WORLD BANK OPERATIONS EVALUATION DEPARTMENT IFC OPERATIONS EVALUATION GROUP U P I NFORMAMO I NFRASTRUCTU RE The World Bank Group's Experience A Joint Operations Evaluation Departmenit- Operations Evaluation Group Review 22512 June 2001 -~xa s.X mm W 1{Sr as~~f ' -' ' /r'; THE WORLD BANK GROUP The World Bank Group consists of five institutions-the International Bank for Reconstruction and Development (IBRD or the Bank), the International Finance Corporation (IFC), the International Development Association (IDA), the Multilateral Investment Guarantee Agency (MIGA), and the International Centre for Settlement of Investment Disputes (ICSID). Its mis- sion is to fight poverty for lasting results and to help people help themselves and their environment by providing resources, sharing knowledge, building capacity, and forging partnerships in the public and private sectors. OPERATIONS EVALUATION DEPARTMENT ENHANCING DEVELOPMENT EFFECTIVENESS THROUGH EXCELLENCEAND 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. OPERATIONS EVALUATION GROUP The Operations Evaluation Group (OEG) was set up in 1995 as an independent evaluation unit to introduce systematic procedures, a broader evaluative framework, and improved instruments for corporate accountability and learning. It has a broad mandate to review IFC activities, strategies, and policies. Its reports provide independent review and analysis of mature projects, including those self-evaluated by operations staff. OEG assesses results and identifies lessons learned. To ensure independence, OEG reports to IFC's Board through the World Bank's Director General, Operations Evaluation. WORLD BANK OPERATIONS EVALUATION DEPARTMENT MOE N) IFC OPERATIONS EVALUATION GROUP UED INFORMATION I NFRASTRUCT[JRE The World Bank Group 's Experience A Joint Operations Evaluation Department- Operations Evaluation Group Review Alain Barbu Rafael Dominguez William Melody 2001 International Finance Corporation Washington, D.C. The World Bank http://www.ifc.org/oeg Washington, D.C. http://www.worldbank.org/html/oed Copyright © 2001 The International Bank for Reconstruction International Finance Corporation and Development/THE WORLD BANK 2121 Pennsylvania Avenue, N.W. 1818 H Street, N.W. Washington, D.C. 20433, USA Washington, D.C. 20433, U.S.A. All rights reserved Manufactured in The United States of America First printing June 2001 1 234 04 0302 01 The opinions expressed in this report do not necessarily represent the views of the World Bank Group or its member governments. The World Bank Group does not guarantee the accuracy of the data included in this publication and accepts no responsibility whatsoever for any consequence of their use. The boundaries, colors, denominations, and other information shown on any map in this volume do not imply on the part of the World Bank Group any judgment on the legal status of any territory or the endorsement or acceptance of such boundaries. The material in this publication is copyrighted. The World Bank Group encourages dissemination of its work and will normally grant permission promptly. Permission to photocopy items for internal or personal use, for the internal or personal use of specific clients, or for educational classroom use is granted by the World Bank Group, provided that the appropriate fee is paid directly to the Copyright Clearance Center, Inc., 222 Rosewood Drive, Danvers, MA 01923, U.S.A., telephone 978-750-8400, fax 978-750-4470. Please contact the Copyright Clearance Center before photocopying items. For permission to reprint individual articles or chapters, please fax your request with complete information to the Republication Department, Copyright Clearance Center, fax 978-750-4470. All other queries on rights and licenses should be addressed to the Office of the Publisher, World Bank Group, at the address above, or faxed to 202-522-2422. ISBN 0-8213-4969-4 Library of Congress Cataloging-in-Publication data has been applied for. 4 Printed on Recycled Paper CONTENTS v Acknowledgments vii Foreword xi Executive Summary xxvii Abbreviations and Acronyms I 1. Introduction 1 Study Objectives and Rationale 1 Study Scope and Methods 3 2. Overview of Bank Group Assistance by Sector 3 Recent Sector Developments 4 Implications for the Bank Group's Development and Poverty Reduction Agenda 4 Bank Assistance in the 1990s 9 IFC Assistance Since 1993 13 3. Project and Sector Outcomes 13 Efficacy of Bank Telecommunications Projects-Performance and Outcomes 19 IFC Projects-Outcomes and Performance 22 Main Lessons from Bank Group Project Experience 23 Sector Outcomes 28 Relevance of Bank Group Interventions in the 1990s 31 4. Outstanding Policy and Strategic Issues 31 Rural and Universal Access 32 Holistic Approach to Information Infrastructure Issues 33 World Bank Group Coordination 37 5. Conclusions and Recommendations 37 Successful Projects but Variable Sector Impact 38 An Enhanced Focus on Institutional Reforms 39 Neglected Institutional Options 39 Toward an Integrated Approach to Information Infrastructure Development 39 Selecting Bank Group Instruments 40 Research Capacity and Application of Research Knowledge 40 Recommendations Annexes 43 Annex A: IFC Evaluation Framework and Criteria Attachment A.1: Definition of Framework Terminology Attachment A.2: Telecommunications Project Evaluation 46 Annex B: Lessons Learned from IFC's Telecommunications Operations 48 Annex C: Management Action Record (MAR) 50 Annex D: Report from the Committee on Development Effectiveness (CODE) 53 Endnotes iii INFORMATION INFRASTRUCTURE Tables 7 2.1 Project Objectives Have Shifted from Physical Infrastructure toward Reform and Privatization 13 3.1 Ratings for Completed Telecommunications Projects Have Declined, but Remain Higher than for the Larger Portfolios ... 14 3.2 . . . But Institutional Development and Sustainability Ratings Are on an Up- ward Trend 24 3.3 Competition Is Most Likely for the Youngest Services, 1999 25 3.4 Low-Income Countries Lag in Teledensity Figures 5 2.1 Bank Telecommunications Projects Are Increasing in Number, Decreasing in Funding 6 2.2 The Regional Distribution of Projects and the Instruments Used Have Changed 11 2.3 IFC Annual Telecommunications Commitments and Approvals, 1986-99 20 3.1 Performance Dimension Rating of FY93-96 Telecom Approvals 26 3.2 Telecommunications Investments Have Accelerated in Some Developing Countries but Lagged in Most Low-Income Countries ... 27 3.3 ... As Has the Percentage of Private Sector Financing 33 4.1 Key Dimensions of a National Information Infrastructure Strategy Boxes 9 2.1 The Global Gateway Internet Partnership for Development' 16 3.1 The Impact of Peru's Telecommunications Reform 19 3.2 infoDev-An Interim Evaluation 24 3.3 Resistance to Institutional Adjustment 32 4.1 Rural Telecommunications Focus May Be Increasing 32 4.2 Grameen Phone: Model for a New Approach to Rural Telecommunications? 38 5.1 Building Human Capital-The El Salvador Experience 41 5.2 IBRD and IFC Leverage Their Comparative Advantages-The Sri Lanka Experience iv ACKNOWVLEDGMIENTS T his report was prepared by Alain Barbu, ships and Knowledge Programs Group (OEDPK), task manager, OED; Rafael Dominguez, under the guidance of Elizabeth Campbell-Page task manager, OEG; and William Melody, (task team leader), including Caroline McEuen consultant, with contributions from Margaret (editor) and Juicy Qureishi-Huq (administrative Davis, Anupama Dokeniya, Kurumi Fukaya, assistant). Communications Development pro- Adele Cao Han, Jean-Albert Nyssens, and Do- vided graphic, layout, and editorial support. minique Zwinkels. William Hurlbut provided editorial support, and Soon-Wok Pak and Yvette Director-General, Operations Evaluation: Robert Picciotto jarencio, administrative support. Director, Operations Evaluation Department: Gregory Ingram This publication was prepared by the Dis- Director, Operations Evaluation Group: W E. Stevenson semination and Outreach team of the Partner- Task Managers: Ala in Barbu, Rafael Dominguez V FOREWORD PREFACIO AVANT- - - E PROPOS This joint OEDIOEG study fol- Este estudio conjunto del De- Cette etude conjointe de l'OED lows up OED's 1993 review of the Bank's partamento de Evaluaci6n de Opera- otde l'OEG fait suite a la revue par l'OED, experience in telecommunications. It ciones (DEO) y el Grupo de Evaluaci6n en 1993. de l'exp6rience de la Banque assesses how World Bank Group as- de Operaciones se Ileva a cabo en se- dans le secteur des telecommunica- sistance from 1993 onward has influ- guimiento del examen de la experien- tions. Elle examine la fa,on dont I'aide enced the development of information cia del Banco en materia de du Groupe de la Banque Mondiale, de- infrastructure (the mix of telecommu- telecomunicaciones efectuado en 1993 puis 1993, a influence le developpement nications networks, computing hard- por el DEO. Se analiza en el la medida des infrastructures d'information (c'est ware and software, and services y la forma en que la asistencia pres- a dire l'ensemble constitue par les re- required forthe efficienttransmission of tada por el Grupo del Banco Mundial seaux de telecommunications, 1'equi- information, together with the related desde 1993 en adelante ha influido en pement en materiel informatique, les policy, legal, and institutional frame- el desarrollo de la infraestructura de logiciels, les services necessaires pour work) in developing countries. la informaci6n (el conjunto de redes de une transmission efficace de l'informa- It finds that the recommendations telecomunicaciones, computadoras, tion et le cadre strategique, juridique et of the 1993 review have generally programas informaticos y servicios institutionnel du secteur) dans les pays been heeded with: (a) the adoption necesarios para la transmisi6n efi- en developpement. of a "new" private-sector-led agenda ciente de informaci6n, junto con el L'etude conclut que les recom- (emphasizing privatization, compe- marco de politica, juridico e institu- mandations de la revue de 1993 ont tition, and independent regulation) cional conexo) en los paises en et6 generalement appliquees grace under Operational Policy (OP) 4.50 desarrollo. aux mesures suivantes: a) I'adop- (1995); (b) the incorporation of the Se Ilega a la conclusi6n de que se tion d'une nouvelle politique domi- 'new" agenda in most recent Bank han seguido en general las reco- nee par le developpement du lending and non-lending interven- mendaciones del examen de 1993 secteur prive (qui met l'accent sur la tions; and (c) the increasing share of mediante: a) la adopci6n de un privatisation, la concurrence et un IFC in total Bank Group funding "nuevo" programa encabezado por pouvoir regulateur independant) commitments for telecommunica- el sector privado (en que se hace dans le cadre de la Politique Op& tions (from 8 percent in 1986-92 to hincapie en la privatizaci6n, la com- rationnelle 4.50 (1995); l'integration 32 percent in 1993-99). It also finds petencia y la reglamentaci6n inde- de cette nouvelle politique dans la that both the Bank and the IFC pendiente) con arreglo a la politica plupart des operations financieres et telecommunications portfolios have operacional OP 4.50 (1995); b) la in- non financieres de la Banque; et c) continued to perform better than corporaci6n del "nuevo" programa l'augmentation de la part de la SFI average but cautions that existing en las actividades crediticias y no dans le total du financement des t6- measures do not capture the bulk of crediticias mas recientes del Banco, leconmmunications par le Groupe de Bank Group information infrastruc- y c) el aumento de la participaci6n la Banque (de 8% en 1986-92 a 32% ture (II) activity-the telecommuni- de la CFI en los compromisos de fi- en 1993-99). L'etude constate egale- cations components of multisector nanciamiento totales del Grupo del ment que la performance du porte- projects, the estimated $2 billion per Banco para el sector de telecomu- feuille de la Banque et de la SFI dans year worth of IT components in nicaciones (de 8% en 1986-92 a 32% le secteur des telecommunications Bank projects, and the wide-ranging en 1993-99). Se determina tambien continue d'&re superieure a la analytical and advisory services and que las carteras de telecomunica- moyenne, mais note que ces indica- partnerships-which for the most ciones del Banco y de la CFI han se- teurs ne couvrent pas une grande vii INFORMATION INFRASTRUCTURE part is not subject to system- guido teniendo un desem- partie des activites du Groupe atic performance monitoring. - peno mejor que el prome- de la Banque dans le domaine At the same time, the study dio, pero se advierte que las des infrastructures d'infor- points out that, at a time - medidas actuales no captan mation - les composantes when the information revo- el grueso de la actividad del telecommunications des ope- lution presents developing Grupo del Banco en materia rations multisectorielles, les countries with far-reaching de infraestructura de la in- composantes technologie de opportunities and risks, the formaci6n-los componen- l'information des projets de la Bank Group's ability to play a global tes de telecomunicaciones de los Banque (dont le montant total est es- policy leadership role has been proyectos multisectoriales, los tim6 a environ $2 milliards par an) et hampered by its benign neglect of US$2.000 millones por ano aproxi- une multitude de travaux analytiques, the sector at both the strategic and madamente de componentes de tec- de services de conseil et de partena- country management levels, as well nologia de la informaci6n en los riats - qui pour la plupart ne sont pas as a fragmented internal organiza- proyectos del Banco, y la amplia soumis a un suivi systematique de la tion. As a result, the number of coun- gama de asociaciones y servicios de performance. tries where it has had a real impact analisis y de asesoramiento-que En meme temps, M&tude souligne is limited. In a majority of develop- en su mayor parte no estan sujetos qu'a une epoque oui la revolution de ing countries, the sector reform a un seguimiento sistematico del l'information offre aux pays en de- agenda remains unfinished, tele- desempefio. veloppement des opportunites density remains far below the min- Al mismo tiempo, en el estudio se considerables, mais comporte aussi imum threshold required to join the sefiala que en un momento en que quelques risques, la capacite du information revolution, private in- la revoluci6n de la informaci6n Groupe de la Banque de jouer, a vestment in most lower-income ofrece a los paises en desarrollo 1'echelle mondiale, un r6le de leader countries has been lagging, and grandes oportunidades y riesgos, la sur le plan des politiques sectorielles connectivity gaps have been grow- capacidad del Grupo del Banco de est affectee par l'oubli relatif du sec- ing within countries (between urban desempefiar un papel de liderazgo teur dans les strategies et les ope- and rural areas and between higher a nivel mundial en las polfticas se ha rations par pays, comme d'aiileurs and lower income groups). visto menoscabada por la falta de par la fragmentation de l'organisation The study recommends that: atenci6n prestada al sector tanto al interne. Le r6sultat est que l'inter- (a) the Bank Group restate its strat- nivel estrategico como al nivel de vention de la Banque n'a eu un v6- egy in the broader information infra- gesti6n en los paises, y tambien por ritable impact que dans un nombre structure, transcending its traditional una organizaci6n interna fragmen- limite de pays. Dans la plupart des focus on telecommunications, with a tada. Como resultado de esto, el nu- pays en developpement, les re- particular focus on optimizing the mero de paises en que ha tenido un formes sectorielles sont restees in- use of its instruments (lending and impacto real es limitado. En la ma- achevees, la densite du reseau de nonlending) and expert skills; and yoria de los paises en desarrollo, el telecommunications est tres inferieure (b) gaps in the existing monitoring programa de reforma del sector esta au seuil minimum indispensable pour and evaluation systems be filled, at inconcluso, la teledensidad sigue participer aux benefices de la revo- both the project and global level, to muy por debajo del umbral minimo lution de l'information, les investis- provide the necessary framework to necesario para participar en la re- sements prives progressent lentement assess the future effectiveness of the voluci6n de la informaci6n, la in- dans la plupart des pays a faible re- revised strategy. versi6n privada en la mayorfa de venu et les gaps de communication los paises de ingreso bajo es insufi- s'aggravent dans ces pays (entre les ciente y las deficiencias de conecti- zones urbaines et rurales et entre vidad han ido en aumento dentro de les riches et les pauvres). los paises (entre zonas urbanas y ru- L'6tude conclut par les recom- rales, y entre grupos de altos y bajos mandations suivantes: a) le Groupe de ingresos). la Banque doit depasser les frontieres viii FOREWORD En el estudio se reco- de ses activites traditionnelles mienda: a) que el Grupo del ax6es sur les telecommunica- Banco reformule su estrategia tions et reaffirmer ses objectifs relativa a la infraestructura de strategiques qui concernent la informaci6n en sentido am- l'ensemble des infrastructures plio, mas alla del foco de d'information, mettant l'accent atenci6n tradicional en las sur l'emploi optimal de ses telecomunicaciones, concen- instruments (financiers et non trandose especialmente en la opti- financiers) et de ses competences; mizaci6n del uso de sus instrumentos b) les deficiences presentes des (tanto crediticios como no crediti- systemes de suivi et d'evaluation cios) y sus conocimientos especiali- doivent etre corrigees - l'echelle zados, y b) que se subsanen las des projets et de facon globale - deficiencias en los sistemas de se- afin de mettre en place un cadre guimiento y evaluaci6n existentes, pour l'evaluation future de l'efficacite tanto al nivel de los proyectos como de la nouvelle strategie. al nivel global, a fin de proporcionar el marco necesario para evaluar la efi- cacia futura de la estrategia revisada. Robert Picciotto Director-General, Operations Evaluation ix EXECUTIVE RESUMEN RESUME SUMARY - With the globalization of the Con la globalizaci6n de la eco- Avec la globalisation de l'eco- world's economy, communications is nomia mundial, las comunicaciones nomie mondiale, le secteur des com- increasingly critical to economic son cada vez mas importantes para el munications joue un role de plus en growth and development Indeed, the crecimiento y el desarrollo econ6mico. plus important dans la croissance et le convergence of information and com- De hecho, se ha dicho que la conver- developpement economique. On dit, en munication technologies and the ex- gencia de las tecnologias de la infor- effet, que l'alliance destechnologies de plosion of the Internet are said to be maci6n y de las comunicaciones y la I'information et des communications et launching an information revolution explosi6n de la Internet han puesto en 1'explosion de l'lnternet sont en train de with the potential to reshape society marcha una revoluci6n de la informa- declencher une revolution de l'infor- and commerce. The extent to which ci6n que podria alterar fundamental- mation capable de transforner de fa,on developing countries benefit from this mente las modalidades de la sociedad fondamentale le commerce et los rela- revolution will largely depend on the y del comercio. La medida en que los tions sociales. La participation des pays development of their 'information in- paises en desarrollo se beneficiaran de en developpement aux benefices de frastructure'-the increasingly inte- esta revoluci6n dependera en buena cette revolution dependra beaucoup du grated mix of telecommunications parte del desarrollo de su "infraes- developpement de leurs ainfrastruc- networks, computing hardware and tructura de la informaci6n", a saber, tures d'informationm (I'ensemble de plus software, and value-added services el conjunto cada vez mas integrado de en plus integre que constituent les re- required for the efficient transmission redes de telecomunicaciones, equipo seaux de telecommunication, les equi- of information, together with the re- y programas de computadoras y ser- pements informatiques, les logiciels et lated policy, legal, and institutional vicios de valor anadido conexos ne- les services complementaires neces- frameworks. While this study updates cesarios para la transmisi6n eficiente saires a une transmission efficace de the 1993 OED telecommunications re- de la informaci6n, junto con el marco l'information), ainsi que du cadre stra- view, it also assesses how the World juridico, institucional y de politicas tegique, juridique et institutionnel re- Bank Group's assistance has ad- correspondiente. Este estudio actualiza gissant le secteur. La presente etude dressed the challenge brought about by el examen del sector de telecomuni- met a jour la revue de l'OED de 1993 sur the convergence of telecommunica- caciones realizado en 1993 por el De- les telecommunications et evalue com- tions and information technology (IT). partamento de Evaluaci6n de ment le Groupe de la Banque Mondiale It stops short, however, of a detailed Operaciones (DEO), y tambien evalua a releve le defi de l'alliance des tele- evaluation of IT components of Bank la forma en que la asistencia del Grupo communications et de la technologie de and IFC projects, given their sheer del Banco Mundial ha hecho frente al I'information. Cependant l'etude ne number and the absence of detailed desafioplanteadoporlaconvergencia comprend pas l'evaluation detaillee monitoring and evaluation (M&E) data. de las telecomunicaciones y la tec- des composantes (technologie de While recognizing the Bank's nologia de la informaci6n (TI). Sin em- I'information)) integrees dans des pro- achievements in the sector, the 1993 bargo, no pretende ser una evaluaci6n jets de la Banque et de la SFI; en effet OED telecommunications review detallada de los componentes de TI cescomposantessonttropnombreuses questioned the relevance of Bank de los proyectos del Banco y de la CFI et la Banque ne dispose pas des don- interventions through much of the en raz6n de su gran nuimero y de la au- nees n6cessaires en raison de l'ab- 1980s and early 1990s, pointing to sencia de datos completos para fines sence d'un systeme ad6quat de suivi the "disconnect" between the appar- de seguimiento y evaluaci6n. et d'evaluation. ent success of Bank projects (with Aunque se reconocian los logros La revue de l'OED de 1993 sur les telecommunications standing out as del Banco en el sector, en el examen telecommunications reconnaissait xi INFORMATION INFRASTRUCTURE the top-performing sector in de las telecomunicaciones les resultats obtenus par la the Bank) and the sector's per- - realizado por el DEO en 1993 