coherence and reduces transaction erable adaptation. Regional research centers costs for African NARS and SROs? 1 1 6 Section III The Organizational Effectiveness of the CGIAR A s noted in Section II, the CGIAR is a unique instrument of international cooperation that has demonstrated the genius of its framers and made its mark in the agricultural research and development continuum span- ning basic, strategic, applied, and adaptive research and technology transfer. It has succeeded mostly because of its emphasis on strategic research of a global or regional public goods nature, the benefits of which spill across national boundaries and cannot easily be obtained through private or national re- search, and because of its practical, problem-solving focus on bringing the best of known science to address the problem of food security. But the CGIAR also displays symptoms of "mission cessfully restructuring the CGIAR to meet current creep" under pressure from various interests, and challenges and assessing how well the reform has diluted its original science-based character. It process is addressing issues raised in previous has failed to keep pace with the evolving techno- evaluations requires an understanding of the logical and institutional environment. While each evolution of the CGIAR governance, manage- of its Centers reports a number of partnerships, the ment, and financial systems over the past decade. small-scale and short-term nature of its funding Chapter 15 reviews reforms since the mid-1990s, means that the System as a whole lacks robust, sus- starting with the financial crisis in 1993­94, fol- tainable links to the private sector and the uni- lowed by the "Renewal" of the mid-1990s, the versities of advanced and developing country Third System Review in 1998, and the Federation institutions needed for a global research network. Proposal of the Committee of Board Chairs It has yet to fully tap the opportunities opened up (CBC) and Center Directors' Committee (CDC) by the new information technologies and the im- in 2000, and concludes with an assessment of the proved skills available in some developing countries. continuing relevance of the CGIAR's founding Notwithstanding recent changes instituted principles. Chapter 16 reviews the current reform under the Change Design and Management process that started in 2000 in the light of these Process (CDMP), OED concludes that the gov- experiences. Chapter 17 concludes Section III ernance, management, and financing of the with a review and assessment of the multiple CGIAR have become increasingly cumbersome, roles that the World Bank has played in the and require fundamental changes. However, suc- CGIAR. 1 1 7 15 Past Attempts to Restructure the CGIAR Financial Crisis T he growth in the number of Centers and the growth of System-wide and ecoregional programs in the early 1990s strained the CGIAR strategi- cally and financially. While asking the CGIAR to do more and causing it to expand its agenda to include, among other things, concerns for farming sys- tems, farmer participation in priority setting, and sensitivity to the needs of women farmers, donors were simultaneously cutting resources. Then, when the United States and Canada, pri- Ismail Serageldin, who assumed the chair- marily for domestic reasons, cut their funding to manship in 1994, strongly believed that consol- the CGIAR by $24 million and $6 million, re- idating Centers in a period of weakness would spectively, between 1992 and 1994, this precip- convey the wrong message to the development itated a financial crisis. While this was offset to community at a time when more resources were some extent by increasing contributions from needed for research.1 Serageldin confirmed in an Denmark, Japan, and the Netherlands, these interview with the meta-evaluation team his de- were not enough to fill the gap. sire for the CGIAR's programs to drive its budget, The two chairmen whose terms coincided and not the other way around. Knowing that with the financial crisis responded in different this would require a redoubled financial com- ways. First, CGIAR Chairman V. Rajgopalan opted mitment to the CGIAR by traditional and new to consolidate the two livestock Centers located contributors, he embarked in May 1994 on a in Africa, and commissioned TAC to explore con- program of political and financial mobilization solidation and restructuring options through- that he called "Renewal." out the System. The TAC Restructuring Review (1994) called for centralization of a variety of The Renewal Process functions carried out by the Centers; rationali- A ministerial-level meeting in Lucerne, Switzer- zation of central services such as reviews, as- land, in February 1995 endorsed the Renewal sessments, and evaluations; and reorganization process and reaffirmed donor support for the of research on cereals, roots and tubers, forestry CGIAR. Renewal represents an important ex- and agroforestry, policy, and NRM to increase ample of how the Bank's leadership has made it integration within each research area. possible for the CGIAR to open doors at the 1 1 9 T H E C G I A R AT 3 1 highest levels of government in both developed the Global Forum for Agricultural Research and developing countries to ensure continued (GFAR) to better reflect the views of NARS and political and financial support.2 The meeting re- the other components of the global research sulted in some key decisions regarding changes system. in programmatic, management, governance, and Renewal also brought fundamental, interre- financial allocation rules. But, while "Renewal" lated changes in resource allocation practices. was adopted by the membership, largely be- First, the CGIAR instituted a resource allocation cause they viewed this as a way of mobilizing ad- matrix as a tool to increase budgetary trans- ditional funds, it had no formal status and it parency, accountability, and predictability.4 Sec- lacked an enforcement mechanism beyond the ond, in order to create incentives for Centers to Bank's own contributions to Renewal, moral mobilize additional funding and to accommo- suasion, and selected incentives. date donors, the "agreed research agenda" of Under Renewal, the chairman actively so- the System was broadened to include both "core" licited and expanded developing country mem- and "complementary" activities. Previously, Cen- bership and increased their ownership of the ters' research was termed either core or com- System. He also established committees that re- plementary--"core" being largely the high-return flected diverse viewpoints, including those of global and regional public goods research, and the private sector and the NGO community. He "complementary" or "noncore" being the donor- successfully persuaded the Bank and others to funded, mostly downstream activities that TAC did sustain and even increase their levels of funding, not consider as high a priority--and within the while also introducing important policy changes, complementary category, no formal distinction such as redefining the "agreed research agenda" was made between activities outside the CGIAR's to create incentives for Centers to mobilize ad- mandate and activities within the mandate but ditional funding and to accommodate the outside the core funding envelope. This distinc- donors. tion between core and complementary activities In great part due to the tireless efforts of Is- has essentially disappeared. mail Serageldin, Renewal stabilized contribu- Third, the mechanism for allocating the World tions to the CGIAR. Renewal also brought a Bank's annual contribution was changed from a heightened acknowledgment of the need for "donor of last resort" model to a "matching grant" partnerships and paved the way for more direct model. Under the former, the Bank's contribution interaction with the private sector and civil so- had been used to fill gaps between the System's ciety.3 While neither the private sector nor the research priorities as articulated by TAC and the NGO community became full members of the financial contributions to those priorities by other CGIAR, they did acquire a seat at the table, and donors. Under the matching grant model, the the CGIAR began to contribute financially to the Bank's contribution indiscriminately matched operations of the resultant NGO and Private Sec- funding from other donors, whether in support tor Committees. These changes fundamentally of System-wide priorities or not. This step was altered the character of the System, from that of taken to address perceived downsides in the a technical organization of scientists and donors donor of last resort model, including: interested in funding science for the benefit of the poor, to the pursuit of a broader agenda in- · An incentive for Centers to classify activities as volving the views of diverse stakeholders. Si- "complementary" to maximize Bank funds for multaneously, the CGIAR--with the initiative of budgetary shortfalls in their "core" budget5 IFAD in 1994, and · A stifling of Center initiatives in raising funds Renewal paved the way for with the active sup- · An insulation of some Centers from economic more direct interaction with port of bilateral realities donors and the · An over-reliance on Bank funds by some Cen- the private sector and civil World Bank and ters, making "internal exit" by the Bank from society. FAO--established some Centers or programs difficult.6 1 2 0 PA S T AT T E M P T S T O R E S T R U C T U R E T H E C G I A R But the change in the allocation formula to matching the fund- TAC's influence and the role address these concerns has had significant con- ing of other donors of independent scientific sequences. It is widely acknowledged by CGIAR vary among the dif- advice in the System have stakeholders that the shift away from the donor ferent stakehold- diminished. of last resort model effectively eliminated the ers. (See Chapter 4 only mechanism to ensure adherence to TAC's for a detailed list.) priorities and resource allocation recommen- One major result of these changes in the mid- dations. As a result, TAC's influence and the role 1990s was that CGIAR expenditures on increas- of independent scientific advice in the System ing productivity declined by 6.5 percent annually have diminished. Second, the Bank's financial in real terms between 1994 and 2001, while those contribution is now indiscriminately matching for improving policies and protecting the envi- funding by other donors, whether in support ronment grew.8 The result is an inefficient System of global and regional public goods or not, un- with a fragmented research agenda that lacks wittingly reinforcing a tendency for "funding to long-term, strategic focus--a trend that has been drive programs" rather than "priority program- reinforced by the acceleration in the growth of ming to drive funding." Largely determined by restricted funding since 1998.9 the pattern of funding of other donors, the Bank's financial contribution is no longer nec- The Third System Review (1998) essarily supporting the long-term, strategic ele- A major finding of the TSR was that the CGIAR's ments of the System. governance structure no longer allowed it to In addition to these policy changes, the Bank make timely, responsive, and effective decisions. forgave the CGIAR's $5.6 million debt to the In order to improve System-level decisionmak- Bank that had resulted from advances on its an- ing, the TSR recommended, among other things, nual allocations, and Serageldin persuaded Bank that the CGIAR restructure itself as a legal entity President Lewis Preston to provide an additional built along the lines of a corporate model. one-time contribution of $20 million to make up Yet the TSR engendered little ownership from for the overall financial shortfall.7 All these meas- the membership, and hence had little impact on ures served as a sufficient incentive to other the organization of the System. CGIAR stake- donors to increase their contributions. But they holders criticized the review process for being too also greatly increased the Bank's role in System costly and for not involving them adequately-- governance and management. even though the review had involved substantial As a further indication of the Bank's com- consultations with the Center directors, boards, mitment to agricultural research, Serageldin of- donors, and members--and criticized the end fered up to $2.5 billion of combined IBRD and product.10 The TSR's recommendations on gov- IDA resources over five years to support devel- ernance and management clashed with the Sys- oping countries' agricultural research and ex- tem's culture of consensus decisionmaking and tension systems (Serageldin 1996). But with a few with vested interests resistant to change. Based exceptions and for the reasons discussed more in part upon members' apprehension about the extensively in Chapter 17--which relate to the declining influence and opportunities for con- increased decentralization of the Bank's lending sultation, particularly given the accompanying operations--demand from developing countries recommendation to eliminate the Mid-Term for investment in agricultural research did not Meeting, the Group decided not to reform its materialize as expected. A significant opportunity governance based on a centralized board as rec- to improve the capacity of developing country ommended by the NARS was lost. TSR (CGIAR Secre- The result is an inefficient tariat 2000a). System with a fragmented The Mixed Impacts of Renewal Ultimately, the research agenda that lacks Renewal had a variety of unanticipated conse- recommended re- quences on the System. The results of the Bank's forms made slow long-term, strategic focus. 1 2 1 T H E C G I A R AT 3 1 progress, among other reasons, because the proposal by the CBC/CDC for a "Federation of Bank and other donors lacked a systematic mech- Centers" as a legal entity comprising the 16 Cen- anism to consider and follow-up on the TSR ters, a Federation office, and a centralized board. findings and recommendations. Moreover, the The proposed Federation would perform six TSR made a very large number of recommen- functions: (1) strategic planning and scientific dations--29 recommendations and 104 subrec- quality enhancement, (2) resource mobilization, ommendations--and many were not actionable. (3) public awareness, (4) science advocacy, (5) With a few minor exceptions, the Group rejected Federation undertakings, and (6) providing com- the recommendations concerning governance mon services to Centers, donors, and the CGIAR and most stakeholders perceive the TSR as hav- chairman. The Centers generally viewed the Fed- ing had minimal impact. eration proposal as "devolving" certain powers In retrospect, however, one can see that the from the Centers to the Federation, moving TSR did have a significant effect on the CGIAR some of TAC's strategic planning functions, in- by exposing the System to a new set of high-level creasing the voice of NARS and the developing science managers and private sector actors in in- regions in the priority-setting process, and uni- dustrial countries, by reaffirming donor support fying the CGIAR Secretariat with the Federation for the System, by formally broadening the on grounds that the CGIAR Secretariat had not CGIAR mission to include integrated natural re- served the Centers well. But a few key donors op- source management research, and by offering posed the Federation proposal as being "bottom- process lessons that guided the consensus-build- up" and as an attempt to take away the donors' ing approach adopted during the current CDMP. prerogatives on the issues of strategic planning and System-level accountability. Reportedly, some The Federation Proposal (2000) developing countries also objected to the Fed- After the TSR's recommendation of a corporate eration idea.12 model was rejected by the membership in 1999, Once again, in the absence of a Group con- the Committee of Board Chairs (CBC) and the sensus, the Federation Proposal was dropped. Center Directors' Committee (CDC) offered an This remains the closest that an internal set of alternative configuration for the CGIAR in 2000-- actors has come to acknowledging the need for a decentralized model of a Federation of Centers. a legal entity with a centralized board much like Despite the differences between the TSR and the the TSR had earlier recommended. But it envis- Federation proposal with respect to the degree aged the Centers to be in the driver's seat instead of centralization, their common threads were of the donors and the existing CGIAR organiza- an acknowledgment of (a) the need for the for- tional units. mation of a legal entity with a centralized board to address System-level issues and (b) the need The CGIAR's Founding Principles Need for a more effective network-based approach to Revisiting developing technologies.11 The six founding principles that underlie the Discussion of Herdt's paper (2000) on re- CGIAR governance structure (see box 4.1) were structuring options for the CGIAR and other adopted when the System consisted of fewer ideas discussed at the Mid-Term Meeting 2000 in Centers and a less diverse constituency, and set- Dresden, Germany, and an electronic confer- ting priorities to achieve impacts on poverty was ence held by TAC to collect ideas for reforms, led relatively simple. For a considerable period, when to a proposal for the unified management of the System was smaller and more focused, the certain CGIAR functions relating to germplasm, CGIAR could be managed effectively within the intellectual property, and public awareness. The framework of these founding principles. Even It envisaged the Centers to be virtual confer- when the number of members and Centers in the ence was fol- CGIAR grew substantially during the System's in the driver's seat instead lowed by a first two decades, the basic governing structure of the donors. more ambitious did not change. But today, with its wider research 1 2 2 PA S T AT T E M P T S T O R E S T R U C T U R E T H E C G I A R agenda, its expanding membership, and its po- litically driven authorizing environment, the CGIAR's ability to address its mission is con- strained by its six founding principles. The System came under chronic pressures in the 1980s and early 1990s to broaden its mission, to be- come "environmentally more correct and socially more sensitive," and to seek new constituencies to increase funding.13 The acquisition of four Centers and the establishment of a new one in the early 1990s (after considering 10 existing "non-associ- ated" Centers for possible inclusion)14 and the concurrent introduction of the System-wide and ecoregional programs represented responses to these pressures.15 The pressures were both sub- stantive (concerns about resource degradation) with undefined and overlapping functions, and and political. The political pressures began in the without clear lines of responsibility and ac- 1980s and grew during and after the 1992 Earth countability. The TSR concluded that the struc- Summit (the United Nations Conference on Envi- ture did not serve the CGIAR well in responding ronment and Development--UNCED). These in- effectively to the rapidly evolving external chal- cluded pressure from the forestry community to lenges. Although many reviews carried by TAC establish a consultative group for forestry, and then had previously recommended streamlining in- a decision to include forestry in the CGIAR fol- dividual components of the System, there was lowing strong resistance from Canada to merging neither a consensus nor a clear mechanism to the forestry agenda into the agroforestry Center, address the broader, System-level reforms. which Canada had been instrumental in estab- The Change Design and Management Process lishing.16 This was reinforced by European sup- (CDMP) initiated in 2001 is addressing some of port for the establishment of a stand-alone forestry the issues related to the diffused decisionmak- Center, by donors' impatience with the old ing processes, but it is not challenging the six un- germplasm approach, by their constituencies' de- derlying governance principles on which the mand for the CGIAR to demonstrate more direct CGIAR was founded, and which the TSR found and quicker impacts on poverty and to be more en- to be either no longer appropriate or no longer vironmentally sensitive, and by the Conway report effective. OED concurs. This is particularly true on a vision for the CGIAR (Conway 1997). of the relative "informality" of the CGIAR's gov- Thus, from its relatively straightforward be- ernance. Unlike more recent global programs, ginnings, the CGIAR System became increas- such as the Global Environment Facility or the ingly complex, especially during the 1990s. It newly established Global Fund for AIDS, TB, and went through several periods of growth in the Malaria, the CGIAR System has no formal or legal number of Centers and more recently in inter- persona, written charter, or even a memorandum Center initiatives. And in an effort to accom- of understanding. The only legal entities in the modate its expanding agenda and increasingly CGIAR are the 16 Centers. Responsibilities and diverse constituencies, the CGIAR's governance accountabilities remain ill-defined as a consid- structure grew with the addition of several com- erable superstructure of committees has evolved mittees with responsibilities for programmatic over the years to oversight, finance, impact assessment, and ge- deal with a complex The governance structure netic resources policy. The expansions greatly set of stakeholders. had become cumbersome taxed the System's governance structure. As a result, greater with undefined and By 1998, the TSR panel found that the gov- responsibility for ernance structure had become cumbersome managing the over- overlapping functions. 1 2 3 T H E C G I A R AT 3 1 all System has accrued by default over time to the reforms, including the Challenge Programs World Bank and the CGIAR Secretariat. (CDMT 2001). The CGIAR Secretariat has noted, "on the question of consolidating Centers, the The Challenge of Reconfiguration CGIAR prefers an evolutionary rather than a top- With the 1994 consolidations, CGIAR Centers now down approach.... It is expected that the de- number 16. Consolidation means different things velopment of Challenge Programs will contribute to different people. Stakeholders, scientific com- to consolidation."21 But there is little agreement mittees of TAC, and the CGIAR's own internal re- among stakeholders on whether the Challenge views have proposed consolidations of various Programs are an appropriate or a sufficient mech- natures from as early as 1994, including mergers anism for restructuring. The meta-evaluation of Center boards or management, closure of phys- team sees reconfiguration through the Chal- ical facilities, mergers of programs, mergers along lenge Programs as an unnecessarily painful regional or research lines, mergers of commodity process that would not necessarily result in re- Centers, ecoregional or regional consolidation, configuration based on the System's long-term and consolidation by problem areas (such as com- strategic interests but, rather, could result in re- mon property resources).17 A large majority (79 configuration based on Centers' abilities to raise percent) of CGIAR stakeholders surveyed by OED funds and appeal to donors' shorter-term inter- agree that consolidation in the number and func- ests. Reflections on the ILCA/ILRAD merger also tions of Centers is advisable (box 4.2). suggest that once agreement is reached in prin- There have been several mergers in the CGIAR ciple, consolidations should be carried out as since 1994.18 The System's most comprehensive quickly and painlessly as possible and they should merger experience, the 1994 ILCA/ILRAD merger, be based on strategic interests (box 15.1). has never been evaluated and is often perceived The integrated natural resources manage- to have been disruptive and of limited value. ment program and the System's response to the The meta-evaluation team's investigation with re- 2001 International Treaty on Plant Genetic Re- gard to this merger has led it to conclude that sources are good examples of System-wide col- the merger has helped position the CGIAR to ad- laboration. But in general, consolidations are dress global livestock issues from a position of less popular than expansions. They have never strength, while its many short-term financial and been systematically evaluated. Where consoli- political costs could have been handled better dations have occurred, these have been per- (box 15.1). A setback in making ILRI a truly global ceived as cost-cutting measures that signal research institution has been that funding for ILRI weakness rather than as a way to strengthen declined by about 10 percent after the merger, programs as part of a well-articulated, long-term despite TAC's recommendation for more re- strategy for enhanced productivity and higher re- sources. With the benefit of hindsight and a turns from the System in terms of impacts on the System-wide strategy, important positive and poor. Thus, consolidation remains a priority negative lessons on how--and how not--to fraught with difficulties. achieve mergers could be learned. The CGIAR has passed up several opportu- Collective Action Problems Are Daunting nities for broad, more systemic consolidation, for Resistance to consolidation demonstrates the example, after the TAC report in 1994, after the Olsonian collective action dilemma. "Unless the Third System Review, after the CGIAR board number of individuals in a group is quite small, chairs/Center directors' proposal for a Federation or unless there is coercion or some other spe- in 2000,19 and most recently in the context of the cial device to make individuals act in their CDMP.20 common interest, rational, self-interested in- In 2001, the Group endorsed the CDMT's dividuals will not act to achieve their common recommendation for an "evolutionary" approach group interest" (Olson 1965, p. 52). There is a to restructuring, in which consolidation or other broad consensus that the increasingly complex reconfiguration should emerge from the other environment in which the CGIAR operates is no 1 2 4 PA S T AT T E M P T S T O R E S T R U C T U R E T H E C G I A R T h e U n e v a l u a t e d I L C A / I L R A D M e r g e r B o x 1 5 . 1 E x p e r i e n c e The CGIAR's examination of its strategy toward livestock re- if the Centers had been more cooperative in explaining the search culminated in the 1994 merger of the two livestock Cen- merger to the hosts. The legal agreements with the hosts were ters--the International Laboratory for Research on Animal even more intractable than the political costs. Some have es- Diseases (ILRAD) and the International Livestock Center for timated financial costs associated with lost research produc- Africa (ILCA)--into the International Livestock Research Insti- tivity during the transition to be as high as $35 million, but this tute (ILRI). The merger stemmed from a desire to craft a more ef- estimate is disputed. Tension and a loss of staff morale in the fective, visionary, and global strategy for livestock research, a short term, while unavoidable due to the uncertainty among re- motivation heightened by the CGIAR's 1993­94 financial crisis. search staff about their programs, employment, and manage- The general consensus within the CGIAR seems to be that the ment, may have been greater because of the inadequate handling resulting institution is far better positioned to address the huge of the different institutional cultures and the appointment of a global challenge of livestock development than were the two director general from one of the two merging Centers to lead unconnected Centers at the two ends of the research and de- the new Center. velopment continuum. The long-term return to this consolidation Several lessons emerge from the ILCA/ILRAD experience. is believed to be substantially larger than the immediate finan- First, consolidation should be undertaken for strategic reasons cial savings in Center administration and management for the and long-term positioning, not for short-term cost savings. Sec- merged Center of some $1.2 million in 1998, compared to the com- ond, there must be a clear a priori consensus regarding the bined costs of ILCA and ILRAD in 1994. The end product is bet- long-term benefits of consolidation. Third, consolidation re- ter and more comprehensive livestock research and quires strong, high-level leadership with credibility. Fourth, it is development, leading to improved coherence and effectiveness important to bring the host countries, Center staff, and other af- of science and the prospect of contributing to improved livestock fected parties into the picture at the outset. Fifth, a new direc- production throughout the developing world. This could result tor general should be brought in to lead the new, consolidated in rates of return similar to those that have been realized in Center. Sixth, the process should be transparent and account- germplasm improvement. able, and consolidations should be carried out as expeditiously However, the political and psychological costs and the cost as possible. Seventh, the policy and operational decisions in car- of disruption to the research programs could have been better rying out the consolidation should be in the hands of a small, handled. At the time, those costs were underrated. First, polit- knowledgeable, genuinely concerned and sensitive group of ical costs for the host countries could have been better handled people. Source: Interviews with V. Rajagopalan, Robert Havener, Hank Fitzhugh, Neville Clarke, Robert Herdt, Alex McCalla, Cees de Haan, and Akke Vanderzijpp. longer compatible with 16 fully independent re- fect their interests, as do Center scientists, be- search Centers. African countries as a group cause of the inevitable uncertainties. have called for consolidation to reduce the trans- These examples illustrate that collective action action costs of dealing with 16 Centers. Box 15.2 problems are daunting. The short-run financial lists some of the many other reasons that have costs of reduced Center productivity resulting been offered to justify consolidation. And OED's from consolidation can be considerable, partic- investigations suggest that developing countries ularly if handled poorly, and this, too, is fre- as a whole might well prefer consolidation to re- quently offered as a reason for maintaining the duce their transactions costs in dealing with status quo. Yet the long-run benefits of consol- Centers. But individual countries that host Cen- idation will undoubtedly be significant. OED be- ters oppose consolidation. Donors, Center chairs lieves that fragmentation in the System should and boards, and Center directors oppose con- be reduced through consolidation, streamlin- solidation of "their" Centers or activities that af- ing, and absorption of marginally effective Cen- 1 2 5 T H E C G I A R AT 3 1 ters.22 This should be based on a management also helped reveal some long-standing prob- review of the organization of Centers, programs, lems with the System's governance structure, and science quality from the viewpoint of gen- such as the profound differences of views among erating global and regional public goods. CGIAR members and donors on its mission and Members debate whether a single centralized strategies. The TSR and other System-level analy- board for the System, along the lines suggested ses recognized many of these problems, yet by the TSR and the Federation proposal, would daunting collective action problems have led increase System coherence and effectiveness. the System to avoid the more fundamental re- The System is being pulled in two opposite di- forms in the governance, organization, man- rections. On the one hand, the CGIAR Centers agement, and financing of the System that a are not conducting sufficiently coordinated re- wide set of CGIAR stakeholders and analysts ac- search on the highly decentralized nature of knowledge are needed. Getting key stakehold- NRM research, which calls for effective partner- ers to agree that the time has come for both real ships with NARS to produce regional and na- reform and organizational change remains a tional public goods in NRM. On the other hand, major challenge. OED believes that the Bank the System is not sufficiently centralized to deal needs to use its good will to work with key with advances in the biological sciences and donors who largely finance the System, together IPRs, which call for a more unified approach to with experienced and knowledgeable repre- research strategies and policies. sentatives of developing countries and the While the financial crisis and Renewal period CGIAR System, to address this collective action created additional challenges for the System, it dilemma. V e t e r a n s C i t e M u l t i p l e R e a s o n s B o x 1 5 . 2 f o r C o n s o l i d a t i o n 1. Donors seeking to implement special projects behave strate- 9. Outside actors, such as the private sector, seek to collabo- gically and "play" Centers against each other. rate with the CGIAR, but the System does not provide a sin- 2. Centers compete against each other for resources from tra- gle contact with the necessary authority or capacity to ditional donors. negotiate. 3. Centers duplicate administrative and other functions. 10. Opportunities to debate issues of interest to Centers and 4. Centers are unable to efficiently exploit scientific advances donors are often displaced by CGIAR meetings that consist in molecular biology. of large plenary sessions and reports filled with platitudes. 5. Centers cannot individually afford to build or mobilize the 11. Opportunities for exchange between researchers and donors capacity necessary to address IPR issues. are often displaced by the CGIAR superstructure of secre- 6. Centers impose costs on clients by having individual policies tariats, special interest committees, and others that neither toward international regulations on germplasm exchange, provide funds nor conduct research. and different procedures for that exchange. 7. The CGIAR has inadequate capability to assist countries with Points 3, 10, and 11 are being addressed by the CDMP with the international negotiations on germplasm and genetic resources. establishment of ExCo, formation of the System Office, discon- 8. Formal and informal linkages between Center and NARS sci- tinuation of Mid-Term Meetings, and streamlining of the com- entists are fragmented and duplicative within the System. mittee structure. Source: Herdt 2000. 1 2 6 16 Reform in the CGIAR (2000­Present) A ccording to Chairman Ian Johnson, the CDMP has greater ownership because it has been driven by insiders, unlike the TSR, which was led by eminent outsiders. Furthermore, he considers the current approach part of an evolutionary process for the CGIAR, rather than a revolutionary change as suggested by the TSR or by the CGIAR board chairs and Center di- rectors' Federation proposal.1 The Change Design and Management Team (CDMT), led by Margaret Catley-Carlson, the former President of CIDA, began with an examination of the process and substance of the TSR recommenda- tions, and drew on the paper prepared by TAC, A Food Secure World for All: Toward a New Vision and Strategy for the CGIAR (2000f). The examination of the experience of the TSR the tolerance of the System to absorb the pro- was intended to produce lessons on the process posed changes" (CDMT 2001b). An important out- needed to achieve internal consensus and reform come of the input it received from its Steering in the CGIAR. Accordingly, the CDMT consulted Group (consisting of stakeholders led by Chairman widely to cultivate ownership of its work and rec- Johnson) was the decision not to directly address ommendations within the CGIAR, although, like the issue of System-wide consolidation of pro- the TSR, which the CGIAR membership had crit- grams and Centers contained in the original terms icized, the CDMT also did not conduct any sys- of reference of the CDMT. The CDMT issued its re- tematic analysis of the System's problems and port in April 2001 for consideration at the Mid-Term challenges. Meeting in May 2001.2 The team's work, as agreed The CDMT acknowledged that "although there by the Group, was to focus on: is general agreement on the goals of change, some of these nevertheless are characterized by unre- · A restructuring action plan for the entire Sys- solved issues, and matters known to be contentious. tem, with a clear rationale for program inte- The CDMT therefore needs guidance on the ac- gration and/or consolidation of Centers ceptability of the concepts it is developing, and on (including analysis of options) 1 2 7 T H E C G I A R AT 3 1 · A governance plan that streamlines CGIAR de- CGIAR's previously uncoordinated and inde- cisionmaking and clarifies the roles of all com- pendent administrative and management units. ponents and brings net efficiency gains Third, it proposed to transform TAC into a Sci- · A business plan for increasing efficiency in the ence Council in order to improve the quality of provision of common services, coordinating science. Fourth, it has established Challenge System-wide programmatic activities, and re- Programs to, in the words of the chairman, "el- ducing overheads in order to transfer more re- evate the game to address issues of global and re- sources to research. gional significance." The CGIAR also eliminated its Mid-Term Meeting, dissolved many of its com- Within its broader terms of reference, the mittees and then reconstituted some of them,3 Steering Group also asked the CDMT to respond and upgraded the position of CGIAR director to several specific issues, including how to en- from executive secretary, giving it more executive hance System-wide synergies, strengthen re- powers. These issues were briefly described in gional priority setting, adopt new arrangements Chapter 5; they are dealt with here in greater de- to deal with patents and intellectual property tail. (Some material is repeated here for pur- rights and the private sector, strengthen NARS- poses of clarity and continuity.) CGIAR relationships, address problems of in- ternal inefficiency (overlap, transaction costs, The Executive Council and the like), and improve alignment among ExCo members and others interviewed by the strategy, financing, management systems, and meta-evaluation team at the Annual General organizational structure. Meetings in 2001 and 2002 are optimistic that this Notwithstanding these ambitious goals, seek- is a step in the right direction, but they are re- ing "internal tolerance" and a desire for "quick serving judgment until they see how ExCo op- wins," the CDMT avoided some of the most con- erates. The fundamental challenge is to tentious issues in its terms of reference, includ- appropriately balance legitimacy and efficiency. ing a restructuring action plan for the entire That ExCo is a stakeholder committee enhances System based on a clear rationale for integrating its legitimacy. All stakeholders are represented, programs and/or consolidating Centers. Drop- both developed and developing countries, the ping the idea of merging Centers or creating a three cosponsors, foundations, the Center chairs "Federation of Centers," suggested by Center and directors, TAC/SC, and GFAR, as well as civil directors and Center chairs, it opted for an "evo- society and the commercial private sector.4 While lutionary approach" in which restructuring would ExCo members are selected through caucuses of emerge from other reforms, particularly the stakeholder groups and can make decisions only Challenge Programs. Given the past difficulty of on matters delegated to them by the member- changing the System, the reforms are signifi- ship, they are not formally accountable to those cant. Yet many of the CGIAR stakeholders sur- groups or obliged to solicit the views of their veyed by OED believe they do not go far enough "constituent groups" before decisions are made. (box 4.2). In addition, most developed-country groups are At the Mid-Term Meeting in May 2001, the more organized than developing-country groups, Group adapted the CDMT's proposals into four and only members "in good standing" (whose an- actionable areas, each of which has become one nual membership dues of $500,000 are paid in pillar of the current reform effort. First, the CGIAR full) are eligible for ExCo membership. These fac- established an Executive Council to improve tors limit the effectiveness and extent of devel- the efficiency of de- oping-country membership.5 The CDMT opted for an cisionmaking and One initial concept for the Executive Council "evolutionary approach" in implementation. was a committee of investors, with groups such Second, it initiated as the Science Council, the NGO committee, and which restructuring would a System Office to Private Sector committee serving in an ex officio emerge from other reforms. bring together the capacity. At the Annual General Meeting 2001, 1 2 8 R E F O R M I N T H E C G I A R ( 2 0 0 0 ­ P R E S E N T ) however, CGIAR members opted to have these ning" under its new Policies need to be designed groups represented on ExCo as full members, and enhanced role, to minimize potential with the body being a stakeholder committee it is also not clear conflicts of interest within rather than a shareholder committee. In the light who currently plays the membership. of a new private foundation (Syngenta Founda- this role. It is not tion) with commercial interests in new agricultural clear if the situation is being addressed satisfac- technologies joining the CGIAR in October 2002, torily through the regular monthly meeting of the the issues of potential conflicts of interest need CGIAR chairman and director with Center di- to be addressed. Similar questions have arisen in rectors and the representation of the Center di- the case of NGOs, since developing countries rectors' chair on ExCo. This issue will need to be and farmers' groups have raised questions about monitored. whose interests the NGOs represent. Policies A third and related concern expressed was that need to be designed to minimize potential con- ExCo itself lacks either scientific capacity or the flicts of interest within the membership.6 Fur- necessary scientific inputs. For well over a year, thermore, if ExCo is to remain a stakeholder the System has been operating with an interim rather than shareholder committee, and since Science Council, which has not enjoyed system NGOs have been able to have a seat on the ExCo confidence. Ideally, scientific views should be without being CGIAR members, there is little solicited from the Science Council in the form justification for requiring developing countries to of position papers analyzing various options to have dues paid in full to be eligible to