Banque dans le secteur, mais sistent dismal situation in most se cuestionaba la pertinen- contestait la pertinence des Bank borrower countries (in - cia de las intervenciones del interventions de la Banque access, demand satisfaction, Banco en gran parte de los pendant la plus grande par- and quality of service). OED anos ochenta y comienzos tie des annees 1980 et au recommended (a) adoption de los anos noventa, y se debut des annees 1990, sou- and implementation of a sefalaba la "desconexi6n" entre lignant la contradiction entre telecommunications sector policy re- el exito aparente de los proyectos le succes apparent des projets de la flecting the three pillars of the "new" del Banco (el sector de telecomuni- Banque (la performance du secteur private sector-led sector agenda (pri- caciones era el mas exitoso del des telecommunications etant su- vatization, competition, and regula- Banco) y la persistente situaci6n de- perieure a celle de tous les autres tion), while filling outstanding policy soladora del sector en la mayor parte secteurs) et la situation catastro- gaps (particularly regarding universal de los paises prestatarios del Banco phique du secteur dans la plupart access); and (b) a more innovative use (en lo que respecta al acceso, satis- des pays emprunteurs (sur le plan de of Bank instruments, combined with facci6n de la demanda y calidad de l'acces au service, de la satisfaction greater coordination with IFC. los servicios). El DEO recomendaba de la demande et de la qualite). On balance, these recommenda- a) la aprobaci6n y la aplicaci6n de L'OED recommandait donc: a) tions have been heeded with (a) una politica para el sector de tele- l'adoption et la mise en otuvre d'une the issuance of OP 4.50 (1995); (b) comunicaciones que reflejara los politique des telecommunications the incorporation of the "new" tres pilares del "nuevo" programa refletant les trois principaux objec- agenda in most recent Bank lending encabezado por el sector privado tifs (privatisation, concurrence et re- and nonlending interventions; (c) (privatizaci6n, competencia y re- gulation) d'une - nouvelle " politique the significant shifts that have taken glamentaci6n), y la eliminaci6n de axee sur le leadership du secteur place in staff skills; and (d) the in- las deficiencias existentes (en parti- prive, mais corrigeant les defauts creasing share of IFC in total Bank cular con respecto al acceso uni- des systemes existants (notamment Group funding commitments for versal), y b) un uso mas innovador sur le plan de l'acces universel), et telecommunications (from $210 mil- de los instrumentos del Banco, com- b) un emploi plus imaginatif des lion, or 8 percent of $2.6 billion, binados con una mayor coordina- instruments de la Banque et une during FY86-92 to $706 million, or ci6n con la CFI. meilleure coordination avec la SFI. 32 percent of $2.2 billion, during En definitiva, esas recomenda- Dans l'ensemble, ces recomman- FY93-99). However, recommenda- ciones han sido escuchadas con a) dations ont et appliquees, grace aux tions have been only partially im- la emisi6n de la Politica Operacio- mesures suivantes: a) la publication plemented in two areas: (a) the nal (OP) 4.50 (1995); b) la incorpo- de la nouvelle Politique Operation- potential for synergies between IFC raci6n del "nuevo" programa en la nelle 4.50 (en 1995); b) l'integration and the Bank has not been fully mayor parte de los prestamos y ac- des " nouveaux - objectifs dans la taken advantage of (leading to the tividades no crediticias mas recien- plupart des prets et activit6s non fi- recent decision to merge the telecom- tes del Banco; c) el cambio nancieres recentes de la Banque; c) munications units of the Bank and significativo que ha habido en los d'importants changements dans la IFC); and (b) the Bank has not suffi- conocimientos del personal, y d) la structure des competences dispo- ciently diversified the use of its in- creciente participaci6n de la CFI en nibles, et d) l'augmentation de la part struments, with more than half of its el total del financiamiento del Grupo de la SFI dans le financement du sec- post-1993 lending in the form of mul- del Banco para las telecomunica- teur par le Groupe de la Banque (de tisector technical assistance (TA) loans. ciones (de US$210 millones, es decir, $210 millions, soit 8 pourcent d'un The review of "old" agenda Bank el 8% de US$2.600 millones, en los total de $2,6 milliards au cours des projects completed after 1993 con- ejercicios de 1986-92 a US$706 mil- exercices 1986-92, a $706 millions, firms lessons already documented in lones, o sea, 32% de US$2.200 mil- soit 32 pourcent d'un total de $2,2 the 1993 review concerning the in- lones en los ejercicios de 1993-99). milliards au cours des exercices 1993- xii EXECUTIVE SUMMARY adequacy of traditional state Sin embargo, las recomen- 99). Cependant, dans deux monopolies to develop a daciones se han puesto en domaines, les recommanda- modern network infrastruc- practica s6lo parcialmente en tions n'ont et6 que partielle- ture. But it is still too early to dos esferas: a) no se han ment executees: a) le Groupe measure the actual outcomes aprovechado del todo las po- de la Banque n'a pas suffi- (in coverage, prices, and qual- tenciales sinergias entre la samment exploite toutes les ity of service) of reforms pro- CFI y el Banco (lo cual llev6 synergies possibles entre la moted by Bank projects under a la decisi6n adoptada re- Banque et la SFI (d'ou la de- the 'new" agenda. These are still in cientemente de fusionar las unida- cision recente des deux institutions the early stages of implementation des de telecomunicaciones del d'organiser la fusion de leurs unit6s and only a handful have been eval- Banco y la CFI), y b) el Banco no ha de t6lecommunications) ; b) la uated (Argentina, Mexico). Yet the diversificado el uso de sus instru- Banque n'a pas assez diversifie les ins- evaluation of mature IFC projects mentos en la medida adecuada, y truments employ6s (les prets d'assis- and the review of information on que mas de la mitad del financia- tance technique multisectoriels re- ongoing Bank projects already con- miento concedido despues de 1993 presentent plus de la moitie des ope- firm a number of lessons that have ha correspondido a prestamos de rations de la Banque depuis 1993). been amply documented in the re- asistencia t6cnica multisectoriales. L'analyse des projets posterieurs search literature: El examen, en el marco del "an- a 1993 concus sur la base des - an- * Even partial competition (via the tiguo programa", de los proyectos ciennes strategies - confirme les liberalization of mobile teleph- del Banco completados despues de conclusions de la revue de 1993 sur ony) brings about increased in- 1993 confirma las ensefianzas ya l'incapacite des monopoles publics vestments, lower prices, and documentadas en el examen de 1993 traditionnels de developper un r6- better quality of service. en cuanto a la ineficiencia de los mo- seau moderne. Neanmoins, il est * Setting up a regulatory body does nopolios estatales tradicionales para encore trop tdt pour 6valuer l'impact not obviate the need for a well- desarrollar una moderna infraes- reel (acces, prix et qualite du service) articulated government policy and tructura de redes. Pero es todavia des reformes encouragees par les strategy. muy pronto para medir los resulta- projets inspires par la nouvelle stra- * The dearth of local expertise in dos efectivos (en cuanto a cobertura, t6gie. En effet, l'execution de ces poor countries is a major con- precios y calidad de servicio) de las projets vient seulement de com- straint to effective and inde- reformas promovidas por el Banco mencer et tres peu d'entre eux (Ar- pendent regulation. con arreglo al "nuevo" programa. gentine et Mexique) ont deja fait * Transparency in licensing and tar- Esos proyectos aun estan en las pri- l'objet d'une 6valuation. Cependant iff setting (particularly for inter- meras etapas de ejecuci6n y s6lo se l'examen de certains projets plus connection rates), as well as han evaluado unos pocos (Argentina avances de la SFI et l'analyse des clarity about the role of the in- y M6xico). Sin embargo, la evalua- donnees disponibles sur les projets cumbent operator, is critical to ci6n de los proyectos de la CFI prd- en cours de la Banque confirment attracting private investment. ximos a terminarse y el examen de plusieurs conclusions des nom- * Based on Bank experience, private la informaci6n sobre los proyectos breuses recherches faites sur le sujet: monopolies are not always better en marcha del Banco confirman ya * Meme une concurrence partielle than public ones. varias ensefiazas que se han docu- (par la liberalisation de la t61e- While in most cases the ultimate mentado ampliamente en la inves- phonie mobile) stimule l'inves- outcomes of the Bank's refocused as- tigaci6n: tissement, reduit les prix et sistance and the IFC's activities in * Que incluso la competencia par- ameliore la qualite du service. specific countries cannot be fully cial (a trav6s de la liberalizaci6n * La mise en place d'un organisme measured, results already achieved de la telefonfa m6vil) trae apare- regulateur n'est pas un substitut under some recent projects- jado un aumento de las inversio- adequat pour des politiques et des primarily in the few countries where nes, precios mas bajos y una strategies gouvernementales bien Bank Group involvement was sub- mejor calidad de servicio; definies. x iii INFORMATION INFRASTRUCTURE stantial and sustained-are . Que el establecimiento de * Le manque d'experts lo- encouraging (for example, un 6rgano de reglamen- caux dans les pays pauvres significant private investment taci6n no elimina la nece- est un obstacle majeur a commitments and initial re- sidad de una polftica y una une regulation efficace et ductions in prices of both mo- estrategia gubernamental independante. bile and basic telephony in bien articulada; * La transparence dans l'oc- Bolivia, Morocco, Peru, and * Que la escasez de perso- troi des permis d'exploita- Sri Lanka). But this group of nal calificado local en los tion et la fixation des tarifs countries is small, and the current paises pobres es una limita- (en ce qui concerne notamment state of information infrastructure ci6n grave para la reglamentaci6n les tarifs d'interconnexion) et une (II) in the developing world is de- efectiva e independiente; definition precise du r6le de cidedly mixed and presents a sober- | Que la transparencia en la con- l'operateur sont indispensables ing picture. It belies the views of cesi6n de licencias y la fijaci6n de pour la mobilisation d'investisse- many nonsector specialists inside las tarifas (en particular para las ments prives. and outside the Bank Group that tasas de interconexi6n), asi como * L'experience de la Banque most developing countries have en- la claridad en cuanto al papel de montre que les monopoles prives tered the postreform era and that la empresa operadora, son fun- ne sont pas toujours plus effi- recent progress in technology leaves damentales para atraer inversio- caces que les monopoles publics. them well positioned to gain from nes privadas, y Il n'est pas encore possible de me- the information revolution: * Que, sobre la base de la expe- surer l'impact final des nouvelles orien- • While telecommunications sector riencia del Banco, los monopolios tations de l'aide de la Banque et des restructuring is now on the policy privados no siempre son mejores activites de la SFI dans un certain agenda of most developing coun- que los puiblicos. nombre de pays precis; mais les re- tries, countries that have started to Aunque en la mayoria de los casos sultats deja obtenus dans le cadre de implement reform in earnest (that no es posible medir en su totalidad certains projets - surtout dans les nou- is, beyond the relatively "easy" los resultados definitivos de la asis- veaux pays oi les interventions du steps of opening up value-added tencia reorientada del Banco y de las Groupe de la Banque sont impor- services) are in the minority. And actividades de la CFI en paises es- tantes et continues - sont encoura- some of these countries are facing pecificos, los resultados que ya se geants (citons par exemple serious implementation problems han obtenido en algunos proyectos l'importance des investissements pri- and delays (particularly with re- recientes, sobre todo en los nuevos ves et la reduction des tarifs de la te- gard to the privatization of the paises en los que la participaci6n lephonie mobile et des services de basic service provider, the func- del Grupo del Banco ha sido consi- base en Bolivie, au Maroc, au Perou tioning of the new regulatory derable y continuada, son alenta- et au Sri Lanka). Neanmoins, il s'agit body, and the introduction of ef- dores (por ejemplo, importantes seulement d'un petit nombre de pays fective competition). compromisos de inversi6n privada y et la condition presente des infra- • Quality of service has improved reducci6n inicial de los precios de la structures d'information (II) dans le and average growth in teledensity telefonia m6vil y basica en Bolivia, tiers monde est inegale et dans l'en- (both fixed and wireless) has re- Marruecos, Peru y Sri Lanka). Pero semble peu satisfaisante, contraire- cently accelerated in most coun- ese grupo de paises es pequefio y el ment aux vues de nombreux tries, largely due to the rapid rise estado actual de la infraestructura de generalistes - a l'interieur et a l'exte- of mobile telephony. But pene- la informaci6n en los paises en de- rieur de la Banque - selon lesquelles tration rates in a majority of coun- sarrollo es muy variado e invita a la la plupart des pays en developpement tries remain far below 10 lines reflexi6n. Desmiente la opini6n de sont au stade de l'apres-reforme et - per 100 inhabitants, the level that muchos especialistas ajenos al sector grace aux progres recents de la tech- experts consider the basic con- tanto del Grupo del Banco como de nologie - sont bien places pour b& nectivity threshold to join the in- otros sitios de que la mayoria de los neficier pleinement des effets de la formation revolution. pafses en desarrollo han entrado en revolution de l'information: xiv EXECUTIVE SUMMARY * Despite the rise in private la etapa de postreforma y * La restructuration du sec- sector involvement in coun- - que los progresos tecnol6gi- teur des telecommunica- tries that have opened up cos recientes los han dejado tions fait desormais partie their sectors, overall invest- en condiciones de cosechar des politiques de la ment levels in lower-income los frutos de la revoluci6n de plupart des pays en and lower-middle-income la informaci6n: developpement, mais seuls countries are far from ade- * Si bien la reestructuraci6n quelques pays ont verita- quate to reach the connec- del sector de telecomunica- blement commence a met- tivity threshold anytime soon. ciones figura actualmente en el tre en ceuvre la reforme (au dela • Existing international service rev- programa de politicas en la mayor des mesures relativement - faciles enue sources of many developing parte de los paises en desarrollo, - que sont la liberalisation des country telecommunications net- los paises que han empezado re- services auxiliaires); en outre, works are under serious threat almente a aplicar decididamente la dans plusieurs de ces pays, les from cost-avoidance practices reforma (es decir, mas alla de los progres sont lents et rencontrent such as callback and use of the pasos relativamente "faciles" de de serieux problemes d'execu- Internet, and from the impending abrir los servicios de valor afia- tion (notamment en ce qui demise of the international rev- dido) son una minoria. Ademas, al- concerne la privatisation du four- enue settlement scheme. gunos de ellos estan haciendo nisseur des services de base, le * Within countries, connectivity frente a problemas de aplicaci6n fonctionnement du nouvel orga- gaps are growing between urban y demoras graves (en particular, nisme regulateur et l'introduction and rural areas and between con respecto a la privatizaci6n del d'une concurrence effective). higher- and lower-income groups. proveedor de servicios basicos, el * Dans la plupart des pays, la qua- . With very few exceptions, most funcionamiento del nuevo 6rgano lite du service est meilleure et la countries have not embarked de reglamentaci6n y la introduc- densite moyenne du reseau (fixe upon-nor even recognized the ci6n de una competencia eficaz). et mobile) a augmente, grace sur- need for-a rethinking and * La calidad general del servicio ha tout A la croissance rapide de la broadening of their telecommu- mejorado y la teledensidad media telephonie mobile. Cependant, nications policy to reflect link- (fija e inalambrica) ha aumentado dans un tres grand nombre de ages with other potentially en la mayoria de los paises, en pays, les taux de penetration res- II-intensive sectors (such as edu- buena medida debido al rapido tent tres inferieurs au taux de 10 cation, health, and tax and trade aumento de la telefonfa m6vil. lignes pour 100 habitants, que reform). Sin embargo, las tasas de pen- les experts considerent comme Why is there such a disconnect etraci6n en la mayoria de los pa- le seuil minimum de connecti- between an apparently responsive ises siguen muy por debajo de 10 vite pour une participation ef- Bank Group approach and the lineas por cada 100 habitantes, el fective aux benefices de la sobering II picture? nivel que los expertos consideran revolution de l'information. First, the pace of change in the que es el umbral de conectividad * Malgre la croissance des activit6s overall sector environment (both in basico para participar en la re- privees dans les pays qui ont li- technology and in market structure) voluci6n de la informaci6n. beralise le secteur, les investisse- has accelerated, particularly in the * Pese al aumento de la participaci6n ments dans les pays pauvres - et late 1990s with the onset of the In- del sector privado en los pafses dans la tranche inferieure des ternet. The development of broad- que han abierto sus sectores, los pays a revenu intermediaire - sont band access (which itself requires niveles de inversiones globales en tres inferieurs aux volumes ne- minimum levels of basic connectivity) los paises de ingreso bajo y cessaires pour atteindre rapide- rather than mere voice connection mediano distan mucho de ser ment les seuils minimums de has become the new goalpost and the suficientes para Ilegar al umbral de connectivite. prerequisite for developing country conectividad en un futuro pr6- * Dans plusieurs pays en develop- access to the new information soci- ximo. pement, les recettes intemationales xv INFORMATION INFRASTRUCTURE ety. And the rapid conver- * Las actuales fuentes de in- des riseaux de telecommuni- gence of telecommunications gresos de los servicios in- cations sont serieusement me- and IT, and the way they in- ternacionales de las redes nacees par les techniques du creasingly transform all as- de telecomunicaciones de callback, l'emploi de l'Inter- pects of society, has rendered muchos paises en desa- - net et le demantelement pro- obsolete the Bank's traditional rrollo se ven gravemente chain du systeme de focus on a narrowly defined amenazadas por las practi- reglement des communica- telecommunications sector. cas dirigidas a evitar costos, tions internationales. Second, in the face of these new como los servicios de devoluci6n * A l'int6rieur des pays, les dispa- challenges, the Bank Group slipped automatica de llamadas y el uso rites des taux de connectivite into complacency for several reasons: de la Internet, asi como por la in- entre les zones urbaines et les (a) a lack of recognition outside the minente desaparici6n del sistema zones rurales et entre les popu- Bank's 1I community that the efficient de liquidaci6n de los ingresos in- lations riches et pauvres sont en transfer of knowledge-whose critical ternacionales. train de s'aggraver. importance the World Development * Dentro de los pafses, estan au- * Sauf un tout petit nombre d'ex- Report 1998: KnowledgeforDevelop- mentando las diferencias de co- ceptions, la plupart des pays - ment rightly emphasized-requires a nectividad entre las zonas urbanas plusieurs ne sont meme pas foundation of network infrastructure y las zonas rurales, y entre los conscients du probleme - n'ont (that is, levels of connectivity) that grupos de ingreso alto y bajo. pas encore entrepris les revisions most developing countries are still * Con muy pocas excepciones, la et elargissements des politiques very far from having; (b) the mistaken mayoria de los paises no han ini- de telecommunications qui sont perception among large sections of ciado el procedimiento de re- n6cessaire pour integrer les rela- management and staff that opening plantear y ampliar su politica de tions avec d'autres secteurs ca- the sector to private investment, com- telecomunicaciones, ni han re- pables d'une utilisation intensive bined with the introduction of new conocido siquiera la necesidad des infrastructures d'information technologies, would enable poor de hacerlo, para que queden re- (par exemple l'education, la sante countries to "leapfrog" into the infor- flejados los vinculos con otros et les r6formes fiscales et com- mation age with little or no need for sectores posiblemente intensivos merciales). a public sector role-a perception be- en materia de infraestructura de D'ou vient cette contradiction lied by developed countries' experi- la informaci6n (como la educa- entre la situation peu satisfaisante du ence; (c) the apparent health of the ci6n, la salud, y la reforma fiscal secteur des II et l'evolution appare- Bank Group's declining II portfolio- y comercial). ment favorable des priorites du explained in part by IFC's concentra- ,Por que existe esta desconexi6n Groupe de la Banque? tion on generally higher-risk, but entre un enfoque aparentemente Premierement, l'evolution des higher-reward, projects and the inad- adecuado del Grupo del Banco y technologies et de la structure du equacy of the Bank Group's per- este cuadro sombrio de la infraes- marche s'est acceleree, surtout de- formance monitoring systems with tructura de la informaci6n? puis la fin des annees 1990 au mo- regard to components of multisector En primer lugar, el ritmo de cam- ment du lancement de l'Internet. Le projects (now accounting for the bio de las condiciones generales del remplacement des communications bulk of the Bank's telecommunica- sector (por ejemplo, en terminos de vocales par une multitude de ser- tions lending activity and as much as tecnologia y estructura del mercado) vices (qui necessitent un seuil mi- $2 billion in annual IT support) and se ha acelerado, sobre todo a fines nimum de connectivite) est devenu wide-ranging advisory and analytical de los afios noventa con el inicio de le nouvel objectif du secteur; c'est services; and (d) the