serve on help make strategic decisions. But the role of the ExCo. The situation could perhaps be improved Science Council beyond assessing the science by recognizing in-kind contributions from de- quality of individual programs, in such matters veloping-country members, since there appears as the approval of Challenge Programs and the to be little parity between developed and devel- Centers' medium-term and annual plans, is now oping countries in reporting of contributions in in question, as discussed further below. Reliance this regard. on the CGIAR Secretariat poses the risk of cre- A second concern is that ExCo may not be set ating conflicts of interest in resource mobiliza- up to make the best possible decisions. Several tion, policy and strategy formulation, and stakeholders expressed a concern to the meta- resource allocation. Besides, the Secretariat lacks evaluation team that ExCo can make decisions the necessary expertise to perform all these without the benefit of substantive interaction functions, and some have shared a view with with the Centers, particularly with the abolition the meta-evaluation team that it poses the risk of the Mid-Term Meeting, and given the contin- of undue concentration of power in that office. ued importance of restricted funding tied to Apprehensions about the substantive analysis particular programs in individual Centers. The underlying ExCo decisions is perhaps com- Mid-Term Meeting provided an important op- pounded by the concern raised by some stake- portunity for Centers and donors to exchange holders that ExCo meetings are closed to other information and perspectives, enabling negoti- CGIAR members and stakeholders, although the ations on financing Centers' research programs. agendas and the outcomes of the meetings are Now, Centers and donors express a concern that made available to the membership. discussion of scientific issues has diminished. Be- A related issue is that of the cosponsors. The sides, few members of ExCo have support struc- TSR recommended eliminating cosponsor status, tures within their own agencies to provide the and instead giving those institutions permanent necessary analysis underlying decisionmaking seats on its proposed centralized board. The in donor agencies not represented on ExCo, CGIAR did not endorse the recommendation to and the body itself lacks independent intellectual change the status of its cosponsoring agencies. analysis on the pros and cons of complex issues. But in keeping with the TSR recommendation, Other than the CGIAR Secretariat, which has cosponsors have three seats on the new Execu- now been given responsibility for "business plan- tive Committee; this helps to increase the own- 1 2 9 T H E C G I A R AT 3 1 ership of the CGIAR, particularly among devel- reduced.9 There is currently a lack of clarity re- oping countries.7 While they continue to enjoy garding whether ExCo is a decisionmaking body cosponsor status on ExCo, one cosponsor (FAO) or a body that proposes decisions to be consid- has expressed a concern to the meta-evaluation ered by the membership at the Annual General team that its historical role may be diminished Meeting.10 as the reform process proceeds. The role of cosponsors in a rapidly changed context may The System Office have to be addressed more directly. The System Office links 10 independent units in In sum, the formation of the ExCo is a reform order to increase their coordination and ability in the right direction, although, understandably to serve the Centers and the membership.11 Is- at this early stage, there are many new challenges sues of authority, responsibility, and accounta- with respect to its role, representativeness, qual- bility between the CGIAR Secretariat (which is the ity, functioning, and effectiveness. How these coordinating unit), the Centers, and other units challenges are handled will be central to the fu- are still being sorted out. Monitoring is required ture effectiveness of the CGIAR. It is important of the System Office's functions, resources, re- that the CGIAR reforms be fully vetted by the sponsibilities, accountabilities, and performance membership and their advantages and disad- through routine evaluations to ensure its effec- vantages systematically understood, for the CDMT tiveness in serving the members and clients. concluded that the TSR proposal for a centralized Having lost the momentum on divesting certain board had been rejected, among other reasons, Center responsibilities to the System level--as because its advantages and disadvantages had proposed in the Federation proposal--there is never been fully debated.8 Similarly, the Federa- a concern that the System Office may be dupli- tion proposal of the CGIAR board chairs and cating services already performed at the Center Center directors seems to have been rejected level. because it was never fully discussed. OED concurs with the proposal that the TSR The Science Council had made for an executive committee with de- Historically, TAC played a powerful role in the cisionmaking powers and consisting of formally CGIAR's governance and organizational structure elected members accountable to the particular by setting System-level priorities; recommend- groups they represented. Currently, ExCo is nei- ing allocations of resources among Centers, pro- ther a decisionmaking body nor fully represen- grams, and activities; monitoring budgets; tative of the membership, or formally conducting Center-level and System-level reviews; accountable to it. Both conditions are neces- and, more recently, assessing impacts.12 How- sary to increase the legitimacy and effectiveness ever, the TAC's influence declined during the of ExCo. Representation will avoid some of the 1990s (box 4.2). In the view of the meta-evalua- problems concerning quality, responsibility, and tion team, its advisory committee, and certain accountability that the CGIAR's own reviews key donors, the most significant reasons for this have detected in their self-nominating Center decline have been the rise in restricted funding boards. Given the importance of ExCo, an effort and the change to a matching grant formula for should be made to improve the quality of stake- the allocation of the Bank's resources.13 Together holder participation and to monitor its ac- these have allowed donor preferences to drive countability to stakeholders. Without these, the programs and decoupled resource allocation from authorizing envi- TAC's medium- and longer-term priority setting. ExCo is neither a ronment and legiti- Some donors have also acknowledged to the decisionmaking body nor macy of ExCo will meta-evaluation team that their priorities are in- fully representative of the be undermined, fluenced by domestic constituency interests, and the CGIAR's col- which tend be short-term, while research is a membership, or formally lective action prob- long-term endeavor requiring steady commit- accountable to it. lem will not be mentsthatarenotnecessarilypoliticallypopular. 1 3 0 R E F O R M I N T H E C G I A R ( 2 0 0 0 ­ P R E S E N T ) Hence, the TSR recommended strengthening to the Challenge Programs and the new and TAC, and the CDMT recommended transforming emerging Finance and Program Committees are it into a Science Council to "ensure that the sci- unclear, particularly as the permanent Science ence practiced in the System meets world class Council is not in place while major decisions on standards." OED strongly agrees with this objec- Challenge Programs are being made.16 Fifth, Sci- tive. The CGIAR needs a strong, qualified, and in- ence Council members need to be independent dependent Science Council to set overall System and objective, and their findings shared with priorities, to support ExCo, and to ensure the the full membership when they are transmitted quality and impact of all System-level programs, in- to ExCo. Yet the Working Group on the Science cluding the Challenge Programs.14 The decline in Council has not determined the time and fi- independent scientific advice in the CGIAR went nancial resources needed, or the remuneration hand-in-hand with the decline in the strategic na- arrangements for the Science Council members ture of the CGIAR's research during the 1990s. But and its chairman (box 16.2) OED is not convinced that the transformation of In summary, for the Science Council to ef- TAC into a Science Council will achieve the desired fectively play this role, several requisites seem ev- objective for a number of reasons. ident as the Council is brought into full First, this transformation is taking place with- operation: out the TAC having been evaluated while major decisions on resource allocations are being made. · The Science Council would need to have a Second, relative to TAC, the Science Council's strong chair, with the necessary distinction in role is greatly diminished (box 16.1). It is ex- the knowledge of smallholder agricultural de- pected to focus mainly on science quality; to velopment, agricultural policy, and the role of have a more limited (if any) role in priority set- science, and who is widely respected for his/her ting, medium-term planning, and monitoring intellect, has a reputation for independence, is resource allocation; and to play no role in the an- willing to speak his/her mind, and enjoys the nual financial planning process.15 Third, FAO respect of CGIAR stakeholders. has indicated to the meta-evaluation team that · The Science Council's views would need to it is not being adequately consulted about the be available not only to the general member- role of the Science Council. Fourth, the roles and ship, including the donors, but also discussed responsibilities of the Science Council in relation at Annual General Meetings. T h e S c i e n c e C o u n c i l H a s L e s s P o w e r T h a n B o x 1 6 . 1 D i d T A C TAC's Terms of Reference: The Science Council's (SC) proposed responsibil- ities: · Provide independent advice and judgments on strategic issues and on the quality of the scientific programs supported by the · Serve as guardian of the relevance and quality of science in CGIAR. the CGIAR. · Recommend research priorities and strategies to the CGIAR. · Advise the CGIAR on the strategic scientific issues relevant · Ensure the quality of research supported by the Group and its to the Group's goals and mission. relevance to the CGIAR's goals and objectives. · Recommend the allocation of resources among Centers in the context of CGIAR-approved priorities and strategies. Source: CGIAR Secretariat 2000e; 2001e. 1 3 1 T H E C G I A R AT 3 1 I s s u e s i n t h e T r a n s f o r m a t i o n o f T A C t o t h e B o x 1 6 . 2 S c i e n c e C o u n c i l The size of the Science Council (SC) has been a matter of Bank, the TAC Secretariat has been funded by all three cospon- debate and cannot be fully resolved until the Council's role is de- sors--FAO, UNDP, and the World Bank. But the UNDP contri- termined. According to the Working Group on the Science Coun- bution has declined significantly, and FAO has stressed the cil, the SC will consist of six members plus a chair, supported risks it foresees in the near future in relying mainly on FAO funds by four Standing Panels, compared with the 10 to 14 members over time. In the future, all donors should share the costs of the on TAC. TAC had wider responsibilities than the SC, but was un- CGIAR Secretariat, the SC and its Secretariat, and other central derfunded. Only the TAC chair worked close to full time and the bodies in the CGIAR System. others were remunerated on a needs basis. Funding restricted Reporting Arrangements: To perform its strategic advi- the kind of members TAC could attract, since young scientists sory function for ExCo, the CGIAR chairman, and the System, the at the cutting edge of research found it difficult to find the time SC should report to the CGIAR membership as a whole, as TAC for TAC work on a dedicated basis. used to do. SC findings should be shared with the full CGIAR mem- Funding of the SC and its Secretariat: The time, re- bership at the same time as they are transmitted to ExCo. While sources, and payment arrangements for the SC members and this remains the principle, in practice, timely reporting to the chair have not yet been addressed by the SC Working Group. To CGIAR membership broke down in the case of the pilot Challenge obtain the inputs of top-quality, cutting-edge scientists on a de- Programs prior to the Annual General Meeting 2002. manding SC and to enable them to devote the necessary time Independent External Evaluations: Thesevariousissues for managing reviews of major programs will require at least with regard to the role of the Science Council reinforce the one-third of the time of the four or six core SC members (as- need for independent external evaluations of the CGIAR System, suming they have other full-time engagements), and should be including the functioning of all its important units, on a periodic reserved by ensuring them appropriate compensation. Provision basis, and providing for appropriate follow-up of the findings by should also be made to compensate Standing Panel members. the boards of the cosponsoring and funding agencies and the Unlike the CGIAR Secretariat, which is fully funded by the membership at large. · The Science Council would have to have a full- partnerships addressing complex issues of global time core body with sufficient resources and or regional significance. These would increase the support. scope for inter-Center collaboration, facilitate a · A concerted effort at the highest levels would wider range of partnerships, tap new sources of be needed to reverse the trend in restricted funding from current and new donors, and im- funding. prove output accountability. The CDMT and the · A separation would be established between chairman also viewed Challenge Programs as a resource mobilization and resource allocation way of addressing consolidation indirectly by functions generally, and with regard to the al- building new programs and partnerships while, location of Bank funds. The latter would need in the words of some interviewees, letting the to be linked to the CGIAR's long-term priori- weak Centers and programs "die on the vine" ties and strategies established by the Science rather than face politically unpopular consoli- Council, which should be based on global and dation. Their design was also to address three im- regional public goods. In this way the World portant shortcomings that have plagued previous Bank would assure itself that its resources are thematic and ecoregional System-wide programs: leveraging other resources to maximize their insufficient funding, poorly defined timelines, impact on poverty. and a lack of adequate governance and man- agement. In October 2002, after a year-long re- Challenge Programs view process involving both the interim Science The CDMT recommended the creation of Chal- Council and ExCo, the Annual General Meeting lenge Programs as high-visibility, time-bound approved the implementation of the first two 1 3 2 R E F O R M I N T H E C G I A R ( 2 0 0 0 ­ P R E S E N T ) pilot Challenge Programs: "Water and Food" and "Biofortified Crops for Improved Human Nutri- tion."17 While passing scientific review, these two pilot programs do not address the three shortcomings of thematic and ecoregional pro- grams (funding, timelines, and governance) noted above. From among existing proposals for Challenge Program themes, the interim Science Council (iSC) had recommended that CGIAR members should decide at the Annual General Meeting 2001 which should be accelerated. Candidate themes were to include those "on which there has been significant preparatory work and prior discussion within the CGIAR and among the stakeholders (and which have already identified up the System by promising to allocate sub- or have potential for significant additional stantial resources to advanced research institu- funding)" (CGIAR Secretariat 2001g). Because of tions and the NARS of developing countries. Yet a lack of agreement in the membership on pri- only one-third of stakeholders surveyed by OED orities and funding potential at the Annual Gen- believe the Challenge Programs are sufficient to eral Meeting 2001, the CGIAR leadership then open up the CGIAR, to produce the best sci- asked the interim Science Council to screen the ence, or to achieve consolidation (box 4.2). concept notes and make recommendations for OED's interviews with CGIAR members reveal those that should be accelerated. The iSC sub- ambiguity and confusion about the Challenge sequently recommended that the concept notes Programs and a range of concerns about their de- on water, biofortification of crops, and genetic velopment--concerns articulated by members at resources be pursued, and full proposals were the Annual General Meetings in 2001 and 2002-- developed. At its April 2002 meeting, ExCo noted, although some supported moving full speed "the CGIAR is not obligated to fund all three ahead with Challenge Programs.19 proposals that are being developed. Some CGIAR members will face difficulty in supporting CPs fo- Funding. The CGIAR leadership has sought far cused on genomics" (CGIAR Secretariat 2002b)-- larger funds for Challenge Programs than was reflecting the tension between science-based originally understood when the idea was ap- priority setting, on the one hand, and funding po- proved--$82 million for Water and Food, $42 mil- tential and political acceptability, on the other. lion for Biofortification, $69 million for Unlocking Twelve of the 13 concept notes selected for Genetic Diversity, and $100 million for Sub-Sa- pre-proposal development in the first phase of haran Africa. Center directors have stressed to the program were submitted by CGIAR Centers, OED that each fully funded Challenge Program and the ExCo acknowledged that 90 percent of is equivalent to the establishment of an additional the pre-proposals for the pilot phase "lacked new Center. Moreover, the Challenge Programs any meaningful NARS participation" (CGIAR Sec- were to be financed with additional funds, but retariat 2002b). The CGIAR chairman noted that initial pledges for the first two are far below World Bank funds would not be allocated to the their costs.20 If addi- If additional funds are not Challenge Programs if they proved to be captured tional funds are not by the Centers.18 forthcoming, the forthcoming, the Challenge OED believes that, properly developed, the Challenge Programs Programs will either be Challenge Programs could enhance CGIAR's ef- will either be under- under-funded or funded fectiveness and impact. One of their strong and funded or funded positive features is that they are helping to open from existing pro- from existing programs. 1 3 3 T H E C G I A R AT 3 1 grams. Reducing core funding to the Centers sufficiently spell out their methodology and lack has the danger of further undermining the well-defined endpoints.23 Large sums announced CGIAR's proven germplasm improvement pro- by donors for programs involving individual Cen- gram. The allocation of Bank funds to Challenge ters in advance of the review process compromises Programs is a pivotal concern, given that Bank the SC review process for scientific quality. The un- funds are unrestricted and have traditionally certain role of independent scientific advice in re- provided the System with a seal of approval for lation to the Challenge Programs is exemplified science quality and management. Instead of al- by the limited resources allocated to the iSC for locating Bank funds to areas other donors are evaluating concepts, pre-proposals, and propos- most willing to finance, the Bank should use its als for the pilot Challenge Programs.24 Related funds to achieve strategic reforms in the Sys- concerns are the current and future roles of the tem as a whole. SC, and the roles of ExCo, the Finance Commit- tee, and the Program Committee relative to the Strategic Priorities. The tension between topics SC, and of the Challenge Programs themselves that can generate funds and those that are needed within the System (Dalrymple 2002a). Without the to ensure impacts on the largest number of the Science Council's strong and independent ana- poor has been evident from the outset of the lytical input, ExCo, the CGIAR Secretariat, and Challenge Programs. The Challenge Program the Finance and Program Committees are not process is becoming time-consuming, and their equipped to deal with the appropriateness or sci- proposed sequential approval is distracting from ence quality of proposals or with the Challenge System-level priorities and strategies.21 Past re- Program's role in System-wide priorities. Nor is the allocations, based partly on the revealed prefer- membership likely to be able to make informed ences of donors, have led to reductions in decisions without timely, widely shared, high- productivity-enhancing strategic activities of a quality input from a strong SC with well-defined, global or regional public goods nature and in re- transparent procedures. search on commodities of importance to the poor.22 The iSC therefore recommended to ExCo Governance and Management. The governance and in October 2002 that the introduction of new management structures proposed for individual Challenge Programs into the CGIAR research Challenge Programs vary, seem ad hoc, and pose agenda be slowed down to enable the CGIAR to concerns about transparency, accountability, likely (1) review the evaluation criteria designed by science quality, and ultimately about science lead- the Task Force on Challenge Programs and en- ership. Challenge Programs and their evolving re- dorsed by the Group at the Annual General Meet- porting arrangements (as reported at the Annual ing 2001, (2) learn lessons from the pilot process, General Meeting 2002) seem to be autonomous and (3) review the implications of the Challenge from the System. For example, the Biofortification Programs for System-level resource allocations. Challenge Program proposal states that its inde- OED considers this an important signal that pendent Project Advisory Committee will make an should be acted upon before approval of any ad- annual progress report to the SC and that the two ditional Challenge Programs. Researchers are sponsoring Center directors will report to ExCo. trying to convert almost every important research Yet guidelines on reporting arrangements have yet theme into a Challenge Program as a way of rais- to be issued. The ad hoc nature of the gover- ing its profile nance and management arrangements in the pilot The tension between topics that and mobilizing programs are nevertheless establishing prece- can generate funds and those funds. dents in standards and procedures. that are needed to ensure impacts on the largest number Science Quality. Global Public Policy Issues. It is unclear how the The approved Challenge Programs will address the key global of the poor has been evident Challenge Pro- public policy issues associated with intellectual from the outset. grams do not property rights and public-private partnerships 1 3 4 R E F O R M I N T H E C G I A R ( 2 0 0 0 ­ P R E S E N T ) identified in this meta-evaluation. The Biofortifi- manent Science Coun- A Science Council is not in cation Challenge Program has only partially ad- cil with the necessary place while major dressed this. Avoiding undesirable outcomes will qualified members decisions about the System require System-wide priorities, monitoring of in- endowed with a man- are being made. dividual Challenge Programs, and assessing the date and equipped opportunity cost to developing countries of un- with resources is not in place while major deci- dertaking one program relative to another. The sions about the System are being made. It also be- CDMT's recommendation that one-half of CGIAR hooves the CGIAR to update its methodology research be moved to Challenge Programs by 2006, for priorities and strategies, to track expendi- while making those programs responsible for rais- tures by broad categories, and to require rolling ing their needed funds, seems premature. It seems annual and multiyear research priorities, by means prudent not to approve any more Challenge Pro- of a balanced approach of identifying new sci- grams until a new Science Council is in place, and entific opportunities based on analysis and a its roles, responsibilities, accountabilities, inde- broad consultative process, led by the SC, with pendence, reporting arrangements, and sup- scientists, Centers, NARS, stakeholders, and porting resources in relation to the rest of the donors. Such a process should result in a call for System are fully established. The Bank should act competitive proposals rather than assigning re- as a leader in helping to establish high standards sponsibility to individual Centers to come up for the conduct of the Challenge Programs. with a single specific proposal that is funded, as In summary, the Challenge Programs have has been the case with the Challenge Programs. brought into sharper focus the long-standing is- Lack of decisions on the role of the new Science sues the System has been facing, in particular the Council has held back such an important long- classic public finance tension between unrestricted term strategic effort to develop new method- common pool funding to produce a public good ologies for System-level priority setting. The and restricted project funding from specific donors CGIAR's old approach--based on commodities to address particular agendas by supporting spe- and their correspondence with potential impact cific CGIAR Centers. Ideally, the System should on poverty--is clearly no longer sufficient by it- raise as much funding with as few restrictions as self, since the CGIAR has diversified in many possible. It should use strict scientific assess- other directions of research beyond commodities. ments driven by the CGIAR's central mission, but The old approach is unable to assign budgetary largely independent of funding pressures to al- priorities to research themes such as water, NRM, locate resources to certain activities. Even if fund- policy, research management, and so forth in a ing is restricted to specific topics, the use of those scientific fashion. More recently TAC/iSC has fol- resources should be based on objective criteria lowed a new, regionally focused approach in two and their transparent application. A better way of regions, which is potentially too broad, time- handling Challenge Programs would have been to consuming, and onerous. Without sufficient re- solicit the commitment of donors around the im- sources to staff a Science Council and without portance of an idea--for example, to address decisions on consolidation of the System, the water issues. Proposals could then have been in- two key elements of the reform process--the vited on a competitive basis by the iSC, with sci- role of the Science Council and the Challenge Pro- ence quality rather than relevance alone being a grams--pose the risk of diverting limited re- screening device. Instead, full proposals were so- sources and attention away from strategic issues licited by allocating up to $200,000 to all the three of governance, organization, management, and proponents whose pre-proposals were approved finance that a truly global research system should and Centers de facto were given tied funds, put- address at this time. ting the approval process of the SC in jeopardy.25 This point raises the interesting issue of when, Reforms Do Not Go Far Enough by whom, and how the CGIAR System-level pri- On the whole, the reforms of the CDMP are a orities and strategies should be established. A per- meaningful step forward, particularly relative to 1 3 5 T H E C G I A R AT 3 1 other attempts at System reform. At the same sulted. Nevertheless, few CGIAR stakeholders time, the combination of seeking "internal tol- believe that the recent changes go far enough.26 erance" to proposed changes and the desire for The reforms are constrained by the lack of a "quick wins" resulted in a set of reforms that still clear vision for the CGIAR. While TAC undertook do not address some of the System's most un- an effort to update the CGIAR's vision and strat- resolved and contentious issues. Indeed, the egy, which was endorsed by the CGIAR at the Mid- current reforms might be thought of as picking Term Meeting in 2000, this vision still falls short the low-hanging fruit and reached with a mini- of addressing key questions such as whether the mum of strain. Given the historical difficulty of CGIAR will pursue research focusing on improved achieving change in the System, it is easy to ap- agricultural productivity and efficient use of nat- preciate the constraints faced by the CDMT, and ural resources, or whether it must be more op- thus the significance of the reforms that have re- portunistic in its choice of activities.27 1 3 6 17 The World Bank and the CGIAR T he purpose of this chapter is to assess the Bank's performance of its three major roles in the CGIAR partnership: · As founder, cosponsor, and convener: chairing instruments, such as agricultural sector analysis, the System, housing the CGIAR Secretariat, policy dialogue, and lending? and managing the CGIAR trust funds · As donor: one of the two largest donors to Exercising the Bank's Leadership Role the CGIAR, with the most unrestricted funding as Founder, Cosponsor, and Convener · As development partner: the largest lender to Former Bank President Robert McNamara was a agricultural development, committing $85.6 key player in founding the System. Since then, billion to 1,770 projects since 1971, and to each president has been strongly committed to agricultural research and extension, commit- the CGIAR. All the chairmen3 have exercised ting $6 billion to 173 projects in 91 countries their role with energy and dedication. While fa- since 1971.1 miliarity with the management of science and agricultural policy has been the exception rather How has the Bank performed in these roles than the rule, all CGIAR chairmen have been in relation to the three key principles that gov- strong, committed champions advocating the ern the Bank's grant-giving under the Develop- CGIAR cause and mobilizing resources toward its ment Grant Facility (DGF)--an arm's-length accomplishment. Bank support has lent legiti- relationship, an exit strategy from DGF fund- macy to the System in various ways--for exam- ing, and financial subsidiarity?2 For example, as ple, in expanding the System by bringing in new a convener, has the Bank provided the neces- Centers when new social or environmental con- sary intellectual and financial leadership appro- cerns warranted broadening the agenda, in con- priate to changing times? How does the Bank's solidation of the Centers when economizing, relationship with the CGIAR work in practice? As and more recently in reforming governance and donor, should the Bank consider an exit strat- management when needed. egy? If so, what might such a strategy be? And as The Bank's chairmanship has made it possi- a development partner, does the Bank's an- ble for the CGIAR to enjoy access to govern- nual grant to the CGIAR compete with or com- ments at the highest levels in both developed and plement the Bank's other country assistance developing countries, ensuring continued po- 1 3 7 T H E C G I A R AT 3 1 The Bank has been described litical and financial DGF grant of $50 million for the CGIAR Secretariat as the indispensable support and ex- and$0.75millionfortheTACSecretariat.Thishis- guardian of the CGIAR panded member- torical situation has not yet been brought in line ship. The Bank with the new guidelines adopted by the DGF in System, and as the glue that has been de- June 2000, that the Bank should not fund more makes the System coherent. scribed to the than 50 percent of in-house secretariat costs in meta-evaluation order "to avoid a program's overreliance on the team as the indispensable guardian of the CGIAR Bank."6 System, and as the glue that makes the System Several donor representatives to the CGIAR coherent and larger than the mere collection of have indicated that, while donors have been in- 16 international research Centers. Particularly terested in the Secretariat's efficacy (that is, since the 1993­94 funding crisis, the Bank has whether it serves the System well), they have assumed a far larger role than the founders of the been less concerned about its efficiency (at what CGIAR, including Robert McNamara, had envis- cost) because they have not contributed to its op- aged.4 Among the original three cosponsors--the erations. By the same token, if they were to con- Bank, FAO, and the UNDP5--the Bank has ended tribute to the Secretariat costs, they would pay up assuming increased responsibility, some more attention to the running of the Secretariat would say an undue burden, for the governance, and expect greater accountability in costs and financing, and management of the System as performance. Few donors interviewed by OED the roles, responsibilities, and contributions of indicated resistance to the notion of cost-shar- other cosponsors have declined. The Bank's ing of Secretariat expenses. However, the CGIAR leadership role, its financial contributions, and chairman has pointed out several potential draw- its operational support are viewed by other backs to cost-sharing: (1) difficulty for develop- donors as a seal of approval, giving them confi- ing-country members and smaller donors from dence to continue to invest in the System. They industrial countries in participating, (2) a concern also observe that without the Bank, the System that any savings to the Bank from such an likely would not have lasted as an integrated arrangement would likely be returned to the whole, although some of the stronger Centers DGF budget rather than remaining in the CGIAR, would no doubt have survived. (3) unpredictability and instability entailed in The CGIAR Secretariat is an independent cost relying on donor funding, and (4) likely dimin- center in the Bank. Its executive director, who has ished Bank influence.7 recently been given substantially increased pow- The performance of the CGIAR Secretariat in ers, reports to the chairman and, working with conducting its various roles to service the System him, raises funds, serves as secretary to the ExCo, has not been evaluated since 1987. Several key and largely functions as the chief operations of- members indicated in October 2001 that the Sec- ficer. The CGIAR leases office space from the retariat had increasingly begun to serve the needs Bank and benefits from the conveniences of being of the donors, and particularly the chairman, housed in the Bank, including access to the Bank's rather than those of the Centers, a result con- administrative infrastructure and special status firmed by OED's survey of CGIAR stakeholders. as a U.N. specialized agency (G-4 visas, personnel It is too early to assess performance of the cur- work, travel support, and management informa- rent director in view of the major changes in tion). The Bank has been the sole financier of the CGIAR governance. But interviews with several CGIAR Secretariat members of ExCo and the chair of the Center Di- According to new guidelines and contributes to rectors' Committee have since indicated that the adopted by the DGF in June the TAC Secretariat Secretariat's service role has become more ef- 2000--the Bank should not (housed in FAO)-- fective with the appointment of the new direc- in recent years, pro- tor. Vested with more executive powers since fund more than 50 percent viding $4.25 million the inception of the CDMP, the CGIAR director of in-house secretariat costs. out of the annual hastakenaleadershiproleandanumberofsteps 1 3 8 T H E W O R L D B A N K A N D T H E C G I A R to increase clarity of roles and responsibilities, ment, and provide a service by making IAD re- identified issues needing decisions, and acceler- ports publicly available. This would be consistent ated implementation. In 2001, the Secretariat with the objective of developing an informed published its first-ever "business plan" and annual membership, able to make decisions based on report for the general public, improved the timely independent professional analysis. distribution of meeting documents, reduced the Notwithstanding the Bank's large role, the turnaround on the proceedings of meetings, and CGIAR System has succeeded in establishing an has generally begun to actively serve ExCo. Stake- identity for itself that is quite separate from the holders indicate that the different operating styles Bank. Stakeholders do not regard the CGIAR as of the two recent chairmen and executive sec- largely a World Bank program, unlike some other retary/director, combined with the pressure from global programs with in-house secretariats. the Bank management and DGF to reconsider the level of Bank support to the CGIAR, may explain Funding the CGIAR and Exit Strategy the increased responsiveness and greater ac- from the DGF? countability to the System. The Bank's financial contributions to the re- The Bank's administration of the CGIAR trust search agenda of $796 million up to the end of fund is an increasingly important service pro- calendar year 2001 have been more valuable vided to CGIAR members. In 2001, $81.8 million than most because they are completely unre- of member contributions were channeled to the stricted, providing 31 percent of the CGIAR's Centers through trust funds, compared with unrestricted funding in 2001. The Bank has $35.7 million in 1999. This line of business may helped stabilize the System by providing ex- well grow. The Bank acts as a fiscal agent in this traordinary funds, such as the additional $10 regard and does not charge any administrative million in both 1994 and 1995 and by advancing fees, since the Bank provides minimal project funds from its annual contribution during peri- management and reporting services.8 ods of financial difficulty caused by unexpected The Bank used to conduct audits of the Sys- shortfalls in other donors' contributions. tem at regular intervals; the last one was done The CGIAR currently receives 40 percent of in 1995. The 1995 Internal Audit Department the DGF funds going to global programs. It is one (IAD) report was grounded in extensive field of 12 "Window 1" programs for which the Bank's discussions with CGIAR Centers, directors, and Executive Board and the DGF have endorsed leaders of scientific programs. It was of high multiyear funding of more than three years,10 and quality and prophetic with regard to the opera- because it was grandfathered when the DGF tional concerns it raised. It anticipated many of was created in 1998, it has no strategy for a grad- the problems connected with changes in the ual disengagement or eventual exit from DGF Bank's funding formula. Unfortunately, its cir- funding. The reluctance of the Bank's Executive culation was restricted to the CGIAR Secretariat. Board in June 2002 to shift $2 million of DGF re- Had it been available to all partners and widely sources from the CGIAR to new, as yet untested discussed within the Bank, it might have led to global initiatives is understandable. There is lit- an internal debate in the Bank and in the CGIAR tle doubt of the continued necessity of invest- membership before the new matrix manage- ment in agricultural research at the global level, ment reporting system was put in place, and the considerable continued impact of the CGIAR, changes in management practices might have and the difficulty in mobilizing funds for agri- been adopted. There has been no similar audit cultural research. A steady stream of agricultural since 1995, despite the current CGIAR direc- technologies adapted to the varied circumstances tor's request for one upon assuming office.9 In of developing countries will remain critical for line with its fiduciary responsibility for the man- economic growth, poverty reduction, and sus- agement of trust funds, the Bank should un- tainable development. And it is likely that pub- dertake periodic System-level IAD reviews, lic sector support at the global level will be charge a standard fee for trust fund manage- needed for a long time to come. 1 3 9 T H E C G I A R AT 3 1 Yet decisions on DGF allocations to the tive sources of funding. Private and corporate phi- CGIAR tend to transcend normal DGF processes. lanthropy, for example, offered limited oppor- The long-standing allocation to the CGIAR of tunities for the CGIAR and, furthermore, would large amounts of DGF funding and the lack of involve costly cultivation efforts and rethinking an exit strategy for the program have raised of the donor role. Partnerships with the business concerns within the Bank about the different community had considerable potential, but it standards that apply to the CGIAR compared could be difficult to bridge the gap between cor- with other programs--even to the 11 other porate culture and the culture of public good and global programs in Window 1 that have been poverty alleviation. An endowment would allow approved for medium- to long-term funding. the System to plan for the long-term sustain- The relationship between the DGF and the ability of its components, but would also call for CGIAR and the oversight arrangements for the special management expertise. allocation of such large sums of funds in an ob- The funding choices the CGIAR ultimately jective manner need to be clarified. makes will require careful planning of the pub- Some Bank staff are concerned about con- lic awareness and fund-raising strategies, re- tinuing to finance the CGIAR through the DGF porting arrangements for the fund-raisers, and the and undermining the DGF criteria in the process. overall management arrangement. In particular, But others argue that, given the history of lob- who would be in charge? Should it be carried out bying by the CGIAR, if the CGIAR were moved by an NGO reporting to the Centers, such as the outside the DGF, it would become more difficult Future Harvest Foundation? Or should such an to objectively consider the level of resource al- NGO report to the stakeholders, the chairman, location to the CGIAR in relation to other global the CGIAR Secretariat, or the major funders? The programs. This is one reason for OED's recom- options for funding are many. For example, the mendation, in its Phase 1 report on the Bank's donor community could choose to: involvement in global programs, that continued support to the CGIAR, as indeed to other pro- · Fund certain overall objectives of the CGIAR grams above a particular threshold, should in- System (in the manner of the Global Environ- volve a triennial appraisal of the partnership, ment Facility). with Board approval as the basis for continuing · Organize periodic replenishments (as in the Bank support. Down the road, during the IDA 14 Global Fund for AIDS, TB, and malaria). replenishment, IDA deputies might also con- · Fund specific programs with tied aid (as is the sider expanding the IDA grant-making criteria to case in the current CGIAR). include support for long-term research as a · Opt for completely untied, performance-based global public good.11 funding through IDA. The report "Towards a Long-Range Financ- · Solicit private large- or small-scale philanthropy ing Strategy for the CGIAR" (The Conservation (in the manner of World Wildlife Find fund-rais- Company 1999), commissioned by former Chair- ing, some of it at the retail level). man Ismail Serageldin, presented guidelines to · Establish endowments for certain CGIAR ac- improve the stability and predictability of fund- tivities (as the Global Conservation Trust Fund ing for the CGIAR research agenda and pro- does). posed appropriate structures and mechanisms. The report consid- All options remain open. Currently the chair- Some limited genuine global ered a number of fi- man, the director of the CGIAR, and the Center public goods requiring long- nancing options directors and boards are all raising funds, while term investments will have and concluded that the Future Harvest Foundation is looking for a to be financed through the official develop- fund-raising strategy around some strategic goals ment assistance for the System. OED believes that some limited provision of IDA-type funds could not be re- genuine global public goods requiring long-term on a larger scale. placed by alterna- investments will have to be financed through 1 4 0 T H E W O R L D B A N K A N D T H E C G I A R the provision of IDA-type funds on a larger scale combined with The Bank vice president who than the Bank's net income can allow. greater executive chairs the CGIAR plays two functions, it might conflicting roles. The Bank's Multiple Roles well be close to a OED observed in its 1998 review of the World full-time job, and it might require someone familiar Bank grant programs that whenever the Bank is with managing science. Moving in this direction-- called upon to handle a combination of leader- and implicitly removing chairmanship from the ship, fund-raising, management, and fiduciary Bank--was rejected by the CGIAR members.15 responsibilities within such a close-knit collab- The Bank's cosponsor representative on the orative arrangement as the CGIAR, there is po- CGIAR (usually the director of Agriculture and tential for lack of objectivity and conflicts of Rural Development), who officially represents the interest that compromise the Bank's conduct Bank in the CGIAR and ExCo, reports to the same of these roles. The present meta-evaluation con- vice president who chairs the CGIAR. He has gen- curs that conflicts of interest in the Bank's roles erally represented the ESSD (Environmentally and and insufficient oversight have compromised Socially Sustainable Development) vice presidency the Bank's ability to provide the necessary strate- on the DGF Council, which is responsible for rec- gic leadership (box 6.1). ommending the annual DGF allocations to senior The Bank vice president who chairs the Bank management and the Bank's Executive CGIAR plays two conflicting roles: representing Board.16 Historically, the Agriculture and Rural the CGIAR to the Bank in advocating DGF fund- Development director (or the Bank's director for ing for the CGIAR, and representing the Bank to agricultural research) also chaired the CGIAR Fi- the CGIAR in ensuring the relevant, efficacious, nance Committee until its dissolution in 2000, and efficient use of the Bank's grant resources and that individual now chairs the new Finance allocated to the CGAIR.12 It is problematic for the Committee of ExCo. The likely adverse incentives chairman to be both advocate and judge--to of the changes in the Bank's funding formula in maintain the political support of the key inter- 1994 were not identified by the Bank's cosponsor est groups and to make the case for continued representative to the CGIAR at the time. How- and even enhanced funding to the Bank and ever, the rural development director who replaced other donors, while also acknowledging the him, and who previously served as the chairman need for and pressing for major reforms. In ad- of TAC, had reservations about the funding formula dition, having a chairman who is not at arm's and acknowledged to the meta-evaluation team the length from the Bank compromises the Bank's conflict in roles of serving as the ESSD represen- ability to press for reforms on the scale or at the tative on the DGF Council.17 speed that might be warranted. Generally speak- The Bank's rural research advisor, who in ing, because of the pressure from other members principle exercises the Bank's oversight of the to deliver the Bank's annual contributions, the CGIAR, reports to the Agriculture and Rural De- experience to date indicates, and the former velopment director (the Bank's cosponsor rep- chairmen acknowledged, that the chairman ends resentative) and the ESSD vice president (who up representing the CGIAR to the Bank more chairs the CGIAR). The role of advisor in the than representing the Bank's concerns to the oversight of the CGIAR has been dysfunctional CGIAR.13 Most donors want the Bank to con- for a number of years, and the present advisor tinue to chair the CGIAR, in part because they has acknowledged the ambiguity of his over- believe Bank funding and other donor funding sight role in a situation where two directors and will decrease if the chairmanship is moved out- a vice president manage the System.18 side the Bank.14 The executive director of the CGIAR Secre- To his credit, during the CDMP, the current tariat, recently elevated to the same rank as the chairman explicitly discussed the merit of a Bank Agriculture and Rural Development director, re- vice president chairing the System. He argued ports to the ESSD vice president who chairs the that the chairman's future role might need to be CGIAR. He is essentially treated as a line manager 1 4 1 T H E C G I A R AT 3 1 within ESSD. He attends all meetings of the convening power to raise additional resources for ESSD management group, including sector board the System than to provide the necessary strate- and senior staff meetings, and his performance gic leadership to help the CGIAR respond to is assessed by the ESSD vice president and the the rapidly changing external and internal envi- managing director responsible for ESSD, illus- ronment. trating the lack of arm's length in the day-to-day Neither the Bank nor other members have management of the Secretariat from the net- been exercising conflict-free, unencumbered work and management that is supposed to over- oversight of the CGIAR as a System--not the see the System. Bank because of the above-mentioned conflicts Thus, as is the case for a number of other of interest, and not the other members because global programs with in-house secretariats, there they indicated to the meta-evaluation team that is currently no effective independent oversight "the Bank is doing so."19 Bank management has within the Bank outside the network of its in- also presumed, as it generally does with other volvement in the CGIAR, nor a mechanism to global programs that are partnerships, that the conduct, to assess, and or to follow-up on Sys- member-partners are exercising adequate over- tem-level evaluations. Given a relative lack of sight over the operations of individual programs. DGF guidelines as to how CGIAR funds are to be On the contrary, other cosponsors and donors allocated, the ESSD vice president has wide lat- have been induced to act as "free riders," abro- itude to determine how the contribution will gating their legitimate responsibilities. Some com- be spent, with no checks and balances. It is un- promises made in delivering the Bank funds while derstandably difficult for the rural research ad- managing the organization for results have been visor to exert oversight over his own manager. to the detriment of the long-term interests of the Beyond sponsoring the CGIAR's annual appli- CGIAR and its developing-country members. cation to the DGF, the Rural Sector Board has In short, both the CGIAR and the Bank face a played little role in monitoring the use of the problem of corporate governance with regard funds from a strategic perspective, and is unlikely to global programs. Having grown from an in- to have the will to displease its chairman. formally structured System, unlike most of the While the Bank's unique contribution to the newer global programs housed in the Bank, the CGIAR has been widely acknowledged, the mul- CGIAR has neither a written charter that clearly tiplicity of roles it has assumed has led to (1) ex- delineates the roles, responsibilities, and ac- cessive Bank involvement in the day-to-day countabilities of the various officers and bodies management of the System and dependence of that make up the governance of the System, nor the System on the Bank, (2) a disproportionate an agreed-upon mechanism to reform the gov- share of management responsibility allocated ernance of the System as needed.20 The distinc- to a senior manager already burdened by other tion between oversight and management has heavy managerial responsibilities, and (3) re- become blurred. Collective responsibility rather porting arrangements both for the CGIAR Sec- than checks and balances have characterized retariat and the Bank that are fraught with real CGIAR management. While this has facilitated or perceived conflicts of interest. These features the forging of consensus, it has not been con- limit the capacity of the Bank to provide the ob- ducive to strong leadership. These weaknesses jectivity and leadership needed to position the may have been less of a problem in less complex CGIAR System to address the external challenges times, when the mission-oriented genetic re- that it now faces, and to allocate the Bank's fi- search agenda of four to six individual Centers did nancial contributions not require close relationships across the whole Both the CGIAR and the to the CGIAR in a System, pooling of resources to tackle major Bank face a problem of strategic way. It is not strategic research challenges, or a capacity to surprising that the forge partnerships with private sector firms. corporate governance with Bank has been more The current situation raises a corporate gov- regard to global programs. successful in using its ernance responsibility issue in the management 1 4 2 T H E W O R L D B A N K A N D T H E C G I A R of global programs also raised in OED's Phase 1 programmatic linkages with the CGIAR in various Report on Global Programs. Who in the Bank forms: linkages with the CGIAR through Bank can provide the necessary intellectual and man- country assistance loans; interaction among Bank agerial oversight to a large, intellectually de- and CGIAR/Center staff (for example, through manding, and hierarchically challenged global Bank staff serving on review panels or Center program? In its Phase 1 Report on the World boards, or Center staff providing input to Bank Bank's Approach to Global Programs (World activities such as economic and sector work or Bank 2002a), OED recommended designating country assistance strategies); and institutional- responsibility to an appropriately staffed and level discourse on issues of common interest. mandated secretariat linked to a managing di- Indeed, many of these currently take place. For rector or existing vice presidential unit. The sec- example, about 20 Bank staff are former em- retariat would have two interacting units: (1) a ployees of Centers and several Bank staff serve think tank to monitor and anticipate changes on CGIAR Center boards, and in the conduct of and emerging opportunities in the global envi- External Program and Management Reviews ronment for global programs, draw partnership (EPMRs) and other thematic reviews. implications for the Bank, and provide intellec- But operationally, at the level of Bank policy tual leadership, and (2) a separate operational unit dialogue and lending to agriculture or agricultural to provide oversight of internal management of research, there has been little impact. The Agri- global programs and ensure that the Bank has a culture and Rural Development Department and coherent strategy across networks for global pro- the CGIAR are currently putting together an in- grams such as the CGIAR. The challenges inher- ventory of the programmatic linkages between ent in the Bank's relationship with the CGIAR the Bank's lending to agricultural research and reinforce the OED conclusion contained in its the CGIAR. The meta-evaluation team's inter- Phase 1 report on the Bank's involvement in views of Bank staff confirm the conclusions of the global programs: that the level of strategic co- previous OED review of the CGIAR that "the herence required in this area could only be Bank, for all its virtues with respect to the CGIAR, achieved if responsibility for it were assigned to has been less than fully effective in building links a single senior Bank manager to coordinate ac- between its own programs in agriculture and tivities on a Bankwide basis and to ensure selec- natural resources, particularly loans for agricul- tivity and quality. Without the intellectual capital, tural research, and the CGIAR Centers" (Ander- the oversight function will be a bureaucratic hur- son and Dalrymple 1999). After reaching a peak dle (a concern some have expressed in discussing in the 1986­88 period, when it represented more the recommendations of the Phase 1 Global Re- than 25 percent of the Bank's new commitments, view). Without the hierarchical clout, it will be dis- Bank agricultural lending has declined precipi- missed. Hence the oversight of the CGIAR, as tously (figure 6.1). indeed of other global programs supported by the New commitments in agricultural research Bank, should be considered in the context of and extension peaked at $665.7 million in 1992.22 the larger Bank management response to the Commitments to agricultural research and ex- strategic and programmatic management of the tension have approached this amount only once Bank's global portfolio.21 since--$420.1 million in 1998--with the ap- proval of a large agricultural research project to Enhancing the Bank's Role as a India (for $196.8 million). Sub-Saharan Africa Development Partner and Latin America and the Caribbean have been Linkages between the CGIAR and the Bank's agri- significant parts of this lending, accounting for cultural development strategies were strong when 24.9 percent and 24.0 percent, respectively, of the CGIAR was established, especially in Asia. Al- new commitments in agricultural research and though the CGIAR Secretariat is located in the extension since 1971. South Asia and East Asia Bank, the Third System Review recommended and the Pacific have accounted for 19.7 percent that the cosponsoring agencies strengthen their and 18.4 percent respectively (figure 17.1). 1 4 3 T H E C G I A R AT 3 1 W o r l d B a n k L e n d i n g t o A g r i c u l t u r a l F i g u r e 1 7 . 1 R e s e a r c h a n d E x t e n s i o n , 1 9 7 1 ­ 0 2 Millions of 2000 U.S. dollars 1000 900 800 700 600 500 400 300 200 100 0 Sub-Saharan East Asia & Europe & Latin America Middle East & South Asia Africa the Pacific Central Asia & the Caribbean North Africa Volume of new commitments in each time period 1971­78 1979­86 1987­94 1995­02 With respect to quality, the OED outcome rat- from the policies pursued by developing coun- ings on 57 projects completed by 2001 have been tries themselves. The Bank has cut back lending below the Bankwide average; only two-thirds (38 to agricultural research partly because most out of 57) were rated satisfactory. Only 18 out of countries have not been committed to the re- 57 were rated as likely to be sustainable, 19 as un- forms needed to make effective use of the funds: certain, 15 as unlikely, and 5 not rated. The lack putting more resources into public research or- of a sufficient treatment of agricultural develop- ganizations in many countries is just postponing ment in the country assistance strategies and their demise. Today, more attention is being the decline in Bank lending to agriculture led to given to sector work and policy dialogue to open an intense and healthy debate in the context of up options for support to research.23 The Bank's the preparation of the new rural development Rural Strategy identifies agricultural research as strategy between management and the Bank's ex- a high priority. However, OED agrees with Agri- ecutive directors. While eschewing rigid lending culture and Rural Deveopment research staff targets, it was agreed that it was appropriate to that in a demand-led Bank, the task of convinc- analyze lending patterns and trends as indica- ing policymakers in developing countries of the tors of the Bank's role, impact, and responsive- importance of agricultural research and agri- ness to country demand under the new strategy. cultural development must come from the re- A businesslike monitoring system is being set up gional vice presidents and country directors, as to ensure effective oversight of the strategy and was the case in India in the 1960s and 1970s. Lack of a sufficient regular reporting to How should the Bank develop a vision, strat- the Board about egy, management, and staff incentives for an im- treatment of agricultural progress. proved approach to agricultural development development and the decline The Bank's role, and greater linkages between it and the CGIAR? in Bank lending to impact, and re- And what does experience tell us about what is sponsiveness to needed to convert the vision into action? agriculture led to an intense country demands A better approach than the alternatives of and healthy debate. cannot be isolated lending targets or monitoring of effort as the new 1 4 4 T H E W O R L D B A N K A N D T H E C G I A R rural development strategy proposes is the one for agricultural re- But country directors and pursued in India in the 1970s between the search. Working with regional vice presidents are CGIAR, the Bank, and the other major donors, Bank staff, they per- not convinced of the central at that time chiefly the United States. This ap- suaded the Bank's importance of agriculture proach recognized the distinction between coun- country operations try demand for Bank borrowing and county to become engaged. and rural development. needs, since the two do not always coincide. Given the dramati- Under McNamara's leadership, the Bank, work- cally changing global environment, they seemed ing jointly with donors, particularly USAID and to be aware of the need for reforms in their the CGIAR, played a major role in bridging this own research systems.24 Regretably, few devel- gap between wants and needs and lending sub- oping countries have such a long-term and stantial resources for investment in irrigation strategic view of development. research, seed production and distribution, agri- The Bank's regional departments need--but cultural credit, and price support programs. Sim- do not have--the commitment, the budgets, or ilar efforts did not work in Africa (Lele and the qualified and experienced staff in the agri- Goldsmith 1989; Lele 1994). cultural sector to develop and make a convinc- Poor-performing loans in the agricultural and ing case to the governments of the importance rural sector have discouraged the Bank from in- of investment in agricultural research and de- vesting in this sector. The concurrent shift to a velopment. The situation is no different in demand-driven Bank and the role of the gov- ESSD--the Bank lost 60 of its 140 technical staff ernments known for their urban bias, combined in the agricultural sector. The number of Bank with the controversies surrounding investments staff with experience in agricultural research and in dams, the safeguard issues related to the treat- technology issues, including ex-CGIAR staff, is de- ment of minorities, and, not least important, clining very rapidly through retirements. Most Re- the wholesale questioning of the roles of the gions lack capacity to lead dialogue on public sector--even when markets for inputs agricultural research and related policy and agri- are altogether absent--has not helped. For ex- cultural development issues. The situation is es- ample, fertilizer consumption in Nigeria has pecially serious in Africa.25 Although the Bank is dropped from 450,000 tons in 1993­94 to 100,000 hiring local staff, it lacks the international staff tons in 1999­00 (see Chapter 14). with the requisite global experience in agricul- Poverty reduction strategies (PRSPs) need to tural research and development. The situation recognize the fundamental importance of agri- is expected to get worse with more retirements. culture and rural development in countries The new Rural Strategy and the Rural Sector where most of the employment, exports, GDP, Board's staffing plans do not sufficiently address and food comes from agriculture. But country this issue. Without qualified, internationally re- directors and regional vice presidents are not spected staff with the necessary technical knowl- convinced of the central importance of agricul- edge and operational experience, the Bank ture and rural development. Hence, more visi- cannot expect to provide leadership in agricul- ble, quick-yielding, and politically popular tural research and development. investments in education, community-driven In short, the Bank needs to re-equip itself to development, and the like receive priority. Ismail provide the needed leadership to the CGIAR Serageldin's public announcements in 1994, that it has historically provided. Given the rap- committing the Bank to lend $500 million an- idly changing external and internal environment nually to agricultural research for the following for the CGIAR and its global importance in con- five years (Serageldin 1996b, p. 8), worked only tributing to sustainable development, the Bank in Brazil, India, and later in China, mainly needs to deal effectively with the challenges fac- because these countries were convinced of ing the CGIAR in its governance, financing, and the importance of borrowing from the Bank management. 1 4 5 18 Conclusions Science and Strategy T he CGIAR's productivity-enhancing research has had sizable im- pacts on reducing poverty. Its research in germplasm improvement and related areas of natural resource management (such as integrated pest management) has made important contributions to agricultural pro- ductivity and poverty reduction. This type of research is a global or regional public good whose production is ideally suited to a publicly funded global or regional network. The research benefits the poor directly through inherently different in character from germplasm increased production of subsistence foods, em- improvement research. To be relevant, effective, ployment, and income generation, and indirectly fine-tuned, and have impact, policy research on through reduced prices for food-deficit house- constraints to technology adoption in develop- holds. The research also generates positive en- ing countries should be conducted close to the vironmental impacts such as large savings in developing countries whose policy environments land used, increased production possibilities on are the subject of the research. Most NRM re- resource-poor lands, and improved strategies search must also be conducted on site. The for the use of water, soils, and pesticides. CGIAR's policy research will yield higher returns if it addresses commonly observed policy failures But the CGIAR is less focused than it used to be; its cur- in developing countries, while helping to build rent mix of activities reflects neither its comparative the capacity of developing country NARS to con- advantage nor its core competence. Its research ex- duct their own country-specific policy research. penditures on increasing productivity have de- The comparative advantage of a global research clined by 6.5 percent annually in real terms since system in NRM research lies in bringing to bear 1992, while those for policy research and on advanced multi-disciplinary methods and protecting the environment (largely related to processes that developing countries do not pos- NRM) have increased by 3.1 percent over the sess, and in conducting research with the po- same period. Both policy and NRM research are tential for wide applicability. 1 4 7 T H E C G I A R AT 3 1 A complex combination of factors explains the chang- model. Under the former, the Bank's contribution ing research mix. First, germplasm improvement was used to fill gaps between the System's research and associated biological research has been un- priorities as articulated by TAC and the financial con- popular in the constituencies of some key donors tributions to those priorities by other donors. Under because of negative perceptions of the Green the matching grant model, the Bank's contribu- Revolution--that it made the rich richer and the tion indiscriminately matches funding from other poor poorer and caused environmental damage. donors, whether in support of System-wide prior- Second, the CGIAR has correctly responded to the ities or not. Second, to create incentives for Cen- genuine second-generation environmental pres- ters to mobilize additional funding and to sures on soils and water created by the radical accommodate donors, the CGIAR expanded the def- change in farming systems during the Green Rev- inition of its "agreed research agenda" to include olution, an area where research continues to be both the former "core" agenda (largely the high- needed. Third, the rise of environmentalism, the return global and regional public goods research) 1992 United Nations Conference on Environment and the "noncore" agenda (donor-funded, mostly and Development (UNCED) in Rio de Janeiro, downstream activities that TAC did not consider as and growing environmental advocacy in donor high a priority). Third, donors have collectively in- countries led to growing demands on the CGIAR creased their degree of restricted funding from 36 to respond to environmental concerns. percent of total funding in 1992 to 57 percent in Fourth, the failure of governments of devel- 2001. Most of the increase in restricted funding oping countries and their donor supporters to has occurred since 1998. make the necessary investments in developing countries' own research, education, and devel- CGIAR funding has not kept up with the System's opment systems limited their ability to adapt broader mission. Overall CGIAR funding has stag- CGIAR technologies to their own farming sys- nated in nominal terms, declined in real terms, tems. This led CGIAR donors to turn to the Cen- and become increasingly restricted over the past ters to fill the national and local public goods decade. Overall contributions grew at an average gaps. Fifth, the biotechnology revolution, the annual rate of 0.7 percent in nominal terms and emergence of intellectual property rights issues, declined by 1.8 percent per year in real terms be- and the associated rapid increase in private sec- tween 1992 and 2001. It appears that the in- tor investments have challenged some CGIAR creased expenditures on policy and NRM donor constituencies. Thus, to avoid contro- research have come at the cost of germplasm versy among Centers and within its member- conservation and improvement. However, it is ship, which makes decisions by consensus, the also possible that, because of the principles of CGIAR leadership has not explored the full im- donor sovereignty and Center autonomy under plications of the biotechnology revolution for the which the CGIAR operates, germplasm research System--a response reinforced by the auton- would not have received the funding provided omy of the 16 Centers. to policy and NRM research. Three changes in the funding processes of the CGIAR since the mid-1990s have also increased the Strong NARS are critical to ensure the CGIAR's impacts. influence of individual donors (and their domes- The CGIAR's impacts result from the joint out- tic constituencies) on the research expenditures puts of CGIAR Centers and NARS in developing of the CGIAR. First, in response to a funding cri- countries. The NARS generate location-specific sis in 1993­94, the Bank changed the allocation technologies and are essential for testing, adapt- of its own finan- ing, and disseminating the products of CGIAR re- Changes in the funding cial contribution search. Yet the rate of growth of investments, processes have also increased from a "donor which would enable NARS to undertake such of last resort" research and which was quite rapid until the the influence of individual model to a mid-1980s, has slowed for well over 15 years, and donors. matching grant has even become negative in Africa. 1 4 8 C O N C L U S I O N S The meta-evaluation has identified three im- African agricultural The CGIAR has not kept pace portant issues with regard to the NARS. First, in productivity chal- with the changing and some regions of the world, most notably Africa, lenge is to be ad- highly divergent needs of underinvestment in NARS has adversely affected dressed. Many argue NARS. their strength and capacity. This weakness has convincingly that been compounded by economywide and sector- Africa's excessive donor dependence has reduced level policy and institutional failures that inhibit the incentive for countries to invest in their own the development of rural infrastructure, input de- institutions, create domestic constituencies for re- livery systems, and output markets. Therefore, search, and ensure long-term stability and na- technology adoption has not been accompanied tional priorities. They believe that recent proposals by effective delivery of services and increases in (by Canada, the EU, the World Bank, and the productivity and incomes. This failure has led the Challenge Program on Sub-Saharan Africa) to in- CGIAR to conduct, adapt, and disseminate re- crease donor support to the region may waste re- search, extension, and information at the na- sources without providing high-priority long-term tional level. research of regional significance and without cre- Second, the CGIAR has not kept pace with ating long-term domestic capacity and a political the changing and highly divergent needs of NARS. or professional constituency for research. The Large and small NARS alike acknowledge that overall approach to improving agricultural pro- the CGIAR has made major contributions to their ductivity and reducing poverty in Africa, includ- growth. But the capacity of NARS in large and mid- ing the role of the CGIAR, requires fundamental dle-income developing countries now substan- collective rethinking. tially exceeds that of the CGIAR. NARS argue that the CGIAR has not kept pace with their increas- The CGIAR faces a number of new challenges and ingly complex needs and does not consider them opportunities. These include (1) the growth of equal partners. They lament the decline in re- modern biotechnology and bioinformatics, (2) search collaboration and the limited efforts of the effective genetic resource conservation and man- CGIAR to draw on their expertise and experi- agement of the CGIAR's 600,000 accessions of ge- ence to build capacity in smaller, less-advanced netic material, (3) the growing importance of NARS. Smaller and weaker NARS are concerned intellectual property rights in agricultural re- that the CGIAR's training and collaborative re- search, and (4) the growth of private agricul- search have not kept up with their emerging tural research. Responding to these challenges needs. Both the large and small NARS consulted requires a System-level approach (a) to develop by the meta-evaluation expressed interest in the a System-level policy and strategy on intellectual large NARS working with the smaller NARS, since property and public-private partnerships so the their agroecologies and development conditions CGIAR can speak with one voice and become a are often similar. Capacity building efforts of the powerful force in international negotiations, and NARS under the CGIAR umbrella could be sub- (b) to forge active partnerships with the private stantially augmented, based on clear business-like sector, universities in advanced countries, and the agreements and international financial support national systems of developing countries. for such south-south cooperation. Third, the CGIAR faces structural issues in The CGIAR is responding, but slowly. To maintain Africa. Eighty percent of Africa's agricultural re- its genetic material, the CGIAR has worked with searchers are concentrated in 13 large countries. the SDC and FAO to establish the Global Con- The other 30 plus small countries face disec- servation Trust and helped launch a fund-raising onomies of scale in organizing and managing effort. The prospects for raising the conserva- their own agricultural research and lack the ca- tively estimated $260 million endowment--the pacity to negotiate with the 16 CGIAR Centers. interest would support the gene banks--are un- This highlights the importance of relying on sub- clear at this stage. By November 2002, commit- regional research organizations in Africa if the ments of $60 million had been obtained. These 1 4 9 T H E C G I A R AT 3 1 steps are appropriate, but are unlikely to be suf- Governance, Organization, Finance, ficient. The mandate of the System-wide Ge- and Management netic Resources Program (SGRP) applies only to the Centers fulfilling their obligations under The CGIAR's founding principles are unsuited to en- the FAO agreement, and does not extend to the suring poverty impacts in a changed environment. program on the conservation, management, and The six founding principles that underlie the enhancement of genetic resources, which is the CGIAR--donor sovereignty, Center autonomy, responsibility of individual Centers. consensus decisionmaking, independent tech- nical advice, informal status of the System, and The CGIAR lacks a System-level strategy for public- nonpolitical nature--were adopted when the private partnerships. It needs to mobilize the best System consisted of fewer Centers and less di- practical System-level expertise in global pub- verse constituents, and setting priorities to lic policy, law, and ethics, and to develop, mon- achieve poverty impact (through its governance, itor, and report regularly on effective management, and financing processes) was rel- partnership arrangements to the System as a atively simple. But in today's more politically whole, while actively developing strategies and driven authorizing environment, and with a wider policies that advance its poverty alleviation mis- research agenda and expanding membership, sion. It needs to document and learn concrete the CGIAR's ability to address its mission is now lessons from its own experience in public-pri- undermined by these six founding principles vate partnerships and management of IPRs at that exacerbate the System's collective action the System level. Regaining focus requires bold problem. In particular, unlike more recent global reforms not only in the CGIAR System, but also programs, such as the Global Environment Fa- in shaping the environment in which the CGIAR cility or the Global Fund for AIDS, TB, and operates. With a System-level policy framework Malaria, the CGIAR System has no formal or legal that provides authoritative, transparent, and persona, or written charter, or even a memo- accountable System-level responses to the randum of understanding. The only legal entities changing technological and institutional envi- in the CGIAR are the 16 Centers. Because a con- ronment, the CGIAR would be a powerful force siderable superstructure of committees has in global negotiations backing the interests of evolved over the years to deal with a complex set developing countries. In its response to the of stakeholders, responsibilities and account- 2001 International Treaty on Plant Genetic Re- abilities are not well defined. Greater responsi- sources for Food and Agriculture (ITPGR), the bility for managing the overall System has CGIAR showed that it can a develop an appro- accrued by default over time to the World Bank priate System-level response. But Center au- and the CGIAR Secretariat. tonomy and System-level governance weaknesses have prevented the CGIAR from Collective action problems are daunting. The mid- addressing other challenges. 1990s Renewal demonstrates how changing the funding rules can have unintended consequences. The System is being pulled in two opposite directions. The corporate model recommended by the Third On one hand, the CGIAR Centers are not con- System Review (TSR) to formalize decisionmak- ducting sufficiently coordinated research on the ing, transparency, and accountability was rejected highly decentralized nature of NRM research, for being top-down and contrary to the CGIAR's which calls for effective partnerships with NARS founding principles. A subsequent Federation to produce regional and national public goods proposal by the CGIAR board chairs and Center in NRM. On the other hand, the System is not suf- directors was opposed because it was bottom-up ficiently centralized to deal with advances in the and would have increased Center control at the biological sciences and IPRs, which call for a expense of donors and the CGIAR Secretariat. more unified approach to research strategies But, while the last two proposals differed in the and policies. degree of decentralization, both acknowledged 1 5 0 C O N C L U S I O N S the need for a legal entity with a centralized board groups or obliged to solicit the views of their to enable System-level responses. "constituent groups" before decisions are made. In addition, certain factors limit the effectiveness Consolidation is needed but resisted. Since 1994, and extent of developing-country membership. stakeholders, scientific committees of TAC, and OED concurs with the proposal that the TSR the CGIAR's own internal reviews have proposed had made for an executive committee with for- many types of consolidation. A large majority mally elected members that are accountable to (79 percent) of CGIAR stakeholders surveyed the particular groups they represent. by OED agree that consolidating the number and functions of Centers is advisable. African The System Office has promise, although it is a work countries as a group have called for consolida- in progress. The System Office links 10 inde- tion to reduce the transaction costs of dealing pendent units in order to increase their coordi- with 16 Centers. And the CGIAR has some suc- nation and ability to serve the Centers and cessful examples of consolidation, which in the membership. Issues of authority, responsibility, livestock sector have positioned the CGIAR to ad- and accountability between the CGIAR Secre- dress global issues. Yet individual countries that tariat (which is the coordinating unit), the Cen- host Centers oppose consolidation, as do donors, ters, and other units are still being sorted out. Center chairs and boards, scientists, and Center Monitoring is required of the System Office's directors for "their" Centers. Getting key stake- functions, resources, responsibilities, account- holders to agree that the time has come for both abilities, and performance through routine eval- real reform and organizational change remains uations to ensure its effectiveness in serving the a major challenge. members and clients. Having lost the momentum The current CGIAR Chairman, Ian Johnson, on divesting certain Center responsibilities to the initiated a Change Design and Management System level, as proposed in the Center directors' Process in 2000 to effect key reforms. As a result Federation proposal, there is a concern that the of widespread stakeholder consultations, some System Office may be adding to the duplication members and observers feel that there has been of services already performed at the Center level. greater ownership of both the process and the outcomes of the current reform process com- The transformation of the Technical Advisory Com- pared with the TSR in 1998. Given the historical mittee (TAC) into a Science Council (SC) raises many difficulty of achieving change in the System, the questions. OED is not convinced that the trans- reforms are significant, but they do not go far formation of TAC into a Science Council will enough. achieve the desired objective of ensuring "that the science practiced in the System meets world The Chairman gets high marks for establishing the class standards" for a number of reasons. First, long-overdue Executive Council (ExCo). ExCo mem- this transformation is taking place without the bers and others interviewed by the meta-evalu- TAC having been evaluated while major deci- ation team at the Annual General Meetings in sions on resource allocations are being made. 2001 and 2002 are optimistic that this is a step Second, relative to TAC, the Science Council's in the right direction, but they are reserving role is greatly diminished. It is expected to focus judgment until they see how ExCo operates. mainly on science quality; to have a more limited The fundamental challenge is to appropriately (if any) role in priority setting, medium-term balance legitimacy and efficiency. That ExCo is a planning, and monitoring resource allocation; stakeholder committee rather than a shareholder and to play no role in the annual financial plan- committee enhances its legitimacy. But while ning process. Third, FAO has indicated to the ExCo members are selected through caucuses of meta-evaluation team that it is not being ade- stakeholder groups and can only make decisions quately consulted about the role of the Science on matters delegated to them by the member- Council. Fourth, the roles and responsibilities of ship, they are not formally accountable to those the Science Council in relation to the Challenge 1 5 1 T H E C G I A R AT 3 1 Programs and the new and emerging Finance and substantial resources to advanced research in- Program Committees are unclear, particularly stitutions and the NARS of developing countries. as the permanent Science Council is not in place Yet only one-third of stakeholders surveyed by while major decisions on Challenge Programs are OED believe the Challenge Programs are sufficient being made. Fifth, Science Council members to open up the CGIAR, to produce the best sci- need to be independent and objective, and their ence, or to achieve consolidation. OED's inter- findings need to be shared with the full mem- views with CGIAR members reveal much bership when they are transmitted to ExCo. Yet ambiguity and confusion about the Challenge the Working Group on the Science Council has Programs and a range of concerns about their de- not determined the time and financial resources velopment, including (1) proposed levels of fund- needed, or the remuneration arrangements for ing that are far larger than originally understood, the Science Council