illusion of ac- la Internet. El desarrollo del acceso aussi la condition necessaire pour la tion created by the sponsoring of de banda ancha (que requiere en si participation des pays en develop- myriad trust-funded activities, initia- mismo niveles minimos de conecti- pement a la nouvelle societe d'in- tives, partnerships, and "microevents" vidad basica) en lugar de una sim- formation. Enfin, les methodes related to the II sector, with little co- ple conexi6n de voz, se ha traditionnelles de la Banque - la xvi EXECUTIVE SUMMARY ordination or monitoring of convertido ahora en la nueva priorite accord6e au seul sec- their design, funding, and im- meta y el requisito previo teur des telecommunications pact. A noteworthy excep- para el acceso de los paises au sens etroit du terme - sont tion is infoDev, an innovative - en desarrollo a la nueva so- d6sormais depassees, du fait Bank-sponsored public-pri- ciedad de la informaci6n. de la convergence croissante vate grant program, which Ademas, la rapida conver- des telecommunications et proved its agility in launching gencia de las telecomunica- des technologies de l'infor- a quick and effective Y2K as- ciones y la tecnologia de la mation et des transformations sistance program and has the po- informaci6n, y la manera en que qu'elle implique pour tous les as- tential to play a more central role in ambas estan transformando cada pects de la vie sociale. the Bank Group's II assistance. vez mrns todos los aspectos de la Deuxiemement, en face de ces Third, the Bank Group's frag- sociedad, han dejado obsoleta la nouveaux dlfis, le Groupe de la mented organization in the sector atenci6n que el Banco ha prestado Banque a relache ses efforts pour les (with many units active, and some- tradicionalmente a un sector de te- raisons suivantes: a) peu nombreux, times competing, in the provision of lecomunicaciones definido en forma en dehors des specialistes des in- II-related assistance), combined with muy especifica. frastructures d'information de la distorted intemal incentives created by En segundo lugar, ante estos nue- Banque, sont ceux qui reconnaissent the Bank's budget and work plan- vos desafios, el Grupo del Banco qu'un transfert efficace du savoir - ning systems, has hampered the Bank cay6 en la complacencia por varias ra- dont le Rapport sur le D6veloppe- Group's ability to optimize the use of zones: a) la falta de reconocimiento ment dans le Monde: Connaissance its available instruments and expert fuera de la comunidad que se ocupa au Service du Developpement, de skills. In particular, task management de la infraestructura de la informaci6n 1998, soulignait a juste titre l'im- arrangements and budget limitations en el Banco de que la transferencia portance - doit etre base sur une in- associated with the widespread use of eficiente del conocimiento, cuya im- frastructure de reseau (c'est a dire multisector loans by the Bank have portancia critica se puso de relieve des taux de connectivite) que la plu- prevented optimal deployment of spe- con raz6n en el Informe sobre el de- part des pays en developpement cialized anchor staff. And the absence sarrollo mundial 1998/99: El conoci- sont encore loin de posseder; b) of a clearly recognized and funded in- miento al servicio del desarrollo, beaucoup d'experts et de cadres su- formatics thematic group has further requiere como base una infraestruc- perieurs de la Banque s'imaginent a contributed to this fragmentation in IT- tura de redes (es decir, niveles de tort que la simple ouverture du sec- specific issues. Similarly, fragmentation conectividad) que la mayoria de los teur a l'investissement prive et l'in- in the administration of Bank-manlaged paises en desarrollo estan muy lejos troduction des nouvelles trust funds and internal pricing dis- de tener; b) la percepci6n err6nea de technologies permettront aux pays tortions for advisory and analytical grandes segmentos de directivos y pauvres de sauter dans l'ere de l'in- services (for example, between eco- del personal de que la apertura del formation, sans intervention signi- nomic and sector work, trust-funded sector a la inversi6n privada, combi- ficative du secteur public - notion technical assistance, and the fee-based nada con la introducci6n de nuevas que dement l'experience passee des services of IFC's Corporate Finance tecnologias, permitiria a los paises pays developpes; c) la sant6 appa- Services Department) have often re- pobres "saltar" a la era de la infor- rente du portefeuille - en declin - sulted in poor coordination in the maci6n, sin ninguna o con muy poca du Groupe de la Banque dans le sec- provision of Bank Group advice to necesidad de participaci6n estatal; teur des infrastructures d'informa- countries. In tum, these constraints esta percepci6n ha sido desmentida tion s'explique en partie par la have affected the design of Bank in- por la experiencia de los paises de- concentration des financements de terventions, which too often have fo- sarrollados; c) la aparente solidez de la SFI sur des projets a haut risque cused primarily on inputs and la decreciente cartera de proyectos de et a haut rendement et par l'ineffi- processes (privatization transactions, infraestructura de la informaci6n del cacite des systemes de suivi du hiring of legal or regulatory consult- Grupo del Banco, que se explica en Groupe de la Banque en ce qui ants) without sufficient attention to parte por la concentraci6n de la CFI conceme d'une part les composantes xvii INFORMATION INFRASTRUCTURE substantive policy and strate- en proyectos generalmente de II des projets multisectoriels gic issues, arguably the area mayor riesgo pero de mayor (qui constituent aujourd'hui where the Bank can add the rendimiento y por la defi- 1'essentiel des prets de la most value. And the lack of a - ciencia de los sistemas de se- Banque aux telecommunica- strong institutional framework guimiento del desempenio del - tions et representent pres de for effective coordination be- Grupo del Banco con respecto deux milliards de dollars tween the Bank and IFC may a los componentes de pro- d'appui annuel aux techno- have resulted in forgone op- yectos multisectoriales (que logies de l'information), et portunities. Meanwhile, high-quality, constituyen actualmente la mayor d'autre part les nombreuses activites high-impact Bank Group activities parte de los prestamos para de conseil et autres interventions were found to be associated with telecomunicaciones del Banco y equi- non financieres du Groupe; enfin cases where these disincentives were valen a un apoyo de US$2.000 mi- d) une illusion d'activite a et cr6e overcome by individual staff initia- llones por ano para la tecnologia de par le lancement d'une multitude tives or a country director's strategic la informaci6n), y a los variados ser- d'initiatives, de partenariats et de vision of II. vicios de asesoramiento y otros ser- microinterventions soutenues par Finally, not until recently was the vicios no crediticios, y d) la apariencia des fonds fiduciaires, sans coordi- need recognized for an integrated de acci6n que se cre6 al apoyar una nation et suivi effectif de la concep- operationalstrategy (at the global, re- gran cantidad de actividades, inicia- tion, du financement et de l'impact gional, and country levels) that goes tivas, asociaciones y "microeventos" de ces initiatives. Une importante beyond the general policy princi- relacionados con el sector de la in- exception est l'infoDev, programme ples in OP 4.50 and reassesses the fraestructura de la informaci6n y fi- original de subventions publiques et Bank Group's role and comparative nanciados con fondos fiduciarios, con privees, appuye par la Banque, qui advantage in the new, broader II escasa coordinaci6n o seguimiento a demontre son agilite par le lance- "sector." IFC's annual "strategies" en cuanto a su diseno, financiamiento ment rapide d'initiatives efficaces have been more akin to detailed e impacto. Una excepci6n notable es pour faciliter le passage a l'an 2000; marketing plans and have lacked a infoDev, un programa innovador pa- infoDev peut etre appele a jouer un well-articulated policy and strategic trocinado por el Banco con dona- role plus important dans l'aide du framework. These gaps further con- ciones pdblicas y privadas, que Groupe de la Banque au secteur tributed to the fragmentation of the demostr6 su agilidad en el lanza- des II. Bank Group's approach to II in in- miento de un programa de asistencia Troisiemement, l'organisation tres dividual countries, with critical de- rapido y efectivo para la transici6n al fragmentee du Groupe de la Banque cisions too often left to staff, afio 2000 y que tiene posibilidades de (de nombreuses unites interviennent consultants, or not addressed at all desempefiar un papel mas central en dans le secteur des II et se font par- for lack of budget resources (as ap- la asistencia del Grupo del Banco en fois concurrence) et les distorsions pears to be the case in many Bank el ambito de la infraestructura de la dans la structure des incitations in- multisector projects). (For example, informaci6n. ternes (qui resultent des systemes de regarding what advice to give on En tercer lugar, la organizaci6n planification et de budgetisation des sector structure, licensing regimes, particularmente fragmentada del activites de la Banque) n'ont pas per- approaches to universal service, or Grupo del Banco en el sector (al mis au Groupe d'optimiser l'emploi interconnection pricing; how to de- contar con muchas unidades activas, des instruments et des competences cide on the respective roles of the y a veces en competencia, en la disponibles. En particulier le systeme Bank and IFC; or how to establish prestaci6n de asistencia relacionada des chefs de projets et les contraintes linkages with other sectors such as con la infraestructura de la infor- budgetaires resultant de l'emploi fre- health and education.) A case in macion), combinada con incentivos quent des prets multisectoriels ont point is the persisting ambiguity of internos distorsionados creados por rendu impossible l'utilisation opti- the Bank Group's policy on the role los sistemas de presupuesto y de male des principaux specialistes du of the public sector in fostering rural planificaci6n del trabajo del Banco, secteur. De la meme facon, la gestion access, which has led staff to shy ha menoscabado la capacidad del dispersee des fonds fiduciaires et les xviii EXECUTIVE SUMMARY away from promoting inno- Grupo del Banco de optimi- distorsions de prix entre les vative policy options or pro- zar el uso de los instrumen- differentes formes de conseils posing Bank financial support tos disponibles y de sus et de travaux analytiques (par (for example, to private-pub- - conocimientos de expertos. exemple entre les etudes eco- lic partnerships) in that area, * En particular, la organizaci6n - nomiques et sectorielles, les leaving a gap that IFC has al nivel de los jefes de pro- assistances techniques finan- not been able to fill for lack yecto y las limitaciones pre- c6es par les fonds fiduciaires of sufficient investor interest. supuestarias asociadas con et l'aide remuneree de la SFI) Notwithstanding the existence el uso generalizado por el Banco ont souvent eu pour consequence within the Bank Group of substan- de pr6stamos multisectoriales han une mauvaise coordination des acti- tial expertise in areas related to II, be- impedido el despliegue 6ptimo de vites de conseil aux pays du Groupe nign neglect of the sector at both the personal de coordinaci6n especiali- de la Banque. Ces facteurs ont ensuite strategic and country management zado. De manera analoga, como re- influence la conception des op6rations levels has hampered the Bank's abil- sultado de la fragmentaci6n de la de la Banque, trop souvent axees sur ity to play a policy leadership role. administraci6n de los fondos fidu- les produits et les processus (opera- This is at a time when many devel- ciarios gestionados por el Banco y tions de privatisation, recrutement de oping countries are faced with the de las distorsiones internas de los conseillers juridiques ou de consul- prospect of being left out of the in- precios de los servicios analiticos y tants en matiere de reglementation), formation revolution altogether, with de asesoramiento (por ejemplo, aux depens des problemes fonda- potentially irreversible consequences entre los estudios economicos y sec- mentaux de strat6gies et de politiques, for their long-term economic and toriales, la asistencia t6cnica finan- domaines dans lesquels la Banque social well-being. Indeed, in recent ciada con fondos fiduciarios y los est capable d'apporter une contribu- years, the Bank Group missed the servicios basados en el cobro de tion particulierement importante. opportunity to reverse this trend on honorarios que presta el Departa- Enfin, la faiblesse du cadre institu- several occasions. First, in 1995, mento de Cofinanciamiento y Ser- tionnel regissant la coordination entre when the former Telecommunica- vicios de Asesoria Financiera de la la Banque et la SFI peut avoir fait tions and Information Division or- CFI), a menudo la coordinacion de perdre a la SFI des opportunites d'in- ganized the highly successful first la asesorfa del Grupo del Banco a los tervention dans le secteur. En meme Symposium on Information and De- paises ha sido deficiente. A su vez, temps, il apparait que certaines activi- velopment (with high-level repre- esas limitaciones han afectado el di- tes du Groupe de la Banque, dont la sentation from borrowing countries seiho de las intervenciones del qualite et l'impact ont ete remar- and industry leaders) without much Banco, que con mucha frecuencia se quables, ont souvent ete le produit operational follow-up by the Bank's han concentrado fundamentalmente d'initiatives individuelles ou d'une vi- Regional units. Second, in 1996, en los insumos y procesos ( tran- sion strategique du secteur par un when the draft of a World Bank Strat- sacciones de privatizaci6n, contra- responsable de pays, en depit de egy on Information in Develop- taci6n de consultores juridicos o l'obstacle que constitue un systeme ment-itself largely built on the especialistas en reglamentaci6n) sin d'incitations internes deficient. ground-breaking work of a 1990 prestar suficiente atenci6n a las cues- Enfin, jusqu'a une date recente, la Bankwide task force-was prepared tiones de politica y estrategicas sus- haute direction de la Banque n'avait but never finalized. And third, upon tantivas, que constituyen pas pris conscience de la n6cessite the publication of World Develop- posiblemente la esfera en que el d'une strategie operationnelle inte- ment Report 1998. Knowledge for Banco puede ahiadir mas valor. Es gree (a l'echelle globale, regionale Development, which had the poten- posible tambien que la falta de un et des pays) allant au deli des prin- tial to sensitize critical audiences to marco institucional s6lido para una cipes generaux de la Politique Ope- the magnitude of the challenges pre- coordinaci6n eficaz entre el Banco rationnelle 4.50 et comportant une sented to our clients by the infor- y la CFI haya llevado a la perdida de reevaluation du r6le et des avan- mation revolution, but whose main oportunidades. Mientras tanto, se tages comparatifs du Groupe de la impact so far has been the Bank's re- observo que las actividades de alta Banque dans le vaste secteur des in- xix INFORMATION INFRASTRUCTURE cent impetus toward inward- calidad y alto impacto del frastructures d'information. oriented knowledge man- Grupo del Banco estaban Quant aux strategies an- agement efforts. asociadas con casos en que nuelles de la SFI, elles res- But two very recent major estos desincentivos fueron semblent plut6t a des plans initiatives by senior man- superados gracias a la ini- detailles de marketing et ne agement (the Global Gate- ciativa personal de algunos sont pas basees sur un cadre way and the IFC-Softbank funcionarios o a la visi6n de strategies et de politiques Internet venture) may sig- estrategica acerca de la in- clairement definies. Ces de- nal a real breakthrough-as long as fraestructura de la informaci6n del ficiences ont aggrave la dispersion their primary focus on content does director a cargo de las operaciones des activites du Groupe de la not detract from the seriousness of del Banco en un pais determinado. Banque dans les pays; en outre, des the connectivity challenge evi- Por ultimo, hasta hace muy poco decisions fondamentales (concernant denced in this study. Indeed, the tiempo no se habia reconocido la ne- par exemple les conseils relatifs a ongoing preparation of the FY01 cesidad de tener una estrategia ope- l'organisation du secteur, aux sys- Sector Strategy Paper (SSP) on In- racional integrada (al nivel mundial, temes d'octroi des permis d'exploi- formation Infrastructure, together regional y de los paises) que fuera tation, au developpement d'un with the recent management deci- mas alla de los principios generales service universel et aux prix des in- sions to merge the telecommuni- de politica contenidos en la OP 4.50 terconnexions, la definition des r6les cations units of the Bank and IFC y reevaluara el papel y la ventaja respectifs de la Banque et de la SFI and to consolidate the management comparativa del Grupo del Banco en ou l'etablissement de liens avec of all Finance, Private Sector, and el nuevo "sector" ampliado de la in- d'autres secteurs tels que la sante et Infrastructure Network (FPSI) non- fraestructura de la informaci6n. Las I'education) ou bien etaient basees lending services, provides a unique "estrategias" anuales de la CFI han sur les preferences individuelles des opportunity for the Bank Group to consistido mas bien en planes de co- experts ou des consultants - ou, overcome the fragmentation of re- mercializaci6n detallados y han ca- pire, n'etaient pas prises faute de cent years and squarely reposition recido de una politica global y bien ressources budgetaires adequates II on the Bank Group's core de- articulada y de un marco estrategico. (notamment dans le cas des projets velopment agenda. Estas deficiencias contribuyeron tam- multisectoriels). Un bon exemple In this context, the forthcoming bien a la fragmentaci6n del enfoque est l'ambiguite permanente de la SSP should incorporate the follow- del Grupo del Banco de la infraes- politique du Groupe de la Banque ing recommendations: tructura de la informaci6n en los dis- sur le r6le du secteur public dans le * Policy. The forthcoming SSP tintos paises. Muchas veces las developpement des services en mi- should restate the Bank Group decisiones criticas solfan quedar en lieu rural; cette ambiguit6 a decou- objectives in the broader II sec- manos del personal o de consultores, rage le personnel de la Banque de tor in terms of ultimate results o simplemente no se encaraban por promouvoir des initiatives originales (for example, access to informa- falta de recursos presupuestarios ou de proposer des appuis financiers tion, pricing and quality of serv- (como parece ser el caso en muchos (par exemple pour des partenariats ices, fiscal impact) rather than proyectos multisectoriales del Banco). publics et prives), creant un gap just means and intermediate (Por ejemplo, con respecto al aseso- que la SFI n'a pas pu combler en rai- outcomes (private sector in- ramiento que debia darse sobre la son du manque d'int6ret des inves- vestment, competition, or regu- estructura del sector, los regimenes de tisseurs. lation), as OP 4.50 does. It licencia, el enfoque del servicio uni- Bien que le Groupe de la Banque should address outstanding versal o los precios de interconexi6n; ait une bonne expertise dans le sec- policy gaps and differences in a la manera de decidir las respectivas teur des II, le manque d'interet des regulatory reform strategy re- funciones del Banco y de la CFI; o a cadres superieurs et des responsables garding the Bank Group's ap- la forma de establecer vinculos con des pays a entrmin un declin graduel proach to sequencing sector otros sectores, como los de salud y du leadership intellectuel de la liberalization and privatization, educaci6n). Un ejemplo es la ambi- Banque - acquis pendant la periode xx EXECUTIVE SUMMARY as well as to universal serv- guedad persistente de la po- moins compliquee des mo- ice/rural access and its link- litica del Grupo del Banco nopoles publics. Ce declin ages with the Bank's sobre el papel del sector pu-- intervient au moment ou un poverty-reduction agenda blico en el fomento del acceso grand nombre de pays en (including through inter- en las zonas rurales, que ha developpement risquent de - ventions in other sectors llevado al personal a abste- ne pas pouvoir participer a la such as education and nerse de promover opciones revolution de l'information, health). de politica innovadoras o de avec peut-etre des conse- * Institutional options. While en- proponer apoyo financiero del Banco quences irreversibles pour leur pros- suring consistency with the Sep- (por ejemplo, para la creaci6n de perite economique et sociale a long tember 1999 Bank Group private asociaciones entre el sector pilblico terme. Recemment, le Groupe de la sector development (PSD) strategy y el sector privado) en esa esfera, Banque a manque plusieurs occa- and drawing on lessons learned dejando asi una laguna que la CFI no sions de renverser cette tendance. from recent experience in re- ha podido lienar por falta de sufi- Premierement, en 1995, l'ancien de- forming countries, the forthcom- ciente inter6s de los inversionistas. partement de I'Energie, des Mines et ing SSP should provide detailed Pese a la existencia dentro del des Telecommunications a organise, guidance to Bank Group staff on Grupo del Banco de considerables avec succes, le premier Symposium a range of IIpolicy and strategic conocimientos y especialistas en es- sur l'Information et le Developpe- options best suited to various cat- feras relacionadas con la infraes- ment (qui a reuni des responsables egories of borrowers, rather than tructura de la informaci6n, esta falta de haut niveau des pays emprun- a rigid "one-size-fits-all" reform de atenci6n por parte de la admi- teurs et de l'industrie); mais ce sym- model. The SSP could use the nistraci6n superior y de los directi- posium n'a pas ete suivi d'initiatives country segmentation framework vos a cargo de las operaciones del operationnelles importantes de la proposed in a recent infoDev- Banco en los paises ha provocado part des unites regionales de la financed background study. Sim- una erosi6n gradual del liderazgo in- Banque. Deuxiemement, en 1996, ilarly, the SSP should promote the telectual del Banco en el sector, que un projet de Strategie de la Banque use of a broader range of Bank se habia ganado durante la epoca, Mondiale sur l'Information et le D& Group instruments, away from sin duda mas sencilla, de los mo- veloppement - base en grande par- the excessive reliance of recent nopolios estatales. Actualmente, mu- tie sur les travaux remarquables years on multisector TA loans. chos paises en desarrollo se ven d'une equipe de la Banque en 1990 * Regional/country strategies. The enfrentados a la perspectiva de que- - a et prepare mais jamais termin6. SSP should include detailed Re- dar totalmente excluidos de la re- Troisiemement, la Banque, en 1998, gional Bank Group II strategies, voluci6n de la informaci6n, situaci6n a publie le Rapport sur le Develop- building on the recent, very suc- que tendria consecuencias posible- pement dans le Monde: La Connais- cessful exercise carried out for mente irreversibles para su bienes- sance au Service du Developpement; the Europe and Central Asia Re- tar econ6mico y social a largo plazo. ce rapport aurait pu sensibiliser des gion (ECA) telecommunications En los ultimos anios, el Grupo del publics importants sur l'ampleur du strategy. To this end, manage- Banco perdi6, en efecto, la oportu- d6fi de la revolution de l'information ment of the newly merged nidad de invertir esa tendencia en pour les clients de l'institution. Mais, Bank/IFC unit should give top varias ocasiones. En primer lugar, en pour l'instant, le principal benefice du priority to carrying out a rapid 1995, cuando la antigua Divisi6n de rapport a et l'impulsion recente country-level IIpolicy and strat- Telecomunicaciones e Informaci6n donnee a la gestion du savoir a l'in- egy stock-taking exercise cover- organiz6 con gran exito el primer terieur de la Banque. ing every country where the Bank Simposio sobre la informaci6n y el Neanmoins, deux importantes ini- or IFC is involved. This exercise desarrollo (con representaci6n de tiatives tres recentes de la haute di- should not aim at the preparation alto nivel de los paises prestatarios rection (le Portail Universel et le of polished country reports, but y del sector), sin que hubiera un lancement par la SFI d'une Banque rather at providing the basic in- adecuado seguimiento operativo por de Logiciels) sont peut-&tre le signe xxi INFORMATION INFRASTRUCTURE tellectual foundation for the parte de las unidades regio- d'un changement radical - formulation of the Bank nales del Banco. En segundo pour autant qu'elles ne font Group's operational strategy. lugar, en 1996, cuando se pas oublier la gravite du pro- Consideration should be preparo pero nunca se fina- bleme de connectivit6 souli- given to using infoDev re- liz6 el proyecto de una es- gne par l'etude. La 'u sources more systematically trategia del Banco Mundial preparation en cours du do- to support such country-level sobre la informaci6n en el cument de strategie secto- II policy and strategic work. desarrollo, que se elabor6 en rielle sur les Infrastructures * Partnerships. The SSP should pro- gran medida tomando como base la d'Information pour l'exercice 2001 et pose ways to streamline the vast labor innovadora de un grupo de tra- les decisions recentes d'organiser la array of current external partner- bajo establecido en 1990. En tercer fusion des unites de te- ships in II, based on a systematic lugar, en oportunidad de la publi- lecommunications de la Banque et evaluation of their relevance, im- caci6n del Informe sobre el desa- de la SFI et de consolider la gestion pact, and efficiency. The exces- rrollo mundial 1998/99: El de tous les services non financiers du sive fragmentation of their conocimiento al servicio del desa- Reseau Finance, Secteur Prive et In- management within the Bank rrollo, documento que tenia la po- frastructure sont pour le Groupe de Group should be specifically ad- sibilidad de sensibilizar a un puiblico la Banque une occasion unique de dressed, possibly by formally as- de importancia fundamental acerca lutter contre la dispersion des annees signing overall strategic and de la magnitud de los desafios con anterieures et de placer le probleme coordination responsibility to the que se enfrentaban los paises a raiz des infrastructures d'information au manager of the newly merged de la revoluci6n de la informaci6n, centre des objectifs de developpe- Bank/IFC unit. pero cuyo principal efecto ha sido ment de la Banque. • Staffing. The SSP should include hasta el momento el impulso re- Dans ce contexte, le futur docu- a plan to continually update staff ciente por parte del Banco de ini- ment de strategie sectorielle devrait skills to cope with the demands ciativas de orientaci6n interna sobre comprendre les recommandations of this dynamic sector and enable gesti6n de los conocimientos. suivantes: the Bank to maintain policy lead- No obstante, hay dos grandes ini- . Politiques. Le document devra ership. The SSP would build on ciativas muy recientes impulsadas reaffirmer les principaux objectifs the shift that has already taken por la administraci6n superior (el du Groupe de la Banque dans le place in recent years, both in the Portal mundial sobre el desarrollo y vaste secteur des infrastructures Bank and IFC, in response to the la iniciativa de la CFI y Softbank en d'information, qui donnent la new private sector-led agenda. Internet) que pueden significar un priorite aux resultats finaux (par Given the rapid technological avance trascendental, en la medida exemple l'acces a l'information, le changes and the increasing scope en que el principal foco de atenci6n prix et la qualite des services, les of the information infrastructure de ambas iniciativas "a saber, el incidences budgetaires) et non sector, this may include expand- contenido", no reste importancia al pas seulement aux moyens em- ing beyond the telecommunica- desafio que plantea la conectividad, ployes et aux produits interme- tions sector the successful staff segun se comprueba en el presente diaires (investissement prive, exchange program already in estudio. En efecto, la actual prepa- concurrence ou r6gulation), place and setting up a well-struc- raci6n del documento de estrategia comme dans la Politique Opera- tured knowledge sharing pro- sectorial para el ejercicio de 2001 tionnelle 4.50. Le document devra gram with leading private sector sobre la infraestructura de la infor- aborder le probleme des defi- companies. maci6n, asi como la reciente deci- ciences des politiques et des dis- * MVonitoring and evaluation. The si6n de la administraci6n de fusionar parites dans les strategies de SSP should draw on this report to las unidades de telecomunicaciones reforme du cadre reglementaire, summarize lessons from recent del Banco y la CFI y de consolidar inherentes aux conceptions du Bank Group experience. And it la gesti6n de todos los servicios no Groupe de la Banque sur l'en- should include a detailed frame- crediticios de la Red sobre Finanzas, chainement des processus de li- xxii EXECUTIVE SUMMARY work for the future monitor- Sector Privado e Infraestruc- beralisation et de privatisa- ing and evaluation (M&E) of tura (FPSI), brindan una tion, sur " l'universalite " du the revised Bank Group pol- oportunidad uinica para que service, sur l'acces du monde icy and strategy. To this end, el Grupo del Banco elimine rural et sur les liens avec les current M&E processes la fragmentaci6n observada - politiques de reduction de la should be enhanced at three en los ultimos anos y decida pauvrete (notamment grace levels: (a) at the project/ac- volver a colocar a la infraes- aux interventions dans tivity level, by setting M&E tructura de la informaci6n en d'autres secteurs tels que systems for interventions that cur- el lugar que le corresponde en su l'ducation et la sante). rently fall through the cracks programa basico de desarrollo. * Options institutionnelles. Le futur (telecommunications/informat- En este contexto, en el docu- document devra etre compatible ics/IH components of multisector mento de estrategia sectorial de pr6- avec la strategie de developpe- Bank projects and the Bank xima aparicion se formulan las ment du secteur prive de sep- Group's advisory and analytical siguientes recomendaciones: tembre 1999, mais, sur la base de services); (b) for IFC projects, by * Polfticas. El proximo documento l'experience recente des pays re- specifying at appraisal the pro- de estrategia sectorial debe vol- formateurs, il devra aussi donner ject's expected contributions to ver a exponer los objetivos que des directives detaillees au per- sector and development impact se persiguen en el sector de la in- sonnel du Groupe de la Banque objectives and tracking their fraestructura de la informaci6n en sur les politiques et les options progress toward achievement in terminos de los resultados ulti- strategiques dans le secteur des supervision and evaluation re- mos (por ejemplo, acceso a la in- II; au lieu de la rigidite des so- ports; and (c) at the global level, formaci6n, fijaci6n de precios y lutions standards, ces directives by incorporating in the SSP spe- calidad de los servicios, aspectos devront coller aux situations par- cific, relevant indicators of Bank fiscales), y no solamente en los ticulieres de differentes catego- Group effectiveness, including medios y los resultados interme- ries d'emprunteurs. La strategie sector outcomes and develop- dios (inversi6n privada, compe- pourra utiliser les classe- ment impacts. tencia o reglamentaci6n), como ments de pays proposes par une ocurre con la OP 4.50. Debe ocu- etude recente financee par info- parse de las deficiencias en las Dev. De la meme facon, le do- politicas y las diferencias en la es- cument de strategie devra trategia de reforma de las regla- encourager l'emploi d'instruments mentaciones que siguen operationnels diversifies, au lieu existiendo con respecto al enfo- du r6le preponderant recemment que del Grupo del Banco acerca accorde aux prets dassistance de la secuencia de la liberaliza- technique multisectoriels. cion y la privatizacion del sector, * Strategies regionales et de pays. asi como del servicio "universal" Le document de strategie secto- y el acceso de las zonas rurales rielle devra comprendre des stra- y sus vinculos con el programa de tegies regionales detaillees en reduccion de la pobreza matiere d'infrastructures d'infor- (inclusive mediante intervenciones mation, sur la base de l'expe- en otros sectores, como el de la rience reussie de la Region educaci6n y la salud). Europe et Asie Centrale pour la * Opciones institucionales. Al preparation d'une strategie des tiempo que se asegura la cohe- telecommunications. A cette fin, rencia con la estrategia del Grupo la direction de la nouvelle unite del Banco sobre desarrollo del commune de la Banque et de la sector privado, de septiembre de SFI devra donner la plus haute xxiii INFORMATION INFRASTRUCTURE 1999, y aprovechando las priorite a un inventaire ra- ensefianzas obtenidas de la pide des politiques et des experiencia reciente en pai- strategies d'II au niveau des ses que han introducido re- pays, couvrant tous les pays * formas, en el pr6ximo oui operent la Banque et la documento de estrategia sec- SFI. Cet inventaire ne devra torial se debe proporcionar pas aboutir a la preparation orientaci6n precisa al per- de rapports parfaits sur sonal del Grupo del Banco sobre chaque pays, mais il devra etablir diversas opciones estrategicas y les bases intellectuelles pour la de politica en materia de infraes- formulation de la strategie ope- tructura de la informaci6n mas rationnelle du Groupe de la adecuadas para las distintas ca- Banque. Il conviendra d'examiner tegorfas de prestatarios, en lugar la possibilite d'utiliser systemati- de un modelo de reforma rigido, quement les ressources d'info- igual para todos. En el docu- Dev pour appuyer les analyses de mento de estrategia sectorial se politiques et de strat6gie d'II au podria utilizar el marco de seg- niveau des pays. mentaci6n de paises propuesto * Personnel. Pour assurer le main- en un reciente estudio de ante- tien du leadership de la Banque, cedentes financiado por infoDev. le document de strategie devra Del mismo modo, en el docu- comprendre un plan d'ajustement mento se debe promover el uso permanent de l'expertise du per- de una gama mas amplia de ins- sonnel de la Banque aux exi- trumentos del Grupo del Banco, gences d'un secteur dynamique. para no depender excesivamente La strategie sera basee sur le mo- de los prestamos multisectoria- dele des initiatives recemment les de asistencia tecnica, como prises par la Banque et la SFI au ha ocurrido en los ultimos anios. service des nouveaux objectifs * Estrategias a nivel regional y de de developpement du secteur los paises. El documento de es- prive. Compte tenu de la rapi- trategia sectorial debe contener dite des evolutions technolo- las estrategias regionales detalla- giques et de l'elargissement du das del Grupo del Banco en ma- secteur des infrastructures d'in- teria de infraestructura de la formation, le plan pourra 6tendre informaci6n, aprovechando el exi- au dela du secteur des telecom- toso proceso realizado para la es- munications le programme en trategia de telecomunicaciones de cours d'echanges de personnel la regi6n de Europa y Asia Cen- et mettre en place un programme tral. A tales efectos, los directivos bien structure de partage de l'in- de la unidad del Banco y la CFI formation avec d'importantes en- recientemente consolidada debe- treprises du secteur prive. nan asignar la primera prioridad a * Suit4 et evaluation. Le document una evaluaci6n rapida de las po- de strategie pourra utiliser le pre- liticas y estrategias nacionales sent rapport pour un bref rappel sobre infraestructura de la infor- de l'experience recente du maci6n que abarque todos los Groupe de la Banque. Il devra paises en que el Banco o la CFI comporter un cadre precis pour xxiv EXECUTIVE SUMMARY desarrollan actividades. En le suivi et l'&valuation future - esta tarea no deberia preverse des nouvelles strategies du la preparaci6n de informes Groupe de la Banque. A cette acabados sobre los paises, fin, il conviendra de renforcer * sino mas bien el establec- sur trois plans les procedures imiento de una base in- de suivi et d'evaluation : a) au telectual para la formulaci6n niveau des projets et autres de la estrategia operacional activites, en 6tablissant des del Grupo del Banco. Debe consi- systemes de suivi et d'evaluation derarse la posibilidad de utilizar portant principalement sur les in- los recursos de infoDev de ma- terventions echappant actuelle- nera mas sistematica para apoyar ment a ces contr6les (les la labor en materia de politicas y composantes tel&communica- estrategias de infraestructura de la tions/informatique et II des projets informaci6n al nivel de los paises multisectoriels de la Banque et . Dotaci6n de personal. En el do- les services de conseils et autres cumento de estrategia sectorial activites non financieres du se debe incluir un plan de per- Groupe de la Banque); b) en ce feccionamiento continuo de las qui concerne les projets de la SFI, aptitudes del personal para poder en precisant, au stade de l'evalu- satisfacer las necesidades de este ation, la contribution previsible dinamico sector y permitir que el du projet aux objectifs de Banco mantenga su liderazgo en developpement economique et materia de politicas. Se deberia sectoriel et en utilisant les rap- tomar como base el cambio de ports de supervision et d'evalua- orientaci6n que ya se ha produ- tion pour analyser les progres cido en los ultimos aios tanto realises; c) au niveau global, par en el Banco como en la CFI, en l'introduction dans le document respuesta al nuevo programa im- de strategie sectorielle d'indi- pulsado por el sector privado. En cateurs appropries de l'efficacite vista de los acelerados cambios du Groupe de la Banque, y com- tecnol6gicos y del aumento del pris les resultats du secteur et les ambito de acci6n del sector de la effets sur le developpement. infraestructura de la informaci6n, este plan puede contemplar la ampliaci6n a otros ambitos del exitoso programa de intercam- bio de personal que existe en la actualidad en el sector de las te- lecomunicaciones, y el estable- cimiento de un programa bien estructurado de intercambio de conocimientos con importantes empresas del sector privado. . Segutimiento v evaluaci6n. El do- cumento de estrategia sectorial debe basarse en el presente in- forme para resumir las ense- xxv INFORMATION INFRASTRUCTURE ?hanzas recogidas de la ex- - periencia reciente del Grupo del Banco. Asimismo, debe incluir un marco detallado para las futuras actividades de seguimiento y evaluaci6n de las politicas y la estrategia modificadas del Grupo del Banco. Para ello deben perfec- cionarse los actuales proced- imientos de seguimiento y evaluaci6n, en tres niveles: a) al nivel de los proyectos y activi- dades, mediante el establec- imiento de sistemas de seguimiento y evaluaci6n de las intervenciones que actualmente escapan al seguimiento (teleco- municaciones/informatica/com- ponentes de infraestructura de la informaci6n en proyectos multi- sectoriales del Banco y servicios analiticos y de asesoramiento del Grupo del Banco); b) al nivel de los proyectos de la CFI, especifi- cando en la evaluaci6n inicial los aportes previstos de los proyec- tos a los objetivos sectoriales y en terminos de desarrollo, y ha- ciendo un seguimiento de los avances en la consecuci6n de estos en los informes de super- visi6n y evaluaci6n, y c) al nivel mundial, mediante la incorpora- ci6n en el documento de estra- tegia sectorial de indicadores especificos pertinentes de la efi- cacia del Grupo del Banco, in- clusive resultados sectoriales y el impacto en el desarrollo. xxvi ABBREVIATIONS AND ACRONYMS CAS Country Assistance Strategy CFS or CCF Corporate Finance Services Department (IFC) DECRG Development Research Group (WB) ECA Eastern Europe and Central Asia Region EDI Electronic data interchange EMT Energy, Mining, and Telecommunications Department (WB) EMTII Telecommunications and Informatics Division (WB) ERR Economic rate of return ESMAP Energy Sector Management Assistance Program ESW Economic and sector work FIAS Foreign Investment Advisory Services FPSI Finance, Private Sector, and Information Network FRR Financial rate of return GDP Gross domestic product GKP Global Knowledge Partnership GPE Growth of productive enterprise IAD Internal Audit Department IAR/XPSR Investment Assessment Report/Expanded Project Supervision Report ICT Information and communications technology IFC International Finance Corporation II Information infrastructure infoDev Information for Development Program IT Information technology ITU International Telecommunications Union M&E Monitoring and evaluation MIGA Multilateral Investment Guarantee Agency MIS Management information systems OED Operations Evaluation Department (WB) OEG Operations Evaluation Group (IFC) OP Operational Policy PSD Private sector development PSR Project Status Report (WB) PSR Project Supervision Report (IFC) PTO Public telecommunications operator QAG Quality Assurance Group SAL Structural Adjustment Loan SECAL Sector Adjustment Loan SLT Sri Lanka Telecommunications SSP Sector strategy paper TA Technical assistance TAL Technical Assistance Loan TATF Technical Assistance Trust Funds VANS Value-added network services VSAT Very small aperture terminal WBI World Bank Institute WDR World Development Report WTO World Trade Organization xxvii 'I ' 'i '1 *1 '*,* INTRODUCTION Study Objectives and Rationale T he telecommunications and information technology (IT) sectors of the world economy changed dramatically throughout the 1990s, driven by major improvements in technology, the development of new services with potential widespread applications (for example, the Internet), industry restructuring, the shift in the composition of capital flows to de- veloping countries toward private direct investment, and major changes in government policies. The convergence of technologies, services, and mar- kets in these sectors has created what can be best understood as a single information infrastructure (II) sector. The rapid growth of II is expected to further accelerate, offering unprecedented opportunities for economic and human development in poor countries. The advent of e-commerce, in par- ticular, has the potential to alter fundamentally the way commerce and trade are conducted in many countries. Government service delivery, education and training, working, and consumption practices will all be affected. In response to these developments, most de- Agency (MIGA) portfolios; major shifts in the mix veloped countries, and some developing coun- of instruments (lending and nonlending) used by tries, have adopted information society" policies. the Bank; development of new partnerships (for Many developed countries have allocated re- example, infoDev); and staffing and organiza- sources to ensure that II access will be extended tional realignments within the Bank Group. This to all people. Concern is growing, however, that study evaluates the effectiveness of the Bank most developing countries will not be able to Group's response to the challenges of the 1990s. take advantage of II opportunities, furthering the It was also scheduled to contribute to the prepa- gap between rich and poor countries. ration of the Bank Group Sector Strategy Paper The recent rapid changes in the II sector have (SSP) for Information Infrastructure. been accompanied by significant changes in the Bank Group's activities and approach in this Study Scope and Methods sector. Among the changes have been the build- The study was designed as more than an update up of significant International Finance Corpora- of the Operations Evaluation Department's (OED) tion (IFC) and Multilateral Investment Guarantee 1993 telecommunications review' in two major INFORMATION INFRASTRUCTURE respects. First, it is the first joint World Bank database of such projects/components. OED-IFC Operations Evaluation Group (OEG) The review did not carry out any fieldwork evaluation of both Bank and IFC activities in a and drew from the following sources: sector. Second, it attempts to expand the scope * A review of key Bank and IFC policy and strat- of the evaluation beyond the narrow confines of egy documents (OP 4.50 and annual IFC the telecommunications sector-within the con- telecommunications strategies for FY93-00), straints of data availability (see below)-con- project and activity reports (concept papers, sistent with the broader II focus of the SSP. appraisal and completion reports, as well as However, given its limited budget and absence selected supervision reports), and previous of fieldwork, the study did not set out to pro- evaluations. vide fundamentally new insights into the "how * An analysis of selected internal cost data. to" of telecommunications sector reform, a sub- * Interviews of Bank and IFC staff. ject abundantly covered in the recent literature * Telephone calls and exchanges of e-mails (the report does summarize generic lessons on with a sample of stakeholders (borrowers, sector reform and illustrates them with examples private sector and other donors, and multi- drawn from Bank Group experience). Rather, the lateral organizations) and IFC clients selected study's primary focus, and its main value-added, nonrandomly and based on ease of access. is on the relevance, efficacy, and efficiency of * An analysis of evaluation and performance rat- the Bank Group's assistance strategy in a sector ings (OED ratings of the 35 completed projects in which it has been an essentially marginal covered by the study as well as performance player-in financial terms at least. ratings of the 21 IFC investment operations ap- On the World Bank side, the Bank Group ac- proved during FY93-96, using an evaluation tivities reviewed for the study were the 35 framework adapted from the self-evaluation telecommunications projects completed and system).5 evaluated since the 1993 OED review (both More recent Bank projects as well as non- stand-alone telecommunications projects and lending services, for which evaluations of outcome multisector or adjustment loans with telecom- were not yet possible or available, were reviewed munications or II components), the 57 projects for their design characteristics and relevance. Su- that are still active, and Bank-funded sector pervision ratings of active Bank projects were work, as well as some 25 trust-funded advisory used primarily to assess aggregate trends and pat- activities (mostly under the aegis of the Private tems in the current Bank telecommunications Sector Development Department, or PSD) un- portfolio.6 dertaken in the II area since 1993. The study The review of individual Bank and IFC ac- also draws from the findings of two inde- tivities was carried out using the respective pendent evaluations of informatics compo- evaluation frameworks of OED and IFC, which nents of Bank projects and of the Information take into account differences in mandates and for Development (infoDev) program.2 On the project features. In addition, six countries IFC side, the report covers an evaluation of 21 (Brazil, Hungary, Indonesia, Jordan, Sri Lanka, telecom3 investment operations approved since and Tanzania) where both the Bank and IFC FY934 and a review of the design and charac- have been active were selected for joint re- teristics of the three telecommunications advi- views by combined OED/OEG teams. These sory assignments (managed by Corporate country reviews assessed the design, imple- Financial Services, or CFS) and the 27 tele- mentation, and impact of the Bank Group's as- com-related projects financed by the Techni- sistance as a whole. Although not based on a cal Assistance Trust Fund (TATF) program. No randomly selected, representative sample, the systematic review of informatics or II compo- country reviews were helpful in illustrating nents of nontelecommunications IFC opera- good as well as poor examples of Bank Group tions was carried out because there was no synergies. 