members and its chair. (2) a lack of strategic priorities, (3) inadequate procedures for ensuring science quality, (4) the The Challenge Programs are not addressing issues ad hoc nature of the governance and manage- of System-level funding, priority setting, science qual- ment arrangements, and (5) a lack of clarity as to ity, and governance. The Challenge Programs are how the Challenge Programs will address the intended to be high-visibility, time-bound part- key global public policy issues associated with nerships that will address complex issues of public-private partnerships and intellectual prop- global or regional significance, increase the scope erty rights identified in this report. for inter-Center collaboration, facilitate a wider range of partnerships, tap new sources of fund- The Role of the World Bank ing from current and new donors, and improve output accountability. They have also been The World Bank has played multiple roles in the viewed as a way of addressing consolidation in- CGIAR--as convener (founder and cosponsor), directly by building new programs and partner- as donor to the System, and as a lender to de- ships while, in the words of some interviewees, veloping countries for complementary activi- letting the weak Centers and programs "die on ties. As a result of its multiple roles, the Bank has the vine" rather than face politically unpopular been described to OED as the indispensable consolidation. Their design was also to address guardian of the CGIAR, and as the "glue" that three important shortcomings that have plagued makes the System coherent and larger than the previous thematic and ecoregional System-wide sum of 16 research Centers. The Bank's leader- programs: insufficient funding, poorly defined ship role, its financial contributions, and its op- timelines, and a lack of adequate governance erational support are viewed by other donors as and management. In October 2002, after a year- a seal of approval, giving them the confidence to long review process involving both the interim continue to invest in the System. Science Council and ExCo, the Annual General But, while the Bank's unique contribution to Meeting approved the implementation of the the CGIAR has been widely acknowledged, the first two pilot Challenge Programs: "Water and multiplicity of roles it has assumed has led to (1) Food" and "Biofortified Crops for Improved an excessive dependence of the System on the Human Nutrition." While passing scientific re- Bank, (2) a disproportionate share of manage- view, these two pilot programs do not address ment responsibility allocated to a senior manager the three shortcomings of thematic and ecore- already burdened by other heavy managerial re- gional programs (funding, timelines, and gov- sponsibilities, and (3) reporting arrangements ernance) noted above. that are fraught with real or perceived conflicts OED believes that, properly developed, the of interest. These features limit the capacity of the Challenge Programs could enhance the CGIAR's Bank to provide the objectivity and leadership effectiveness and impact. One of their strong necessary for far-reaching reforms of the CGIAR, and positive features is that they are helping to and to allocate the Bank's financial contribution open up the System by promising to allocate in a strategic way. As a result, the Bank has been 1 5 2 C O N C L U S I O N S more successful in using its convening power to mark and demon- The Bank has been more raise additional resources for the System than to strated the genius successful in using its provide strategic leadership to the CGIAR. of its framers. It convening power to raise has succeeded additional resources for the Conflicts of interest and inadequate oversight con- mostly because of strain the Bank's strategic leadership. Many donors (a) its emphasis System than to provide have indicated that they want the Bank to con- on strategic re- strategic leadership to the tinue to chair the CGIAR, in part because they be- search of a global CGIAR. lieve Bank and other donor funding will decrease or regional public if the chairmanship is moved outside the Bank. goods nature, the benefits of which spill across However, it is problematic for the chairman to be national boundaries and cannot easily be ob- both judge and advocate, to acknowledge the tained through private, national, or regional re- need for and to press for major reforms while also search; and (b) its practical, problem-solving making the case for continued funding to the focus on bringing the best of known science to Bank and donors. Conflicting political pressures address the problem of food security. and the need to maintain political support for the But the governance, management, and fi- System can come at the cost of the reforms in the nancing of the CGIAR have become increasingly System. Having a chairman who is not at arm's cumbersome in recent years, and fundamental length from the Bank compromises the Bank's changes are called for. The improved skills avail- ability to press for reforms on the scale or at a able to developing countries and the opportuni- speed that might be warranted, and can increase ties opened up by the new information its exposure and risks. technologies have yet to be fully tapped. The There is currently no effective independent CGIAR has operated in relative isolation without oversight within the Bank of its involvement in strong links to the private sector and developing- the CGIAR (as is also the case for a number of country institutions. It has failed to keep up with other global programs), nor a mechanism to the rapidly evolving technological and institu- conduct, assess, or follow up on System-level tional environment. Its original science-based evaluations. The ESSD vice president has, among character has become more diffuse, and the role other things, wide latitude to determine how of the Bank has become increasingly ambiguous. the Bank's contribution will be spent, given a lack It is time to make the CGIAR System leaner, more of guidance from managing directors or the DGF efficient, and more attuned to today's realities. The on how the DGF grant should be allocated. It is CDMP is a good start, but the Bank should sup- understandably difficult for the rural research ad- port an accelerated pace of reforms while re- visor to oversee his own manager. Beyond spon- examining its own roles and the conflicts between soring the CGIAR's annual application to the them. It is also time for donors (including the DGF, the Rural Sector Board has also played lit- Bank) and developing countries to reverse the de- tle role in monitoring the use of the funds from cline in their complementary investments in agri- a strategic perspective, and is unlikely to have the cultural research and extension at the national will to displease its chairman. Further, while the level, without which reforms in the CGIAR alone Bank used to conduct periodic financial audits will not achieve results on the ground. of the CGIAR: the insightful 1995 audit was the The Bank should fully exercise its leadership last of its kind. role, if necessary by convening the best external expertise to bring In the research and Further reforms are needed. The CGIAR is a unique about additional instrument of international cooperation with a reforms in the development continuum the solid record of achievements. In the research System. If addi- CGIAR has made its mark and and development continuum ranging from basic, tional reforms demonstrated the genius of its strategic, applied, and adaptive research to are achieved, technology transfer, the CGIAR has made its there is a strong framers. 1 5 3 T H E C G I A R AT 3 1 But the governance, argument for in- rapidly growing but underevaluated areas of pol- management, and financing creased funding for icy and NRM research and the CGIAR's capacity of the CGIAR have become the CGIAR, includ- building activities would benefit from an inde- ing exploring the use pendent impact evaluation. Yet it would be more increasingly cumbersome. of grants for the pro- appropriate to conduct such an evaluation after vision of regional public goods, and eventually the more fundamental, longer-term issues re- global public goods, that reduce poverty. lated to System-level strategy, governance, or- OED also concludes, based on the evidence ganization, management, and finance identified the meta-evaluation team has assembled, that the in this report are fully vetted. 1 5 4 APPENDIXES APPENDIX 1: CURRENT ORGANIZATIONAL CHART OF THE CGIAR Subsequent to the organizational changes adopted at the Mid-Term Meeting in May 2001. Source: the CGIAR Secretariat. See also the notes on the following page. 1 5 7 T H E C G I A R AT 3 1 Notes The Executive Council comprises: Chairman Ian Johnson Cosponsors Officer-in-Charge/Sustainable Development Department FAO Kevin Cleaver World Bank Rodney Cooke IFAD CDC Adel El-Beltagy CDC Chair CBC John Vercoe CBC Chair TAC/iSC Emil Javier TAC/iSC Chair GFAR Mohamed Roozitalab GFAR Chair OECD/DAC Americas Jonathan Conly U.S.A. Asia-Pacific Toshinori Mitsunaga Japan Europe Hans-Jochen de Haas Germany Klaas Tamminga The Netherlands Klaus Winkel Denmark Developing Countries Americas Luis Arango Nieto Colombia Sub-Saharan Africa Bongiwe Njobe South Africa Asia-Pacific Dongyu Qu China CWANA Noureddin Mona Syria Regional Fora Abed Al-Nabi Fardous AARINENA Foundations Robert Herdt Rockefeller Foundation Partners Civil Society Temporarily vacant Private Sector Sam Dryden PSC Chair Executive Secretary, ExCo Francisco Reifschneider Secretariat Support Selcuk Ozgediz ExCo currently has two subcommittees not shown in the chart--the Program Committee and the Finance Committee. The System Office comprises 10 units, only the first 3 of which appear on the organizational chart: 1. CGIAR Secretariat 2. Interim Science Council Secretariat 3. Future Harvest Foundation 4. Association of International Agricultural Research Centers 5. Gender and Diversity Program 6. Internal Auditing Unit 7. Central Advisory Service for Intellectual Property Rights 8. CDC Executive Secretariat 9. Chief Information Officer 10. Strategic Advisory Service on Human Resources. The interim Science Council was previously called the Technical Advisory Committee, and is currently being transformed into a Science Council. 1 5 8 APPENDIX 2: META-EVALUATION TERMS OF REFERENCE Rationale tors to be considered in revisiting the relations The meta-evaluation1 of the Consultative Group between CGIAR, national research institutions, on International Agricultural Research (CGIAR) and the private sector. In addition, the rapid is part of a larger OED evaluation of Global Pub- emergence of communications and informa- lic Policies and Programs (GPPP) currently being tion technologies has created opportunities for carried out under the leadership of Uma Lele. improved connectivity within the CGIAR Sys- Among the overarching objectives of the GPPP tem and greater devolution of CGIAR activities evaluation are to assess progress in the imple- to national and regional research agencies. mentation of the recommendations of OED's 1998 internal review of the World Bank grant Terms of Reference programs2 and to derive strategic, programmatic, Given resource constraints and the DGF-pro- and operational lessons and implications for the posed independent review of the CGIAR, this Bank's future involvement in global public pro- meta-evaluation does not aim at an in-depth as- grams and policies in relation to other partners sessment of CGIAR performance. Nor does it and the Bank's comparative advantages. seek to develop detailed recommendations about Given the size and visibility of the CGIAR and the CGIAR's future role. Its main purpose is to its long and well-documented track record, the draw broad lessons of experience for the future meta-evaluation will help OED acquire a sound management of global programs and policies understanding of the Bank's involvement in the Bank supports by tracking implementation global programs. With a $50 million annual DGF of recommendations of OED's 1998 internal re- allocation, the CGIAR accounts for 40 percent of view of World Bank grant programs; undertak- DGF funds available to all global programs. DGF ing a review of the Bank's management of its contributions to new and high-priority needs current portfolio; and, specifically in the case of have been hindered by the direction of 80 per- the CGIAR, other extensive reviews of the CGIAR. cent of DGF funding to 10 programs, including The CGIAR meta-evaluation will be under- the CGIAR, for which no credible exit strategies taken by a small, independent team as a com- have been designed.3 ponent of the OED evaluation. The CGIAR The need for this review of prior evaluations meta-evaluation will largely rely on prior reviews of the CGIAR is rooted in part in the dramatic of the CGIAR and on interviews. It will be carried changes in the technological and institutional en- out in close coordination with the 27 case stud- vironment of global agricultural research. The ies of global programs being undertaken as part evolving policy framework for intellectual prop- of the GPPP review. It will feed into the more erty rights (IPRs), the biotechnology revolution, comprehensive independent review of the the associated issues related to the CGIAR's ge- CGIAR commissioned by Management. netic resources, and the increased research ca- Using OED's evaluation criteria, the meta- pacity of NARS in middle-income developing evaluation will: countries (combined with the continuing chal- lenge of inadequate institutional development in · Evaluate the implementation of recommen- small, low-income countries) are among the fac- dations of OED's 1998 internal review of the 1 5 9 T H E C G I A R AT 3 1 World Bank grant programs with respect to CGIAR Change Design and Management Team the CGIAR, including an assessment of sub- from the standpoint of previous evaluations. sidiarity, arm's-length relationship with the · Identify issues confronting the CGIAR from a Bank, and exit strategy. forward-looking perspective, and, hence, those · Review the quality, standards, and treatment of issues that need to be addressed by the inde- the key issues described above as they relate to pendent evaluation. the CGIAR's structure, its governance and fi- · Draw lessons for the Bank's overall GPPP strat- nancing mechanisms, and its scientific strategy egy and the CGIAR, including the role of pub- by all relevant previous evaluations of the CGIAR lic sector funding (particularly that of the World System as well as those of selected Centers. Bank) for developing and disseminating tech- · Analyze and assess the coverage, quality, and nologies for agriculture and resource man- recommendations of previous evaluations. agement to reduce poverty and achieve · Consider the recent recommendations of the sustainable development. 1 6 0 APPENDIX 3: META-EVALUATION TEAM AND ADVISORY COMMITTEE: BIOGRAPHICAL INFORMATION Meta-Evaluation Team Economics for the U.S. Department of Agricul- ture, a Fellow, and later President, of the Amer- Christopher B. Barrett, a tenured Associate Profes- ican Agricultural Economics Association. Dr. sor in the Department of Applied Economics Gardner has been a member of numerous in- and Management of Cornell University, holds a ternational delegations and missions, primarily dual Ph.D. in Agricultural Economics and Eco- in Eastern Europe, and has published exten- nomics from the University of Wisconsin-Madi- sively on U.S. agricultural policies. son and a Master's in Development Economics from the University of Oxford. Dr. Barrett has Chris Gerrard, a Canadian national, is a senior agri- served as associate editor for the American Jour- cultural economist (M. Phil., Oxford, and Ph.D., nal of Agricultural Economics and Agricultural Minnesota). Before joining OED in 1999, he was Economics journal, has won numerous awards the task manager of the World Bank Institute and fellowships, and has worked as a consultant training program on "Policy and Institutional Re- for the World Bank, USAID, and the OECD. He form for Sustainable Rural Development," a world- is currently collaborating with ICRAF on a re- wide program with a special emphasis on Africa. search project. Before joining the Bank in 1994, he was a pro- fessor of agricultural economics at the University Carl Eicher, a University Distinguished Professor of Saskatchewan, Canada. Emeritus at the Michigan State University, De- partment of Agricultural Economics, holds de- Lauren Kelly earned her B.A. in Political Science grees from both Harvard and Michigan State from the University of Rochester and her M.A. University. He has been a longstanding scholar in Development Economics/Conflict Manage- of African agriculture and has written exten- ment at the Johns Hopkins School of Advanced sively on the subjects of agricultural research and International Studies. She was a contributing development. In addition to having worked in member to OED's Forest Policy and Imple- Nigeria, Zimbabwe, and the United States, Dr. mentation Review. Before joining the World Eicher has been a member of several World Bank in 1998, she co-founded and staffed a Bank and USAID missions to various African United States Congressional Briefing Series on countries. A former visiting senior research fel- security and defense policy issues, cospon- low of ISNAR (1988), Dr. Eicher has also held a sored by bipartisan members of the U.S. Con- small number of consultancies with other CGIAR gress, the Center for Strategic International Centers. Studies, and which was funded by several foun- dations in support of peace and security aims. Bruce L. Gardner, who obtained his Ph.D. in Eco- Ms. Kelly has worked as a legislative aid in the nomics from the University of Chicago, is Dis- U.S. Congress in the field of U.S. Foreign Pol- tinguished University Professor and Chair of the icy and in the European Parliament (Brussels Department of Agricultural and Resource Eco- and Strasbourg), where her research centered nomics at the University of Maryland. He previ- on security and defense policy in Europe and ously held positions as Assistant Secretary for EU-U.S. relations. 1 6 1 T H E C G I A R AT 3 1 Uma Lele is currently Senior Advisor, Operations and Latin America. Ms. Perkins also worked on Evaluation Department. She has held various OED's review of the World Bank's forestry policy. research, operational, managerial, and advisory positions in the World Bank, with experience in Saeed Rana has 45 years of practical experience South and Southeast Asia, Sub-Saharan Africa, working on water resources planning, imple- and Latin America. She has served as visiting mentation and management of projects and pro- professor at Cornell University, as Graduate Re- grams. Dr. Rana joined the World Bank in 1981 search Professor and Director of International where he has worked on a variety of water and Studies at the University of Florida, and as Di- agriculture projects in the Middle East and North rector of former President Carter's Global De- Africa and South Asia Regions. He has also velopment Initiative at the Carter Center in worked in the resident mission in Bangladesh. Atlanta, Georgia. Her work with the CGIAR has Before joining the Bank, Dr. Rana worked for 25 included serving on TAC in 1993­95, the Conway years in the public sector in Pakistan and the pri- Panel on the CGIAR vision and strategy in 1994, vate sector in Turkey, Indonesia, and the United serving as a founding board member of CIFOR States. Since his retirement from the Bank, he has in 1992, and working with the CGIAR and the been involved in QAG panels for supervision Bank during 1995­98 to build trilateral partner- and QAE assessments in the field and at head- ships between the CGIAR Centers, the World quarters. He has served as a member of the Bank's agricultural research support to devel- Water Policy Advisory Group of the Government oping countries, and the science of advanced of Bangladesh since 1998. Dr. Rana has also lent countries. She recently completed a review of the his expertise to OED's reviews of the Bank's World Bank's forest strategy and is currently in water and forest policies. Dr. Rana received his charge of the evaluation of the Bank's global training in water resources engineering at uni- portfolio of 70 programs in its various networks. versities in Pakistan and the United States. Uma Lele is Fellow of the American Agricultural Economic Association. Mandivamba Rukuni, a Program Director for the W.K. Kellogg Foundation, received a Ph.D. from William H. Lesser, who holds a Ph.D. in Agricultural the University of Zimbabwe and an M.Sc. in Trop- Economics from the University of Wisconsin, is ical Agricultural Development from the Univer- currently the Professor of Marketing in the De- sity of Reading (Pennsylvania). Prior to his current partment of Applied Economics and Manage- position he was a professor of Agricultural Eco- ment at Cornell University. He has chaired the nomics at the University of Zimbabwe and visit- Food and Agricultural Marketing Policy Section ing professor at Michigan State University. Dr. of the American Agricultural Economics Associ- Rukuni has consulted for the World Bank, the ation, published extensively on intellectual prop- USDA, USAID, CIDA, the Ford Foundation, and erty rights and agriculture, and assisted the the German Agency for Technical Cooperation International Academy of the Environment in (GTZ) as well as ISNAR and CIMMYT. He is cur- Geneva, Switzerland, in establishing a program rently a member of the Board of Trustees of in biodiversity and biotechnology. IFPRI and Chair of the Proposal Evaluation Com- mittee for the IFPRI 2020 Vision Network for Karin Perkins has a Master's Degree in Develop- East Africa, and has previously been a member ment Policy from Cornell University, with a back- of several TAC panels as well as the CGIAR ground in Agricultural and Resource Economics. Change Design Management Team. Among other professional positions, she has worked with the Third External System Review of David J. Spielman received his M.Sc. in Develop- the CGIAR, Cornell's International Institute for ment Studies from the London School of Eco- Food, Agriculture and Development, the Foreign nomics and Political Science in 1993 and his B.A. Agricultural Service of USDA, and World Resources in International Relations from Tufts University Institute. She has lived and worked in Indonesia in 1992. He is currently working toward his Ph.D. 1 6 2 A P P E N D I X 3 : M E TA - E VA L U AT I O N T E A M A N D A D V I S O R Y C O M M I T T E E in Economics at American University, Washing- various institutions: Institute of Development ton, D.C. Mr. Spielman has most recently worked Studies, University of Sussex, and All Souls Col- with the Economic Research Service of the U.S. lege, Oxford. He also served as a Program Di- Department of Agriculture in Washington, D.C., rector of the Food Consumption and Nutrition and with the Aga Khan Development Network in Program at the International Food Policy Re- northern Pakistan. He has lived and worked in search Institute. As Senior Policy Adviser at the Egypt, Pakistan, the U.S., and Zimbabwe, and World Bank, he was involved in the Task Force has conducted research on a range of topics in- established to study the impact of the World cluding agricultural seed markets, rural micro- Bank's activities on poverty groups, among oth- finance, and rural education systems. ers. Professor Lipton holds an M.A. degree from Oxford University and D. Lit. from the University Advisory Committee of Sussex. He has received various awards and distinctions, including the Jenkyns Prize and the Yujiro Hayami holds a B.A. in Liberal Arts from George Webb Medley Prize in Economics. He the University of Tokyo (1956). He was one of the was Rockefeller Fellow at the Massachusetts In- first Japanese to earn a U.S. Ph.D. in agricultural stitute of Technology. He has extensively re- economics after World War II, at Iowa State in searched agricultural and rural development, 1960. He began his professional career in the particularly technical choices, land distribution National Research Institute of Agricultural Eco- poverty impact, the economics of poverty re- nomics in Japan, followed by two decades' serv- duction, and demographic nutrition econom- ice at the Tokyo Metropolitan University. He has ics, among others. He also has extensive been a Professor of International Economics, experience in African and Asian studies. School of International Politics, Economics and Business, Aoyama-Gakuin University, 1986­2000; Harris Mule received his B.Sc. in Business Ad- Professor of Economics at Tokyo Metropolitan ministration in 1962 and M.A. in Economics in University, 1966­86; and the Lee Teng Hui Pro- 1963 from the University of Denver, Colorado, fessor of World Affairs at Cornell University, and his M.P.A. in Development Planning from 1995­96. He has also served as visiting professor Harvard University in 1967. Mr. Mule was Per- at the University of Minnesota and as an econo- manent Secretary, Ministry of Finance/Planning mist with the International Rice Research Insti- (1978­86), and Deputy Permanent Secretary, tute in the Philippines. Dr. Hayami has won Ministry of Finance and Planning (1972­78), in awards for a number of outstanding works in the Government of Kenya. Mr. Mule is a qualified agricultural economics. He holds honorary life- and experienced development economist and time membership in the International Association policy analyst. He supervised and made major in- of Agricultural Economists and is a fellow of the puts in the formulation of Kenya's Development American Agricultural Economics Association. In Plans, Sessional Papers, and other policy docu- 1999 he received a Purple Medal (Shiju Hosho) ments between 1964 and 1986. Noteworthy from the Government of Japan for his outstand- among these was the policy framework that ing work. formed the basis for development of ASAL in the 1979­83 Development Plan. More recently, Michael Lipton is currently Research Professor Mr. Mule was assistant president in charge of and Director of the Poverty Research Unit at the economic planning at IFAD. He has served the University of Sussex. He has been Professor of World Bank, the African Development Bank and Development Economics at the Poverty Research the International Food Policy Research Institute Unit since 1994. Before this, he was a Fellow to in advisory capacities and/or as a board member. 1 6 3 APPENDIX 4: METHODS AND TOOLS The assessment of CGIAR is assisted by its tra- sisted of the CGIAR's "core collection" of docu- dition of conducting periodic reviews of its Cen- ments, including reports mentioned above. Par- ters (called External Program and Management ticular attention was given to the report of the Reviews--EPMRs) and by System-wide thematic Third System Review,1 impact studies, and EPMRs, reviews commissioned by the Technical Advi- and reports that assessed follow-up on the rec- sory Committee and managed by the CGIAR ommendations of these reviews, various OED Secretariat. In addition, there are many papers reviews of the CGIAR, the Environmentally Sus- on the work of the Centers and the System by tainable Development Agricultural Research and outside analysts. And there is a vast literature on Extension Group (ESDAR), and agricultural and impact analysis and independent periodic ex- rural development project investments. (See Bib- ternal System reviews, although the most re- liography.) The quality of analysis and the rec- cent of these, the Third System Review, was ommendations of these reviews were assessed in carried out after a 17-year hiatus. Finally, OED and terms of their actual and expected impacts on the other donors that support the CGIAR have car- CGIAR System. Unresolved issues related to the ried out independent reviews. mission, substance of research, and organization and management of the System were considered Analytical Tools and Methods and implications for the proposed independent The meta-evaluation used several analytical tools review were identified. and applied the related methodologies itera- tively. These consisted of the following. Preparation of major issue papers. Detailed papers focused on CGIAR policy research, the biotech- Comprehensive desk review of relevant documents. nology revolution, the evolving policy frame- The meta-evaluation team has reviewed over 700 work for intellectual property rights, the scope documents (box A4.1). The main source con- and impact of CGIAR's research on plant and an- D o c u m e n t s R e v i e w e d f o r t h e B o x A 4 . 1 M e t a - E v a l u a t i o n External System-wide Reviews 3 World Bank Documents 60 External System-level Thematic Reviews 21 Academic Articles and Publications 212 External Center-level Reviews 22 Other Documents 80 TAC Strategy Papers and Commentaries 42 CGIAR Meeting Documents and Proceedings 41 Total 737 CGIAR Committee Documents 63 Center Research Studies 99 Note: Does not include additional materials referenced in Back- Center Documents (General) 94 ground Papers. 1 6 5 T H E C G I A R AT 3 1 imal genetic resources, work on strengthening Questionnaire survey of the views of 235 stakehold- the research capacity of NARS, and research on ers and clients. A detailed questionnaire elicited natural resources management. In addition, one views on key issues facing the CGIAR, and on regional study on Africa and four country case the changes recently introduced in governance studies on Brazil, Columbia, India, and Kenya ob- and programs, such as the formation of the Ex- tained developing-country perspectives on the ecutive and Science Councils and the proposed CGIAR. All these papers were extensively peer- launching of (global) Challenge Programs. reviewed by knowledgeable external experts. Quantitative analysis of survey data and a syn- (See Appendix 5 for the list of Working and Back- thesis of qualitative write-in comments helped ground Papers, authors, and peer reviewers.) the team sharpen its key findings and conclu- sions. Extensive consultations with stakeholders. Team The tools and methods used enabled the members conducted interviews with well over team to conduct a comprehensive survey of 100 stakeholders and attended the CGIAR's an- past reviews of the CGIAR and to probe deeply nual general meeting in Washington, including into issues regarding the System's current the first meeting of its newly created Executive and future mission, strategy, governance, and Council, and interviewed major stakeholders: effectiveness. Besides conducting a thorough donors, representatives of NARS, current and assessment of available documents and prepar- former chairs of the CGIAR and its committees, ing issues papers, the team has extensively chairs of Center Boards, and Center Directors and consulted with key stakeholders and knowl- staff. (See List of People Consulted.) In addi- edgeable experts. The methodology has thus tion, written comments were received from a provided a firm foundation for the conclu- wide range of knowledgeable partners and ob- sions and recommendations in the final report. servers of the CGIAR. 1 6 6 APPENDIX 5: LIST OF WORKING AND BACKGROUND PAPERS, AUTHORS, AND PEER REVIEWERS Working Papers Lesser, William. 2002. Reviews of Biotechnology, Genetic Resource and Intellectual Property Barrett, Christopher B. 2002. Natural Resources Rights Programs. Management Research in the CGIAR: A Meta- Peer Reviewers: Ronnie Coffman, John Dodds, Evaluation. Robert Evenson, Brian Ford Lloyd, Anatole Peer Reviewers: Jock Anderson, Derek Byer- Krattiger, Steve Kresovich lee, Dana Dalrymple, Hans Gregersen, Ted Henzell, John Lynam, Vernon Ruttan, Mere- Spielman, David. 2002. International Agricultural dith Soule, Joachim von Braun, Usha Barwale Research and the Role of the Private Sector. Zehr Macedo, Jamil, Marcio C. M. Porto, Elisio Contini, Eicher, Carl K. and Mandivamba Rukuni. 2002. The and Antonio F. D. Avila. 2002. Brazil Country CGIAR in Africa: Past, Present, and Future. Paper for the CGIAR Meta-Evaluation. Peer Reviewers: Malcolm Blackie, Dana Dal- rymple, Bob Herdt, Alain de Janvry, Romano Katyal, J.C., and Mruthyunjaya. 2002. CGIAR Ef- Kiome, John Lynam, Eric Tollens, Geoffrey fectiveness--A NARS Perspective from India. Mrema, Wilfred Mwangi, Cyrus Ndiritu, Emmy Simmons, Moctar Touré Background Papers Gardner, Bruce. 2002. Global Public Goods from (Available upon request) the CGIAR: Impact Assessment. Peer Reviewers: Jock Anderson, Dana Dal- Ndiritu, Cyrus. 2002. CGIAR-NARS Partnership: rymple, Osvaldo Feinstein, Paul William The Case of Kenya. Glewwe, Hans Gregersen, George Norton, Scott Rozelle, Vernon Ruttan, Sara Scherr, Romano, Luis. 2002. Colombia Country Paper for Sudhir Wanmali the CGIAR Meta-Evaluation. 1 6 7 APPENDIX 6: LIST OF PEOPLE CONSULTED World Bank Name Title Department Robert McNamara Former President Ian Johnson Vice President and Chairman of CGIAR ESDVP David Hopper Former CGIAR Chairman World Bank V. Rajgopalan Former CGIAR Chairman World Bank Ismail Serageldin Former CGIAR Chairman World Bank Wilfried Thalwitz Former CGIAR Chairman World Bank Jock Anderson Adviser ARD Shawki Barghouti Adviser ARD Hans Binswanger Sector Director, Environmental, Rural & AFTRS Social Development Derek Byerlee Lead Economist ARD Luc Christiaensen Economist AFTP2 Cornelis De Haan Consultant ARD Thomas Duvall Chief Counsel LEGCF Stephen Eccles Retired World Bank Sushma Ganguly Sector Manager, Rural Development ARD Paul Hubbard Manager DGF Motoo Kusakabe Vice President RMCVP Mohamood Abdi Noor Senior Agricultural Specialist AFTR1 Eija Pehu Senior Adviser ARD George T. K. Pitman Senior Evaluation Officer OEDST Amedee Prouvost Director and Chief Financial Officer MIGFR Randall Purcell Senior Partnership Specialist DGF Alassane Sow Senior Agricultural Economist AFTR2 Kalanidhi Subbarao Lead Economist AFTHD Eugene Terry Adviser ARD Robert Thompson Senior Adviser ARD Moctar Touré Executive Secretary AFTR1 Achim E. von Heynitz Director CRMDR 1 6 9 T H E C G I A R AT 3 1 CGIAR Secretariat Name Title Francisco Reifschneider Director Alexander von der Osten Former Executive Secretary Selcuk Ozgediz Management Adviser Ravindra Tadvalkar Lead Financial Officer Shey Tata Senior Financial Officer Manuel Lantin Scientific Adviser Ernest Corea Consultant Sarwat Hussain Senior Information Officer Feroza Vatcha Resource Management Analyst CGIAR Members Name Title/Department Organization (Country) Ian Bevege Australia Robert Clements Director Australian Centre for International Agriculture Alberto Duque Portugal President Empresa Brasileira de Pesquisa Agropecuaria (Brazil) Iain MacGillivray Senior Adviser, Agriculture and Canadian International Natural Resources Policy Branch Development Agency Klaus Winkel Head of Department Danish International Development Assistance Phillippe Vialatte European Commission European Commission Jacques Eckebil Assistant Director-General, Sustainable Food and Agriculture Development Department Organization Gilles Saint-Martin Chargé de Mission coordination DRIC ­ Ministère de la international de la recherche Recherche/Ministère de l'Education Nationale (France) Hans-Jochen de Haas Head, Rural Development and Food Federal Ministry of Security, Agricultural and Rural Economic Cooperation Development Division and Development (Germany) Rodney Cooke Director, Technical Advisory Division International Fund for Agricultural Development Shantanu Mathur Technical Adviser, Economist International Fund for Agricultural Development Panjab Singh Secretary to the Government of India Department of Agriculture and Director General, Indian Council and Cooperation, Ministry of Agricultural Research of Agriculture (India) Brendan Rogers Counsellor, Development Department of Foreign Cooperation Division Affairs (Ireland) Wilfred Mwangi Permanent Secretary Ministry of Energy (Kenya) 1 7 0 A P P E N D I X 6 : L I S T O F P E O P L E C O N S U LT E D CGIAR Members (continued) Name Title/Department Organization (Country) Romano Kiome Director Kenya Agriculture Research Institute (Kenya) Klaas Tamminga The Netherlands Ruth Haug Director of Research NORAGRIC (Norway) Ruben Villareal Director SEAMEO Regional Center for Graduate Study and Research in Agriculture (Philippines) Robert Herdt Vice President, Program Rockefeller Foundation Administration John Lynam Senior Scientist Rockefeller Foundation Peter Matlon Deputy Director Rockefeller Foundation Dr. Akinwumi A. Adesina Senior Scientist Rockefeller Foundation Eva Ohlsson Research Officer, Division for Thematic Sida/SAREC (SWEDEN) Programmes, Department for Research Cooperation Christine Grieder Senior Adviser, Natural Resources Swiss Agency for Develop- and Environment ment and Cooperation Issam El-Zaim Minister of State for Planning Ministry of State for Affairs Syria Planning Affairs (Syria) Joseph Mukiibi Director General Natural Agricultural Research Organization (Uganda) Andrew Bennett Executive Director Syngenta Foundation for Sustainable Development Robert Carlisle Department for Interna- tional Development (U.K.) Emmy Simmons Director, Center for Economic United States Agency for Growth and Agricultural Development International Development Jonathan Conly Deputy Assistant Administrator, United States Agency for Center for Economic Growth and International Development Agricultural Development Felipe Manteiga Director, Office of Agriculture and United States Agency for Food Security International Development Rob Bertram Chief, Multi-Lateral Programs Division, United States Agency for Center for Economic Growth and International Development Agricultural Development Dana Dalrymple Research Adviser United States Agency for International Development Merideth Soule International Affairs Specialist United States Agency for International Development Roberto Lenton Director (Former), Sustainable Energy United Nations Develop- and Environment Division (SEED) ment Programme Dietrich Leihner Director, Research, Extension and Food and Agriculture Training Division Organization 1 7 1 T H E C G I A R AT 3 1 CGIAR Members (continued) Name Title/Department Organization (Country) Douglas Wholey Technical Adviser, Agronomy, International Fund for Technical Advisory Division Agricultural Development Technical Advisory Committee Name Title Emil Javier Chair, Technical Advisory Committee Shellemiah Keya Executive Secretary, Technical Advisory Committee Guido Gryseels Former Deputy Executive Secretary, Technical Advisory Committee Michael Cernea Member, Technical Advisory Committee Hans Gregersen Member, Technical Advisory Committee and Chair, Standing Panel on Impact Assessment Amir Kassam Senior Research Officer, Technical Advisory Committee Partnership Committees Name Title, Committee Monica Kapiriri Co-Chair, NGO Committee Ann Waters-Bayer Co-Chair, NGO Committee Sam Dryden Co-Chair, Private Sector Committee CGIAR Centers Name Title Timothy Reeves Director General, CIMMYT Alex McCalla Board Chair, CIMMYT David Hoisington Director, Applied Biotechnology Center, CIMMYT Prahbu Pingali Director, Economics Program (India) , CIMMYT Wilfred Mwangi Principal Scientist, Economist (Kenya) , CIMMYT Robert Havener Board Chair, ICARDA John Dodds Special Adviser, Director General's Office, ICARDA Meryl Williams Director General, ICLARM Kurt Peters Board Chair, ICLARM Pedro Sanchez Director General, ICRAF Per Pinstrup-Andersen Director General, IFPRI Peter Hazell Division Director, IFPRI Chris Delgado Senior Research Fellow, IFPRI Philip Pardey Research Fellow, IFPRI Keijiro Otsuka IFPRI Associate, Foundation for Advanced Studies on International Development (Japan) Jim Ryan Research Associate, IFPRI Sudhir Wanmali Coordinator, 2020 Vision Network for West Africa, IFPRI Lukas Brader Director General, IITA 1 7 2 A P P E N D I X 6 : L I S T O F P E O P L E C O N S U LT E D CGIAR Centers (continued) Name Title Enrico Porceddu Board Chair, IITA Carlos Sere Director General, ILRI Hank Fitzhugh Former Director General, ILRI John Vercoe Board Chair, ILRI Bruce Scott Director, Corporate Services, ILRI Geoffrey Hawtin Director General, IPGRI Coosje Hoogendoorn Deputy Director General, IPGRI Jan Engels Director, Genetic Resources Science and Technology Group, IPGRI Jane Toll Senior Scientist/Coordinator, SGRP, IPGRI Anthony Kalm Fundraising Consultant, IPGRI Angeline Kumba Board Chair, IRRI Michael T. Jackson Director for Program Planning and Coordination, IRRI Stein Bie Director General, ISNAR Moise Mensah Board Chair, ISNAR Joel Cohen Project Leader, Managing of New Technologies for Agricultural Research, ISNAR Howard Elliott Principal Research Officer, ISNAR Francis Idachaba Project Leader, Policies for Institutional Innovation for Agricultural Research, ISNAR Doug Horton Project Leader, Linking Research Organizations & Stakeholders, ISNAR Victoria Henson-Apollonio Project Manager, Central Advisory Service on Intellectual Property, ISNAR Frank Reijsberman Director General, IWMI Partner Organizations, Non-Member Countries, and Others Name Title/Department Organization (Country) Martin Abel Agricultural Economic Consultant I.P. Abrol Board Member Centre for Advancement of Sustainable Agriculture John Antle Professor Montana State University Samuel Bruce-Oliver Director General National Agricultural Research Institute (The Gambia) Balu Bumb Research and Development Division International Fertilizer Development Center Neville Clarke Program Coordinator Texas A&M University Erick Fernandes Assistant Professor Cornell University J.M. Haki Tanzania Seyfu Ketema Vice Chairman, Committee of Directors Association for Strengthen- ing Agricultural Research in Eastern and Central Africa 1 7 3 T H E C G I A R AT 3 1 Partner Organizations, Non-Member Countries, and Others (continued) Name Title/Department Organization (Country) Jean-Claude Legoupil Southern African Centre for Cooperation in Agricul- tural and Natural Resources Research and Training Klaus Leisinger President and Executive Director Novartis Foundation for Sustainable Development Ndiaga Mbaye Executive Secretary West and Central African Council for Agricultural Research and Develop- ment Geoffrey Mrema Executive Secretary Association for Strengthen- ing Agricultural Research in Eastern and Central Africa Rebecca Nelson Associate Professor Cornell University Rajendra S. Paroda Chairman Global Forum on Agricultural Research Donald Plucknett President and Principal Scientist Agricultural Research and Development International Thomas Reardon Professor Michigan State University Scott Rozelle Associate Professor University of California at Davis Ashok Seth Director Agriculture and Rural Development Consultants (U.K.) Maurice Strong Chairman The Earth Council M.S. Swaminathan Chairman, M.S. Swaminathan Research Foundation Adama Traore President, Executive Committee West and Central African Council for Agricultural Research and Develop- ment Norman Uphoff Professor Cornell University 1 7 4 APPENDIX 7: MANAGEMENT RESPONSE Management recognizes and appreciates the contributing more comprehensively to the work that went into the production of the meta- achievement of Millennium Development evaluation report and its annexes. The report Goals--a move fully supported by the investors provides a comprehensive and valuable analysis and clients alike. It did so by adopting natural re- of past evaluations and offers a set of views on source management (NRM) as a twin pillar of its issues confronting the Consultative Group on In- research efforts (complementing germplasm re- ternational Agricultural Research (CGIAR), using search) 12­15 years ago, in full recognition of a forward-looking perspective. We concur with changing paradigms on agricultural and rural most of the report's recommendations, and note development, and enhanced concerns with re- that many of these reflect actions already taken gard to protecting the environment and the or planned by the CGIAR. We differ with the re- long-term sustainability of agriculture. The view team on some points, which are highlighted change was made at the urging of several exter- below, and detailed in the attached draft Man- nal panels made up of eminent experts. That agement Action Record. said, Management does see a need to continue to focus on productivity research and to be alert Continued Relevance of the CGIAR. We are to the risk that in increasing the focus on natu- pleased to note that the review argues for the ral resources the centers may dilute their con- continued relevance and indispensability of in- tribution to global public goods--including ternational agricultural research as a global pro- natural resource management--if it results in gram.1 As the report recognizes, "the CGIAR's their becoming more involved in local develop- productivity-enhancing research has had sizable ment activities. We agree more strongly that re- impacts on reducing poverty"2 and "large and stricted funding by some donors allocated to small NARS [National Agricultural Research Sys- projects preferred by those donors has diverted tems] alike acknowledge that the CGIAR has some activities of the Centers from their core re- made major contributions to their growth."3 search programs. This is an issue deserving the Donor confidence in the CGIAR System is high, attention of the entire CGIAR system. as reflected by sustained funding and the deci- sion by Executive Directors last year to recom- CGIAR Reform Program. The report ac- mend the maintenance of World Bank knowledges that the ongoing Reform Program is Development Grant Facitlity (DGF) resources "a good start," finds some of the reforms "cred- for the CGIAR at recent levels. itable" and others in need of "revisiting."5 It also calls for further reforms,6 which we agree may be Strategic Focus of the CGIAR. Management necessary, but which need to be explored and sees the report's claim that "the CGIAR is less fo- fleshed out in more detail and, equally important, cused"4--an assertion on which several of the vetted by the entire CGIAR membership. It is specific recommendations rest--as potentially imperative not to slow down the current reform overly sweeping. It is true that the CGIAR has effort or significantly shift its direction. That ef- broadened its strategic focus, to enhance the fort has been planned with care by the CGIAR relevance and impact of its research, thereby leadership, and approved by the membership. 