2 OVERVIEW OF BANK GROUP ASSISTANCE BY SECTOR Uw tntil the late 1980s, economies of scale and scope, together with the public utility characteristics of telecommunications, kept ownership, control, operation, and policymaking squarely in the public sector. While public monopolies in developing countries attempted to expand their telecommunications networks and to fulfill social welfare objectives through such instruments as cross-subsidization, high levels of inefficiency kept them from constructing more than the rudiments of a national telecom- munications system. Recent Sector Developments have generated most of the pressures for telecom- The 1993 OED review documented pressures for munications reform. Relatively high levels of net- change in the sector since the early 1990s. These work penetration and a mature and diversified pressures came from rapid technological inno- market have resulted in reform objectives that vation in the industries supplying equipment to emphasize competition to lower prices and pro- public telecommunications operators (PTOs), vide a diversified portfolio of VANS to a de- advances in the technologies of supply, and in- manding customer base. In contrast, in developing novations in network use that were driven pri- countries, poor network penetration, less- marily by value-added network service (VANS) developed domestic capital markets, limited tech- suppliers. Rapid convergence of telecommuni- nological expertise, and constraints on raising cations and information technology (IT), in- resources by government departments have made creasing differentiation and diversification in foreign private investment indispensable to ex- value-added services, and heightened competi- panding the telecommunications infrastructure. tion based on continual innovation and access Capital-starved PTOs, burdened with inefficient to global markets have now rendered monop- management and budgetary and fiscal constraints, oly control of teleconmmunications infrastructure have increasingly found themselves unable to deal inefficient. with the needs of technological advances or in- in developed countries, technological inno- vestment in large-scale infrastructure rollout. One vations, carried out by large corporate users, way to overcome this has been through policy 3 INFORMATION INFRASTRUCTURE reforms to open the economy and reduce con- izing access to information and knowledge, straints on private and foreign investment. fostering political participation, and enabling But liberalization and privatization, in devel- the bypassing of failing domestic institutions. oped and developing countries alike, have ex- However, a recent study commissioned by posed complex policy and regulatory issues. infoDev for the forthcoming sector strategy paper These include the form that restructuring of the (SSP)Y offers a stark warning: many countries- PTO will take, terms of privatization, market the so-called latecomers, characterized by low structure in competitive segments, legal arrange- connectivity rates, moderate or low literacy lev- ments, functioning of regulatory agencies, com- els, and low incomes-may be unable to par- patibility between political considerations of ticipate in this revolution. This will lead to an universal access and employment, and economic increasing gap in social and economic well- considerations of efficiency, competitiveness, being between the information "haves" and and innovation. Perpetual and rapid changes in "have-nots" of the world. technology, network architecture, and service Given the magnitude and still-evolving nature provision raise complex issues and conflicts in of this challenge, the Bank Group's future role the industry that need to be resolved. Using will have to be redefined in light of its main com- policy and regulatory decisions to do this strains parative advantages, including: the institutional capacity of even the most de- * Its access to and ability to act as a catalyst for veloped countries. global experience and knowledge - Its experience, through IFC, in financing pri- Implications for the Bank Group's vate sector projects in developing countries Development and Poverty * Its ability to engage borrowers on key strate- Reduction Agenda gic and policy issues The development since the 1993 OED review * Its ability to act as an honest broker among with the most profound implications for devel- often-competing donors and private industry oping countries is the growing convergence of standards or agendas telecommunications and information technol- * Its ability to deal with the cross-cutting ogy, currently exemplified by the Internet rev- dimensions of the new information infra- olution, with its attendant opportunities and structure sector (particularly within a Com- risks. Predicting the impact of this revolution on prehensive Development Framework) the developing world is a perilous exercise, par- * Its capability to mobilize substantial resources ticularly as the revolution is disrupting the ob- for the piloting and scaling up of promising served historical relationships between network grassroot initiatives. development and socioeconomic parameters, making if difficult to use econometric techniques. Bank Assistance in the 1990s Nonetheless, a growing body of research points to three main areas in which networking can ac- Policy and Strategic Framework celerate social and economic development: The relative telecommunications policy vacuum * It fosters growth through improved efficiency that the 1993 OED review identified has been for- and competitiveness (including by attracting for- mally filled by OP 4.50 of May 1995. The con- eign investment) and better opportunities to ex- tent of this policy statement is in line with the ploit low factor costs in international markets. recommendations of World DevelopmentReport * It permits a more efficient delivery of social sewv- 1994: Infrastructure for Development. Specifi- ices (education, health care, public adminis- cally, OP 4.50 advocates "(a) largely private pro- tration), particularly to the most isolated and vision of telecommunications services; (b) a poorest segments of the population, such as in regime of open entry and competition; and (c) rural areas. Government responsibility for sector policy and * It provides opportunities to increase social regulation, reconciling commercial and social capital by reinforcing social cohesion, equal- equity objectives." Accordingly, the OP makes 4 OVERVIEW OF BANK GROUP ASSISTANCE BY SECTOR lending to state enterprises contingent on sector Telecommunications in Bank Projects reform and mandates that "the Bank, IFC, and The 1993 OED review documented trends and MIGA work together in formulating joint assis- patterns in Bank telecommunications lending tance packages, making flexible and innovative through the 1980s and early 1990s. It found a use of their various instruments." The OP, how- predominance of relatively large investment ever, has not yet been given operational form loans (on average five per year, accounting for through either a global Bank strategy or a set of 2 to 3 percent of total Bank lending) and an in- regional teleconmmunications strategies (except for crease in lending to the Europe and Central Asia a June 1999 strategy for the Europe and Central Region, mostly from 1990 onward. It also noted Asia Region). Nor does the Bank have a similar the gradual appearance, mostly from 1990, of policy statement, or operational strategy, for IT lending instruments such as sector adjustment or the II sector. loans and multisector loans (including techni- World Development Report 1998: Knowledge cal assistance loans and structural adjustment for Development sowed the seeds for such a lending) that had telecommunications compo- broader approach, however, and the forthcom- nents or conditionality. The OED review saw ing Bank Group sector strategy paper on in- the use of the new lending instruments-to- formation infrastructure (FY01) is expected to gether with an increased emphasis in project de- fill outstanding gaps. The sector strategy paper sign on broader issues of sector structure, is expected to reassess existing telecommuni- competition, and private sector participation- cations policies in the context of the convergence as evidence that the Bank had already em- of telecommunications and IT, take an inte- barked on the "new" telecommunications sector grated Bank Group approach, and operational- agenda that was subsequently formalized in ize the Bank Group policy. OP 4.50. Figure 2.1 Bank Telecommunications Projects Are Increasing in Number, Decreasing in Funding USS million Number of projects 800 _ _ 12 700 - 10 600- 500- 400 -_ 6 - - - 300- 200 - 2 100 O | * L * * * * * * * | * 0 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 * US$ million - Number of projects - - - Linear (US$million) - - Linear (Number of projects) 5 INFORMATION INFRASTRUCTURE Indeed, changes in the scope and form of Bank (TA) loans compared with other instruments- involvement in the sector since 1993 have accel- half the total number of operations, mostly for erated dramatically. The number of projects is in- multisector projects in public enterprise re- creasing, while the total commiitments are declining form, privatization, and regulatory reform. (figure 2.1). Furthermore, the regional distribution * A heavy concentration of activities in the of lending has changed and there has been a shift Africa and Latin America and the Caribbean toward multisectoral lending instruments rather Regions, comprising almost exclusively mul- than individual investments (figure 2.2). tisector loans, which together account for 34 The characteristics of the FY93-99 period are: of the 54 projects. * A significant increase in the total number of approved projects involving telecommunica- Project Objectives and Design tions to an average of eight per year between The trend in the objectives and components of FY93 and FY99. the telecommunications projects reviewed by * A reduction of 40 percent in the estimated vol- the study has followed the overall lending trend ume of Bank funds dedicated to telecommu- in infrastructure. Since 1993, Bank telecommu- nications activities. The new figure is about nications projects have clearly shifted toward the $210 million per year on average-excluding sector reform and privatization objectives em- information technology components of Bank bodied in OP 4.50 and away from physical in- projects-that is, less than 1 percent of total frastructure, commercialization, and capacity Bank lending.2 building (table 2.1 and Annex C). * A complete reversal in the respective shares of Physical objectives. Most projects through telecommunications-only versus multisector the 1980s focused on financing investment to projects, with the latter accounting for as much modernize and expand physical plant and net- as two-thirds of the total during that period. works. In the 1990s, however, the content of the * The predominance of technical assistance physical components changed to spectrum man- Number of evaluated projects Number of evaluated projects 60 Sub-Saharan 60 - Africa 50 _ Middle East and 50 _ Telecommunications North Africa investment Latin America and Telecommunications the Caribbean 40 technical assistance Telecommunications 30 - & Europe and 30 - adjusted Central Asia Multisector technical 20 East Asia 20 - assistance and Pacific 10 _ Africa 10 Multisector other I | Multisector adjusted 0 0 FY86-92 FY93-99 FY86-92 FY93-99 Source: World Bank data. 6 OVERVIEW OF BANK GROUP ASSISTANCE BY SECTOR Project Objectives Have Shifted fro m Table 2.1 Physical Infrastructure toward Reform and Privatization Pre-1993 approvals Post-1993 approvals Project objective Number Percent" Number Percent8 Physical infrastructure 23 42 Rural 7 13 Information infrastructure 4 7 Commercialization or corporatization 18 26 Capacity building 34 62 Sector reform 47 82 Universal service X O Legislation or regulation 37 67 Privatization 42 76 a Does not add to 100 percent Source Wor d Bank and IFC data agement equipment. Competition in mobile Privatization of the incumbent mnonopolt. telephone markets and the increasing demand While most projects before 1993 emphasized re- for mobile telephone services have intensified form and operational efficiency of the monop- emphasis on procurement ancd installation of oly enterprise, subsequent projects have included hardware and software for radio spectrum plan- private sector participation as an objective. Usu- ning, managenment, and monitoring systems. ally this has involved partial or complete divest- Agency-level instituitioni building. In the 1980s, ment of the assets of the state, at least as a projects sought to strengthen the organization long-term goal. and management of state-owned telecommu- lintrodutctioni of'competition. Technological nications enterprises-what the 1993 OED re- advances have made continued monopoly con- view called "institutional strengthening at the trol of the telecommunications sector untenable. agency level." In such projects, TA to help com- And both Bank policy and recent projects have mercialize the monopoly and improve its func- emphasized the entry of additional operators in tioning included training and consultancy in the various segments of the telecommunications financial management, or modernizing ac- sector as an essential condition of sector re- counting and financial systems and procedures; form. Except in Latin America and the Caribbean, commercial management, or billing and collec- most projects focused on introducing competi- tion systems: human resources management; tion in mobile telephone operations, viewing management information systems (MIS), or im- competition in basic wireline service as more of proving project management; and auditing pro- a long-term goal. cedures, materials management, and operations Tariffs and interconnection fees have also and maintenance. been important to liberalization, but have been Regulatorlfiramework. A majority of the proj- mentioned in comparatively fewer projects. ects approved after 1993 (67 percent) had sig- Where tariffs or interconnection have been men- nificant components to set tup a sector-specific tioned, it has usually been in the form of tariff regulatory agency. Projects emphasized the im- studies conducted by external consultants. portance of independent regulation and a legal framework to regulate the sector. Assistance to Economic and Sector Work and Trust the regulatory agency usually consisted of train- Fund-Financed Technical Assistance Activities ing for regulatory staff and, sometimes, special- As in its lending interventions, the Bank has rad- ized consulting services. ically altered the focus of its telecommunications 7 INFORMATION INFRASTRUCTURE nonlending activities. Whereas most of the Bank's shares of projects with informatics components (29 economic and sector work before 1993 was for- percent and 20 percent, respectively) as did the mal country or Regional sector reports and mem- telecom, public sector management, and educa- oranda, none has been issued since 1993, with the tion sectors (respectively, 19 percent, 11 percent, exception of a 1996 discussion paper on telecom- and 10 percent). Few projects had a primary munications reform in Mexico, a 1997 discussion focus on IT; those that did were mostly in the area paper on telecommunications policies for Sub- of tax or trade reform. Saharan Africa, and the 1999 Europe and Central Asia Region telecommunications strategy. Instead, infoDev the Bank's effort has consisted of research papers infoDev is a global grant program that began in or "viewpoints" on generic issues of sector regu- September 1995 at the Bank's initiative. Its ob- lation and competition. In addition, a large research jective is to promote innovative projects that use project on telecommunications reform in Africa was information and communication technology (ICr) recently undertaken by the Bank's Development for economic and social development. The pro- Research Group, with partial funding from the U.S. gram is managed by the Bank, which also pro- Agency for International Development. vides contributions in kind, and it receives funding In addition to Bank-funded nonlending serv- from several bilateral and multilateral donors as ices, 11 telecom-related activities totaling about well as a handful of private enterprises and non- $300,000 were funded in the FY94-98 period profit foundations. infoDev activities include under the Privatization Trust Fund (administered workshops, assessments, demonstration projects, by the Bank's Private Sector Development and and feasibility studies. Privatization Group), and 14 more activities, to- As of December 31, 1999, infoDev had ap- taling about $2.8 million, were funded by the Pol- proved a total of 113 projects (outside the Y2K icy and Human Resources Development Fund Initiative) and had provided grants totaling $19.7 between 1991 and 1996. However, the infor- million, pointing to a relatively small average mation available on the size and scope of these project size of about $174,400. Grants can be activities reveals that they have only partly filled provided only to outside partners that have an the gap in country-focused policy and strategy arn's-length relationship with the World Bank work. Instead, many have concentrated on sup- Group. The largest number of activities involved porting individual privatization transactions or information sharing and dissemination (30.7 per- providing support to Bank project preparation. cent), followed by local services (28 percent), global rule setting (17.3 percent), capacity build- Information Technology in Bank Projects ing and training (17.3 percent), advocacy (4 per- The latest inventory of Bank projects with IT cent), and research and statistics (2.7 percent). components3 identifies more than 1,000 such Thirty-five percent of the projects were interre- projects in the portfolio. Those components have gional or global, while 30 percent were in Africa, included (a) investments in communications in- 16 percent in Latin America and the Caribbean, frastructure and information technology; (b) tech- 9 percent in Europe and Central Asia, and only nical assistance in the set-up of institutional 2 percent in the Middle East and North Africa. In management systems, including project operations addition to these activities, infoDev managed the systems (for example, for health, education, or mi- separately funded Y2K Initiative launched in Feb- crocredit projects) and financial management sys- ruary 1998, which had approved 136 grants in 74 tems (for example, for budget formulation, billing, International Development Association countries, and the like); and (c) related training. Total fund- for a total of $25.6 million as of December .1999. ing for the components is estimated to have av- eraged more than $2.0 billion per year in recent Other Partnerships and Initiatives years, 84 percent for the purchase of goods and in Information Infrastructure 16 percent for consulting services. Africa and The infoDev program is the largest and most Latin America and the Caribbean had the largest visible of numerous Bank-sponsored initiatives 8 OVERVIEW OF BANK GROUP ASSISTANCE BY SECTOR and pairtnerslhips broadly related to the infor- Connectivity Initiative, "ICO/21st Centurv for mation infrastructtire (It) sector. The most re- Africa" and "Rural Telecommunications.' cent initiative is the proposed Global Gateway Kniol lecgejbr-Donelceopm7lesitPartlersb-hip. The (box 2.1). While two initiatives-Africa Con- World Bank Institute (WBI) has recently initiated nectivity Initiative andl Knowledlge for Devel- a forum to foster cooperative activities that pro- opment-deserve special mention because of mote access to, and effective use of, knowledge their dedicated staff and funding.' many oth- andl informiiation as tools of sustainable dlevel- ers are microactivities sponsored by various opment. WBI also provides information dis- parts of the Bank without muclh apparent co- semination, partnership management ancl ordination or monitoring of their design, fund- outreach, and activity coordination for the GKP- ing, and impact. While no complete inventor-y an evolving, inforrmal collaboration of some 50 could be found, a partial list of client-focused, organizations, inclulding in/fbDev. working to knowledge-based initiatives at the World Bank promncote knowledge for development. puLblished on the Global Knowledge Partner- ship's (GKP') Internet site lists 42 activities, IFC Assistance Since 1993 most of which were apparently inspired by the VlPbrldDevelopinentReport 1998: Knoivledgec Policy and Strategic Framework fbr Development andl the Bank's recent em- IFC promotes grovwth in the developing world by phasis on knowledge management. financing private sector projects and providing AJrica C'onnectivityInitiativ'e. In collaboration technical assistance and advice to governments wvitlh other tlnited Nations agencies and multi- and businesses. IFC participates in a project only lateral development organizatiolns the Internet when it can make a special contribution that Connectivity Initiative in the Africa Region pro- complemiients the role of market operators. It motes the dlevelopment of connectivity activities plays a catalytic role. stimulating and moobilizing in Sub-Saharan Africa. The initiative has organ- private investments. By investing in viable com- ized a major conference on the theme in col- mercial enterprises. IFC accepts the same risks and laboration witlh the Economiiic Commission for expects to earn the same returns as private paLr- Africa and the International Development Re- ticipants in the same risk categories. The prof- search Centrem identifiecd four pilot countries- itahility of IFC's investments is important in Ghalna, Malawi, Mozambiqum, ancl Senegal-for bringing market dliscipline to its projects and in accelerated preparation of connectivity comllpo- recluoing the financial burden on its shareholders. nents; and producecl a handbook on connectivity. Profitable IFC investments have been vital in inbof)ev is funding two activities of the Africa demonstrating to private investors that well- B o x 2 . 1.........-...........- fBor Development The World Bank Group has recently launched an effort to a three-year business plan and budget to construct a global build a Global Development Gateway on the World Wide Web gateway and help developing countries create national gate- in support of worldwide poverty alleviation. The Gateway ways in conjunction with the Comprehensive Development will be built by and serve developing country governments, Framework. the international donor conmnunity, the media, civil society, The gateways will support countries in their own initiatives and the private sector. it will help people. agencies, and com- to use the benefits of the Internet fortheir development. Several panies by connecting them to new ideas, funding, and op- are already being iniplemented. The first phase (a proof of con- portunities. It will exchange knowledge and information with cept) was scheduled for launch in the first half of 2000. Major governments and theirpartners that will promote national, re- private companies have been invited to contribute to planning gional, and local development The Bank Group is preparing and funding this progressive venture. 9 INFORMATION INFRASTRUCTURE structured, careful investments can offer attractive equity investments are in eight countries that, risk/reward opportunities. in 1997, had telephone densities9 ranging In the telecommunications sector, IFC policy is from 2.47 (Indonesia) to 30.42 (Hungary). governed by OP 4.50, and its projects are held to * From the traditional basic networks, IFC's in- the same financial, economic, and environmental vestments recently have been moving to the policy standards as other IFC projects.5 IFC's mobile telephone subsector. As of end-FY99, telecommunications sector strategies, which are up- IFC's net investment commitments to mobile dated annually, reflect IFC's mandate to support telephone projects amounted to $298 mil- financially and economically viable projects where lion, or 32 percent of the total. IFC also has private capital is otherwise not available on rea- investments in basic and mobile projects sonable terms and where other investors are pre- which, when combined with "mobile," ac- pared to provide at least 75 percent of the count for 61 percent of total net investment financing. IFC's annual telecommunications sec- commitments, or $559.3 million. tor strategies have generally been focused on * The telecommunications portfolio is heavily identification of opportunities and on a business weighted by funding toward Latin America and marketing plan to support those opportuni- and the Caribbean, which accounts for $373 ties. Given their business and marketing focus, they million, or about 41 percent of net commit- generally have not articulated IFC's development ments, as of the end of FY99. However, re- impact objectives and strategic framework nor cent committed approvals show a shift toward has there been a process to formally discuss them Eastern Europe and Sub-Saharan Africa. From with Bank sector management. 6 its peak of 66 percent in FY96, committed ap- provals on new projects in Latin America and Telecommunications Investments the Caribbean dropped to 30 percent in FY97 IFC made its first investment in telecommuni- and to none in FY98 and FY99. cations in FY70, a $4.5 million loan to help fi- * Three quarters, or $706 million, of net com- nance a $180 million fixed-line expansion project mitments are for projects approved after the in Asia. The next investment was approved in Telecommunications Division was created in FY86. As of end FY99, IFC's cumulative gross FY93. investment commitments7 amounted to $2.2 bil- * The decline in the dollar volume of commit- lion, of which 43 percent, or $0.92 billion, was ments and approvals reflects the recent shift for IFC's own account and the balance of $1.3 from large post-privatization wireline invest- billion was for the account of participant com- ments in the mid-1990s to relatively smaller mercial banks. These commitments were made mobile telephone investments in higher-risk for 41 projects in 30 countries, with total proj- countries. The drop in commitments is due to ect costs of $8.5 billion (figure 2.3). These trans- the difficulty in concluding agreements in late to a mobilization rate8 of 9.3x compared higher-risk countries where the regulatory with 5.6x for IFC's total portfolio in FY99. Av- framework is less developed. erage net investment is $23 million, while av- * The decline in both number and dollar erage project cost is $208 million. These are amounts of approvals in recent years is also significantly larger than IFC's portfolio aver- due to IFC's increased focus on difficult mar- ages. At end-FY99, IFC's exposure in telecom- kets, where projects are few and smaller and munications represented 8 percent of the total where private sector interests are limited. disbursed portfolio. Given the success and demonstration effects Other features of the telecommunications of early IFC and subsequent private sector portfolio include the following: telecommunications projects, along with the * Since FY70, IFC has $280 million total equity overall improved outlook for these liberalized or quasi-equity investments in 28 projects. markets, private flows to these markets have These investments represent 30 percent of net increased. This prompted IFC to move on to commitments. About three-quarters of the newer and riskier markets where IFC's pres- 10 OVERVIEW OF BANK GROUP ASSISTANCE BY SECTOR Figure 2.3 IFC Annual Telecommunications Commitments and Approvals, 1986-99 Amount ($ millions) Number 250 -9 7Fn~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ _~~ ~ ~ ~ ~~~~~~~ *, _8 E~~~~~~~ 200 - L. 150 - 2!SSl. 1% I~~~~~~~~~~~~~~~~~~~~~~~ 10 0 _ Y ea 100 Ie Nb of c t * - * S o u l data, 3 50 ,,r N^ _ ~_ I1 1986 1987 1988 1989 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 Fiscal Year * IFC net at commitment U1 IFC net at approval --U--- Number of commitments -U--Number of approvals Source: IFC data. ence can have a larger impact in stimulating privatization of the state-owned telephone com- private sector investments. pany. One assignment included advice on the li- - By number, 63 percent of the commitments censing of a second national operator. Two are in low- and lower-middle-income coun- assignments were undertaken in FY95 and one tries. But by funding, only 40 percent of com- in FY96. No Foreign Investment Advisory Service mitments are in these countries, reflecting (FIAS)Y activities were undertaken specifically the small size of projects. in the telecommunications sector. * By number, 71 percent of the commitments Technizcal assistance. IFC provides TA fund- are in high- and high-medium-risk countries.'0 ing using grants from bilateral and multilateral By funding, 69 percent of commitments are donors under the Technical Assistance Trust in these countries. Funds (TATF) Program. The TA assignments under the TATF program are developed and Other IFC Activities-Telecommunications task-managed by IFC's operations staff. Tech- Advisory services. IFC's Corporate Finance Ser- nical assistance under this program includes fea- vices Department (CFS) provides fee-based ad- sibility and prefeasibility studies, project visory services that facilitate private sector identification studies, strengthening of the en- participation in state-owned-and-operated en- vironment for private sector development, ca- terprises. It has undertaken three telecommuni- pacity building for private businesses and cations advisory assignments (Ecuador, Haiti, government agencies, and privatization. IFC and Uganda) since its establishment in FY89 (4 approved 27 telecommunications sector TATF percent of the total CFS advisory assignments projects from FY91 to FY99 for a total of $1.9 through FY99). All three assignments dealt with million. 11 INFORMATION INFRASTRUCTURE Among the notable features of the telecom- As of the end of FY99, IFC's investment com- munications TA approvals are: mitments in this area amounted to only $120 * Telecommunications TA approvals in the 1990s million, or 13 percent of the total in the sector. averaged 2.9 percent of the total TA approvals. IFC made its first investment in this area in FY92, * Sub-Saharan Africa accounted for 57 percent of an $8 million investment in a project to manu- the TA projects by funding, followed by Asia (17.5 facture digital telecommunications switches. Other percent), Latin America and the Caribbean (12 IFC investments had the following components: percent), Europe (8.3 percent), and Central Asia, * Upgrade of earth (satellite) station facilities to the Middle East, and North Africa (5.4 percent). introduce and link an electronic data inter- This reflects IFC's more proactive telecommu- change (EDI) service for business-to-business nications strategy in high-risk countries. electronic commerce; support an existing, ex- * Up to 52 percent of the countries with TATF panding Intemet subscriber support base; and assignments are low income. Approvals in enable the creation of private data networks. these countries account for 72 percent of * Establish paging systems; a nationwide satel- funding. By number, 81 percent of the coun- lite-based tiny aperture terminal (VSAT) net- tries with TATF assignments are high and work for data and occasional voice; and high-medium risk. By funding, 95 percent of nationwide e-mail and fax store-and-forward approvals are in these countries. services. . While Sub-Saharan Africa had only 2 percent of * Manufacture, launch, and operate a com- IFC's committed telecommunications invest- mercial satellite for domestic and regional ments, more than half of TA assignments were communications and TV broadcasting appli- in that Region. Latin America and the Caribbean cations. Expand a cable TV network. had 41 percent of IFC's telecommunications in- vestments, although it had only 12 percent of Information Infrastructure Components TA assignments. This is probably due to the dif- of Nontelecommunications Projects ference in the level of reform process and lib- Any project in any sector can have telecommu- eralization between the two Regions. nications and II components. For instance, gen- • Telecommunications TA included such as- eral manufacturing projects can include signments as structuring a privatization strat- telecommunication and II facilities. A hospital egy for the state-owned telecommunications project can have a telemedicine component, operator; providing preprivatization and proj- while an education project can include Internet ect preparation assistance (preparation of fea- infrastructure. However, IFC does not keep a sibility studies, organizational strategies, and separate database of nontelecommunications market assessments), and reviewing the do- projects with II components. mestic telecommunications sector. Investments in telecommunications and II are • Through FY99, IFC made investment com- also made through investment funds. IFC has five mitments to five telecommunications projects committed technology and infrastructure funds that featured TATF assignments related to that are likely to have significant subprojects in project preparation. II. Only two of them have significant invest- * All four TATF assignments on regulatory re- ments in telecommunications and telecom-related form were connected with CFS privatization businesses. One equity fund has a third of its ap- advisory activities. proved investments by value in a wide-area net- work systems integrator. The other fund has Other IFC Activities in the Information made nearly half of its investments in eight Infrastructure Sector telecommunications projects. IFC's investment Whereas its telecommunications investments commitments in these two equity funds repre- have shifted from basic telephone to mobile tele- sent 70 percent of the total investment commit- phone networks, IFC has made only a few non- ments made by IFC in those five technology telecommunications investments in the II sector. and infrastructure funds. 1 2 PROJECT AND SECTOR OUTCOMIES Effticacy of Bank Telecommunications Projects-Performance and Outcomes Telecommunications projects have generally performed better than the overall Bank portfolio or the infrastructure portfolio, but most ratings have declined. Table 3.1 shows OED ratings for the 35 projects com- pleted since 1993 and ratings of projects completed before 1993. The pre- 1993 population was exclusively representative of "old agenda" projects approved in the 1970s and early 1980s. About a fourth of the post-1993 proj- ects reviewed for this study were approved after 1990 and reflect, at least partly, the "new agenda." Completion Rat in gsap pprN ed after 1990) received sub)stantia,lly het- A closer look at the post-1993 P)roject ratings re- ter suistainability and instituitional deveclopment veals that this deterioration miav hav~e been tern- rating,s than those approv-ed in the late 1980s porary . Thei 12 most recent p-rojects (thiose (table 3.2). This is notew~orthy considering that Ta blIe 3 .1 Telecommunications Infrastructure All sectors Rating factor Pre-1993 Post-1993 Pre-1993 Post-1993 Pre-1993 Post-1993 Satisfactory outcome 90 83 82 68 76 89 Likely sustainability 88 72 61 49 55 47 Substantial institutional development impact 30 45 30 33 28 32 Satisfactory Bank performance n.a 85 n.a. 74 n.a. 72 Satisfactory borrower performance n.. 63 n.a. 65 n.a. 618 Sou,rce urn data. 1 3 INFORMATION INFRASTRUCTURE this latest set of projects had more ambitious in- objectives for expanding networks and intro- stitutional development objectives. ducing new technology. In the few instances where the objectives could not be met, the cause Supervision Ratings was poor management or the weak institutional Assessing ongoing projects on the basis of su- capacity of the public telecommunications op- pervision self-ratings is a perilous exercise in eration (PTO), as in Papua New Guinea. Signif- any sector, but for telecommunications the icant project achievements included setting up problem is compounded because the majority of fiberoptic backbones (Pakistan); deploying of the active portfolio (39 of 57 projects) con- digital technology (Senegal, Togo, Western sists of multisector projects whose overall rat- Samoa); increasing transmission capacity through ings may not reflect the performance of the network expansion (C6te d'Ivoire, India, In- telecommunications component. The small donesia, Laos, Nepal, Pakistan, Senegal); in- portfolio (18) of stand-alone telecommunica- creasing switching capacity (Benin, Laos, Poland); tions projects continues to perform well above improving service quality (Fiji, Pakistan); and ex- the Bank average, with only 7 percent rated at- panding telecommunications services to rural risk and no problem projects (compared with areas (Morocco, Nepal). 19 percent and 14 percent for the Bank as a Agency-level institution building. When the whole). However, a review of the latest Pro- institutional development objectives of projects ject Supervision Reports (PSRs) and an FY99 an- are aimed at capacity building in state-owned nual review of portfolio performance rating enterprises, they were only partly successful. for the 39 active multisector projects involving While management information systems (MIS), telecommunications reveals some worrisome billing systems, and rationalized management findings: practices were introduced, the state-owned en- • Almost a third of the PSRs did not comment terprise usually remained ineffective, under- on the telecommunications component, a scoring the inadequacy of existing institutional clear indication of insufficient attention to arrangements for the sector and precipitating sec- telecommunications in the supervision of tor reform. In Nigeria (1990 project), for exam- multisector projects. ple, few of the physical, institutional, and policy - Only seven PSRs rated the telecommunications objectives were met as project implementation component separately (as allowed under the suffered from procurement delays due to Niger- new format). ian Telecommunications Ltd.'s weak imple- * Some 18 percent of the 39 active multisector mentation capacity, excessive government projects were rated as at-risk overall and 15 interference, and rapid management turnover. percent as problem projects. In Uganda (1989), the primary physical objec- tives were met but fell substantially short of Institutional Development Impact and targeted improvements in operational efficiency Sustainability of 'Old Agenda" Projects and quality of service, mainly because of Uganda Physical objectives. Most completed projects Posts and Telecommunications Corporation's achieved-and often exceeded-their physical (now Uganda Telecom Ltd.) weak implemen- ~~~~~~~~~~~~~~aI v a - a Rating factor Pre-1990 approvals Post-1990 approvals Likely sustainability 71 75 Substantial institutional development 35 58 Source: OED data. 14 PROJECT AND SECTOR OUTCOMES tation capacity. However, most projects did fective. Some projects have sought to strengthen achieve the objective of reorganizing the PTO the regulatory structure and the environment on commercial lines, usually by converting it into for private investment even after the institutions a state-owned corporation (C6te d'Ivoire, had been established (as in Colombia and, more Guinea, Hungary, Indonesia, Morocco, Pak- recently, the Dominican Republic). But setting istan, Senegal, Sri Lanka, Togo). up an independent regulator has not necessar- ily been the solution to sector problems. In Institutional Development Impact and many instances (Guinea, Hungary, Poland) the Sustainability of "New Agenda" Projects sector has attracted private investment even Progress on key dimensions of sector reform (as without a strong regulator. In other instances, set- embodied in OP 4.50) by a sample of Bank bor- ting up a regulatory agency has been inade- rowers includes the following. quate stimulus for sector reform. For example, Privatization. The inherently difficult and Tanzania has had limited success despite having contentious nature of privatization has made an independent regulator (although it has at- both project documents and Bank strategy cau- tracted a global mobile telephone operator whose tious about privatization. While almost all proj- operations became a financial success, it has ects after 1994 mention private sector failed to attract private investment in the PTO). participation and restructuring of government Furthermore, long-term sustainability of the agencies, privatization has only become an ex- regulator often remains fragile due to poor in- plicit objective in the most recent projects. In centive structures, overambitious regulators (Tan- most projects aimed at sector reform, Bank ad- zania), or undue government influence. In other visory services have helped the borrower to cases, however, the Bank's help was central to draw up contracts for privatizing the monopoly. the successful strengthening of regulatory ca- Full or partial privatization was successfully com- pacities (Bolivia). pleted with the Bank's help in a number of Competition. Most recent Bank projects have countries (Argentina, Bolivia, the Czech Repub- included assistance with drawing up licenses and lic, Indonesia, Mexico, Peru, Sri Lanka, the Brazil- creating conditions for competition. There has ian state of Rio Grande do Sul). And privatization been a notable difference in the degree and phas- is under way in several others (Bulgaria, ing of competition between fixed and mobile te- Cameroon, Honduras, Madagascar, Moldova, lephony projects. In mobile telephony, it is more Nicaragua). But privatization has been a slow common to have a faster phase-in of competition process and has run into political (Uruguay) because companies are new, and so are man- and other problems: license terms were unac- agement, labor, and technology. Customers are ceptable to private investors (Ecuador), investors less price-sensitive and are therefore able to af- lacked confidence in the institutional environ- ford cost-based tariffs. And the costs of build-out ment, especially given the early experience of obligations are substantially less. For these reasons, private investors (Tanzania). mobile telephony was almost always opened to Regulation. In some countries, Bank efforts competition in the initial stages of reform, with to set up regulatory bodies have foundered on spectacular results on coverage and prices. lack of country capacity, even when projects in- Fixed telephony projects, in contrast, usually cluded training and consultant services. In Mex- involve the acquisition, modernization, and