1 7 5 T H E C G I A R AT 3 1 However, additional reforms, building on those statement that "conflicts of interest and inade- under implementation, will no doubt be neces- quate oversight constrain the Bank's strategic sary. Management would like to emphasize, how- leadership."7 The Bank has provided strategic ever, that the CGIAR is a partnership. While the leadership including the force behind the cur- Bank may wield considerable influence over the rent reform program. That said, Management partnership, it does not control CGIAR decision- does see the advantage of making a clearer dis- making. Therefore, in responding to OED con- tinction between Bank oversight of its contri- cerns and recommendations regarding CGIAR bution to the CGIAR and its role in CGIAR reform, Management can only promise to do its management. In doing this, Management does utmost to influence the CGIAR to move in di- not wish to encourage additional reporting lay- rections it believes are best to enable the CGIAR ers, which would increase bureaucracy and to achieve its objectives efficiently and effectively. could remove the link between responsibility As with all global partnerships, the Bank must pe- and accountability. Instead, starting from exist- riodically check the alignment of Bank objec- ing structures, Management will establish a clear tives with the objectives of its partners and assess division between: (1) oversight and agreement to what extent the compromises inherent in on activities for which CGIAR may use DGF working in this kind of a partnership are more grants, and review of the results achieved than counterbalanced by the advantages of col- (through a Bank-wide mechanism as yet to be laborative action. At the present, Management be- determined); and (2) the Bank's role as part of lieves (and we believe that OED concurs) that the management of the CGIAR, including imple- interests of the Bank are well-served by its par- mentation of activities using the Bank's funds. ticipation in the CGIAR, even if its views may not Management has reviewed the issue and will always prevail in deliberations with major donors act before the next round of DGF allocations is and other CGIAR members. considered by Executive Directors. OED's ob- servations on the Bank's lending role will be ad- The Bank's Role in the CGIAR. Regarding the dressed in the context of the implementation of responsibilities within the World Bank for the the agriculture and rural strategy recently en- CGIAR, Management disagrees with the strong dorsed by Executive Directors.8 Summary of OED Recommendations with Management's Response OED recommendation Management response 1. Focusing the World Bank's Re- Management agrees with the overall thrust of this recom- sponsibilities mendation, but cannot be bound by the entire detailed list of As a convener and donor to the actions that OED sees as required. Management sees other CGIAR and lender to developing steps as equally or more likely to reach the objective stated so countries, the Bank should ad- well in this OED recommendation, specifically regarding the dress its corporate governance nature of the fundamental reforms to be pursued (see below). responsibilities in the manage- Management also wants to reiterate that the CGIAR is a global ment of the CGIAR and exer- program in which the Bank has considerable influence but does cise a degree of oversight not control. It is a collective effort of many donors. The Bank consistent with the major roles can only promise to utilize its influence to move the CGIAR that it plays in the CGIAR. in directions it believes are best to enable the CGIAR to achieve This will require: its objectives efficiently and effectively. 1 7 6 A P P E N D I X 7 : M A N A G E M E N T R E S P O N S E OED recommendation Management response Leading a concerted effort at the Management agrees on the need for a high-level effort on re- highest level, much as when the form and has already embarked on such an effort. CGIAR was established, to achieve fundamental reforms in the organi- Management agrees on the need to reverse the increasing zation structure, finance and man- trend to restricted funding, recognizing that there is a con- agement of CGIAR, particularly to tinuum between fully restricted and fully unrestricted funding. encourage donors to reverse the Reversing the existing trend is a complicated task because trend in restricted funding, and to es- donors: (a) need to demonstrate "value for money" to their tablish clear targets for an increased own constituencies; and (b) feel that targeted restricted fund- share of unrestricted funding. ing increases accountability for specific outcomes and, there- fore, many donors prefer this approach. However, the CGIAR Chairman (ESSD VP) and CGIAR Director are engaged in tar- geted efforts to increase unrestricted funding. Agreed Action. The CGIAR Chairman and CGIAR Director will continue to work within the CGIAR framework to encourage donors to increase unrestricted funding. Separating oversight and manage- The CGIAR membership strongly supports the role of a sen- ment functions within the Bank to ior Bank official as CGIAR Chair. At the inception of the CGIAR, address the conflicts of interest that the Bank undertook to provide the Group with its Chairman. currently exist among the roles of This responsibility has been consistently performed for 31 the ESSD Vice President, the ARD years. By mutual agreement between the Bank President and Director, the Research Advisor, and the CGIAR membership, the Chairman is a Bank Vice President. the CGIAR Director. While Management does not concur that there is a conflict of interest in the current arrangements, it is important to ensure an appropriate balance and separation between responsibil- ity for implementation and responsibility for oversight. Management agrees on the need for independent oversight of the Bank's financial contribution, ensuring that the use of the Bank's contribution is in line with Bank objectives, by a Bank Manager outside of the management chain directly oversee- ing the CGIAR. Agreed Action. Management agrees on the need to articulate a clear division of responsibilities between strategic planning and oversight of the Bank's grant and the grant implementation fa- cilitation role. Management will undertake to clarify this arrangement before the next round of DGF allocations is dis- cussed by Executive Directors. Abandoning the current matching Management agrees on the principle that the Bank's resources grant model; reporting to the Board should be allocated more strategically than in the past, and that on a regular basis on the impact of is already beginning to happen. Some funds must continue to the allocation of the Bank's re- be allocated in support of the Reform Program, to ensure the sources on the incentive structure long-term effectiveness of the System, and for research that 1 7 7 T H E C G I A R AT 3 1 OED recommendation Management response of the System; and ensuring that has the greatest potential for long-term impact. Management the Bank's resources are allocated agrees to phase out the matching grant allocation mechanism strategically in support of global and in favor of an allocation mechanism based on strategic prior- regional public goods that con- ities and center performance. Management also agrees to in- tribute to agricultural productivity volve the Bank's Agriculture and Rural Board to guide the and poverty reduction, based on allocation of the Bank's grant in support of the reform program long-term priorities as articulated and strategic priorities. by the Science Council. This will re- quire the Bank to assure itself that Agreed Action. The CGIAR Director will include in his/her reg- a strong, qualified, and independent ular reports to the Bank's Agriculture and Rural Board and the Science Council is established and independent manager with oversight responsibility for the vested with the role and resources Bank's financial contribution progress in phasing out the to establish System-wide priorities, matching grant allocation mechanism. policies, and strategies, and to mon- itor and report to the membership Management agrees on the key role of the Science Council, on the uses and allocations of CGIAR as guardian and guarantor of the System's scientific relevance, resources toward fulfilling the pri- excellence, and impact. The Science Council's mission was orities, policies, and strategies. adopted at the Annual General Meeting, in advance of the coun- cil being established. Its mission statement reads: To enhance and promote the quality, relevance and im- pact of science in the CGIAR, to advise the Group on strategic scientific issues of importance to its goals, and to mobilize and harness the best of international sci- ence for addressing the goals of the international agri- cultural research community. An independent Selection Committee chaired by the Execu- tive Director, Third World Academy of Science, is currently seek- ing the best possible talent for the Science Council. Agreed Action: Management will continue to encourage the CGIAR to work for early establishment of a strong Science Council, replacing the current interim council, by the end of 2003. Carrying out independent triennial Management agrees. External and center-specific Internal As- appraisals of the CGIAR, with Board sessments, ex-ante and ex-post, are conducted regularly. These approval as the basis of continuing assessments will feed into the new appraisals process. Bank support. Agreed Action. With completion of the meta-evaluation in 2002, the recommended appraisal will be launched in 2005. The re- sponsibility will be given to an independent third party. 2. Reforming the CGIAR Management agrees generally with the thrust of the recom- The strategic priorities of the mendations regarding reform of the CGIAR system. Again, CGIAR should respond more Management would like to reiterate that the CGIAR is a global actively to changes in the global partnership. Bank Management will do its utmost to influ- 1 7 8 A P P E N D I X 7 : M A N A G E M E N T R E S P O N S E OED recommendation Management response research context, giving more ence the CGIAR to move in directions it believes are best to prominence to basic plant enable the CGIAR to achieve its objectives efficiently and ef- breeding and germplasm im- fectively, but it is up to the partnership to agree on and un- provement, and reshaping nat- dertake reforms. ural resource management research to focus tightly on productivity enhancement and sustainable use of natural re- sources for the benefit of de- veloping countries. This will require: Postponing the approval of new Chal- Management agrees that the CGIAR should proceed cautiously lenge Programs (beyond the first two with Challenge Programs, but cannot accept the recommen- already approved) pending the in- dation that the approval of the Challenge Programs already in stallation of the new Science Coun- the pipeline for full preparation be postponed, since this is a cil, an assessment of System-level key element of the current Reform Program. Such a step would priorities, and a thorough review of represent a break between the CGIAR System and its partners. the design and approval process of the first two programs to learn les- Action (including funding) on Challenge Programs, from con- sons for the selection, design, se- ception and planning to implementation, is based on the re- quencing, and phasing of future view, assessment, and recommendation of the interim Science Challenge Programs in the context of Council. This responsibility will move to the Science Council System-level priorities and strategies. when it is established in 2003. Additionally, as requested by the Group, all Challenge Programs are to be assessed within a strategic framework that the interim Science Council is cur- rently developing. Increasing funding for conventional Management agrees on the need to emphasize genetic resource germplasm enhancement and plant conservation and improvement. However, poverty reduction re- and animal breeding research in quires healthy agroecosystem resources and policies that enable which the CGIAR possesses a com- the rural poor to participate in decisionmaking and innovation. parative advantage; conducting an An independent review of the impacts and the international and independent review of NRM, policy, regional public good elements of NRM activity is about to begin and social science research from a under the auspices of the Science Council's Special Program on global and regional public goods Impact Assessment. The results of this review will serve as the basis perspective to help it address coun- of the Group's decisions on future NRM activities. try- and regional-level issues con- straining productivity enhancement Management acknowledges the need to sustain the unique con- and the sustainable use of natural tribution of the CGIAR Centers to global public goods and resources; and devolving that por- global challenges facing agricultural productivity and natural tion of the CGIAR's applied and resources. We will expand the effort to mobilize funds for the adaptive NRM research program system for these purposes and efforts to build up the Global that does not constitute global or Conservation Trust will continue. Management notes that two regional public goods research to pilot Challenge Programs involve germplasm improvement: national and regional agencies, sup- "Biofortified Crops for Improved Human Nutrition" and "Un- ported by larger funding for na- locking Genetic Diversity in Crops for the Resource-Poor." 1 7 9 T H E C G I A R AT 3 1 OED recommendation Management response tional and regional agricultural re- Agreed Actions. A major share of Bank funding will go to en- search and development from both hancing Centers' gene bank collections in 2003, key to developing-country governments germplasm enhancement and breeding efforts. Management and donors. will also work within the CGIAR to encourage devolution of NRM research to NARSs, provided that the international and regional public good dimensions of such collaboration can be held in the global domain. In addition, Management will participate in a continuing effort to create congruence between CGIAR and Center policies and the global agenda, including MDGs, WSSD targets, Bank's Rural Development Strategy. Developing System-wide strategies Management agrees. These actions/activities are already a part and policies that facilitate busi- of the CGIAR agenda, but will be expanded. nesslike partnerships with NARS, agricultural research institutions, Agreed Actions. Management will work within the CGIAR to en- NGOs and the private sector; courage the CGIAR to: strengthening the management and use of intellectual property and ge- · Expand efforts to work in partnership with NARS, ARIs, netic resources; and using new sci- NGOs, and the private sector. entific areas such as biotechnology · Seek means of implementing the Declaration on private/pub- and bio-informatics to complement lic sector collaboration, signed by CEOs of major international its conventional research. agribusinesses on the eve of WSSD at the initiative of the CGIAR Private Sector Committee. · Implement the decision at the recent CGIAR Annual General Meeting that the CGIAR Executive Council should evaluate civil society relationships with the CGIAR (at the System level and in the field) and submit its report to the Group; and develop code of conduct for joint activities with private sector. · Continue use of new developments in biological research and bio-informatics, subject to measures concerning safety, and in keeping with the expressed views of intended beneficiar- ies in developing countries. · StrengthenexistingCentralAdvisoryServiceonIntellectualProp- erty,andSystem-wideBiosafetyProgram(dataanalyses,andcase studies of biotechnology introductions in Part 2 countries). Enhancing collaborative research Management agrees on the need for collaborative research and as a means of capacity building and capacity building, and that these activities need strengthening training; and engaging qualified de- beyond the first steps included in the current Reform Program. veloping country NARS to provide similar services to smaller and Agreed Actions. Management agrees to encourage the CGIAR to: weaker NARS. · Facilitate more NARS--NARS collaboration and assistance. · Expand emphasis on collaborative research, which lies at the heart of the Reform Program, but on which more can be done. 1 8 0 A P P E N D I X 7 : M A N A G E M E N T R E S P O N S E OED recommendation Management response · Encourage Centers to continue their capacity building efforts. · Continue to emphasize the role of NARS in Challenge Pro- grams and other collaborative efforts, including NARS-- NARS effort. 3. CGIAR Governance Management agrees on the need for greater efficiency, tougher The governance of the CGIAR priority setting, and scientific excellence in the CGIAR. should be reconfigured to pro- mote greater efficiency, tougher Agreed Actions. Management agrees to encourage the CGIAR to: priority setting, and scientific excellence without sacrificing · Broaden and deepen the Reform Program, the goals of which legitimacy and ownership. are efficiency and effectiveness in all aspects of the CGIAR. This will require: · Strengthen the efforts of the ExCo, including its Finance Committee and Program Committee, and the Science Coun- cil to ensure that these goals are met. Adopting a written charter that Management agrees on the need to continue moving toward clearly delineates the roles, re- greater clarity in roles and responsibilities of CGIAR bodies, while sponsibilities, and accountabilities maintaining flexibility and decentralized mode of operations. of the various officers and bodies that govern the System, and ana- Agreed Actions. Management will work within the CGIAR to sup- lyzing examining the advantages port: and disadvantages of establishing all or part of the CGIAR as a sepa- · The consolidation of existing ToRs and charters of CGIAR bod- rate legal entity attuned to deal with ies into a single document that will serve as basis for devel- today's realities on partnerships. oping an integrated Charter. · The development of criteria and procedures for CGIAR mem- bership as agreed at recent Annual General Meeting. While recognizing that CGIAR members have rejected the idea of a single legal entity in the past, in upcoming meetings Man- agement will press key donors and other members to again give serious consideration to the creation of a legal entity covering CGIAR's central oversight and fund allocation functions. Making ExCo members more fully Management agrees that ExCo members should continue to be representative and accountable to fully representative and accountable to the CGIAR and agrees with the CGIAR membership, and having and will work for the principle of cost-sharing for the cost-effi- donors share in the costs of the cient running of the System Office, including the Secretariat. CGIAR Secretariat, the Science Council and its Secretariat, and Agreed Actions. Management agrees to work to encourage the other central bodies of the CGIAR CGIAR to: System. · Ensure the representative nature of ExCo, by continuing current system of ExCo members being selected by con- stituencies. 1 8 1 T H E C G I A R AT 3 1 OED recommendation Management response · Work with all constituencies to develop consultative practices, such as those followed by the European constituency, that enable each ExCo member to serve as an authentic repre- sentative of his/her constituency. · Maintain accountability of ExCo to the Group, through cur- rent and future reporting mechanisms. · Support (through the CGIAR Chairman and Director) and monitor (through the Agriculture and Rural Sector Board and an overseeing Bank body) cost-sharing efforts and practices. Other donors, such as IFAD, have announced at the recent AGM their financial support to the System Office. Increasing the efficiency of the Sys- Management agrees on the need for efficiency gains--an im- tem from the viewpoint of gener- portant objective of the Reform Program. ating global and regional public goods, through appropriate consoli- Agreed Actions. Management agrees to work to promote within dation, decentralization, stream- the CGIAR: lining, organizing of Centers, programs, and science quality. · A continuation of the evolutionary approach to efficiency gains that have already resulted in the following consolidations: ILCA-ILRAD, INIBAP-IPGRI, IBSRAM-IWMI, and CIAT-TSBF. · The effort launched at the recent Annual General Meeting to establish a restructuring team that will propose decentralization, downsizing,outsourcing,andothernecessarychangesatISNAR. Ensuring there is a body that re- Management agrees on the need for enhanced accountability ports to ExCo with responsibility and transparency, but cannot accept the need for an addi- for (a) annual System-level audits tional layer of bureaucracy, which would be contrary to the and System-level external reviews of Bank's efforts to assist the CGIAR to streamline operations. the CGIAR every three to five years in consultation with the Science Each CGIAR Center is a legal entity in the host country. Each Council, and (b) enhancing trans- center conducts an annual audit of its operations according to parency in the reporting of the Sys- international standards and the Center Boards have the over- tem's expenditures to ensure the sight. Copies of these Audit reports are normally sent to the strategic public goods nature of CGIAR Secretariat. CGIAR research. Management will support continued objective external reviews commissioned by Science Council and/or ExCo, on behalf of the CGIAR. A system-level internal audit function is being developed, and already serves several Centers and the CGIAR Secretariat. Agreed Actions. Management will support within the CGIAR: · Due diligence activities by ExCo's Finance Committee and Pro- gram Committee. · Improvements in annual System-level financial reporting. 1 8 2 APPENDIX 8: CHAIRMAN'S SUMMARY: COMMITTEE ON DEVELOPMENT EFFECTIVENESS Main Findings · The Bank should address its corporate gover- The meta-evaluation is part of OED's review of nance responsibilities in the management of the Bank's involvement in global programs and the CGIAR, separate oversight from manage- is the first evaluation of the CGIAR to be dis- ment of the CGIAR, and exercise oversight cussed by CODE. The focus of the review is on consistent with its role in the CGIAR. the Bank and on the strategic role it has played · The Bank should lead a concerted effort at the and might play in the future to help sustain the highest level to achieve fundamental reforms CGIAR's development effectiveness. The evalu- in the CGIAR, particularly to encourage donors ation concludes that the CGIAR's productivity- to reverse the trend in restricted funding. enhancing research has had sizeable impacts · The Bank should ensure that its financial re- on reducing poverty, that further improvements sources are allocated strategically in support of in agricultural productivity are critical to meet- global and regional public goods by ensuring ing the goal of halving poverty by 2015, but that a strong, qualified and independent Science the CGIAR is less focused on enhancing agri- Council is established. cultural productivity than it used to be and that · The CGIAR should adopt a written charter that its current mix of activities reflects neither its delineates the roles, responsibilities and ac- comparative advantage nor its core competence. countabilities of the officers and bodies that Several factors have influenced the changing govern the System, as well as a mechanism to research mix, including: the biotechnology rev- reform the System's cumbersome governance. olution, genetic resource management, the It should also analyze the merits of establish- growing importance of intellectual property ing all or part of the CGIAR as a separate legal rights (IPRs), the growth of private sector re- entity. search, the rise of civil society, increasing con- · The CGIAR's strategic priorities should give cerns about natural resource degradation, policy more prominence to increasing agricultural and social issues, and the increased differenti- productivity and should reshape natural re- ation between the large and strong national source management research to focus tightly agricultural research systems (NARS) in Brazil, on productivity enhancement and sustainable China and India and the smaller and increasingly use of natural resources. weaker ones, particularly in Africa. Together with changes in funding arrangements, the cu- Bank Management recognizes that the report mulative results have been to reduce the influ- provides a comprehensive and valuable analysis, ence of independent scientific advice, a shift and concurs with most of its recommendations. from being science-driven to being donor-driven It agrees with OED that productivity should be and a shift from producing global and regional the focus, but stresses that it should be pro- public goods toward providing local and na- ductivity that is both environmentally responsi- tional services. ble and socially sustainable. It does not believe The evaluation makes a number of recom- that the CGIAR is less focused, but rather that it mendations to the CGIAR and the Bank, in- has broadened its agenda to meet the new chal- cluding: lenges it faces. Management also highlights that 1 8 3 T H E C G I A R AT 3 1 it does not "manage" the CGIAR; any reform of of closer cooperation between the CGIAR System the CGIAR needs to be agreed by the CGIAR and the National Agricultural Research Systems membership. It also disagrees that "conflicts of (NARS), with each concentrating on its com- interest and inadequate oversight constrain the parative advantage, the CGIAR at the global and Bank's leadership" and believes that the Bank has regional level and NARS at the national and local provided strategic leadership including the force level. They added that the new challenges for behind the current reform program. agricultural research, particularly issues like in- tellectual property rights, would require new Conclusions and Next Steps approaches. The Committee commended OED for an excel- The Committee agreed with the OED report lent report. Members praised the CGIAR's solid that the Bank should lead an international effort track record of achievements in providing global to reform the CGIAR, including in the area of and regional public goods and increasing agri- funding. A member suggested that the Bank be cultural productivity in developing countries. engaged at the highest level and perhaps or- The Committee endorsed OED's observation ganize an event to bring together relevant play- that the CGIAR faces many new challenges and ers to address the issues and set the long-term generally endorsed the recommendations for agenda. Several members said the CGIAR's role reform of the System in several areas. It also in agricultural research needed to be comple- made a number of observations about the rec- mented by strong support for agricultural de- ommendations, as highlighted below. Members velopment, and the Bank's agricultural strategy asked Management to take the views of CODE should inform its role in the CGIAR. They en- to the next meetings of the Executive Council in couraged the Bank to continue its efforts to pro- May and the Annual General Meeting in the Fall, mote agriculture. to work for less restricted and more strategic al- location of funds to the System, and to provide Funding. Members agreed that the issue of re- CODE with an update. stricted funding was important but complex. Many countries represented on the Board were Issues contributors to the CGIAR. While expressing The main points raised in discussion by the Com- support for more unrestricted funding and agree- mittee included: ing that the Bank should encourage donors to provide less restricted funding for the System, CGIAR Agenda and Reform Process. The Committee the committee made a distinction between "re- generally considered that while the shift in the stricted funding" as defined in the report and CGIAR's attention to natural resource manage- some donor "requirements," e.g. the priority for ment, biodiversity, etc. was linked to agricultural Africa in the use of some donors' funding. Some productivity, the focus on productivity itself donor requirements may be consistent with de- should be increased with more emphasis on the velopment needs. In this light, members shared provision of global and regional public goods. the evaluation's concern that donor restrictions Global public goods centered on productivity and have created a non-transparent marketplace for efficiency in resource use continue to be vitally global public goods research and changed the needed in developing countries and were the composition of the overall program. Another CGIAR's area of demonstrated comparative ad- member asked if the CGIAR's increased focus on vantage and core competency. Some members natural resource management types of issues suggested the need for a clearer statement of the and less on other productivity issues and CGIAR's vision and on how CGIAR activities strengthening NARS was the result of donor could better respond to the needs of develop- preferences and stated that the CGIAR's research ing countries and to the imperatives of increas- agenda should ensure that the priorities of de- ing agricultural productivity and enhancing food veloping countries were considered. The Com- security. The committee stressed the importance mittee generally agreed with the OED 1 8 4 A P P E N D I X 8 : 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 recommendation that the Bank should promote ing pilots and also to allow the Science Council within the CGIAR the implementation of the to get established before proceeding with more agreed phasing out of the matching grant fund- Challenge Programs. One member suggested ing system. that the provisionally approved pilot Challenge Program 3 (on genetic diversity) should go The Need for a Strong and Independent Science Coun- ahead; another member felt that Challenge Pro- cil. The committee generally expressed support grams 3 and 4 (on Africa), which were works in for the OED recommendation that a strong and progress, should be continued. Topics chosen for independent Science Council be established and the Challenge Programs, one member said, vested with the role and resources to establish should reflect the needs and priorities of clients, System-wide policies and strategies. Some mem- with better participation and consultation with bers, while expressing support for a stronger the Science Council in identifying topics for the Science Council, cautioned that difficult issues Challenge Program. such as accountability, oversight, and how far pri- ority-setting should be assigned to the Council Establishing the CGIAR as a Separate Legal Entity. would need to be looked at. Some members The Committee noted the advantage of a writ- said the Science Council and the Executive Coun- ten charter delineating clear roles and respon- cil should meet regularly to discuss issues of sibilities for various actors. Bank Management relevance to both groups, including a clearly de- informed the committee that establishing the fined division of responsibilities. CGIAR as an autonomous legal entity was a con- troversial issue, and generally not supported by Decentralization versus Centralization. CODE noted CGIAR members. Some members encouraged that the new challenges for agricultural research, further analysis and debate by CGIAR members including IPRs and public-private partnerships, on the pros and cons of the proposal, as sug- would require new System-level approaches; a gested by Management. One member noted that completely decentralized system could not de- for over three decades the CGIAR had achieved liver on these challenges. Members agreed that major successes without a legal charter of its a balance was needed to make the System work own; he asked about the relative merits and de- effectively, and that the approach taken should merits of continuing the status quo. depend on the activity involved. Several mem- bers said that a more centralized approach was Results Measurement and Reporting on the Work of advisable in setting clear priorities for strategic the CGIAR. The Committee discussed the impor- research at the global and regional levels, while tance of results measurement. It noted that Bank development and dissemination activities that Management was working on identifying an im- had their impact at the local level were best fo- proved performance measurement system for cused there. One member cautioned that a cen- each center in order to be able to allocate re- tralized CGIAR management could lead to sources more strategically. One member said accountability issues. Some members stressed performance should be benchmarked as much the importance of ensuring that the needs of the as possible, in particular for the proposed Sys- users of the research--the developing coun- tem-wide functions. Management agreed to share tries--should be central in developing a man- such performance indicators with the Board. agement structure for the CGIAR. Members supported Management's suggestion to have an external independent evaluation of the Challenge Programs. The Committee generally work of the CGIAR every 3 years, and urged agreed with OED that it was advisable to slow Management to update CODE on progress made. down new Challenge Programs. They noted that They noted that the luncheons held during the two pilots were ongoing, and that some others Annual Meetings were useful in acquainting the were in the pipeline. Several members said it was Board with the general work that the CGIAR important to see the initial results of the ongo- was undertaking, but did not respond to the 1 8 5 T H E C G I A R AT 3 1 Committee's desire for more issues-focused in- vide some proposals to address this issue before formation on substantive issues confronting the the next round of DGF allocations. They ex- CGIAR. pressed satisfaction that the Senior Vice President and Chief Economist would be responsible for Separating the Bank's Management and Oversight the oversight function and asked that the Senior Functions. The Committee supported OED's rec- Vice President's assessment of key issues be re- ommendation to separate oversight and man- flected in future Management recommendations agement functions within the Bank, and agreed on DGF funding of the CGIAR. that it was important to avoid any perception of conflict of interest. They welcomed the an- Finn Jonck nouncement by Management that it would pro- Chairman 1 8 6 ENDNOTES Chapter 1 strategies required to achieve its goals and objectives. 1. This includes Center-generated income of $15.7 [The CGIAR] therefore needs to spend more time million and cosponsor support of $6.0 million. See Ap- on building consensus [on mission and strategies] or pendix 1 for the current organizational chart of the build a narrow coalition and move ahead . . . rather CGIAR. than risk paralysis." 2. Interview with David Hopper, January 17, 2002. 7. See the Preface for the reasons for conducting 3. "Germplasm" is the hereditary material of plants a meta-evaluation. See the Preface and Appendixes 4, and animals that is capable of being transmitted from 5, and 6 for the sources of material on which this one generation to the next. DNA by itself is not overview, the Technical Report (Part 2), and the An- germplasm; it is only germplasm when it has the ca- nexes (Part 3) are based. pability of being transmitted. 4. "Biotechnology" has been defined by the Office Chapter 2 of the Technology Assessment (OTA) of the U.S. Con- This section is based on Chapters 8, 9, 11, 12, 13, and gress as "any technique that uses living organisms or 14. substances from those organisms, to make or modify 1. Lower-bound estimates of rates of return from a product to improve plants or animals, or to develop Maredia, Byerlee, and Anderson 2000, presented at a microorganisms for specific uses." (OTA 1989). It en- workshop of the Standing Panel on Impact Assessment compasses both "traditional biotechnology," which of the CGIAR's Technical Advisory Committee, Rome, includes well-established technologies used in com- May. Upper-bound estimates from Alston and others mercially useful operations such as biological control 2000. For a full discussion of the rates of return lit- of pests, conventional breeding of plants, animal vac- erature, estimation methodologies, biases, and other cine production, and cell and tissue culture tech- issues, see Gardner 2002. niques, as well as "modern biotechnology," which 2. The CGIAR acquired the International Center for includes recently available tools for expediting selec- Living Aquatic Resources Management (ICLARM), the tion and breeding ranging from the use of recombinant International Center for Research in Agroforestry DNA, monoclonal antibodies, molecular markers, and (ICRAF), the International Water Management Insti- transgenic techniques, to genetically engineer live or- tute (IWMI), and the International Network for Im- ganisms used to modify a variety of characteristics in provement in Banana and Plantain (INIBAP) and host plants and animals such as productivity en- created the Center for International Forestry Research hancement, growth cycles, and resistance to a variety (CIFOR). INIBAP was subsequently merged with the of environmental or genetic stresses. International Board on Plant Genetic Resources 5. The World Bank's Board of Executive Directors (IBPGR) and renamed the International Plant Genetic recently reemphasized the importance of incorpo- Resources Institute (IPGRI), leading to a net addition rating agriculture in the Bank's Country Assistance of four new Centers. Strategies. 