ex- ico (1990), for instance, limited public pansion of recently privatized networks. These administration budgets, noncompetitive gov- networks usually come with an inefficient man- ernment salaries, and burdensome procurement agement, a bloated workforce, outdated tech- made it difficult to upgrade the capabilities of a nology, and non-cost-based tariffs. Faced with the line ministry to meet the regulatory challenges challenge of transforming an inefficient operation of such a technically advanced sector. In other into a profitable company while committing to countries (Bolivia, Indonesia, Panama, Peru, a capital-intensive build-out investment program, Sri Lanka), the regulator has become more ef- project sponsors view limited exclusivity as one 1 5 INFORMATION INFRASTRUCTURE way of lowering risks. FLrtliermore, countries for instance, works on a public-private principle, have seen the granting of exclusive licenses as with private sector participation taking a variety an easy way to maximize privatization proceeds. of forms in collaboration withl the domestic car- On the whole. where initial exclusivity periods rier, TELKOM, and the international carrier, IN- were granted, as in Argentina, Bolivia, Mexico, DOSAT. Private operators are allowed into the and Peru in the mid-1990s, they were initially kept fixed-services market through joint venttires with relatively slhort (or were subsequently shortened these companies. thtrough negotiation) and were offset by obliga- Impact of reform. Many new agenda projects tions to invest in low-density, particularly rural are still in their initial phase of reform imple- and low-income urban, areas. Some observers he- mentation and it is still too early to measure ul- lieve, however, that many cotintries were too timate outcomes-with the exception of mobile eager to grant exclusive licenses without first telephone penetration, which in most cases in- truly "testing" the market (as did, for instance, creased exponentially immediately after the in- Ghana, which in 1998 opened up the fixed-te- troduction of competition (in El Salvador the lephony segment to outright two-party compe- number of mobile telephone lines surged from tition withotut granting even limited exclusivity to 50,000 to 150,000 in less than a year). And even its just-privatized fixed-line operator). Alternative in less-recent projects, dletailed data on ultimate approaches were pursued in some Asian coun- impact-in increased penetration (by areas and tries. The Indonesia telecommunications sector, income groups), improvements in quality of t V of its t m i st T por was s to m i t under. byrtwo Ym W Ban lonslb Prvtzto AdutenLa (---In pric cap regime. thePrvaiztin A oa (38) btpho approve; in F; 9 New0 The000000000 mobil teehn marke was0 patal lbrlzeih laws encte n 193 end00 tOM0;g prov00000ided00 fo the; privaiatno awa and is000000 nowlally00 competitive0 Iwit0 three curetlieses t:CPT and intel(te w stteowned u00V0tiVltes ad h estab icudngBllSot cUb UnfX00X0 f iEtedSlates). Folowing the exira- linen ofa6needn euao,teSpriigAec ino h xlsvt eidi 90 e prtr nee h PROJECT AND SECTOR OUTCOMES service, decreased prices by type of service, tainability of reforms. In Indonesia, an important changes in government fiscal revenues-are not component w.as the organization of seminars, systematically included in supervision reports, workshops, and studies on restructuring and re- particularly for multisector projects. Available form. Where there was resistance to liberaliza- data on selected countries point to clearly pos- tion, a more pragmatic and cautious approach itive impacts in mobile telephone penetration sometimes proved successful. In the early stages (but generally concentrated in middle-to-higher- of liberalization in Sri Lanka, for instance, the income urban groups), lower prices for mobile Bank's pragmatism in making the original ob- telephone and international calls, and significant jectives of the Second Telecommunications proj- improvements in the quality of fixed-telephony ect less ambitious in anticipation of labor service. At the same time, significant variations problems contributed to a more successful out- exist across countries with regard to local call come and enabled the reform process to proceed. pricing, penetration in low-income and rural Bazk factors that contributed to project suc- areas, and the fiscal impact of reform. Peru ap- cess included staff continuity, sequencing of pears to be one of only a few countries where sector assistance, and a long-term presence. the impact of reform has been fully monitored Bank performance in Nigeria (1990 project) suf- and where it has been clearly positive on most fered from excessive staff turnover during su- counts (box 3.1), thus hinting at a possible best pervision and the absence of a full midterm practice model. review. In contrast, in Sri Lanka and Indonesia, continuity of task managers has helped to cre- Factors in Prolect Performance ate good working relationships between the Global marketpressures and borrower participation Bank and country officials and has elicited both in international agreements and negotiations such commitment and support from the borrowers. as the General Agreement on Tariffs and Trade and When projects have not succeeded, it has often the World Trade Organization were powerful in- been because the lessons of previous projects centives for client countries to heed Bank advice were not heeded. In Guatemala (1984), for in- on sector reform. In particular, liberalization, which stance, concern at appraisal about Guatemala in some countries dismantled cross-subsidies and Telephone Company's (GUATEL) implementation introduced competition in value-added network capacity was well founded but did not produce services (VANS), brought price decreases that in- sufficient skepticism regarding the setting of the duced large corporate users-who provide the project implementation schedule. In Ecuador bulk of PTOs' revenues-to migrate to reformed (1995), the Bank was slow to acknowledge that markets that offered better opportunities. This in- the loan was troubled and slow to cancel it. creased pressure on traditional PTOs to follow the reform path themselves. Information Technology Components of Bank Country factors have also had a major influ- Projects ence on the performance of projects, and largely While Bank lending in information technology (IT) explain the divergence of project outcomes. is estimated at more than $2 billion a year, recent Where country commitment was strong, as in In- (1996. 1998) Internal Audit Department reviews donesia and Sri Lanka, Bank projects have been concluded that the sector lacked institutionwide very successful. Gaining support for the reform leadership and focus and that the Bank had not process from the political and btureaucratic lead- recognized the importance of a clearly organized ership was key. In Sri Lanka, for example, the con- informatics practice. The reviews recommended tinuity of officials within the political structure and greater attention to IT as a significant component in the bureaucracy, who have been strong sup- of development and a clear strategic plan.to or- porters of the reform program, has been critical ganize the informatics practice within the Bank. to the relative success of the reform program. The 1998 review also recommended integrating IT Consensus among all major stakeholders activities more closely with other programs related proved crucial to both the success and the sus- to information infrastructure (11), such as inzfoDev. 17 INFORMATION INFRASTRUCTURE Bank Performance-Process Aspects project in FY99, have been excessively low, and Choice of Instruments both in absolute terms and compared with the Standard input indicators point to above-average average figure for telecom-only projects levels of process efficiency in the sector: over the (again, with a few exceptions, such as past five years, lending costs for stand-alone Cameroon, Ecuador, and Gabon). telecommunications projects have averaged These findings confirm the opinion expressed $317,000 per project, with a supervision inten- by Bank staff during interviews that de facto sity of $63,000 per year (compared to overall budget constraints on the amount of time spe- Bank averages of $350,000 and $62,000, re- cialized staff can spend on telecommunications spectively). This is impressive when looked at components of multisector projects have nega- in conjunction with the above-average per- tively affected the quality, and thus the impact formance record of these projects. However, of the Bank's contribution wherever its inter- this does not take into account the recent reliance vention has been limited to this instrument. In on multisector loans and nonlending activities. contrast, the more "standard" levels of lending and Information compiled by OED raises serious supervision resources assigned to telecom-only concerns about the use of multisector loans to projects are associated with better-than-average address the sector reform objectives pursued by performance. In other words, the Bank's attempts the Bank since 1993: to foster broad sector reform "on a shoestring" * Overall supervision ratings for multisector may have been shortsighted. Unfortunately, Bank projects are significantly worse than those of management, particularly country managers, con- telecom-only projects. And available data on tinues to believe that the Bank can use a single the status of sector reform in Bank borrow- multisector operation to both trigger and support ers shows that, on balance, progress has been the implementation of complex structural reform slower where the Bank's involvement was lin- in several sectors at once-and do so within ited to components of multisector projects standard budget coefficients. (with some noteworthy exceptions, such as Assessing the cost-effectiveness of the Bank's Bolivia and Peru). numerous nonlending activities in the 1I sector * Bank supervision under multisector projects would require a minimum amount of data and appears to be below par: almost a third of the information on both their cost and their impact most recent PSRs for such projects do not even (beyond anecdotal "evidence"), which is not mention telecommunications. Of those that do, currently available. However, their sheer num- only a few rate the telecommunications com- ber and relatively small unit size, the large ponent separately. Supervision efforts for amount of scarce staff time they absorb, and the these projects appear to focus excessively on apparent lack of intra-Bank coordination raise se- short-term process aspects (such as procure- rious concerns about overall efficiency. Indeed, ment of consultants), partly because task the recent external review of infoDev (box 3.2) management is assumed by nonsector spe- highlights the program's lack of focus and high cialists, leaving little time and resources for ratio of operating costs to grants as weaknesses substantive Bank inputs on sector policy and to address. Considering that infoDev is widely strategy. considered a highly effective program, it is doubt- * Also, Bank resources assigned to prepara- ful that other Bank-sponsored or -executed non- tion/appraisal of the telecommunications com- lending activities are faring much better. ponents of these projects, at an average of Inefficiency in the deployment of nonlending $30,000 per project,' have been far below services has also been exacerbated by the distorted those spent on telecom-only projects (with a internal "pricing" regimes that result from the va- few noteworthy exceptions, such as in Bolivia, riety of associated funding arrangements, partic- Ecuador, and Kenya). And Bank resources ularly the various tied and untied trust funds spent on the supervision of these compo- administered by different units of the Bank Group. nents, at an average of 2.2 staff-weeks per Occasionally, this has led various parts of the Bank 18 PROJECT AND SECTOR OUTCOMES itDev is the only It noatending pmoram or activity to ha inomtion and communications technology a high rati of ad- been indepen4denft evaluated. A 1998 review of the program mnistraon costo gras (35 percent). and a mfbetween concluded tht it had a high lovel of success in achieving its ma- the magnitde of Y2a issues and the prograds limted capabl- date of aiding the development of iRfoOmSolO 8and commanica- ities. It recommended greater attention to inlomaton dissemiv tions technology, as well as suppoing private sector and nation and evaluation. The program has taken steps to remedy insftutional development h also pointed to some weaknsses, the shorteomings ideified by the review and implement some however, such as inadequate mechanisms to disseminale prop of the recommendations. In particular, overhead costs hove been oct expeiiences, complicated management prcedwes, Insuffi- brought down to about 15 percent which Is in line with compa- cient emphasis on social and developmental dimensions of rable grant programs. Group to offer conflicting advice to a borrower mitments versus 38 percent in the later approvals. (Tanzania), which seriously undermines credi- Because of the significant differences in charac- hility. Such 'pricing' ancl organizational disin- teristics of the evaluation population from the centives wvcre overcome whenever individual later approvals, the report could not use the find- staff took the initiative to informally coordinate ings from the evaluation population to draw sta- their activities at the working level or when the tistically valid inferences about the later approvals. country manager took a strategic approach to Fourteen of the 21 operations achieved a sat- the Bank's involvement in the sector, based on isfactory-or-better development outcome rating.2 quality rather than internal pricing criteria. This is consistent with the 67 percent success- ful-or-better development outcome rating of the IFC Projects-Outcomes and Performance 1996-98 Inv cstment Assessment Report (IAR) population covering all IFC sectors. Figure 3.1 Development Outcome shovw s the summary rating results for each per- The evaluation population consisted of the 21 formance dimension. Operations wvith satisfac- mature operations approved from FY93 to FY96. tolry-or-better project and company business Approvals since FY97 have not reaclhed sufficient success generally had satisfactory-or-better de- maturity for evaluation. The clevelopment out- velopment outcomes, consistent w ith the 1996-98 come analysis was done using a modifiecl IAR/Expanded Project Supervision Report (XPSR) IAR.iXPSR evaluation framework (clescribed in pattern. The performance was better based on Annex A). wNeighted average (by net investmiient commit- The characteristics of the evaluation population ments). w ith 81 percent of the 21 operations oh- clo not closely resemblne the later approvals. The taining a satisfactory-or-better rating compared evaluation population has 18 percent of investment wvith 68 percent achieved by the 1996-98 commitments by amotnts in the mobile telephone IAR/XPSR population. sector. The Later approvals have 83 percent. The Pr)olect business success. Thirteen of the 21 evaluation population lias 47 percent of investment projects evaluated had satisfactory-or-better proj- commilitlmients by amounts in Latin America and the ect blusiness success. These 13 projects represent Caribbean Region, wlhile only 18 percent of the 7' percent of net IFC commitments on the 21 op- later approvals are in that Region. ITp to 64 per- erations. MIajor contributors to project success cent of the evaluation population's investment were a transparent and conducive regulatory coimmitments enjoyed limited exclusivity. Only environment (limitedl exclusivity or duopoly), 21 percent of the later approvals have the samne faster-than-expecteci demand growth, and strong status. Greenfield projects accotuint for 43 percent management. Reasons for poor performance in- of the evaluation population's investment coin- clided cutthroat competition, delayed imple- 1 9 INFORMATION INFRASTRUCTURE i !, . t i ii li iliSi . S Percentage of operations with rating of satisfactory or better 120- 100 80 - 60 40 20 0 Development Project Company Growth of Growth of the Living outcome business business private enterprises economy standards * Telecommunications El 1996-98 Investment * Telecommunications * 1996-98 Investment unweighted Assessment Reports and weighted by net IFC Assessment Reports and Expanded Project Supervision commitment Expanded Project Source IFC data. Reports, unweighted Supervision Reports, weighted mentation, overexpansion, high capital costs, 91 percent of the net IFC commitments on the 21 lack of economies of scale, and interconnection operations. Most of the projects were first or early revenue-sharing issues. entrants in the telecommunications sector and Company business success. The evaluation their success has inspired further private sector par- population scored lowest in this performance ticipation. Some projects have stimulated existing dimension; only 12 out of 21 were rated satis- players and subsequent entrants to improve op- factory or better. These 12 projects with satis- erations in order to stay competitive. Other proj- factory-or-better company business performance ects enabled existing private enterprises to expand account for 67 percent of net IFC commitments their businesses and operate more efficiently be- on the 21 operations. Most companies achieved cause of access to reliable telecommunications fa- a satisfactory-or-better rating because of the suc- cilities. Telecommunications network projects cess of the projects. One operation with a satis- created downstream linkages such as mobile factory-or-better rating on project business success phone dealers and service centers, and rural com- was less than satisfactory on company business munity-operated payphone centers. success.3 This is consistent with IFC's experi- Contribution to growth of the economy. Sev- ence from the 1996-98 IAR/XPSR population, enteen projects (representing 85 percent of net IFC which implies that project success does not guar- commitments on the 21 operations) out of 21 re- antee company success. ceived a satisfactory-or-better rating. Most projects Contribution to growth of productive private provided telecommunications services that are enterprise. Sixteen of the 21 projects were rated worth more to customers than the price they satisfactory or better. These 16 projects comprise pay. The projects helped stimulate economic 20 PROJECT AND SECTOR OUTCOMES growth by providing efficient and less-costly com- fected by weak strategic partners or none at all, munication infrastructure. They have also re- an unclear regulatory environment, and eco- moved the financial burden from the government nomic slowdown. of operating telecommunications companies. Asia was the worst performer: only one of four Given their generally good operating performance, operations achieved a satisfactory-or-better de- these projects have provided substantial tax rev- velopment outcome. Regulatory and sponsor is- enues to the government. sues adversely affected project performance. In Contribution to improving livinzg standards. two markets, an unexpected issuance of addi- The evaluated operations performed best in this tional operating licenses put extreme pressure on category. Twenty projects out of 21 were rated the viability of the less-than-strong existing op- satisfactory or better. These 20 projects account erators. The sponsor's lack of commitment and for 97 percent of net IFC commitments on the weak financial resources brought down one proj- 21 operations. The projects' contribution to liv- ect. A project in a country heavily affected by the ing standards included job creation, skills train- Asian crisis was able to respond effectively by ing, access to telephones, reduced waiting time scaling down project scope. Consequently, this for new telephone connection, lower telephone project appears likely to weather the crisis. tariffs, improved service, clearer connection, and a higher percentage of telephone calls get- Results by Subsector ting through and completed. Nonbasic telecom- The report classified the projects into four sub- munications projects provided increased access sectors: mobile telephone4 only, basic (fixed to timely information. In all cases, that project line), mobile and basic, and others (cable tele- economic rates of return inclusive of consumer vision, satellite, telecommunications peripherals, surplus almost certainly exceeded their formal and other miscellaneous telecommunications rates of return indicates that projects have ben- services). Because of the small size of the eval- efited the economies at large more than they have uation population, the "basic and mobile" and benefited the project financiers. 