3. The CGIAR's annual financial reports define "in- 6. CGIAR Committee of Board Chairs and CGIAR creasing productivity" as comprising germplasm im- Center Directors' Committee 2000, p. 43, acknowl- provement research and production systems research edged, "the issues reflect profound differences of on crops, livestock, fisheries, and forestry. The clas- opinion about the mission of the CGIAR and . . . sification of expenditures among activities in CGIAR 1 8 7 T H E C G I A R AT 3 1 reports must be interpreted carefully, since the Sys- 2001, p. 15). Restricted funds are defined by the tem as a whole has not maintained expenditure data CGIAR as those that are restricted either by attri- that can be used for strategic planning and manage- bution (to a particular research program or region) rial System-level decisions due to the autonomous na- or by contract (to a project, subproject, or activity). ture of the Centers and of the funding arrangements An allocation to one Center as opposed to another for the CGIAR. Policy and NRM research can also be is not restricted funding by the CGIAR's definition. important ingredients in increasing productivity, but While a Center's funds can be allocated to any pro- their impacts have not been assessed. gram or cost according to a particular Center's in- 4. NRM research is primarily undertaken in four stitutional needs or priorities, these cannot be specialized NRM Centers and four ecoregional Cen- switched from one Center to another. Different de- ters. About half of the policy research is conducted in grees of restriction result in part from donors allo- IFPRI. The balance is spread among the other 15 Cen- cating funds to the Centers from their various units ters in support of specific commodities, agroecologies, and budgets-- from their headquarters, regional, and micro-community or micro-household-level social and country-level budgets, from their agricultural and science research undertaken jointly with natural sci- environmental budgets, etc. Allocations to the new entists. Challenge Programs will likely be classified as un- 5. For example, changes in global trade flows linked restricted, unless restricted by attribution or con- to a national or regional water crisis or the loss of bio- tract. The World Bank is the only donor that allocates diversity of global significance due to poor local nat- resources to the overall System and not to Centers. ural resource management. Other donors allocate their funds to individual Cen- 6. See Barrett 2002. Yet research on smallholder ters (either directly or through a Bank-administered farming has demonstrated time and again that indi- trust fund). vidual farmers and groups adopt innovations if, and 11. That is, while the CGIAR is generally viewed as only if, they ensure sufficient returns. See IFAD 2001. an international program, it is actually, in the words 7. IFPRI has been making methodological efforts of some, a "coordinated bilateral program" in inter- to assess the impacts of policy research and has as- national guise. As the share of restricted funding has sessed the impacts of some of its own research. Yet risen, the ability of Centers to utilize resources in ac- the impacts of either policy or social science research cordance with the priorities recommended by TAC/iSC in the rest of the CGIAR System remain unevaluated, has diminished. The CGIAR needs to publish on a although some research in support of commodities transparent basis the financial and other support (in- is of high quality. cluding in-kind support by developing countries of 8. The 1994 Panel on Vision and Strategy, led by Gor- land and genetic material) to each Center classified by don Conway, recommended that the CGIAR needed to core (global and regional public goods) research, generate thousands of "doubly green revolutions" complementary (national and local) research, dis- suited to highly diverse conditions in developing coun- semination, training, and other activities. tries and sensitive to environmental issues, a theme re- 12. In 2001 the CGIAR had more than 300 donor- inforced by the TSR in stressing the importance of the funded programs in its 16 international Centers. CGIAR "walking on two legs"--productivity growth 13. While the current CDMP is said to be based on and sustainability. See Conway 1997. a vision and strategy statement that resulted from a 9. The "donor of last resort" model also had some two-year long exercise by TAC in 2000, and while the limitations, the primary ones being "a stifling of Cen- seven planks in that statement are incontrovertible, ter initiatives in raising funds and an insulation of they are too general to provide a strategic direction. some Centers from economic realities. As a result, the See TAC Secretariat 2000f. The seven planks are (1) Bank ended up providing more funding for some people and poverty focus, (2) modern science, (3) ge- Centers over a longer period of time than might oth- ographic priorities, (4) regional approach to research, erwise have been the case." (Anderson and Dalrym- (5) new partners in science and development, (6) ple 1999, p. 24.) task force approach, and (7) catalytic role. Further- 10. OED is using the same definition of restricted more, the statement has not yet been followed by funding as the CGIAR. (See CGIAR Financial Report agreed-upon long-term research priorities, now 1 8 8 E N D N O T E S needed even more urgently in the context of the 5. A detailed IFPRI study for the SGRP of the costs rapid commitment of resources to Challenge Pro- of conserving and distributing the current holdings grams discussed in Chapter 5. of the CGIAR's 11 gene banks in perpetuity estimates 14. The current CGIAR chairman has argued, and a required endowment of $149 million, although OED concurs, that there are degrees of restrictiveness plausible variations in interest rates and regeneration in funding. Thus, the allocation of the funds by the cycles of genetic materials cause this estimate to Government of Canada to Africa is not as restrictive range from $100 to $325 million. (See Koo, Pardey, as an allocation to a specific project or activity. How- and Wright 2002.) Informed sources argue that $400 ever, tying reduces the marginal returns of funds if it million to $500 million may be required to maintain increases the amount of resources beyond absorptive the gene banks. (Shands 2001; interview with Alex Mc- capacity, while leaving some important areas of re- Calla, January 2002.) The initial target of $260 million search activity under funded. for an endowment for genetic resources conservation was based on donor prospects at the time of the Chapter 3 2000 feasibility study, when the negotiations of the This section is based on Chapter 10. Treaty were not yet concluded. Moreover, some 1. The management of the genetic resources col- founders and long-time observers of the CGIAR have lections is governed by a standard agreement be- raised concerns about the risks of maintaining these tween FAO and each Center that holds material. The gene banks in developing countries, both because of System-wide Genetic Resources Programme (SGRP) the relatively higher cost of refrigeration in the high- is charged with assisting the Centers in fulfilling their temperature, high-humidity tropics, and the threat obligations under the FAO agreement. The SGRP is ad- posed by conflicts (interview with David Hopper, ministered through IPGRI as the convening Center. January 17, 2002). 2. Commission on Intellectual Property Rights 6. A recent report of the U.S. National Academy of (2002), London, September, chaired by John Barton, Sciences makes a strong case for strengthening science George E. Osborne Professor of Law, Stanford Uni- advice procedures within UN agencies, without which versity. it argues they face the risk of losing both effectiveness 3. These include (1) international negotiation of IPR and credibility. But the report fails to deal with the policies and their implementation to ensure global IP murky territory between science and politics in such systems contribute to the development of developing areas as GM crops. See U.S. National Academy of Sci- countries by stimulating innovation and increasing ences 2000; Norman Borlaug 2003. their access to technology and its products at com- 7. The erosion of the nonpolitical nature of the petitive prices, (2) learning from the international CGIAR is evident in its handling of the politically experience, including understanding the rapidly evolv- charged GMO debate. ExCo decided that the Science ing nature of international public-private partner- Council, which is expected to speak on major scien- ships, (3) assessing the implications of intellectual tific issues of global importance, should not handle property rights for the Convention on Biological Di- GMO issues, but should be left to a special standing versity, (4) strengthening the capacity of international panel on genetic resources established in 1994 to and national institutions involved in intellectual prop- monitor and analyze developments in genetic re- erty rights issues, and (5) supporting the interna- sources policy, internationally and within the System. tional architecture. It is to recommend CGIAR action as necessary, in- 4. The most significant partnerships to date are con- cluding the engagement of the CGIAR in the negoti- centrated in Brazil, China, and India, and are one of ations of the FAO Commission on Genetic Resources five types: basic and applied research led by the pub- for Food and Agriculture that led to the adoption of lic sector; outsourcing of private sector research to the International Treaty on Plant Genetic Resources public institutions; joint public-private ventures in for Food and Agriculture in 2001. ExCo has observed applied research; public partnerships with privately es- that the use of GMOs is not acceptable to certain tablished research foundations; and technology trans- CGIAR members. fer systems between the public and private sectors. See 8. The CGIAR established a private sector com- Spielman 2002; Pray 2001. mittee and an NGO committee that reported to the 1 8 9 T H E C G I A R AT 3 1 group as a whole, and gave the chairs of these two to $2.5 billion of loan/credit funds to support devel- committees a place at the plenary table during In- oping countries' agricultural research and extension ternational Centers Week and the Mid-Term Meet- systems. But, with the lack of effective demand for in- ings. Although the CGIAR membership currently vestment in long-term agricultural research from most does not include the commercial private sector or developing countries, this offer was taken up only in NGOs, their two representatives are full members of a few countries such as Brazil, India, and eventually the Executive Council. According to the CGIAR, its China, countries already committed to agricultural membership consists of "international organizations, research. governments, and private foundations that support 4. Since other donors were unwilling to share the the mission of the CGIAR, participate in policymak- $1.2 million cost of the TSR, which the CGIAR donors ing, and provide support for the conduct of research considered excessive, the Bank paid the full cost from at the 16 international Centers." But what determines its annual DGF grant. The meta-review considers the which members are in good standing, and their roles, cost of TSR in line with similar reviews of major pro- responsibilities, and privileges vis-à-vis non-mem- grams, although it could have benefited from a bers, is not clear--an increasingly important issue in stronger analytical basis. a rapidly growing partnership. The NGOs have raised 5. Adel El Beltagy, "Center Directors' Comments on some important issues about benefit-sharing in the the Draft Meta-Review," and further exchanges with use of genetic resources with developing countries. El Beltagy, Per Pinstrup-Andersen, Hubert Zandstra, However, they recently indicated they are "stepping Andrew Bennett, Dana Dalrymple, and Klaus Winkle. back and freezing their membership" because of the 6. While the 1994 merger of two livestock Centers "refusal of CIMMYT to acknowledge GM [genetically in Africa has been the largest consolidation to date, modified] contamination in the Maize Centre," and it has not been evaluated and is often perceived to the entry of Syngenta as a member of the CGIAR. Dur- have been disruptive and of limited value. OED con- ing their review, they will not accept money from cludes that this merger helped strengthen the CGIAR's the CGIAR or sit on the Executive Council or any ability to address global livestock issues. A setback in other committee. making ILRI a global research institution has been that its funding declined by about 10 percent after the Chapter 4 merger, despite TAC's recommendation for more re- This section is based on Chapter 15. sources. 1. While Renewal was adopted by the member- ship, largely because they viewed this as a way of mo- Chapter 5 bilizing additional funds for the System, it had no This section is based on Chapter 16. formal status due to the informal nature of the Sys- 1. The CGIAR chairman frequently stresses the tem. It lacked an enforcement process beyond the stakeholder ownership of the current reform process. Bank's own contributions to Renewal, moral suasion, Interviews by the meta-evaluation team indicate that and selected incentives. ownership is greater of both the process and the out- 2. Increased developing-country membership and comes of the current reform process compared with ownership has not yet resulted in their increased the TSR, but the meta-evaluation team is less certain contributions to the CGIAR System, or even to fi- of this after the AGM 2002 meeting. nancing the overheads of regional research programs, 2. Particularly the important Finance and Program except for a few exceptions such as Colombia. In this Committees. case the funding is largely to contract out research to 3. The civil society position is temporarily vacant. CGIAR Centers due to the increased weaknesses of See note 8 to Chapter 3. Colombia's national research system. See Romano 4. Only 9 of the CGIAR's 22 developing-country 2002. members met this criterion when the inaugural ExCo 3. The Bank also contributed an additional $10 was established. million in both 1994 and 1995, above its normal fund- 5. Other steps suggested to increase participation, ing, to make up for the shortfall, and the CGIAR Chair- ownership, responsibility, and accountability in ExCo man made clear the Bank's willingness to provide up include: (1) alternating the chairmanship of ExCo be- 1 9 0 E N D N O T E S yond the CGIAR chairman, (2) increasing the tenure which are less appealing politically and are less able of ExCo members beyond two years and designating to generate donor funds. alternative members to ensure participation in ExCo 9. The quality of science must be one considera- meetings, (3) advanced scheduling of ExCo meetings tion, but equally important is the extent to which the three or four times a year until ExCo begins to func- CGIAR is engaged in global and regional public goods tion effectively instead of the current reliance on vir- research with potential for large spillovers and impacts tual communication, and to link ExCo meetings better on poverty, and which only the CGIAR can perform. to the AGM agenda and decisionmaking processes, (4) This requires knowledge of both science and the discussing the agendas of ExCo meetings with ExCo complex policy, institutional, and development envi- members and the Science Council Secretariat prior to ronment. their finalization, and (5) improving preparation for 10. According to OED's survey of stakeholders, ExCo and AGM meetings with the necessary time for there is no agreement on whether the Science Coun- discussion, including position papers prepared by cil should have the lead in priority setting--a huge the Science Council outlining the pros and cons of im- change from the initial years of the CGIAR (box 4.2). portant decisions to be recommended by ExCo to In general, TAC members (86 percent), NARS (83 per- the membership. cent), and the TSR team (67 percent) favor a lead 6. The 10 units are the CGIAR Secretariat, the iSC role for the Science Council, while OECD members Secretariat, the Future Harvest Foundation, the As- (56 percent), Center directors (50 percent), and Board sociation of International Agricultural Research Cen- chairs (45 percent) are more ambivalent. ters, the Gender and Diversity Program, the Internal 11. The Challenge Programs are to have their own Auditing Unit, the Central Advisory Service for Intel- independent governance mechanisms and program- lectual Property Rights, the CDC Executive Secre- related science quality assessments, separate from tariat, the Chief Information Officer, and the Strategic the Science Council. Meanwhile, there seems to be a Advisory Service on Human Resources. consensus that the interim Science Council did a 7. In an interview, Robert McNamara indicated that good job in reviewing the Challenge Program pro- he insisted on endowing the CGIAR with a strong posals in 2002. TAC, led by an eminent Chairman, to guide allocations 12. "Pilot" refers to the CGIAR's decision at AGM of donor resources. He felt most donors would lack 2001 "to accelerate, on a pilot and one-time only basis either the technical knowledge or the dedicated sup- the preparation of up to three Challenge Programs so port to enable them to make informed decisions on that the System can explore ways of improving Chal- complex issues of global science policy, research pri- lenge Program design and implementation." Although orities, allocations, and impacts. Strong TAC leader- it was recommended by the interim Science Council, ship with credibility and independence, he argued, was ExCo did not recommend the third pilot Challenge and must remain the hallmark of the CGIAR. Putting Program, "Unlocking Genetic Diversity in Crops for the TAC in the FAO was intended to increase legitimacy Resource-Poor Areas," for approval to AGM on the in the establishment of research priorities by ensur- grounds that it was insufficiently developed. This pro- ing a voice for developing countries and to balance the gram, along with a fourth that received support at AGM voices of donors with those of clients. 2002 ("Improving Livelihoods and Natural Resources 8. Stakeholders also point to "political correct- Management in Sub-Saharan Africa") will be devel- ness" in TAC representation rather than scientific ex- oped further and resubmitted under the regular cellence, and the fact that CGIAR Chairmen and donors process for Challenge Program approval. have grown weary of a strong, assertive TAC, such as 13. At AGM 2002, Austria, France, Norway, and the existed in the past. But when disputes have arisen on Netherlands strongly supported moving ahead full priority setting and have been referred to TAC, its speed with the Challenge Programs, while Brazil, recommendations have generally been consistent Canada, Denmark, Germany, South Africa, the Philip- with the CGIAR's comparative advantage, e.g., its de- pines, the United Kingdom, the United States, and the cisions with regard to AIDS research, its repeated Rockefeller Foundation favored a more cautious learn- stress on the declining allocations to commodities, and ing-by-doing approach that would review the impli- activities (livestock) of importance to the poor, but cations of System-wide priorities for the choice of 1 9 1 T H E C G I A R AT 3 1 Challenge Programs. As reported in the minutes of financed 100 percent by partners. The CGIAR Secre- AGM 2002, the CGIAR chairman summarized the dis- tariat points out in its FY03 application for DGF funds cussion saying that he "sensed a consensus for sup- that when the entire System overhead is taken into ac- port for CPs as a concept, willingness to approve the count (namely, the entire costs of the System Office two pilots endorsed by ExCo, a desire for prudence currently under creation), the DGF administrative in going forward in terms of priorities within a strate- burden is less than 100 percent. gic framework, and concerns about cannibalizing 2. Besides, the Secretariat lacks the necessary ex- funding." pertise to perform all these functions, and some have 14. The Netherlands has announced $25 million of shared a view with the meta-evaluation team that it new money to Water and Food, the United States has poses the risk of undue concentration of power in that targeted funding for Biofortification, and Canada, office. Sweden, and the U.K. have also expressed interest in 3. During the CDMP, when the current chairman Challenge Programs. explicitly requested that the role of the chair be ex- 15. The minutes of the AGM 2002 report that mem- amined, CGIAR donors rejected the notion of re- bers said "the CGIAR does not have a clear strategic moving the chairmanship from the Bank. This may be framework for evaluating Challenge Programs. The because most donors have reduced their internal ca- `seven planks' of the Vision Paper do not constitute pacity in agriculture and sciences and have delegated a strategy. The CGIAR should not continue appraising greater responsibility to the Bank. That said, the Bank the Challenge Programs in sequence, without a clear has also reduced its capacity in agriculture, having lost strategic framework." See note 13, Chapter 2. 60 of its 140 technical staff in the agriculture sector. 16. The latest Center medium-term plans and pro- The Bank's Regional operations now have a stronger jections posit a continuing decline in activities of the role in lending than the former sector operations, commodity Centers, as also highlighted by the iSC in and both lending to agriculture and agricultural pol- its comments on the 2003­05 Medium-Term Plans icy analytical capacity in Bank's Regional operations considered at AGM 2002. have dwindled. 17. In a news release issued at AGM 2002, the 4. Based on OED's recommendations in World Water and Food Challenge Program declared that it Bank 2002a, Bank management is in the process of in- seeks $100 million in contributions during the first five- stituting reforms in the oversight and management of year phase, with even higher contributions expected its global programs. These will help establish Bank- in a second phase. wide strategies and priorities; increase independent 18. While in transition, the iSC has over a 12-month oversight and accountability; increase the voice of period reviewed 10 pre-proposals in the pilot process, developing countries in global program governance short-listing 3; reviewed and approved the 3 Chal- and management; improve linkages to the Bank's lenge Programs; reviewed an additional 41 ideas and country operations; increase routine quality assur- short listed 13 pre-proposals in the regular process; ance, monitoring, independent evaluation, and follow- and reviewed and endorsed those 13 pre-proposals for up; and strengthen Bank Executive Board and partner full proposal development. inputs. 5. IDA deputies have recently authorized the use Chapter 6 of grants under IDA 13, in the range of 18 to 21 per- This section is based on Chapter 17. cent of IDA 13 resources, in the following five areas: 1. The guidelines go on to state that after no more (1) HIV/AIDS, (2) natural disasters, (3) post-conflict than three years, a decision should be made to either countries, (4) poorest countries, and (5) debt-vul- move the Secretariat out of the Bank, keep it in the nerable countries. While this provides only limited Bank with strong donor support, or discontinue the scope for providing IDA grants to finance regional effort due to lack of donor interest or other reasons. and national agricultural research programs in the In exceptional cases, where there is strong donor in- poorest and debt-vulnerable countries in order to get terest in maintaining an in-house Secretariat in the beyond short-term, small-scale funding of agricul- Bank after three years, then this Secretariat should be tural research, it does open the door. 1 9 2 E N D N O T E S Chapter 8 which more than 100 participated in the CGIAR Busi- 1. Includes Center-generated income of $11.0 mil- ness Meeting itself) has been estimated at more than lion and an estimated $6.0 million of cosponsor sup- $4 million. Second, these accounts do not include port. in-kind contributions of land or buildings from de- 2. See Annex A, Part 3 of the Study (www.world veloping countries for the Centers and their regional bank.org/ped/cgiar), for the specific mandate of each offices, genetic materials, funds devoted to collabo- of the 16 Centers. rative research, and time spent by officials of the 3. CGIAR Secretariat 2000e. The primary role of the NARS on group matters. cosponsors has been to impart international legitimacy 6. Although a finance committee existed, several and an assurance of continuity to the CGIAR. The informed former members of that committee have in- World Bank, FAO, and UNDP have been cosponsoring dicated that it did not function well, and TAC mem- agencies since the CGIAR's founding; IFAD became a bers have pointed out the lack of transparency evident cosponsor in 2001, largely to help assert the CGIAR in decisions made by the finance committee that were poverty agenda. For a short time during the 1990s, the contrary to TAC recommendations. United Nations Environment Programme (UNEP) was 7. A modest (less than 1 percent) proportion of also a cosponsor. UNDP and FAO have not been able these are in-kind contributions, usually technical as- to maintain their financial commitments, UNEP sistance (scientific experts), from industrial countries. dropped out due to funding shortages, and IFAD in- This does not include cosponsor support of $6.0 mil- dicated when it joined that it would not contribute lion from the World Bank, FAO, and UNDP to cover the funds beyond what it already did. This situation has operating costs of the CGIAR Secretariat and TAC/SC, reduced the funding for TAC/iSC, making it more un- nor Center-generated income of $15.7 million in 2001. predictable and more dependent on the World Bank, 8. Although the Centers carry out long-term re- affecting in part the ability of TAC/iSC to perform, search, multi-year commitments are uncommon for whereas funding for the CGIAR Secretariat from DGF unrestricted funding commitments, while restricted has been stable and predictable. FAO's role has recently funding is often a multi-year commitment. Funding of increased as a trustee of the 600,000 accessions of plant Challenge Programs is discussed in Chapter 16. genetic material that the CGIAR Centers hold in trust 9. The 1993­94 funding crisis and the 1994 finan- and with the passage of the International Treaty on cial reforms are discussed in greater detail in Chapter Plant Genetic Resources for Food and Agriculture 15. (2001) (See Chapters 15 and 16 for further discussion 10. These three roles of the World Bank in relation of these organizational and management issues). to the CGIAR are also discussed in greater detail in 4. See Chapter 15 and Annex D of Part 3 for further Chapter 17. details. 11. The CGIAR's annual financial reports define "in- 5. Annex E, table E-1, presents the sources and uses creasing productivity" as comprising germplasm im- of funds for the overall System for calendar year 2001 provement and production systems research on crops, (equal to the fiscal year), which consolidates the Sys- livestock, fisheries, and forestry. The classification of tem-level accounts presented in table E-2 and the expenditures among activities in CGIAR reports must combined Center-level accounts for the 16 Centers in be interpreted carefully, since the System as a whole table E-3. These accounts do not include two additional has not maintained expenditure data that could be sources of funds. First, while the System pays for the used for strategic planning and managerial System-level CGIAR chairman, TAC members, and Center officials decisions because of the autonomous nature of the to attend meetings of the System, such as International Centers and Center-level funding arrangements pre- Centers Week (ICW) and the Executive Council (ExCo), ferred by the donors. Policy and NRM research can also most CGIAR members pay their own way to these be important ingredients in increasing productivity, but meetings, and donate their staff time to the System their impacts have not been assessed. throughout the year. The total collective cost of AGM 12. The CGIAR only provides data on Centers' ex- 2002, held in Manila from October 30 to November 3, penditures by research activity for agreed agenda 2002, which included well over 800 participants (of funding--that is, not including non-agenda funding. 1 9 3 T H E C G I A R AT 3 1 Therefore, the growth rates in figure 8.5 include an 19. World Bank 2002b, p. 110. This definition of core adjustment for the redefinition of agreed agenda and and complementary goods is similar but not identical non-agenda funding over this time period, in order to to the CGIAR's classification of core and comple- facilitate a ready comparison of the growth rates in the mentary research activities up to 1994. figure 8.5 with those in figure 8.4, which include non- 20. For example, India's small surface irrigation sys- agenda funding. tems have widely adopted drip irrigation methods pi- 13. See Kaul, Grunberg, and Stern 1999. OED is also oneered in Israel. Systems in northern Nigeria have addressing these issues in its overall review of global adopted the small surface irrigation systems used in programs. India. 14. The Bank's Development Grant Facility (DGF) 21. See Chapter 6, note 5. was established in 1997 to consolidate the manage- 22. For a detailed treatment of CGIAR evaluations, ment of existing arrangements for grant making under see Anderson and Dalrymple 1999, pp. 35-47. a single umbrella and to ensure that the use of Bank 23. Bruce Gardner, a member of the meta-evalua- grant resources is closely aligned with sector and tion team has observed that the CGIAR has had more Bank-wide priorities. As discussed in more detail reviews, particularly at the Center level, than is typi- below, three main criteria for receiving grants rec- cal of the U.S. agricultural research system, including ommended by OED in 1998, and subsequently for- its land grant college system. See Gardner 2002. malized in OP 8.45, are (1) subsidiarity, (2) arm's-length 24. The Review did not reflect the views of USAID. relationship, and (3) exit strategy. 25. Change Design and Management Team 2001a. 15. The recent controversy over genetically mod- The CDMT was chaired by Margaret Catley-Carlson, for- ified organisms (GMOs) demonstrates the concerns mer president of CIDA. over both positive and negative spillovers that the 26. See Annex J for the composition of CDMT and CGIAR will increasingly have to address. its Steering Group. 16. Dalrymple 2002b cites Byerlee and Traxler's 27. The CDMT's response to how its recommen- findings that economies of scale are common in re- dations address each of these issues can be found in search areas that require substantial fixed investment, Change Design and Management Team 2001a. e.g., in laboratory infrastructure. Economies of scale 28. Change Design and Management Team 2001. are less common in research areas requiring field- work and adaptation to local environmental condi- Chapter 9 tions, such as natural resource management research. 1. For a detailed discussion of germplasm research See also Barrett 2002. in the CGIAR, see Gardner 2002. 17. Henceforth in this report, unless indicated 2. See various papers on why investment in the otherwise, the term "global public goods" includes CGIAR is in danger if returns to research are so high, both international and regional public goods as well. presented at the International Conference on Im- 18. Merit goods are goods whose value derives pacts of Agricultural Research and Development, Feb- not simply from the economic norm of consumer ruary 4-7, 2002, San José, Costa Rica. sovereignty, but from some alternative norm that 3. Lower-bound estimates from Maredia, Byerlee overrides rational choice by individual persons or, in and Anderson 2000. Upper-bound estimates from Al- the case of foreign assistance, individual nations. See ston and others 2000, p. 62. For a discussion of the Musgrave 1987. The concept of merit (or demerit) rates of return literature, estimation methodologies, goods should not be confused with that of public biases and other issues, see Gardner 2002. goods, since it transcends the distinction between 4. Eicher and Rukuni 2002, pp. 26-28. Eicher and public and private goods (based on non-rivalry and Rukuni cite studies that place estimates of the rates of non-excludability). When donors direct development return to (a) hybrid maize research in Kenya at ap- assistance to certain uses, rather than providing pure, proximately 68 percent between 1953 and 1988, and untied assistance to developing countries, they are im- (b) wheat improvement programs in Africa at 23 per- plicitly attaching merit to their own preferences, cent between 1961 and 1991. These studies, while in- whether the assistance is tied to the provision of pub- corporating periods that predate the CGIAR, are lic or private goods. nonetheless assumed to include some level of CGIAR 1 9 4 E N D N O T E S contribution. Still, Eicher and Rukuni's general findings 13. For a discussion of the creation of ILRI, see box on returns to research on these and other food crops 15.1. are somewhat discouraging, fraught with problems 14. Gardner 2002. Empowerment also receives de- such as unreliable data, episodic agricultural stagnation, tailed treatment in IFAD's (2001) report on the chal- selectivity biases, and methodological issues. lenge of ending rural poverty, and places agricultural 5. IFAD 2001; Interview with Michael Lipton, citing research as one of many means to accomplishing this Martin Ravallion, June 24, 2002. end. The IFAD report takes seriously the objections 6. For evidence of these distributional effects in raised by some that technological innovations have ad- countries as varied as China, Colombia, and Kenya, see vantaged the relatively prosperous and well situated Eicher and Rukuni 2002; Romano 2002; Ndiritu 2002; more than the poor, and hedges its discussion of con- interview with Scott Rozelle. servation and environmentally aimed innovations, 7. Interview with Pedro Sanchez; Eicher and Rukuni and commercially oriented technical progress such as 2002. in exported crops. It therefore comes as all the more 8. Mundlak 2001 and Fulginiti and Perrin 1993 are powerful a judgment when the report includes crop among those who have estimated that reduced com- variety improvements as perhaps the key factor in modity prices in developing countries significantly re- rural poverty reduction to date, and points with alarm tard technology adoption and productivity growth. It will to declining rates of increase in yields of staples (ce- be a sad day if OECD countries cut back on funding for reals, roots, and tubers) in the developing world gen- CGIAR research on the grounds that the returns to erally and particularly in Africa. Hence the strong adoption of new technology have fallen, when it is in recommendation of the report for enhancement and substantial degree the commodity policies of the OECD refocusing of agricultural research effort on yield im- countries that are responsible for the fallen returns provement (not distinguishing NARS and CGIAR Cen- (Gardner 2002). Examples of issues arising from trade ters, presumably meaning both). liberalization are discussed in the background papers 15. In this context, the CGIAR Secretariat providing on Kenya (Ndritu 2002) and Colombia (Romano 2002). financing to the gender program addressing issues of 9. For a detailed discussion of assessing the envi- addressing issues of Center staffing and composition ronmental impacts of productivity growth in agricul- for the first time in 2002 may be a regressive step. ture, see Maredia, Pingali, and Nelson 2002. 10. See Angelsen and Kaimowitz 1999 and Lele Chapter 10 and others 2000. However, it is difficult to argue 1. For a detailed discussion of genetic resource strat- against continued research on productivity improve- egy and management in the CGIAR, see Lesser 2002. ment, as it is to argue against trade and investment lib- 2. According to Koo, Pardey, and Wright 2002, the eralization for countries where economic growth and CGIAR collection is housed in 11 Centers, but Shands increased incomes and employment are crucial. 2001, pp. 139­144, lists 12 Centers. Rather, to develop appropriate policies and strate- 3. Wheat and rice research conducted by CIMMYT gies needed to ensure that the most valuable land and and IRRI has generated spillover benefits to U.S. agri- biodiversity are not lost to agriculture nor the sources culture that are estimated to exceed USAID contri- of incomes for the indigenous people who depend on butions to the CGIAR many times over. See Pardey and them, the increased study and understanding of the others 1996. Moreover, land saving attributed to gross impacts of germplasm research and macroeconomic increase in total production from the 1960s to the and sector policies is needed. See also Lele 2002a. 1990s realized by seven CGIAR-mandated commodi- 11. Ladha, Fischer, Hossain, et al. 2000. See also ties (and primarily attributable to germplasm im- Tiongco and Dawe 2002. However, Gardner 2002, provement) are estimated at 100­300 million hectares and Keijiro Otsuka (interview, February 2002) do not in developing countries and 130­340 million hectares find unambiguous evidence of yield growth slow- worldwide, indicating that between 12 and 23 percent downs from recent studies of the topic. of the land savings were realized in regions beyond 12. Maredia, Pingali, and Nelson 2002. See Chap- developing countries. See Nelson and Maredia 2001. ter 17 for discussion of the Challenge Programs in this 4. In responding to the first external program regard. and management review of the CGIAR System-wide 1 9 5 T H E C G I A R AT 3 1 Genetic Resources Program in 1999, the Intercenter IPGRI's influence over the International Undertak- Working Group on Genetic Resources observed, ing, see Sauvé and Watts 2002. "Centers are facing severe financial constraints in 8. See, for example, Shands, in Gerrard, Ferroni, fully meeting international gene bank standards, as and Mody 2001. Also, an interview with Alex McCalla, revealed by the gene bank operations review." Given January 2002. that resources will always be limited for characteri- 9. Interview with David Hopper, January 17, 2002. zation, the CGIAR needs a System-wide prioritization Colombia, Côte d'Ivoire, India, Mexico, Nigeria, Peru, and process in the characterization as well, rather than the Philippines --each a host to a CGIAR Center--have leaving fund-raising for the purpose to individual all experienced armed conflicts or other intrusions. entrepreneurial scientists and the Centers. This prob- 10. See Romano 2002; Katyal and Mruthyunjaya lem is more acute at the national level, where gene 2002. For instance, discussions with Indian and Brazil- banks in developing countries are reported to have ian scientists reveal developing country concerns that greater funding and operational problems. Certainly Centers find bureaucratic ways of constraining access a significant amount of the accessions in the na- as a means of reducing competition among the Cen- tional gene banks are believed to have perished, but ters or between Centers and NARS in the generation exact data are unavailable. In 2002 the Bank in- of improved material. creased its allocation to gene banks, but more is 11. As currently set forth, more funding is expected needed. to go toward a sustainability fund. 5. See comments from the Intercenter Working 12. CGIAR's gene banks qualify for support from Group on Genetic Resources on the Report of the First the Bank's unrestricted funding to the CGIAR. External Review of the System-wide Genetic Resources 13. OED 1995, 1996. U.S. advice was instrumental Programme (1999). in convincing India to undertake a variety of public in- 6. Certain European donors are also concerned terventions in agriculture to address market failures. that supporting genetic resource management in the This included the establishment of the Food Corpo- CGIAR implies support for the use of biotechnology, ration of India to provide price supports, to expand genetic engineering, and research in transgenic crops public sector fertilizer imports and distribution, and and other organisms, a research area of concern to greatly accelerate the supply of agricultural credit to them or their constituents. In fact, sound manage- the farming communities. Some of these were con- ment and use of genetic resources in the CGIAR Sys- ditions of the Bank's lending in support of the Green tem has, to date, been more closely associated with Revolution. The support India received from the conventional plant breeding and germplasm en- Bank--investments in irrigation, balance of payments hancement, an area of comparative advantage as ex- support for fertilizer imports, and other assistance-- plained earlier. was greatly accelerated by having a TAC chair who was 7. In 1994, most of the CGIAR collections were the advisor to the President of the Bank, close ties with placed under the auspices of the FAO International Un- the founders of the CGIAR and the Rockefeller Foun- dertaking on Plant Genetic Resources to constitute dation, and a background in agricultural develop- part of the international network of base collections. ment. Interviews with David Hopper, Robert Picciotto; System needs and procedures were identified in the U. Lele 1994. Global Plan of Action from the Leipzig Conference of 14. A CIMMYT study of maize seed markets ex- 1996. The International Undertaking is a voluntary in- amined seven Asian countries: China, India, Indone- strument and precedes the Convention on Biological sia, Nepal, Philippines, Thailand, and Vietnam. The Diversity. Its re-negotiation to bring it into harmony figures given above exclude China, given the dominant with the Convention ended with the adoption in 2001 role played by state-owned enterprises in China's of the new International Treaty on Plant Genetic Re- agricultural input sector. If China is included in this sources for Food and Agriculture as a legally binding figure, the private sector share of maize seed sales (in instrument. When it comes into force (after the 40th terms of metric tons) falls to 41 percent. See, among ratification), the Treaty will supercede the Undertak- others, Gerpacio 2002. ing, including in relation to the collections held in trust 15. For example, many crops require monitoring by the Centers. See Lesser 2002. For an evaluation of to delay resistance development and possible gene 1 9 6 E N D N O T E S transfer. There is no evidence that the Centers have their chairs in the Consultative Council. The Chair of systematically considered the issue or their role. the CDC indicated, "decision not to pursue a separate 16. Letter to the TAC chairman, September 1998. legal entity for IP followed a series of consultations in 17. "Consistent policies are needed if the CGIAR which the Centers which were generating IP that is to function as a unified System. As it is, the CGIAR might be patented considered different options and currently offers, in the view of the Panel, a bewil- received counsel from many sources. The Centers dering array of policy-making and policy influencing decided to continue their IP audits and complete forums and personalities. These need to be stream- them as soon as possible, and to set up a an advisory lined. The CGIAR is faced with the need for the Cen- service (at ISNAR) to assist Centers on IP issues and ters to move away from a history of autonomy and provide referrals, as required, to qualified legal coun- independent management to a System-wide culture sel and other specialists. In this process, each of the with its attendant structures and partners and chang- Centers involved carried out its own soundings, and ing philosophy that recognizes the value of working the total effort was kept under review by CDC which together to address common goals. The panel is con- did not maintain a record of Center-by-Center con- vinced that for the CGIAR to play its full role in the sultations." genetic resources arena, a System-wide program is 22. John Barton is George E. Osborne Professor of needed... and that the existing SGRP needs to be Law at Stanford University. Other members of the changed, and while there are several options from an commission include Daniel Alexander, Carlos Correa, organizational standpoint, the status quo is not one Ramesh Mashelkar, Gill Samuels, and Sandy Thomas. of them." System-wide Review, p. xi, around recom- 23. A similar proportion of respondents supported mendation 18. a routine System-level monitoring and oversight sys- 18. Centers working with rice and maize have pro- tem, along with a process of annual reporting to gressed in the use of biotechnology applications, CGIAR membership on the implementation of a Sys- partly because of the greater availability of geneti- tem-level IPR policy by Centers. In comments to the cally advanced technologies for these crops worldwide meta-evaluation team, respondents also expressed and relatively better funding for breeding activities for the need for a flexible IPR policy regime that permits these crops in the CGIAR. CIMMYT is testing new va- Centers to address IPR issues in a manner appropri- rieties, but to date no products are ready for the mar- ate and specific to individual products, countries, and ket. CIP is also working with biotechnology, although clients. See Annex Q, Part 3). potato breeding has also experienced significant pri- 24. Indeed, scientists from advanced countries in- vate sector investment. terviewed by the meta-evaluation team assert that re- 19. For example, in view of the thriving private searchers in their countries will only enter risky areas sector seed activity in the hybrid maize market in of enquiry with potentially high returns provided Asia, should the CGIAR reduce its varietal development long-term funding is assured. of hybrid maize or rely on the private sector to com- mercialize its research, thereby allowing the System Chapter 11 to focus more on the open-pollinated varieties used 1. For a detailed discussion of policy research in the by poor farmers? CGIAR, see Gardner 2002. 20. One exception is Gerpacio 2002. 2. Recognition of the centrality of IFPRI is evident, 21. CGIAR Secretariat 2002j. According to the for example, in ISNAR's 4th EPMR, which recom- CGIAR Secretariat at ICW 1998, decisions on imple- mended that its policy research be transferred to menting the TSR recommendations were delegated to IFPRI. the Consultative Council, with the chairman requesting 3. Interview with David Hopper. a number of committees to come up with proposals 4. A special team under the chairmanship of a TAC for implementation. (The committees asked to work member felt strongly that IFPRI should move from its on implementation included the CBC, CDC, the Fi- location in Washington, D.C., to a developing coun- nance Committee, GFAR, the Oversight Committee, try to "place its research staff in an environment which the CG Secretariat, and TAC.) The NGO Committee would be more relevant to the objectives of the in- and the PSC held special sessions for participation by stitute, avoid the perception of IFPRI having a some- 1 9 7 T H E C G I A R AT 3 1 what privileged status in the CGIAR and could also pro- ingly requires close interaction between both functions tect IFPRI from undue donor influences and demands (Barrett, Place, and Aboud 2002). in its analysis of the world food problems" (Baum 3. Ndiritu 2002 stresses the importance of such fi- 1986, p. 139). nance in Africa, but the reluctance of donors to finance 5. CIFOR's policy research was covered in the OED regional and subregional research organizations, de- Evaluation of the World Bank's implementation of spite efforts on financing, is in part the result of the the 1991 forest strategy and is not covered here. See mixed records of the ROs and SROs. Lele and others 2000. 4. In 2002, System-wide initiatives received only 35 6. Ryan 1999b. Despite very substantial efforts in percent of the funding they indicated they required. capacity building, with many tangible products in pa- 5. ICRISAT indicated to the meta-evaluation team pers, training, and advice, the measurable impacts of that despite such successes, recognized in part with the kind estimated for Vietnam were essentially nil in three King Baudouin awards in the past eight years, Malawi. Credit is given for engagement in the food pol- and despite a mandated region with one of the high- icy process at the beginning of the 1990s, but IFPRI's est incidences of poverty, the Center has faced the presence was not apparent in more recent important largest annual reduction in its budget (8.5 percent) rel- policy debates. ative to other Centers. 7. For example, Anderson, Moscardi, and Pardey 6. Definitions used by international and bilateral de- (1994) evaluated the CIMMYT Economic Program velopment agencies and other stakeholders vary, but (CEP), criticizing adoption studies done in the CEP as have converged to include a holistic approach toward not contributing beyond the findings already available sustainable development. Approaches range from meth- in the literature on wheat and maize (p. 3), finding their ods based on rational management of chemical pesti- farm-level adoption studies questionable, and recom- cides to systems based on ecosystem management that mending that CIMMYT cease doing adoption studies include health issues and human capital development. -- while at the same time commending CIMMYT's The remaining disagreements are with regard to the use global impact assessments, e.g., Byerlee and Moya of chemical pesticides in IPM approaches and the role 1993. of GMOs in breeding for pesticide resistance. See Sorby, 8. In 1997 and 2001, IFPRI held conferences in Fleischer, and Pehu 2003, p. ix. which ideas from a broad range of social scientists were 7. See Chapter 16 for further discussion of System- solicited on the question of how to measure the ben- wide priorities, consolidation, and the Challenge Pro- efits of policy-related social science research. Some of grams. these papers contained estimates of impact as well as 8. It is perhaps instructive that recent research discussion of ways to measure such impacts. using WARDA data (Sherlund, Barrett, and Adesina, forthcoming) finds that when one accounts for natu- Chapter 12 ral resource conditions in estimating the technical 1. For a detailed discussion of natural resource efficiency of rice farmers in Côte d'Ivoire, the me- management research in the CGIAR, see Barrett 2002. dian farmer appears to operate on or quite near the 2. In this chapter, national agricultural research production possibilities frontier, achieving maximal systems (NARS) may also be referred to as national output given inputs. By contrast, if one fails to con- agricultural research and extension systems (NARES). trol for environmental production conditions appro- The inclusion of the "E" for extension in NARS is not priately, technical inefficiency estimates rise sharply, meant to imply a broadening of the mandate of the suggesting significant forgone output. Management of CGIAR beyond the research domain. The inclusion of latent variability in natural resource conditions mat- the "E" is meant to reflect both that there have been ters fundamentally to agricultural productivity. some successful Center partnerships with national 9. "The Panel found that some CIFOR researchers extension services that could prove instructive (no- are frustrated with the IPG [International Public Goods] tably by WARDA and IITA in West Africa) and that the criterion because they view it as being associated with model of research-extension interaction is evolving, shallowness of research at the specific site level. The perhaps especially in NRM, where the development Panel wishes to point out that this is (or should be) a and dissemination cycle for best practices increas- false assumption. The IPG nature of CIFOR's work 1 9 8 E N D N O T E S should not make it incompatible with in-depth re- 3. Katyal and Mruthyanjaya 2002. At the same time, search on particular sites. In fact, significant general- many Centers and donors have indicated that NARS izations based upon a profound understanding of the are neither willing nor able to share their technolo- nature of crucial variables at multiple sites is what gies with other actors in agricultural research. To what CIFOR projects should, and in most cases do, seek to extent this is a constraint to collaboration is not known achieve. A misunderstanding of this basic concept will, and needs to be investigated by the CGIAR at the Sys- axiomatically, lead to research that is neither cost-ef- tem level through empirical investigation, identify- fective nor IPG related. The need for cost-effectiveness ing issues and developing possible solutions that are of CIFOR research must not deter in-depth research. widely discussed among members. When expensive senior staff cannot engage in time-con- 4. In interviews, Center directors indicated that a suming field activities, they should recruit and super- significant portion of their resources are intended vise students in the field and develop mutually for the NARS, whereas case studies by NARS under- beneficial links with other appropriate, lower cost and score their lack of financial resources and the un- locality focused partners who can carry out the in- willingness of the Centers to share in the finance. depth field work in the overall IPG context of the re- 5. The CGIAR/NARS partnership in Kenya, as de- search" (TAC Secretariat and CGIAR Secretariat 1998d). scribed by Ndiritu 2002, illustrates the CGIAR's sub- 10. A draft Professional Development Partnership stantive contribution to gains in agricultural research Program is currently under review in the CGIAR Sec- through capacity building, as well as the issues raised retariat. However, the note suggests that this pro- by an evolving, small-country NARS. Kenya's NARS has gram would be heavily focused on developing sought to interact with the CGIAR on a more equal foot- individual scientists, to "enlarge the network of `friends ing, and encouraging steps to this end were observed of the CGIAR'" and "widen and deepen the pool of as some of the CGIAR's Africa-based Centers, most NARS candidates who may be considered for scientist notably ILRI, coordinated efforts in areas such as train- and management positions in the Centers." ing and collaborative research to close the distance be- tween NARS and the Centers. Kenya's NARS has also Chapter 13 benefited from positive experiences with ISNAR in 1. Macedo and others 2002; Romano 2002; Katyal formulating its approach to mobilizing donor funding. and Mruthyanjaya 2002; Ndiritu 2002. Demand surveys Ndiritu (2002) notes that between 1991 and 2000, in- are discussed by Shands, in Gerrard, Ferroni, and ternational funding sources of the Kenyan NARS grew Mody 2001. A strong argument can be made that well- from 19 percent to 61 percent of total funding (and organized, System-wide demand surveys should be World Bank credit grew to 63 percent of outside fund- commissioned regularly by the CGIAR to more clearly ing). Ndiritu also notes that while the CGIAR argues identify client demand for CGIAR services and to in- that they spend 40 percent of their budget in Africa, crease the interest of NARS in identifying their own the NARS believe that the majority of this funding needs, strengths, and weaknesses. goes to CGIAR scientists' salaries and not to support 2. Several NARS scientists commented to the meta- of NARS. evaluation team that while a demand survey approach 6. See Gardner 2002. An exception may be ISNAR's to understanding the NARS issues was an excellent ini- work in developing a methodological framework for tiative, very few NARS have considered their com- evaluating institutional capacity building of NARS parative advantage in relation to other NARS or the through training, networks, and spillovers, and ap- CGIAR Centers and are perhaps unlikely to do so un- plying this framework to several Latin American and less they believe there is a serious effort by their gov- Caribbean case studies. Romano (2002) recommends ernments, the donor community, and the CGIAR to implementing ISNAR's method for this type of eval- address this question at the System level, and unless uation in other Centers. See Horton and others 2000. the results of such analysis are likely to have con- crete impacts on the financing of activities at differ- Chapter 14 ent levels. Several NARS scientists appreciated the 1. For detailed discussions of the CGIAR and Sub- meta-evaluation team's effort to incorporate a de- Saharan Africa, see Eicher and Rukuni 2002 and Ndiritu mand survey approach into its analysis of NARS issues. 2002. 1 9 9 T H E C G I A R AT 3 1 2. In response to the ongoing discussions on the 8. While only 10 percent of the agricultural re- structure, organization, and governance of the CGIAR Sys- searchers in Africa were local nationals at the time of tem, representatives of African NARS leaders met in En- independence (as opposed to 100 percent of the sci- tebbe, Uganda, on 16­17 October 2000, where they entists in India (Lele and Goldsmith 1989), massive relayed an African perspective on this issue. They believed overseas training programs reversed this figure to 90 the current System of 16 global Centers, each working percent by the early 1990s. This capacity building ef- on a specific commodity or problem, was inappropriate fort is an important success story. Yet agricultural re- to meet the needs of NARS and farmers at the subregional search in Africa today is weaker than it should be level. The Centers, they argued, should enable subre- after the initial gains made in human capital devel- gional institutions to become efficient and effective in opment. This is partly because of emigration: today, meeting farmers' needs and to promote increased pro- 45,000 African scientists are working outside the con- duction of food and export crops. They recommended tinent. See Eicher and Rukuni 2002. that CGIAR establish two integrated regional centers 9. However, without question, CGIAR Centers played for Sub-Saharan Africa in lieu of the current structure. a major role in short-term training from 1970 to 1990. 3. Based on regional analyses provided by the 10. In fact, this concentration of status and privi- CGIAR for 1993­2001. This figure includes agreed lege may have stimulated the stinging critique of the agenda funding only, since the CGIAR only classifies CGIAR in the FARA/SPAAR reports (SPAAR/FARA 1999; agreed agenda funding by region. See Annex E. CGIAR Secretariat 2001d). 4. There are several reasons why the impact of 11. Comments provided by Harris Mule, March the CGIAR research in Africa is an underdeveloped area 2002. of study. First, the African database is weak and un- 12. The first review of the CGIAR in 1977 noted, reliable. Second, rate of return studies are biased on "The central thrust of each Center should be....to win-win cases such as hybrid maize in eastern and cooperate with national research and production pro- southern Africa, where commercial farmers helped de- grams to the extent necessary to further the Center's velop the institutional foundation (e.g., seed, credit) own research activities" (CGIAR Secretariat 1989). Yet for subsequent adoption by smallholders. There are it was only in the CGIAR Priorities paper in l987 that few, if any, rate of return studies carried out in coun- "the need to accommodate national priorities when tries such as Chad, Eritrea, Angola, and Zaire. Third, working in a particular country" was explicitly recog- the newness of the CGIAR's natural resources Centers nized (TAC Secretariat 1986). The logjam was broken in Africa make it difficult to evaluate the NRM im- when a CGIAR African Task Force, headed by Guy pacts in Africa. Finally, there is a lack of methodolog- Camus, met eight times from 1986 to 1989 and laid out ical work on the impact of alternative institutions on a pragmatic and politically realistic devolution strat- capacity building, human capital formation, and per- egy. The Camus Task Force reported in 1989, "the formances of NARS (Goldsmith 1993). concept of regional research, particularly where there 5. The large number of mandate crops endorsed are many small countries, is an extremely powerful early by IITA partly reflected the inexperience of the one." The task force recommended regional inter- early IITA managers and scientists. faces driven by the national systems, supported by 6. The litany of constraints to productivity growth donors, and assisted by the CGIAR, for improving the is long, including political instability, lack of stable collaborative process between national and interna- and predictable agricultural policies, dysfunctional tional systems" (CGIAR Secretariat 1989, p. 14). The markets and prices for farm produce, lack of agricul- operative words in the recommendation, "assisted tural credit, poor access roads and extension services, by the CGIAR," furthered the devolution in the 1990s, and the dearth of effective systems of supply of inputs culminating in the preparation of the African Vision such as fertilizers and other agrochemicals. Underly- for Agricultural Research, the Durban Statement, and ing biophysical constraints include depletion of soil the establishment of FARA as the apex organization for fertility, weeds, pests, and diseases. the SROs. To further the devolution, the Centers have 7. Judd, Boyce, and Evenson 1987. Nations in Sub- held a number of useful discussions with SROs and Saharan Africa collectively hired an additional 36,000 NARS. The report of the inter-Center preparatory extension agents from 1959 to 1980. meeting held at ILRI in Kenya alludes to the chal- 2 0 0 E N D N O T E S lenges and realities, stating, "The recognition that ageldin and the CGIAR's Future Harvest Foundation the CGIAR System on its own cannot make a differ- had hoped. A notable exception is Colombia's acces- ence, but can only fulfill its goals through collabora- sion to the CGIAR in 1994, which increased the quan- tive alliances and based on genuine collaborative tity of research contracted to CGIAR Centers because advantage, was central to all discussions" (CGIAR Sec- of the weaknesses of Columbia's national research sys- retariat 2001h, p. 1). At the AGM 2001, ICRAF was as- tem. See Romano 2002. signed the responsibility by the Committee of Center 4. At the Mid-Term Meeting in 1994, Chairman Ser- Directors to facilitate in this regard with respect to east- ageldin stated: "When the System had to cope with ern and southern Africa, the region that has the high- reducing budgets, it became clear that the unpre- est concentration of Centers in Africa. dictability of funding undermined effective manage- 13. Currently, IFPRI's network on agricultural pol- ment of the Centers and undercut the effective funding icy analysis in eastern and southern Africa is sup- of the core research agenda collectively agreed at In- porting policy research and capacity building in ternational Centers Week (ICW). The duplications Ethiopia, Kenya, Malawi, Mozambique, Tanzania, and and overlaps in the System resulting from a lack of clar- Uganda. ity as to whether we are funding programs of work or 14. Three Subregional Research Organizations funding individual Centers has become intolerable. (SROs) were established in Africa in the 1980s and Change must come... These changes must be di- 1990s: SACCAR (southern Africa), ASARECA (eastern rected at introducing predictability in funding and and central Africa), and CORAF/WECARD (western resource management, coupled with transparency and central Africa). and accountability. They must create a system of gov- 15. Under the umbrella of these SROs are initiatives ernance capable of making choices between well-ar- such as the East and Central Africa Programme on Agri- ticulated options and ensuring that the core research cultural Policy Analysis (ECAPAPA), a program of agenda, once arrived at and endorsed, is adequately ASARECA that is working directly with the NARS to funded before resources are diverted to other proj- strengthen their capacity to conduct economic, social, ects." But the reality turned out to be quite different. and policy analysis in collaboration with IFPRI. 5. This concern is underscored in the 1995 audit 16. Regrettably, even though IDA 13 recognized the of CGIAR reporting arrangements. need for IDA grant funding to certain regional activ- 6. Anderson and Dalrymple (1999) provide a de- ities, it does not specify agricultural research as one tailed discussion of the two models. This matching of these activities. grant mechanism is still partially in place today, but is being modified under the CDMP. Technically, only Chapter 15 those projects that were deemed by TAC to meet the 1. Interview with Ismail Serageldin. The need for criteria of the agreed agenda qualified for World Bank more resources for the System is a point that Jeffrey matching funds. Sachs and others have made for both health and agri- 7. Before 1994, the Bank had provided $40 million cultural research. in annual research funding to the Centers (not in- 2. Every chairman of the CGIAR whom OED has in- cluding the Bank's financial support to the CGIAR terviewed has stressed the complexity and onerous re- Secretariat and TAC). In both 1994 and 1995, with sponsibility of chairing the CGIAR while also the additional $20 million grant, the Bank's contri- conducting Bank business as Vice President, which ap- bution increased to $50 million. Although this addi- pears to be insufficiently understood and appreci- tional grant was supposed to be one-time in nature, ated in other parts of the Bank, including among the Bank's support to the research activities of the senior managers. The meta-evaluation team concurs Centers dropped back to only $45 million after 1995. with this assessment. 8. See Chapter 8 and Annex E. Whether measured 3. While increased developing-country membership from 1992 to 2001 or 1994 to 2001, expenditures on and ownership is a favorable aspect of the System's increasing productivity declined by 6.5 percent an- growth, it has yet to result in increased overall con- nually in real terms during both time periods. tributions to the CGIAR or even increased financing 9. The Chairman and the CGIAR Secretariat have of overheads for regional research programs, as Ser- stressed that the nature of restrictions and hence the 2 0 1 T H E C G I A R AT 3 1 degree of flexibility to address strategic issues, vary de- dress the challenges of achieving environmentally pending on the nature of the restricted funding. OED sustainable agricultural production, and how to in- agrees. What the CGIAR has been lacking, however, corporate forestry/agroforestry into the CGIAR, as is a systematic and consistent analysis across Centers the Group had agreed to do in its Canberra Declara- and programs on the nature of restrictions and their tion of 1989. effect on the ability of the System to carry out its 15. The major expansion effort began under Chair strategic mission. David Hopper. The CGIAR acquired the International 10. Since other donors were unwilling to share Center for Living Aquatic Resources Management the $1.2 million cost of the TSR, which the CGIAR (ICLARM), the International Center for Research in donors considered excessive, the Bank ended up pay- Agroforestry (ICRAF), the International Water Man- ing the full cost out of its DGF grant. The meta-review agement Institute (IWMI), and the International Net- considers the cost of TSR in line with similar reviews work for Improvement in Banana and Plantain of major programs, particularly one undertaken after (INIBAP) and created the Center for International 17 years, although it could have benefited from a Forestry Research (CIFOR). INIPBAP was subsequently stronger analytical basis. merged with the International Board on Plant Ge- 11. These debates are consistent with the trends netic Resources (IBPGR) and renamed the Interna- in the business world. Cultivation of competence, its tional Plant Genetic Resources Institute (IPGRI). The diffusion, aggregation, leverage, and renewal are key two subsequent Chairmen indicated in interviews processes in the management of core competencies with OED that they did not agree with the decision in the business world (Doz 1996). In many businesses to expand given the System's impending financial cri- there is currently a simultaneous shift toward core sis, but were essentially powerless to stop it. By 2000, competencies and a close cooperation with partners the System's research portfolio included 21 ecore- in new product development. Experience of multi- gional and System-wide programs. nationals adjusting to globalization suggests that in- 16. In the view of Canadian experts, because novations that depend on tacit knowledge tend to be foresters had a record of being isolated, if the CGIAR more agglomerated, whereas highly localized tech- took on the issues of publicly managed forests it nologies entail either locally imbedded specialization would overwhelm research into the implications for that cannot be accessed elsewhere or that involve farming systems of incorporating trees in farming (in- company-specific global strategies that use the de- terview with David Hopper, January 17, 2002). The ef- velopment of an organizationally complex inter- fect of this was to consider deforestation--the loss or national network for technological learning degradation of the public forest estate--mainly in (Cantwell and Santangelo 1999). the tropical moist forests rather than in other, even 12. Another view is that the Federation proposal more endangered forests or on tree cover in general. was not formally rejected by the Group as a whole, but Although natural forests are being lost, tree cover never given enough time for serious discussion, prior has been expanding outside the public forest estate to the beginning of the Change Design and Manage- in India and China. Taking a more holistic view, the ment Process. World Bank recently revised its forest strategy to in- 13. The CGIAR first expanded its mission in the mid- clude all types of forests and trees and relate the is- 1980s to include the notion of sustainability (TAC sues of deforestation more closely to those of poverty Secretariat 1985). Then, in the late 1990s, as a result alleviation (Lele 2002). The CGIAR, in contrast, con- of the Third System Review, the CGIAR's mission for- tinues to have two forest institutes, one for agro- mally incorporated poverty alleviation and environ- forestry and another for forestry. mental sustainability. See Annex B. 17. It must also be acknowledged that decentral- 14. At the Consultative Group Meeting in May ization, rather than consolidation, may in certain cir- 1988, under the chairmanship of David Hopper, the cumstances be a more appropriate strategy in Group discussed for the first time since 1978 the re- reconfiguring the System, as this meta-review sug- lationship between the CGIAR and other research gests with regard to research on policy and NRM. centers not under its umbrella ("non-associated cen- Thus, the devil would be in the details, depending on ters"). The discussion largely focused on how to ad- the priorities of the System. For a review and analy- 2 0 2 E N D N O T E S sis of previous proposals for reform and reconfigu- 4. The civil society position is temporarily vacant. ration in the CGIAR, see Annex I. NGOs have raised some important issues about ben- 18. In 1994, ILCA and ILRAD merged to form ILRI, efit-sharing in the use of genetic resources with de- and INIBAP was folded into IPGRI. More recently, IWMI veloping countries. However, they recently indicated has absorbed the research programs of the non-CGIAR they are "stepping back and freezing their member- International Board for Soil and Research Management ship" because of the "refusal of CIMMYT to acknowl- (IBSRAM), and the non-CGIAR Tropical Biology and Soil edge GM [genetically modified] contamination in the Fertility Institute (TBSF) has joined CIAT. Maize Centre," and the entry of Syngenta as a mem- 19. One reviewer of this report provides an alter- ber of the CGIAR. During their review, they will not native view that the Federation proposal was in fact accept money from the CGIAR or sit on the Executive an attempt to forestall consolidation. Council or any other committee. 20. System-level consolidation could also have 5. Only 9 of the CGIAR's 22 developing-country been considered when eight Center directors were members met this criterion when the inaugural ExCo being replaced in 2001. While the CGIAR did agree at was established. AGM 2002 to a restructuring of ISNAR, this does not 6. With the increased role of the private sector in represent a System-wide restructuring. global partnerships, most international organizations 21. CGIAR Secretariat's comments on an earlier OED consulted in the course of the global review in- draft of this report, July 31, 2002. dicate they face this issue both with different types of 22. The CGIAR acknowledges the fragmentation in private sector foundations -- e.g., multinational and the System's administration and management, and national, and those with and those without direct steps are being taken to address it in certain areas (e.g., commercial interests associated with the activity of the in human resources policy and in a coordinated, Sys- partnership. Similar questions arise in the case of tem-wide approach to information technology). The representation of NGOs, and are being explored fur- preparation of rules of procedure for some units in ther in the OED review of global programs. the System is likewise a positive change that should 7. The World Bank and FAO, which house the Sys- be accelerated. OED commends the CGIAR for taking tem's two secretariats, have permanent seats. The these steps, including bringing in a management con- third seat will rotate between UNDP and IFAD. sulting firm to advise on the System Office, and notes 8. See Annex J for the CDMT assessment. that these are steps in the right direction. However, 9. For other steps suggested to increase partici- these measures do not address the fragmentation of pation, see note 5, Chapter 5. the CGIAR's research program. Areas where reforms 10. In this context, the elimination at AGM 2002 are urgently needed, such as consolidation of re- of the long-standing CGIAR tradition of maintaining search programs, face resistance due to the current a verbatim record of the business meeting is unfor- organization, management, governance, and financ- tunate. This is an important change for the gover- ing structure. nance of a System that takes pride in reaching decisions by consensus and an example of how long- Chapter 16 established trust among members on consensus 1. Interview with Ian Johnson, August 14, 2002. building can evaporate relatively quickly with a few 2. The CGIAR chairman frequently stresses the stroke-of-the-pen actions. The absence of a verbatim stakeholder ownership of the current reform process. record allows wide scope for interpretation as to Interviews by the meta-evaluation team indicate that what was agreed. It does not allow those who take ownership of both the process and the outcomes of issue with the formal minutes to determine objectively the current reform process may be greater compared who said what, except by going through the notes of with the TSR, but the meta-evaluation team is less cer- individual participants. The CGIAR needs to go back tain about this after the AGM 2002 meeting. to the idea of maintaining verbatim minutes and to 3. Particularly the important Finance and Program ensure these are widely available to members and ob- Committees. See Appendix 1 for the current organi- servers of the process, so that there is transparency zational chart of the CGIAR, subsequent to these or- on which member said what. The same should apply ganizational changes. to ExCo meetings. 2 0 3 T H E C G I A R AT 3 1 11. See note 6, Chapter 5. vices Department, have been a Bank vice president re- 12. See note 7, Chapter 5. sponsible for agriculture and rural development. 13. See note 8, Chapter 5. 4. McNamara did not foresee the World Bank con- 14. See note 9, Chapter 5. tributing more than 10 percent of the resources to the 15. See note 10, Chapter 5. CGIAR in order to ensure its broad-based ownership 16. See note 11, Chapter 5. and sustainability. See Baum 1986, p. 58. 17. See note 12, Chapter 5. 5. UNEP, which had joined the CGIAR in 1974, was 18. Interview with Ian Johnson, August 14, 2002. also a cosponsor from 1995 to 2000, but dropped out 19. See note 13, Chapter 5. because of its inability to contribute any funding or fill 20. See note 14, Chapter 5. a role in environment-related matters. IFAD, which 21. See note 15, Chapter 5. joined the CGIAR in 1979, became a cosponsor in 22. See note 16, Chapter 5. 2001. Its cosponsorship is largely intended to increase 23. See note 17, Chapter 5. political support for the CGIAR's poverty mission 24. See note 18, Chapter 5. rather than to bring additional resources. 25. The United States has indicated interest in fund- 6. The guideline goes on to state that after no more ing some of the Biofortification Program, and the Dutch than three years, a decision should be made to either have pledged at least $25 million for the Water and Food move the Secretariat out of the Bank, keep it in the Program. Canada has announced a doubling of its over- Bank with strong donor support, or discontinue the all contribution, with the additional resources largely effort due to lack of donor interest or other reasons. devoted to Africa. The United Kingdom and Sweden In exceptional cases, where there is strong donor in- have made similar pledges and commitments. terest in maintaining an in-house Secretariat in the Bank 26. In the meta-evaluation team's survey of stake- after three years, then this Secretariat should be fi- holder views, only 22 percent of respondents indicate nanced 100 percent by partners. The CGIAR Secretariat that the reforms go far enough; another 53 believe they points out that when the System's overhead as a whole do not; and 25 percent are unsure (see Annex Q). is taken into account (including the costs of the Sys- 27. Similarly, it was no doubt difficult for the CDMT tem Office currently being established), the DGF bur- to address the issue of intellectual property, which its den is less than 100 percent. OED acknowledges that Steering Group asked it to comment on, without a DGF does not fund 100 percent of the total overhead long-term vision for the kind of research the CGIAR costs of the CGIAR. However, even under the new would be doing. Indeed, the CDMT did not address System Office arrangements, DGF will continue to the IPR issue in its report at all. fully fund the CGIAR Secretariat. Moreover, through the provision of unrestricted funds, DGF indirectly Chapter 17 already finances a portion of the CGIAR's other ad- 1. This comprises the total commitments of those ministrative units as well. projects that were coded as agricultural research or 7. Interview with Ian Johnson, August 14, 2002. agricultural extension in the Bank's coding system up 8. The Bank's current policy is not to charge an ad- to 1989, and the commitments to the agricultural re- ministrative fee for fiscal agency functions that are pro- search and extension components of all projects vided in connection with programmatic trust funds with such components since 1990. This recent change such as the CGIAR. For other types of trust funds the in the Bank's coding system, retroactive to 1990, Bank has recently adopted, effective July 1, 2002, a new has been made in order to adequately capture the policy that will be reflected in the revision of OP/BP greater frequency of multi-component loans in recent 14.40 that is currently under way, and in trust fund years. framework agreements as these are updated to in- 2. See discussion of these three criteria in Chap- corporate other modifications resulting from the trust ter 8. fund reforms. The standard fees are now (a) 5 percent 3. The chair is nominated by the president of the of the trust fund for consultant trust funds and for trust Bank, endorsed by the cosponsors, and approved by funds funding technical assistance or advisory activi- the group. All but the first chairman, Richard H. De- ties, whether executed by the Bank or the recipient; muth, who was Director of the Development Ser- and (b) 2 percent of the trust fund for cofinancing trust 2 0 4 E N D N O T E S funds. VPUs no longer have the discretion to waive or 14. This view was expressed to OED in several in- negotiate trust fund fees case by case. However, for terviews with CGIAR donors in October 2001. large-scale resource mobilization efforts that result in 15. Interview with Ian Johnson, August 14, 2002. trust funds of $30 million or more, customization of 16. Until very recently, the DGF has also lacked in- the fee will be considered by TFO and CRM during the dependent external reviews of proposals presented by Initiating Brief process. different vice presidencies for approval. See World 9. One staff member of the Internal Audit Depart- Bank 2002a. ment has also been seconded to the CGIAR to help 17. Personal communication, Alex McCalla. the System and seven of the Centers meet their in- 18. The oversight roles of the advisers in the net- ternal audit needs. The costs are being shared by the works have diminished considerably with each suc- CGIAR Secretariat and the Centers concerned. cessive reorganization of the Bank. Unlike in the past, 10. The other programs are the Post-Conflict Pro- and except when invited to be peer reviewers by their gram, the Critical Ecosystem Partnership Fund, the colleagues, network advisers are no longer responsi- Small Grants Program, the Global Forum for Health ble for oversight of the quality of Bank investments in Research, UNAIDS, the Special Programme for Re- their sectors, a function now performed Bank-wide by search and Training in Tropical Diseases (TDR), the the Quality Assurance Group. The role of the net- Population and Reproductive Health Capacity Build- work advisers is increasingly that of "knowledge man- ing Program, the Special Programme for Research agement." Few have significant cutting-edge technical and Development in Human Reproduction (HRP), and operational expertise. In any case, the old role of the Consultative Group to Assist the Poorest, the In- quality assurance is in conflict with matrix management formation for Development Program, and the Global since the advisers are expected to provide a consid- Development Network. erable amount of cross-support to the Regions and es- 11. IDA deputies have recently authorized the ex- sentially operate as consultants rather than intellectual panded use of grants under IDA 13, in the range of 18 leaders of their sectors. to 21 percent of IDA 13 resources, in the following five 19. To the best of our knowledge the Rural Sector areas: (1) HIV/AIDS, (2) natural disasters, (3) post-con- Board has played no role in the oversight of the flict countries, (4) poorest countries, and (5) debt-vul- CGIAR. nerable countries. Hence, this currently provides only 20. Steps have been taken to formally outline roles limited scope for providing grants to finance national and responsibilities of various units within the CGIAR, agricultural research programs in the poorest and a laudable development. The CGIAR endorsed rules debt-vulnerable countries, in order to get beyond of procedure for the Executive Council, and a work- short-term, small-scale funding of agricultural re- ing group has proposed rules of procedure for the Sci- search. While IDA resources could currently be used ence Council. The System Office Business Plan, which for regional public goods purposes, this would have is under development, likewise brings clarity to the to be strongly anchored in country-level programs purposes and activities of its component units, as within the region. Based on the agreement with does the CGIAR's informal paper on "Committees donors, IDA could not currently finance the CGIAR it- and Units of the CGIAR." Such clarity is a prerequisite self, nor its member institutions. to establishing a written charter, but not a substitute 12. The Chairman also raises funds from DGF for for a charter. Yet the idea of crafting a charter has been other collaborative programs that fall under his ju- under discussion in the context of the TSR's recom- risdiction, but those programs may not receive as mendation to establish the CGIAR as a legal entity, al- much attention, or the interests of the CGIAR may con- though the latter idea was rejected by the membership. flict with those of other deserving programs. One re- 21. Based on OED's recommendations in "The viewer of this report has suggested that, to the extent Bank's Approach to Global Programs: An Independent that a problematic relationship exists, perhaps DGF Evaluation, Phase 1 Report," 2002, Bank Management decisionmaking processes should be revisited rather is in the process of instituting reforms in the oversight than moving CGIAR chairmanship outside the Bank. and management of its global programs. These will 13. Interviews with Eccles, Hopper, Thalwitz, Raj- help establish Bank-wide strategies and priorities, in- gopalan, Serageldin, and McCalla. crease independent oversight and accountability, in- 2 0 5 T H E C G I A R AT 3 1 crease the voice of developing countries in global in order to provide an objective strategic perspective program governance and management, improve link- on the Bank's role at the global level. (2) DGF criteria ages to the Bank's country operations, increase rou- and the need for subsidiarity: using grants only tine quality assurance, monitoring, independent when the Bank's other traditional instruments of evaluation and follow-up, and strengthen Bank Ex- country assistance could not be appropriately de- ecutive Board and partner inputs. ployed. (3) Exit strategy: avoiding indefinite re- 22. This includes only those projects that have liance on Bank resources without either sufficient been coded as agricultural research in the Bank's effort to mobilize other resources, or by crowding coding system. In recent years, agricultural research them out. (4) Arm's length relationship with has been more often supported as a component of grantees to avoid conflict of interest between the larger multi-component loans. grantor and the grantee. (5) Sound evaluation 23. Regional workshops in the Latin America and process: the need for regular, high-quality inde- Caribbean Region and Europe and Central Asia Region pendent evaluations as an input into strategic re- were conducted in 2001, while a workshop for the sponses to the globalization process. In addition, the South Asia Region was slated for 2002, and the Sub- review recommended that the grant programs be Saharan Africa Region is conducting a major subsec- mainstreamed to the Bank's Regions and networks on toral strategy on agricultural research. Communication a competitive basis so that the DGF Council could from Derek Byerlee. focus its attention on "developing and implementing 24. The idea of using part of the proceeds of Bank a grants policy, reviewing compliance with criteria, ad- loans to make a country's annual contribution to the vancing partnerships with other public and private CGIAR was explored in a loan to Brazil in 1997. Other donors, promoting creative financing and conducting developing countries have been reluctant to borrow ex post evaluations of results." It further recommended IBRD or IDA funds for this purpose, however, though that "management should explicitly develop an Op- they may be more willing to do so now that their par- erational Policy for all grant programs with improved ticipation in the governance of the CGIAR is likely to reporting based on regular external evaluations for all increase. grants, an annual report to the Board on compliance 25. From time to time the Bank has mentioned the with grant criteria and periodic reviews by the Inter- idea of using a portion of the CGIAR funds for proj- nal Audit Department and the Quality Assurance ect preparation as a way of boosting Bank invest- Group." See OED 2001b. ments--for example, in a recent discussion in Nigeria. 3. See OED Precis 224 and the OED Evaluation The ESSD leadership regrets that this idea was put Strategy for a discussion of the issues related to exit forth and will ensure that funds are not used in this strategies. manner. OED contends that this would violate the subsidiarity principle. CGIAR funds should not be Appendix 4 used to develop the Bank's lending program. The 1. The TSR stimulated considerable discussion budget for it must come from its own administrative within the CGIAR and helped define the scope of the sources. current CDMP. For this meta-evaluation, the TSR pro- vided a useful point of reference for identification of Appendix 2 key issues. The report's recommendations were re- 1. The term is used for evaluations designed to ag- viewed, as were the CGIAR Secretariat's detailed in- gregate findings from a series of evaluations. It can also ventory of the extent to which these recommendations be used to denote the evaluation of an evaluation to were implemented and reasons why some were not. judge its quality and/or assess the performance of Reports of the Change Design and Management Team, the evaluators. and various task forces subsequently appointed by the 2. In 1998, OED undertook a review of the Bank's CGIAR, were also reviewed. grant program as administered by the Development Grant Facility and identified five areas that needed to Appendix 7 be addressed: (1) DGF governance: expanding the 1. See The CGIAR at 31: An Independent Meta-Eval- DGF Council to include outside development experts uation of the Consultative Group on International 2 0 6 E N D N O T E S Agricultural Research, Part 1, OED, April 30, 2003, paragraph 1.3. 2. Op. cit., paragraph 2.1. 3. Op. cit., paragraph 2.14. 4. Op. cit., paragraph 2.5. 5. Op. cit., Section 5. 6. Op. cit., paragraph 5.2. 7. Op. cit., paragraph 6.9. 8. See Reaching the Rural Poor--A Renewed Strat- egy for Rural Development (2002), available at http://www.worldbank.org/rural. 2 0 7 BIBLIOGRAPHY Ahmed, M.M., J.H. Sanders, and W.T. Nell. 2001. Asenso-Okyere, W.K., R. Agble, K. Attah-Krah, "New Sorghum and Millet Cultivar Introduc- K. Boakye-Boateng, and others. 1998. 2020 tion in Sub-Saharan Africa: Impacts and Pol- Vision Network for West Africa. Country Note icy Implications." In I. Akintaya and J. Sedego, for Ghana. Accra, Ghana: IFPRI. eds., Towards Sustainable Sorghum Pro- Asia Pacific Association of Agricultural Research duction, Utilization and Commercializa- Institutions. APAARI. June 30, 2002. tion in West Central Africa. Proceedings of a technical workshop of the West and Central ASARECA (Association for Strengthening Agri- Africa Sorghum Research Network. Bamako, cultural Research in Eastern and Central Mali: ICRISAT. Africa). 2002. "Proceedings of the First Alston, J.M., and P.G. Pardey. 1996. Making Sci- ASARECA CD Retreat." Mombasa Continental ence Pay: The Economics of Agricultural Resort, Mombasa, Kenya, February 19­22, R&D Policy. Washington, DC: American En- 2001. Entebe, ASARECA (Draft). terprise Institute. ------. June 30, 2002. About ASARECA. Alston, J.M., G.W. Edwards, and J.W. Freebairn. 1988. "Market Distortions and Benefits from Association of Agricultural Research Institutions Research." American Journal of Agricultural in the Near East and Northern Africa. AARI- Economics 70: 281­88. NENA -- General Information. July 20, 2002. Alston, J.M., C. Chan-Kang, M.C. Marra, P.G. Pardey, and T.J. Wyatt. 2000. A Meta-Analysis Barrett, C.B. 2002. "Natural Resources Man- of Rates of Return to Agricultural R&D: Ex agement Research in the CGIAR: A Meta- Pede Herculem? IFPRI Research Report 113. Evaluation." OED Working Paper. Washington, DC. Washington, DC. Anderson, J.R., and D.G. Dalrymple. 1999. "The Barrett, C.B., F. Place, and A.A. Aboud, eds. World Bank, the Grant Program, and the 2002. Natural Resources Management in CGIAR: A Retrospective Review." OED Work- African Agriculture: Understanding and ing Paper Series No. 1. Washington DC. Improving Current Practices. Walling Pond, ------. 1995. "Improving the Quality and Con- U.K.: CAB International. sistency of the CGIAR's External Center Re- Baum, W.C. 1986. Partners Against Hunger: The views." Washington, DC: CGIAR Secretariat. Consultative Group on International Agri- Anderson, J.R., and P.G. Pardey, with E. Moscardi. cultural Research. Washington, DC: World 1994. Report of the 1994 Internally Managed Bank. External Review of the CIMMYT Economics Baur, H., M. Bosch, S. Krall, T. Kuby, A. Lobb-Rabe, Program. Mexico: CIMMYT. P. T. Shütz, and A. Springer-Heinze. 2001. "Es- Angelsen, A., and D. Kaimowitz. 1999. "Re- tablishing Plausibility in Impact Assessment." thinking the Causes of Deforestation: Lessons GTZ, Eschborn, Germany. from Economic Models." The World Bank Berdegue, J.A., and G. Escobar. 2000. Electronic Research Observer 14 (1): 73­98. Conference on CGIAR Governance, Organi- 2 0 9 T H E C G I A R AT 3 1 zation and Structure: Final Synthesis. Chile: ------. 2002d. Summary Record of Proceed- RIMISP. ings, Second Meeting of the Executive Coun- Bertram, R., and D. Dalrymple. 2000. "Toward a cil, London, April 16­17, 2002. Washington, More Strategic Allocation of World Bank DC. Matching Funding to CGIAR Centers," Draft ------. 2002e. "CGIAR System Office -- Business No. 3. World Bank, Washington, DC. Case, Draft 3.0." May 8, 2002. Washington, DC. Binswanger, H.P. 1994. "Policy Issues: A View ------. 2002f. "Terms of Reference for the ExCo from World Bank Research." In J.R. Ander- Working Group on the Establishment of a son, ed., Agricultural Technology: Policy Is- CGIAR Science Council." Second Meeting of sues for the International Community. the Executive Council, April 16­17, 2002. Wallingford, UK: CAB International. Washington, DC. Borlaug, Norman. 2003. "Science and Hysteria." ------. 2002g. "Strengthening the Participation Wall Street Journal, January 22. of Farmers' Organisations and of NGOs in Byerlee, D., and C.K. Eicher, eds. 1997. Africa's Global Agricultural Research for Develop- Emerging Maize Revolution. Boulder, CO: ment (ARD)." Global Forum on Agricultural Lynn Rienner. Research, Second Meeting of the Executive Byerlee, D., and C.K. Fischer. 2000. "Accessing Council, April 16­17, 2002. Washington, DC. Modern Science: Policy and Institutional Op- ------. 2002h. "Program Committee of the tions for Agricultural Biotechnology in De- CGIAR Executive Council: Terms of Refer- veloping Countries." AKIS Discussion Paper, ence." Second Meeting of the Executive Coun- Washington, DC. cil, April 16­17, 2002. Washington, DC. Byerlee, D., and P. Moya. 1993. Impacts of In- ------. 2002i. "Outline of 2003 CGIAR Financ- ternational Wheat Breeding Research in the ing Plan." Second Meeting of the Executive Developing World, 1966-1990. Mexico, D.F.: Council, April 16­17, 2002. Washington, DC. CIMMYT. ------. 2002j. "Implementation of Recommen- Cantwell, J., and G. D. Santangelo. 1999. "The dations from the Third System Review of the Frontier of International Technology Net- CGIAR." Draft. Washington, DC. works: Sourcing Abroad the Most Highly Tacit ------. 2002k. "Finance Committee of the CGIAR Capabilities." Information Economics and Executive Council: Terms of Reference." Sec- Policy 11 (1): 101­23. ond Meeting of the Executive Council, April Cernea, Michael, and Amir Kassam. 2002. "The 16­17, 2002. Washington, DC. CGIAR Social Research Conference: Report to ------. 2002l. "ExCo Decisions on Challenge the Chairman of the Interim Science Council." Programs." Second Meeting of the Executive In Report from the iSC Chair. Rome: iSCS/FAO. Council, April 16­17, 2002. Washington, DC. The Conservation Company. 1999. "Toards a ------. 2002m. "CGIAR Nominees on Center Long-Range Financing Strategy for the Boards -- Need for Reform." April 4 Draft, Sec- CGIAR." Philadelphia, PA. ond Meeting of the Executive Council, April CGIAR Secretariat. 2002a. "Updated Analysis of 16­17, 2002. Washington, DC. Future Harvest Center Demographics." CGIAR ------. 2002n. "CG System Office Workshop, Gender and Diversity Program, Second Meet- March 13­14, 2002, Summary of Workshop ing of the Executive Council, April 16­17, Discussions and Agreed Action Plan." Second 2002. Washington, DC. Meeting of the Executive Council, April 16­17, ------. 2002b. Second Meeting of the Execu- 2002. Washington, DC. tive Council, Summary Record of Proceed- ------. 2002o. "8 February 2002 ExCo Briefing." ings. Washington, DC. Briefings from the CGIAR Director to ExCo, ------. 2002c. "Financial Guidelines Series Num- Second Meeting of the Executive Council, ber 4: CGIAR Resource Allocation -- Devel- April 16­17, 2002. Washington, DC. oping and Financing the CGIAR Research ------. 2002p. "5 March 2002 ExCo Briefing." Agenda." Washington, DC. Briefings from the CGIAR Director to ExCo, 2 1 0 B I B L I O G R A P H Y Second Meeting of the Executive Council, ------. 2001d. "Report of the Private Sector April 16­17, 2002. Washington, DC. Committee." Mid-Term Meeting, Durban, ------. 2002q. "29 March 2002 ExCo Briefing." South Africa, May 21­25, 2001. Washington, Briefings from the CGIAR Director to ExCo, DC. Second Meeting of the Executive Council, ------. 2001e. "Draft IEC Recommendations on April 16­17, 2002. Washington, DC. CGIAR Reform -- An Integrated Proposal." In- ------. 2002r. "23 January 2002 ExCo Briefing." terim Executive Council, Washington, DC. Briefings from the CGIAR Director to ExCo, ------. 2001f. CGIAR Annual Report 2001. Wash- Second Meeting of the Executive Council, ington, DC. April 16­17, 2002. Washington, DC. ------. 2001g. Summary Record: Third Meeting ------. 2002s. "2003 Planning Guidelines." Sec- of the CGIAR Interim Executive Council (IEC). ond Meeting of the Executive Council, April October 29, 2001. Washington, DC. 16­17, 2002. Washington, DC. ------. 2001h. "Integration of CGIAR Activities ------. 2002t. "20 November 2001 ExCo Brief- in Eastern and Southern Africa: Report of the ing." Briefings from the CGIAR Director to Inter-Center Preparatory Meeting held at ILRI, ExCo, Second Meeting of the Executive Coun- Kenya, March 14­16." Nairobi: CGIAR. cil, April 16­17, 2002. Washington, DC. ------. 2000a. System-Wide Reviews in the CGIAR: ------. 2002u. "20 December 2001 ExCo Brief- Concepts, Options and Recommendations, ing." Briefings from the CGIAR Director to Report of a Study Team Commissioned by ExCo, Second Meeting of the Executive Coun- the Oversight Committee (Martin Pineiro, cil, April 16­17, 2002. Washington, DC. Panel Chair). Washington, DC. ------. 2002v. "19 March 2002 ExCo Briefing." ------. 2000b. Summary of Proceedings and Briefings from the CGIAR Director to ExCo, Decisions, CGIAR Mid-Term Meeting, Dresden, Second Meeting of the Executive Council, Germany, May 26­30, 2000. Washington, DC. April 16­17, 2002. Washington, DC. ------. 2000c. Summary of Proceedings and De- ------. 2002w. "Call for Ideas: CGIAR Challenge cisions, CGIAR International Centers Week, Programs." August 20, 2002. Washington, DC, October 23­27. Washington, DC. ------. 2002x. "Process and Guidelines for De- ------. 2000d. Report of the Twelfth Meeting of veloping and Implementing Challenge Pro- the CGIAR Private Sector Committee. Mid- grams." August 20, 2002. Term meeting, Dresden, Germany, May 19­20, 2000. Washington, DC. ------. 2002y. "Who We Are: History of the ------. 2000e. "Committees and Units of the CGIAR." February 15, 2002. CGIAR: Roles, Responsibilities and Proce- dures." Washington, DC. ------. 2002z. "Structure and Governance." May ------. 2000f. CGIAR Annual Report 2000: The 15, 2002. http://www.cgiar.org/who/ wwa_ Challenge of Climate Change: Poor Farmers governance.html at Risk. Washington, DC. ------. 2001a. Summary Record of Decisions ------. 2000g. Background Paper: CBC/CDC and Proceedings, Mid-Term Meeting, Dur- Retreat: The Hague, Netherlands, 2­3 Sep- ban, South Africa, May 21­25, 2001. Wash- tember 2000 and Report of the CBC/CDC Re- ington, DC. treat (September 2000). Washington, DC. ------. 2001b. Summary of Proceedings and ------. 1999a. Summary of Proceedings, Mid- Decisions, CGIAR Annual General Meeting, Term Meeting, Beijing, China, May 24­28, Washington, DC, October 30­November 1. 1999. Washington, DC. Washington, DC. ------. 1999b. Summary of Proceedings and De- ------. 2001c. "Summary of Discussions: CGIAR cisions, CGIAR International Centers Week, Private Sector Committee Teleconference." October 25­29, 1999 (Reducing Poverty Through May 8, 2001. Washington, DC. Cutting-Edge Science), Washington, DC. 2 1 1 T H E C G I A R AT 3 1 ------. 1999c. Inter-Center Working Group on CGIAR (David R. MacKenzie, Panel Chair). Genetic Resources (ICWG-GR) on the Report of Washington, DC. the First External Review of the System-wide Ge- ------. 1996c. Summary of Proceedings, Mid- netic Resources Programme. Washington, DC. Term Meeting, Jakarta, Indonesia, May ------. 1999d. CGIAR Annual Report 1999: Sci- 20­24, 1996. Washington, DC. ence for the Poor and the Environment. Wash- ------. 1995a. Summary of Proceedings and ington, DC. Decisions, CGIAR Ministerial-Level Meeting, ------. 1999e. "Evolution of the CGIAR." Wash- Lucerne, Switzerland, February 9­10, 1995. ington, DC. Washington, DC. ------. 1998a. The International Research Part- ------. 1995b. Summary of Proceedings and nership for Food Security and Sustainable Decisions, CGIAR International Centers Week, Agriculture: Third System Review of the Con- Washington, DC, October 30­November 3, sultative Group on International Agricultural 1995. Washington, DC. Research (Maurice Strong, Panel Chair). Wash- ------. 1995c. CGIAR Mid-Term Meeting. ington, DC. Nairobi, Kenya, May 22­26, 1995. Washing- ------. 1998b. Summary of Proceedings, Mid- ton, DC. Term Meeting, CGIAR 1998 Mid-Term Meet- ------. 1995d. "Improving the Quality and Con- ing, Brasilia, Brazil, May 25­29, 1998. sistency of the CGIAR's External Center Re- Washington, DC. views." October 24. ------. 1998c. Summary of Proceedings and De- ------. 1994a. Report of the Study Panel on the cisions, CGIAR International Centers Week, CGIAR's Long-Term Governance and Fi- Washington, DC, October 26­30, 1998. Wash- nancing Structure (Klaus Winkel, Panel ington, DC. Chair). Washington, DC. ------. 1998d. Main Conclusions Reached and ------. 1994b. Summary of Proceedings and Decisions Taken. Consultative Group Meet- Decisions, Mid-Term Meeting, New Delhi, ing, May 16­20, 1998. Washington, DC. India, May 23­27, 1994. Washington, DC. ------. 1998e. Third System Review of the ------. 1994c. Summary of Proceedings and De- CGIAR. CGIAR System Review Secretariat. cisions, CGIAR International Centers Week, Washington, DC. ington, DC. ------. 1997a. Summary of Proceedings and De- ------. 1994d. CGIAR's Governance and Orga- cisions, CGIAR International Centers Week, nization: Is There a Need for Change? (Over- Washington, DC, October 27­31, 1997. Wash- sight Committee). Mid-Term Meeting, May ington, DC. 23­27, 1994, New Delhi, India. Washington, DC. ------. 1997b. "Report of the NGO Committee ------. 1993a. Summary of Proceedings and De- to the CGIAR, International Centers Week, cisions, CGIAR International Centers Week, October 24­25, 1997." Paper prepared for dis- Washington, DC, October 25­28, 1993. Wash- cussion during International Centers Week ington, DC. 1997, Washington, DC. ------. 1993b. Summary of Proceedings, Mid- ------. 1997c. Summary of Proceedings, Mid- Term Meeting, San Juan, Puerto Rico, May Term Meeting, Cairo, Egypt, May 26­30, 1997. 23­28, 1993. Washington, DC. Washington, DC. ------. 1992a. Summary of Proceedings and ------. 1996a. Summary of Proceedings and De- Decisions, CGIAR International Centers Week, cisions, CGIAR International Centers Week, Washington, DC, October 26­30, 1992. Wash- Washington, DC, October 28­November 2, ington, DC. 1996. Washington, DC. ------. 1992b. Summary of Proceedings, Mid- ------. 1996b. Report on the Inter-Center Review Term Meeting, Istanbul, Turkey, May 18­22, of Root and Tuber Crops Research in the 1992. Washington, DC. 2 1 2 B I B L I O G R A P H Y ------. 1991a. Summary of Proceedings and CORAF/WECARD. June 3, 2002. Decisions, CGIAR International Centers Week, Washington, DC, October 28­November 1, Dalrymple, D. 2002a. "Challenge Programs: Some 1991. Washington, DC. Cross-Cutting Implementation Issues." Draft ------. 1991b. Summary of Proceedings, Mid- Discussion Notes, September 21. Term Meeting, Paris, France, May 21­23, ------. 2002b. "International Agricultural Re- 1991. Washington, DC. search as a Global Public Good: A Review of ------. 1990. Summary of Proceedings and De- Concepts, Experiences, and Policy Issues." cisions, CGIAR International Centers Week, Submitted to Food Policy, May 31. Washington, DC, October 29­November 2, Delgado, C.L., C.B. Courbois, and M.W. Rosegrant. 1990. Washington, DC. 1998. Global Food Demand and the Contri- ------. 1989. Final Report, CGIAR Task Force on bution of Livestock as We Enter the New Mil- Sub-Saharan Africa. Washington, DC. lennium. MSSD Discussion Paper No. 21. ------. 1986. "CGIAR Africa Task Force Meetings, Washington, DC: IFPRI. June 22­24." Washington, DC. Delgado, C.L, J. Hopkins, and V.A. Kelly. 1998. ------. 1984. CGIAR Annual Report 1984. Wash- "Agricultural Growth Linkages in Sub-Saha- ington, DC. ran Africa." IFPRI Research Report 107. Wash- CGIAR Committee of Board Chairs and CGIAR ington, DC. Center Directors' Committee. 2000. "Towards Doz, Y. L. 1996. "Managing Core Competency a Federation of Centers: The Report of the for Corporate Renewal: Towards a Manager- CBC/CDC Retreat." The Hague, September ial Theory of Core Competencies." In F. 2­3, 2000. Washington, DC. Malerba and G. Dosi, eds., Organization and CDMT (Change Design and Management Team). Strategy in the Evolution of the Enterprise. 2001a. Designing and Managing Change in London: Macmillan. the CGIAR: Report to the Mid-Term Meeting ECAPAPA (Eastern and Central Africa Programme 2001. Washington, DC. for Agricultural Policy Analysis). June 30, 2002. ------. 2001b. "Change Design and Manage- Group Guidance." ISNAR, The Hague, Feb- Eicher, C.K. 1999. "Institutions and the African ruary 6­7, 2001. Washington, DC. Farmer. Third Distinguished Economist Lec- ------. 2001c. "Designing and Managing Change ture." Mexico, D.F.: CIMMYT. in the CGIAR: An Interim Paper for the Steer- Eicher, C.K., and M. Rukuni. 2002. "The CGIAR ing Group and Stakeholders." March 13. Wash- in Africa: Past, Present, and Future." OED ington, DC. Working Paper. Washington, DC. CIMMYT (International Maize and Wheat Im- EIARD (European Initiative in Agricultural Re- provement Center). 2002. Conference Pa- search for Development) Working Group. pers: International Conference on Impacts 2000. "Contribution of the EIARD Working of Agricultural Research and Development. Group on Restructuring of the CGIAR." Brus- Costa Rica, February 4­7, 2002 [CD-ROM]. sels. CGIAR. FAO (Food and Agriculture Organization of the Conway, G. 1997. The Doubly Green Revolution -- United Nations). 2001. International Con- Food for All in the 21st Century. Har- vention for the Protection of New Varieties of mondsworth, UK: Penguin. Plants. Rome. Commission on Intellectual Property Rights. Farrar, C. 2000. The Consultative Group for In- 2002. Integrating Intellectual Property Rights ternational Agricultural Research. Case Stud- and Development Policy. Report of the Com- ies for the UN Vision Project on Global Public mission on Intellectual Property Rights. Policy Networks. London. 2 1 3 T H E C G I A R AT 3 1 Ferroni, Marco, and Ashoka Mody, eds. 2002. In- Global Forum for Health Research. 2001. Mon- ternational Public Goods: Incentives, Mea- itoring Financial Flows for Health Research. surement, and Financing. Boston: Kluwer. Geneva. Fitzhugh, H., and L. Brader. 2002. "Core Fund- Guiterrez, A., and H., Waibel. 2002. "Review of the ing for System-wide and Ecoregional Pro- CGIAR System-wide Programme on Integrated grams." Draft, March 31, 2000. CGIAR, Pest Management (SP-IPM)." Report for the Washington, DC. Technical Advisory Committee of the CGIAR. FONTAGRO (The Regional Fund for Agricultural CGIAR, Washington, DC. Technology). June 30, 2002. Hayami, Y., and V.W. Ruttan. 1971. Agricultural Development: An International Perspective. Fulginiti, L.E., and R.K. Perrin. 1993. "Prices and Baltimore, MD: Johns Hopkins Press. Productivity in Agriculture." Review of Eco- Hayami, Y., M. Kikuchi, and K. Morooka. 1998. nomics and Statistics 75:471­82. "Market Price Response of World Rice Re- Gardner, B. 2002. "Global Public Goods from search." Agricultural Economics (3):333­43. the CGIAR: Impact Assessment." OED Work- Hazell, P., and L. Haddad. 2001. Agricultural Re- ing Paper. Washington, DC. search and Poverty Reduction. Food, Agri- Gautam, M. 2000. Agricultural Extension: The culture, and the Environment Discussion Kenya Experience. Operations Evaluation Paper 34. Washington, DC: IFPRI. Department Study Series. Washington, DC: Heisey, P., M. Lantican, and H.J. Dubin. 2002. World Bank. Assessing the Benefits of International Wheat Gerpacio, R. 2002. "Impact of Public Sector Ver- Breeding Research in Developing Countries, sus Private Sector in R&D and Technology Gen- 1966­97. Mexico, D.F.: CIMMYT. eration: The Case of Maize in Asia." Presented Herdt, R.W. 2000. "Thoughts on the CGIAR." at International Conference on Impacts of Agri- CGIAR Mid-Term Meeting 2000, May 21­26, cultural Research and Development, San José, Dresden, Germany. CGIAR, Washington, DC. Costa Rica, February 4­7 2002. Hewitt, Adrian, Oliver Morrissey, and Dirk Willem Gerrard, Christopher D., Marco Ferroni, and te Velde. 2001. Financing International Pub- Ashoka Mody, eds. 2001. Global Public Poli- lic Goods: Options for Resource Mobiliza- cies and Programs: Implications for Fi- tion. London: Overseas Development nancing and Evaluation--Proceedings from Institute. a World Bank Workshop. Washington, DC: Horton, D., R. MacKay, A. Andersen, and L. Du- OED, World Bank. pleich. 2000. Evaluating Capacity Develop- Gerstner, L., Jr. 2002. Who Says Elephants Can't ment in Planning, Monitoring, and Dance? Inside IBM's Historic Turnaround. Evaluation. Research Report 17. The Hague: New York: HarperCollins. ISNAR. GFAR (Global Forum on Agricultural Research). Instituto Interamericano de Cooperacion Para La 2002. "E-GFAR -- About Us." July 10, 2002. Agricultura. June 17, 2002. page.shtml> IFDC (International Fertilizer Development Cen- ------. 2000. Latin America and the Caribbean ter). 2001. "Agricultural Input Markets in Nige- (LAC) Regional Forum, Global Forum for Agri- ria: An Assessment and a Strategy for cultural Research Meeting, May 20­23, 2000, Development." IFDC, Muscle Shoals, AL. Dresden, Germany. Blessing? Washington, DC: IFPRI. Gladwin, C.H., J.S. Peterson, and A.C. Mwale. IFAD (International Fund for Agricultural De- 2000. "The Quality of Science in Participatory velopment). 2001. Rural Poverty Report: The Research: A Case Study from Eastern Zam- Challenge of Ending Rural Poverty. Oxford, bia." World Development 30 (4):523­43. U.K.: Oxford University Press. 2 1 4 B I B L I O G R A P H Y iSCS (Interim Science Council Secretariat). 2002a. State of the Planet Conference, Columbia Uni- "Assessment of Challenge Programme Concept versity, N.Y., May 13, 2002. Lesser, W. 2002. "Reviews of Biotechnology, Genetic Lele, U. 2002a. Managing a Global Resource: Resource and Intellectual Property Rights Pro- The Challenge of Forest Conservation and grams." OED Working Paper. Washington, DC. Development. New Brunswick, NJ: Transac- Lipton, M. 1999. "Reviving Global Poverty Re- tion Press. duction: What Role for Genetically Modified ------. 2002b. "Science, Dollars and Sustain- Plants?" 1999 Sir John Crawford Memorial ability: Why Should We Care?" Presented at the Lecture, CGIAR International Centers Week, 2 1 5 T H E C G I A R AT 3 1 October 28, 1999. CGIAR Secretariat, Wash- grave Dictionary of Economics. London: ington DC. Macmillan. ------. 1994. "Food Production and Poverty." In Ndiritu, C. 2002. "CGIAR-NARS Partnerships: The Food: Multidisciplinary Perspectives, B. Har- Case of Kenya." OED Background Paper, Wash- riss-White and Sir R. Hoffenberg, eds. Oxford, ington, DC. UK and Cambridge, MA: Blackwell. Nelson, M., and M.K. Maredia. 2001. Environ- Lipton, M., with R. Longhurst. 1989. New Seeds mental Impacts of the CGIAR: An Assessment. and Poor People. Baltimore, MD: Johns Hop- TAC Secretariat Report No. SDR/TAC:IAR/ kins University Press. 01/11, May 2, 2001. Washington, DC: CGIAR Macedo, J., M.C.M. Porto, E. Contini, and A.F.D. Secretariat. Avila. 2002. "Brazil Country Paper." OED Work- Oehmke, J.F., and E.C. Crawford. 1996. "The Im- ing Paper. Washington, DC. pact of Agricultural Technology in Sub-Saha- Maredia, M.K., and D. Byerlee, eds. 1999. The ran Africa." Journal of African Economies 5 Global Wheat Improvement System: Prospects (2):271­92. for Enhancing Efficiency in the Presence of OTA (Office of Technology Assessment, U.S. Con- Spillovers. CIMMYT Research Report No. 5. gress). 1989. New Developments in Biotech- Mexico, D.F.: CIMMYT. nology: Patenting Life. Special Report Maredia, M.K., D. Byerlee, and J.R. Anderson. OTA-BA-370. Washington, DC.: U.S. Govern- 2000. "Ex Post Evaluation of Economic Im- ment Printing Office. pacts of Agricultural Research Programs: A Olson, M. 1965. The Logic of Collective Action: Tour of Good Practice." Presented at a work- Public Goods and the Theory of Groups. Cam- shop of the SPIA (Standing Panel on Impact bridge, MA: Harvard University Press. Assessment) of the CGIAR's Technical Advisory OED (Operations Evaluation Department). 2002a. Committee, Rome. CGIAR, Washington, DC. "Grant Programs: Improving Their Effective- Maredia, M.K., D. Byerlee, and P. Pee. 2000. "Im- ness." OED Précis 224. Washington, DC. pacts of Food Crop Improvement Research: ------. 2002b. "The World Bank's Approach to Evidence from Sub-Saharan Africa." Food Pol- Global Programs: An Independent Review." icy 25:531­59. Washington, DC. Maredia, M.K., P. Pingali, and M. Nelson. 2002. "As- ------. 2001. The World Bank and Global Pub- sessing Environmental Impacts of Productiv- lic Policies and Programs: An Evaluation ity-Enhancing Crop Research: Concepts, Strategy. Washington, DC. Evidences, and Challenges." Presented at the ------. 1996. "The Seed Industry in South Asia." International Conference on Impacts of Agri- Operations Evaluation Department Précis & cultural Research and Development, February Briefs. Washington DC. 4­7, 2002, San José, Costa Rica. ------. 1995. Bangladesh, India and Pakistan-- Mrema, G.C. 2001. "Interaction Between Inter- Seed Projects. OED Project Evaluation. Wash- national Centers, National and Regional Agri- ington, DC. cultural Research Organizations in Africa." Ozgediz, S. 1995. "Strengthening Evaluation in Presented at the Conference on the Role of the CGIAR: Needs and Options." Draft. CGIAR Research in the Development of African Agri- Secretariat, Washington, DC. March 10. culture, with a Special Focus on the Interna- ------. 1994. "Reflections on Reforming Gover- tional Agricultural Research Centers. CGIAR, nance and Management of the CGIAR: Back- Washington, DC. ground Note for the Study Panel on the Mundlak, Y. 2001. "Production and Supply Re- CGIAR's Long-term Governance and Financing sponse." In B. Gardner and G. Rausser, eds., Structure." August 19, 1994. Washington, DC. Handbook of Agricultural Economics. New Pardey, P.G., and N.M. Beintema. 2001. Slow York: Elsevier Science. Magic: Agricultural R&D a Century after Musgrave, R.A. 1987, "Merit Goods." In J. Eatwell, Mendel. Food Policy Statement No. 36. Wash- M. Milgate, and P. Newman, eds., The New Pal- ington, DC: IFPRI. 2 1 6 B I B L I O G R A P H Y Pardey, P.G., and J.E. Christian. 2002. The Pro- Ryan, J.G. 1999b. "Assessing the Impact of Policy duction and Diffusion of Policy Knowledge: Research and Capacity Building by IFPRI in A Bibliometric Evaluation of the Interna- Malawi." Impact Assessment Paper No. 11. tional Food Policy Research Institute. Impact IFPRI, Washington, DC. Assessment Discussion Paper No. 14. Wash- SACCAR (Southern African Center for Coopera- ington, DC: IFPRI. tion in Agricultural and Natural Resources Re- Pardey, P.G., J. Roseboom, and N.M. Beintema. search and Training). June 1, 2002. 1997. "Investing in African Agricultural Re- search." World Development 25 (3):409­23. ------. "Long-Term Strategy and Five-Year Plan, Pardey, P.G., J.M. Alston, J.E. Christian, and S. 2000­2004." July 20, 2002. Fan. 1996. Hidden Harvest: U.S. Benefits from IFPRI. Sanchez, P. 2002. "Soil Fertility and Hunger in Paul, S. 2002. "Governance Issues in the CGIAR." Africa." Science Magazine 295:2019­20. Informal paper prepared for the OED Meta- Sanders, J.H., J.G. Nagy, and S. Ramaswamy. 1990. evaluation of the CGIAR. OED, Washington, DC. "Developing New Agricultural Technologies Petit, M.J., and others. 1996. The Emergence of for the Sahelian Countries: The Burkina Faso a Global Agricultural Research System: The Case." Economic Development and Cultural Role of the Agriculture Research Extension Change 39:1­22. Group (ESDAR). Washington, DC: World Bank. Sandler, T. 2002. "On Financing Global and In- Pinstrup-Andersen, P. 2001. "Overview of the ternational Public Goods." In M. Ferroni and World Food Situation." Paper prepared for A. Mody, eds., International Public Goods: In- the CGIAR Annual General Meeting, Wash- centives, Measurement, and Financing. ington, DC, October 30. CGIAR, Washington, Dordecht, NL. DC. Sauvé, R., and J. Watts. 2002. "An Assessment of Pitman, G.T.K. 2002. Bridging Troubled Waters: IPGRI's Impact on International Policy-Mak- Assessing the World Bank Water Resource ing: A Case Study of the International Un- Strategy. Operations Evaluation Department dertaking on Plant Genetic Resources Study Series. Washington, DC: World Bank. (1996-2001)." Presented at International Con- Pray, C.E. 2001. "Public-Private Sector Linkages ference on Impacts of Agricultural Research in Research and Development: Biotechnology and Development, San José, Costa Rica, Feb- and the Seed Industry in Brazil, China and ruary 4­7, 2002. India." American Journal of Agricultural Serageldin, I. 2000. "CGIAR Chairman's Farewell Economics 83 (3):742­47. Comments, May 27, 2000." CGIAR Mid-Term Romano, L. 2002. "Colombia Country Paper for Meeting 2000, Dresden, Germany. the CGIAR Meta-Evaluation." OED Back- ------. 1996a. Milestones of Renewal: A Journey ground Paper. Washington, DC. of Hope and Accomplishment--CGIAR Policy Rosegrant, M. 2002. "Water Constraints and En- Statements by Ismail Serageldin. Washing- vironmental Impacts of Agricultural Growth." ton, DC: CGIAR Secretariat. American Journal of Agricultural Econom- ------. 1996b. "Crisis Confronted: The Journey ics 84(3):832­38. Begins." Chairman's Opening Address at the Rosenzweig, C., and D. Hillel. 1998. Climate CGIAR Mid-Term Meeting, New Delhi, India, Change and the Global Harvest: Potential May 23, 1994. (In Serageldin, Milestones of Re- Impacts of the Greenhouse Effect on Agri- newal: A Journey of Hope and Accomplish- culture. New York: Oxford University Press. ment: CGIAR Policy Statements by Ismail Ryan, J.G. 1999a. "Assessing the Impact of Rice Serageldin. Washington, DC: CGIAR Secre- Policy Changes in Vietnam and the Contri- tariat.) bution of Policy Research." Impact Assess- Shands, H. 2001. "The CGIAR at a Crossroads?" ment Paper No. 8. IFPRI, Washington, DC. In C.D. Gerrard, M. Ferroni, and A. Mody, 2 1 7 T H E C G I A R AT 3 1 eds., Global Public Policies and Programs: sight Committee, Sonning, England, 4­8 Oc- Implications for Financing and Evalua- tober 2000 (Emil Q. Javier, TAC Chair). Rome. tion. Proceedings From a World Bank Work- ------. 2000b. "Structure and Governance Im- shop. Washington, DC: World Bank. plications of the New CGIAR Vision and Strat- Sherlund, S.M., C.B. Barrett, and A.A. Adesina. egy." Presentation at Synthesis Group Meeting 2002. "Smallholder Technical Efficiency: Con- Organized by the CGIAR Oversight Commit- trolling for Environmental Production Con- tee, Sonning, England, 4­8 October 2000 ditions." Journal of Development Economics (Emil Q. Javier, TAC Chair). Rome. 69(1):85­101. ------. 2000c. "Report from TAC on its Review Sorby, Kristina, Gerd Fleischer, and Eija Pehu. of 2001 Centre Financing Plans." FAO, Rome. 2003. "Integrated Pest Management in De- ------. 2000d. "CGIAR Research Priorities for velopment: Review of Trends and Imple- Marginal Lands." FAO, Rome. mentation Strategies." World Bank Agriculture ------. 2000e. "CGIAR Priorities and Strategies and Rural Development Working Paper 5. for Resource Allocation during 1998-2000 and Washington, DC. Centre Proposals and TAC Recommenda- SPAAR Secretariat. 2000. "SPAAR/FARA Vision of tions." FAO, Rome. African Agricultural Research and Develop- ------. 2000f. A Food Secure World for All: To- ment." SPAAR/FARA Plenary Session, Conakry, ward a New Vision and Strategy for the CGIAR Guinea, April 9­14. Washington, DC: SPAAR, (Emil Q. Javier, TAC Chair). FAO, Rome. World Bank. ------. 1999a. "The CGIAR and National Agri- ------. 1997. "Report of the Second External cultural Research Systems" (Hans Gregersen, Program and Management Review of the Spe- Consultant). FAO, Rome. cial Program for African Agricultural Research ------. 1999b. "Study of CGIAR Commitments (SPAAR)." Washington, DC: SPAAR, World in Latin America and the Caribbean" (Dr. Lucio Bank. Reca, Panel Chair). FAO, Rome. ------. 1995. "Lessons Learned from Imple- ------. 1999c. "Review of System-wide Pro- mentation of the Framework for Action grammes with an Ecoregional Approach" (Ted (FFAs)." Consultant's Report. SPAAR, World Henzell, Panel Chair). FAO, Rome. Bank, Washington, DC. ------. 1999d. "Report of the First External Review Spielman, D.J. 2002. "International Agricultural of the System-wide Genetic Resources Pro- Research and the Role of the Private Sector." gramme" (Dr. Jaap Hardon, Panel Chair). FAO, OED Working Paper. Washington, DC. Rome. TAC Secretariat. 2001a. System-wide Review of ------. 1999e. "Report from TAC on its Review Plant Breeding Methodologies in the CGIAR. of 2000 Centre Financing Plans." FAO, Rome. Rome: FAO. ------. 1998a. "Report of the 16th Meeting of the ------. 2001b. "Progress Report on Regional Oversight Committee." FAO, Rome, October Approach to Research." FAO, Rome. 1998. ------. 2001c. "ILRI Centre-Commissioned Review ------. 1998b. 1999 "Financing Plan -- TAC Ob- of the System-wide Livestock Programme."(Jock servations." Rome: FAO. Anderson, Panel Chair). FAO, Rome. ------. 1997a. "TAC: Priorities and Strategies ------. 2001d. "NRM Research in the CGIAR: A for Resource Allocation in the CGIAR" (Don- Framework for Programme Design and Eval- ald Winkelmann, TAC Chair). FAO, Rome. uation." FAO, Rome. ------. 1997b. "Report on the Inter-Center Review ------. 2001e. TAC's "Comments on the 2002 of Root and Tuber Crops in the CGIAR" (Dr. Centre Financing Plans." FAO, Rome. David MacKenzie, Panel Chair). FAO, Rome. ------. 2000a. "TAC's Views on Implications of the ------. 1997c. "Policy and Management and In- New CGIAR Vision and Strategy for Structure stitution Strengthening: Research and Ser- and Governance." Paper Prepared for the Syn- vice in the CGIAR" (Dr. Alain de Janvry and Dr. thesis Meeting Organized by the CGIAR Over- John Nickel, Panel Chairs). FAO, Rome. 2 1 8 B I B L I O G R A P H Y ------. 1997d. "1998 Financing Plan -- TAC Ob- ------. 1985. Report of the Thirty-Seventh Meet- servations." FAO, Rome. ing of the Technical Advisory Committee. ------. 1996a. "The Future Role of the CGIAR in Rome: FAO. Development of National Agricultural Re- TAC Secretariat and CGIAR Secretariat. 2002a. search Systems: A Strategic Study of Institu- Fourth External Program and Management tion Strengthening Research and Service" Review of CIP. Rome. (Donald L. Winkelmann, TAC Chair). FAO, ------. 2002b. "Fourth External Programme and Rome. Management Review of ISNAR." Draft. Rome. ------. 1996b. "Perspectives on Policy and Man- ------. 2001a. Fourth External Programme agement Research in the CGIAR." TAC Sec- and Management Review of ICARDA. Rome. retariat Report No. SDR/TAC:IAR/95/26.1, April ------. 2001b. Fifth External Programme and 1996. FAO, Rome. Management Review of IITA. Rome. ------. 1996c. Priorities and Strategies for Pol- ------. 2000a. Second External Programme icy, Public Management and Institution and Management Review of IWMI. Rome. Strengthening Research and Service in the ------. 2000b. Fourth External Programme CGIAR. TAC Secretariat Report No. and Management Review of WARDA. Rome. SDR/TAC:IAR/96/4.1, April 1996. Rome: FAO. ------. 2000c. First External Programme and ------. 1995a. "Report of the Study on CGIAR Management Review of ILRI. Rome. Commitments in West Africa" (John McIntire, ------. 2000d. Fifth External Programme and Panel Chair). FAO, Rome. Management Review of CIAT. Rome. ------. 1995b. Implementing CGIAR Priorities ------. 1998a. Third External Programme and and Strategies: 1998-2000. Rome: FAO. Management Review of IFPRI. Rome. ------. 1994a. "TAC Restructuring Review" ------. 1998b. Second External Programme (Alexander F. McCalla, Panel Chair). FAO, Rome. and Management Review of ICRAF. Rome. ------. 1994b. "Stripe Study of Genetic Re- ------. 1998c. Fourth External Programme sources in the CGIAR" (Henry L. Shands, Panel and Management Review of CIMMYT. Rome. Chair). FAO, Rome. ------. 1998d. First External Programme and ------. 1994c. Review of CGIAR Priorities and Management Review of CIFOR. Rome. Strategies. Rome: FAO. ------. 1998e. Fifth External Programme and ------. 1994d. CGIAR Medium-Term Resource Management Review of IRRI. Rome. Allocation 1994-98: Analysis and Recom- ------. 1997a. Third External Programme and mendations. Rome: FAO. Management Review of ISNAR. Rome. ------. 1993a. Priorities and Strategies for Live- ------. 1997b. Fourth External Programme stock Research in the CGIAR. Rome: FAO. and Management Review of IPGRI. Rome. ------. 1993b. Investment in Rice Research in ------. 1996. Fourth External Program and the CGIAR: A Global Perspective. Report of the Management Review of ICRISAT. Rome. Inter-Centre Review of Rice (Grant Scobie, ------. 1995. Report of the Second External Panel Chair). Rome: FAO. Programme and Management Review of ------. 1993c. "International Livestock Research ICLARM. Rome. in the CGIAR: Report of the Steering Com- Tiongco, M., and D. Dawe. 2002. "Long-term mittee." FAO, Rome. Evolution of Productivity in a Sample of Philip- ------. 1992. "Review of CGIAR Priorities and pine Rice Farms: Implications for Sustain- Strategies: Part II" (Draft). FAO, Rome. ability and Future Research." World ------. 1991. An Ecoregional Approach to Re- Development 30 (5): 891­98. search in the CGIAR. Rome: FAO. Watson, R.T. 2002. "Draft Issues Paper on Con- ------. 1990. A Possible Expansion of the CGIAR. sultations on the Role of Science and Technol- Rome: FAO. ogy in Food Security." ESSD Notes, June 2002. ------. 1986. CGIAR Priorities and Future ------. 2001. "Developing an Authorizing Envi- Strategies. Rome: FAO. ronment for an International Assessment on 2 1 9 T H E C G I A R AT 3 1 the Role of Science and Technology in Food ------. 2002b. Global Development Finance: Security." ESSD Notes, July 2001. Building Coalitions for Effective Development U.S. National Academy of Sciences. 2000. Knowl- Finance. Washington, DC: World Bank. edge and Diplomacy: Science Advice in the ------. 2002c. World Development Report 2002: United Nations System. Washington, DC: The Building Institutions for Markets. Washington, National Academies Press. DC: World Bank. WECARD/CORAF (Conférence des Responsa- ------. 2002d. Can Africa Claim the 21st Cen- bIes de Recherche Agronomique en Afrique tury? Washington, DC: World Bank. de l'Ouest et du Centre). 2000. "Strategic Plan ------. 2002e. "Sub-Regional Organizations." for Agricultural Research Cooperation in West July 20. Winkelman, D.L. 1994. "A View of Quintessential ------." 2002f. SPAAR." July 10. International Agricultural Research." In J.R. icy Issues for the International Community. ------. 2002g. "SPAAR Transition." July 20. Wallingford, U.K.: CAB International. Pilot Analysis of Global Ecosystems: Agroe- ------. 2002h. Reaching the Rural Poor--A Re- cosystems. Washington, DC: World Resources newed Strategy for Rural Development. Institute and IFPRI. World Bank. 2002a. The World Bank's Approach ------. 1992. World Development Report 1992: to Global Programs: An Independent Eval- Development and the Environment. New uation, Phase 1 Report. Washington, DC: York: Oxford University Press for the World World Bank. Bank. 2 2 0 OPERATIONS EVALUATION DEPARTMENT ENHANCING DEVELOPMENT EFFECTIVENESS THROUGH EXCELLENCE AND INDEPENDENCE IN EVALUATION The Operations Evaluation Department (OED) is an independent unit within the World Bank; it reports directly to the Bank's Board of Executive Directors. OED assesses what works, and what does not; how a borrower plans to run and maintain a project; and the lasting contribution of the Bank to a country's overall development. The goals of evaluation are to learn from experience, to provide an objective basis for assessing the results of the Bank's work, and to provide accountability in the achievement of its objectives. It also improves Bank work by identifying and disseminating the lessons learned from experience and by framing recommendations drawn from evaluation findings. TMxHSKIMBy356456zv":;:':/:& THE WORLD BANK ISBN 0-8213-5645-3