'others" subsectors are underrepresented and therefore could not be used to predict with con- Results by Region fidence the results of similar operations. The evaluation population is too small to draw The eight investment operations with "mobile any conclusion on regional trends. For instance, telephone only" projects scored high on all per- there are only two projects in Sub-Saharan Africa formance dimensions: seven out of eight had a and one in Central Asia, the Middle East, and North satisfactory-or-better development outcome rating, Africa. This section of the report, therefore, illus- accounting for 99 percent of net IFC investment trates fFC's experience in the different Regions and commitments on the eight mobile telephone does not draw conclusions on regional trends. operations. Mobile telephone projects were either Evaluated operations in Sub-Saharan Africa pioneers or early entrants in the sector and en- and Central Asia, the Middle East, and North joyed fixed-term exclusivity or duopoly. The only Africa had a perfect satisfactory-or-better rating mobile telephone project that did not do well suf- on all development outcome performance di- fered from cutthroat competition that pushed mo- mensions, despite the relatively high-risk rating bile telephone tariffs in that market to one of the of the countries in these two Regions. Latin Amer- lowest in the world. "Basic and mobile" operations ica and the Caribbean ranked second, with six were the second-best performing subsector with out of seven receiving a satisfactory-or-better de- two out of three achieving a satisfactory-or-better velopment outcome, although only three out of development outcome. The poor-performing proj- seven of the operations had a satisfactory-or- ect was hit by an economic slowdown that damp- better rating on company business success. Four ened demand and by the foreign technical partner's out of seven operations in Europe achieved a sat- withdrawal from the country. isfactory-or-better development outcome rating. Four out of seven of the "basic" investment The poor-performing projects were adversely af- operations achieved a satisfactory-or-better 2 1 INFORMATION INFRASTRUCTURE development outcome. These four operations itive operating efficiency, strong billing op- account for 57 percent of IFC net commitments erations, and good customer services. on the seven "basic" operations. All three op- * Operating license. Pioneering/first entrant, erations with less-than-satisfactory performance limited exclusivity (mostly for fixed-line te- had problems with, or had no strong and com- lephony) or duopoly status. mitted, strategic partners. In one operation, the Among the reasons for less than satisfactory sponsors' conflicting interests stalled the proj- investment performance (and development out- ect. Another operation was a rural telephony come) are: network where the original sponsor abandoned * Regulatory issues. Unfavorable and nontrans- the project after concluding that growth po- parent regulatory environment, unbalanced rev- tential was limited given the small population enue sharing with the national operator, high of the concession area. The third operation license fees, short license periods, overcrowded had difficulty meeting build-out requirements market, subsidized state-owned competitors, because of no strategic partner and weak fi- and unfavorable interconnection agreements. nancial resources. * Sponsor/managementproblems. Weak spon- Investment operations in the "other" sector sor commitment, expertise, experience, co- performed the worst, with only one out of three ordination, control and/or resources, and rated satisfactory or better in development out- working relationship with local partners; lack come. One less-than-satisfactory project overbuilt of management expertise. capacity and was unable to induce a propor- * Macro environment. Currency depreciation, tionate increase in demand due to a slowdown financial crisis, and economic slowdown. in the economy. The other project that did not do well experienced significant delays in im- Main Lessons from Bank Group Project plementation. Demand for these peripherals or Experience nonbasic services proved to be income-elastic It is too early to measure the full impact of reforms and therefore more vulnerable to macroeco- promoted by Bank projects under the "new" nomic swings. agenda as most are in their early stages of im- plementation and only a handful have been eval- Performance Drivers and Obstacles uated. Nonetheless, the evaluation of mature IFC Operations that achieved a satisfactory-or-better projects and the review of ongoing Bank projects development outcome are closely associated with offer several lessons regarding sector reform ap- satisfactory-or-better investment performance. Of proaches (lessons specific to IFC telecommuni- the 14 operations with a satisfactory-or-better de- cations operations are presented in Annex B): velopment outcome, 12 have satisfactory-or-bet- * Even partial competition, through liberaliza- ter investment performance. IFC operations that tion of mobile telephony, can bring about in- achieved satisfactory-or-better investment per- creased investments, lower prices, and formance (and development outcome) generally improved service quality. feature the following characteristics.5 * Establishing a regulatory body does not ob- * Markets and marketing. The ability to respond viate the need for a well-articulated govern- quickly to emerging market opportunities and ment policy and strategy. threats, such as accelerating expansion capi- * Weak institutional capacity in poor countries tal expenditures to meet a rapidly rising de- is a major constraint to effective regulation. mand, introducing prepaid calling cards and Two steps are crucial to regulatory sustain- other innovative payment plans to address ability: (a) adequate capacity building within declining customer credit quality, and build- the regulatory body to ensure its effectiveness ing brand loyalty in anticipation of new or in- and efficiency; and (b) attention to the broader creased competition. institutional environment that would deter- * Physicalproducts and services. Superior net- mine the regulatory structure's ability to func- work quality, advanced technology, compet- tion with transparency, accountability, and 22 PROJECT AND SECTOR OUTCOMES independence from undue political interfer- task managers (five task managers in six ence. Ghana demonstrates the importance of years), leading to a lack of consistent support this: although the Bank helped in drafting leg- to the borrower. islation and setting up the regulator, the reg- * The choice of Bank Group instruments- ulator has not operated successfully. lending and nonlending-and their effective Transparency in licensing and tariff setting (par- coordination has had a demonstrable effect on ticularly for interconnection rates), as well as the Bank Group's effectiveness in the sector. clarity regarding the role of the incumbent op- erator, are critical to attracting private investment. Sector Outcomes In two areas, however, Bank and IFC ex- Broad sector outcomes need to be differentiated perience and views have occasionally diverged: from the outcomes of specific Bank Group in- (a) the degree to which, in some countries, ex- terventions because: (a) Bank Group interven- clusivity periods need to be granted initially tions have been limited to a relatively small in order to attract private investment; and (b) subset of developing countries, and (b) the out- whether the privatization of the incumbent come of Bank Group projects was assessed operator should be completed and an inde- against stated objectives that were sometimes pendent regulatory body fully established be- narrow in scope (such as specific geographical fore opening the sector to private investment areas or parts of the whole network) or focused (in some countries, Poland among them, pri- exclusively on means (such as initiation of the vatization was undertaken relatively success- privatization process, enactment of a law) rather fully, even in the absence of an independent than results. Three sector-specific trends merit regulator). particular analysis: institutional reform in the Several lessons relate to the Bank Group's own telecommunications sector, improvement of na- performance in the sector: tional telecommunications systems, and devel- = An objective assessment of government com- opment of an II capacity. The information mnitment to reform is essential. As the Bank's presented here derives mainly from recent Bank Second Telecommunications Project in Hungaty and International Telecommunications Union demonstrated, when there is political will and (ITU) data. commitment to the reform process, even sup- porting the public PTO through investment Institutional Adjustment and Sector Reform lending can be highly instrumental in the ac- Ptrvatization. A recent ITU report finds that 88 tual reform and privatization. In contrast, when countries have fully or partly privatized their political will and institutional capabilities for re- PTOs.7 In the majority of countries, the PTO is form were lacking, reform lagged: in Tanzania still government-owned, but if the countries that many of the critical elements-capital restruc- have plans to privatize do so, government- turing and tariff increases-were finally owned PTOs will become a minority by 2003. achieved only after the International Devel- However, most poor countries have not priva- opment Association threatened to suspend tized. The Arab states and 20 to 30 percent of processing of the third project. African countries have not yet even accom- * Indonesia and Sri Lanka have demonstrated plished the old agenda objective of separating the advantage of long-term Bank involve- postal from telecommunications activities. Every ment and staff continuity. In these countries, Region has a significant number of countries the Bank sequenced its reform support- that have not acted on the ownership issue. In- from physical investment, to institutional deed, even in Western Europe, the majority of strengthening and corporatization of the mo- countries have not yet privatized. nopoly, to introduction of competition and Regulation. The trend in regulation tends to regulation. In Papua New Guinea, Bank su- follow the trend in ownership. By August pervision was fragmented and deficient pri- 1999, 84 countries had established separate marily due to an unusually high turnover of telecommunications regulators. By the end of 23 INFORMATION INFRASTRUCTURE Service Countries Local 32 Long distance 26 International 26 Leased lines 40 Cellular mobile 66 Internet 80 Source: International Telecommun cations Uninn 2000, another 15 countries were expected to easiest to implement and resistance to them was act. As with ownership, poor countries are relatively low (see box 3.3). Thus, the institutional lagging behind in establishing independent adjustment process for developing countries is regulatory agencies. still at an early stage, andl for most poor coun- Competition. Competition is most common in tries it is just beginning. the youngest services, mobile telephone and For countries that have embarked on institu- Internet (table 3.3). Internet service providers typ- tional adjustments, continuing attention and sup- icallv lease telecommunications capacity to pro- port to capacity building is reqJuired to ensure vicle value-added services and compete directly their long-term sustainability. There is already with a PTO if the PTO establishes its owvn Internet enough evidence in some countries of rushed services. Competition in the mobile telephone ancd failed privatizations, misguicecl regulation subsector has largely been driven by private in- programs, and misdirected competition policies vestment in new operators. to lemonstrate that institutional acljustment is not The ITU does not distinguish in its data be- a one-time country injection to aclhieve change. tween developed and developing countries. The most important investments in institutional Thus, the data for some regions look better than adjustment projects are those made in local others because they include levelopecl countries human capital. that undertook institutional adljustments some time ago. If about 50 high-income countries are Improvement of the National removed fromii the lata, then the progress per- Telecommunications System centages for institutional adjustment decline sig- In the aggregate. significant amounts of private nificantly. The first countries to adlopt adjustments investment have been committecd to national were those where institutional adljustments were telecommunications systems, andl there is a gen- 2 4 PROJECT AND SECTOR OUTCOMES 1998. excluding major Income group 1995 1998 Increase urban centers Low-income countries (61) . 1.6 / 1.1 Lower-middle-income countries (60) 8.2 . 8 6.6 Upper-middle-income countries (33) ¶$ 16.5 .; 13.8 High-incomc countries (52) -1 - . 56.1 iAA 47.8 Africa -L.. 2.2 05 14 Asia --4 7.3 - 6.3 Source International Telecommun cat ons Union. eral impression that in many dex-eloping coon- mal national cov erage. Most experts consider that tries the systems have improved significantly. cotuntries need to reach 10 before they have a The principal measures of improvement in na- foundation for developing a national II of any tional systems are increased connectivity, in- conse(luence for the domestic cconomy. vestment in infrastructure, andl improved service Despite a smzall increase in main line penetra- quality. In all tliree areas, however, low-inconme tion rates between 1995 and 1998, low-income countries are faring far w orse than countries in countries are a long wvay from having developedl higher income catcgories, national telecomnmunications systems, let alone Gon nectivi t'. National telecoimnunications sys- Ils. At the present growth rate, it xvvill take nearly tems provide the connectivity essential for any a decade for those countries to reach an average kind of electronic commnUnication or inform1ation penetration rate of 5, and about 15 years to reach exchange. Althotigh the prnmcuay purpose of the sys- 10. As a significant majority of low-income coun- tems now in place is voice communication, the It tries have penetration rates well below the aver- now being developed in many countries can trans- age, 35 to -tO countries will fall well hehind even mit any form of comm-nunication or information ex- at this rate of development. Moreover, the gap he- change-voice, clata, image, 1iUSiC, video. Those tween the low-incorme countries and the other whco clo not have access to a network connection countries has risen dramatically and wvill likely are denied access to all services. For those who are continue to widen througlh 200) and heyond. The connected, the range of service possibilities is de- absolute increase in penetration rates for middle- terminecl by the capacitv and capabilities of the net- incomiie countries is 6 times that of the lowv-income work. It may be only voice telephone: it may be countries. Surprisingly, it is the high-incomiie coun- all II services; or, more likely, it may he a range of trics, those that already have universal service, that services between these extremnes. have the largest increase in main lines per 100 in- The most commonly used indicator of telecom- habitants of any incomiie group. This suggests that municarions s-stem development in a coUntry is the grow th in secondl lines in households to pro- total main lines per 100 population. Table 3.4 sum- vide for contintious Internet access in many higih- marizes [TtJ data for countries hy income group. income countries may he faster than the growth Countries with main line penetration rates of in first lines in developing cotintries. 50 or higlier are regardledl as having a tiniversal Remioving the data for major urban centers af- service coverage of at least 90 percent of house- forcds a better indication of gcographical cover- holds ancd reasonable access to a telephone in age. Th e data show lower numbers across all all inhabited areas. Countries with penetration countries: 83 countries have penetraLtion rates rates below 5 typically do not have even mini- below 4 and 56 countries have rates below l. 25 INFORMATION INFRASTRUCTURE Investments. The trend in investment helps ex- The contrast between investment patterns of plain why many countries have not been able middle-income and low-income countries is also to develop national telecommunications sys- striking when comparing recent trends in private tems more rapidly. More recent (1998) ITU data sector financing (figure 3.3). put telecommunications investments in high- This reflects the fact that, as for other infra- income countries at $134 per capita. In contrast, structure sectors, the recent increase in private in at least 95 countries, investment was less financing of telecommunications has been heav- than $10 per capita and in at least 21 countries, ily concentrated in a few, mostly upper-middle- it was less than $1 per capita (if nonreporting income, Latin American and Asian countries. candidate countries are included, the number Quality of senice. Notwithstanding the limited would increase to about 45). More significantly, and uneven progress achieved on overall in- figure 3.2 shows that investments, as a per- vestment levels, improvements in the quality of centage of GDP, have increased more rapidly in service have been substantial and widespread, middle-income and upper-middle-income coun- as illustrated by the statistics on faults per line. tries than in OECD countries since 1995. This This improvement has been most pronounced provides some hope that the gap with OECD and sustained in countries that have carried out countries will gradually be closed. Meanwhile, deep and early sector reform (such as Hungary, however, investments have remained flat in Indonesia, Mexico, Peru, Sri Lanka). low-income countries (with the important ex- ception of China, which has spent a remarkable Information Infrastructure Development 3.3 percent of GDP on the telecommunications Given the enormous challenges of building na- sector in the past five years). tional telecommunications systems to supply L i - t1 * L ! I Az1!' S_. Percent of GDP 1.4 - 1.2 - , 1.0 _ 0.6 .4 0- -, 0.4 - 0.2 0.0 I I I I 1991 1992 1993 1994 1995 1996 1997 1998 1999 Year - Lower income - _ Lower-middle _ Upper-middle - OECD lexcl. China) income income Source: Pyramid Research. 26 PROJECT AND SECTOR OUTCOMES basic services in developing countries, adding the lic-private partnerships (for example, to foster even more difficult challenge of establishing na- low-income and/or rural access). tional Ils might seem unrealistic. If high-income Every country will have a program, however countries are undertaking investments per capita large or small, to improve its national telecom- in the telecommunications sector that are 12 munications system. By considering specific op- times those of lower-middle-income countries portunities for II improvements where benefits and 54 times those of low-income countries, greatly exceed additional costs, these programs there is little hope that most developing coun- can move along a more productive path at every tries will participate in the global information stage of development. By integrating telecom- economy of the next century. munications system and II development into a The challenge-and the opportunity-lie in single program, a country can generate additional the ability of II to make a national telecommu- resources that may have an accelerator affect on nications system significantly more efficient, the telecommunications sector and the econ- more capable, more valuable, and more prof- omy. If developing countries and development itable to individual users, firms, industries, and institutions see II development as a step in a lin- the government. But that will not happen auto- ear process that must follow telecommunica- matically, nor will it happen without careful re- tions system development, few poor countries search, planning, and implementation. will ever get to the II step. But if II development Governments and regulators should establish is seen as a catalyst to stimulate a more rapid an II policy and regulatory framework with a building of the national telecommunications sys- view to maximizing efficient private sector par- tems of developing countries, chances of success ticipation, supplemented as appropriate by pub- increase measurably. Figure 3.3 . . As Has the Percentage of Private Sector Financing Percent 100 90 _ 80 _ 70- 60- 50- 40- 30- 20- 10 1991-94 1995-99 1991-94 1995-99 Lower income Middle income * Private Ii Public Source: Pyramid Research. 27 INFORMATION INFRASTRUCTURE Relevance of Bank Group Interventions terest and IFC's development agenda, IFC's in- in the 1990s vestee companies are heavily concentrated (up The relevance of Bank Group interventions can to 63 percent) in low-income and lower-middle- be assessed by answering four questions: income countries, although they account for only * Were the stated objectives of the interventions 40 percent of investment volume. This also re- in line with existing Bank Group policies? flects the relatively small size of investments in - Were the interventions focused on the coun- poorer countries. More investments have been tries most in need of assistance? made in poorer countries in recent years * Were the instruments used suited to the stated (Moldova, Tanzania, Uganda, and Zambia). objectives and to individual country circum- Appropriateness of Bank Group instruments, stances? The Bank's use of lending and nonlending in- * Have Bank Group policies been appropriate struments in the sector has relied excessively on given the evolving sector environment? "inefficient" multisector loans and a fragmented Consistency with existing policies. Detailed approach to advisory and nonlending services. analysis of the objectives and design of Bank in- The countries where the Bank has had the most terventions and IFC activities shows that they have impact, Indonesia and Sri Lanka, are those where been broadly consistent with the principles in OP it took a long-term view of reform and sup- 4.50. In the few countries where Bank investment ported initiation and implementation with a mix lending for the telecomnmunications sector con- of instruments, including economic and sector tinued (Indonesia, Jordan, Sri Lanka), it was ex- work, trust-funded activities, and investment plicitly tied to firm government commitments on and technical assistance (TA) lending. The Bank sector reform. IFC, given its private sector man- Group's organizational structure has not fos- date, has focused on supporting the "private tered a coordinated approach to the use of Bank provision" aspect of the policy. The Bank, for its and IFC instruments. Of particular concern, with- part, has tended to emphasize competition and out clear criteria on their priority use there is po- regulation. This difference in focus is evident in tential for overlap (or even conflict) between the occasional IFC financing of projects in coun- Corporate Finance Services advisory services tries with clearly suboptimal regulatory and pol- directed to the government (including the at- icy environments (for example, no independent tendant Technical Assistance Trust Funds regulator and the main telephone operator re- assignments) and similar advisory activities pro- maining state-owned, as in El Salvador and vided or funded by the Bank.8 The choice of Uganda), while the Bank is alleged to have taken Bank Group instrument in funding the main in- an excessively dogmatic view and blocked po- cumbent operator before privatization has been tentially beneficial IFC preprivatization loans in another ambiguous area, again due to the lack some cases (as in Bulgaria and Romania). While of clear criteria: Bank funding has been criticized understandable, these different perspectives need for risking the perpetuation of state dominance to be reconciled at the country level via an agree- of the sector, while IFC preprivatization loans ment on a joint sector strategy. have been viewed by some as creating percep- Country selection. The review of sector out- tions of conflict of interest (that is, between the comes suggests that lower-income countries were Bank Group's dual role as advisor and investor). (and are) in more urgent need of assistance than To address this concern, IFC does not participate others, given the lag in sector reform, the grow- in privatization projects where it has had an ad- ing gaps in investments and teledensity, and visory role. their lesser access to foreign private capital. The Relevance of Bank Group policy and strat- breakdown of Bank interventions by country in- egy to sector environment. Overall, the policy come groups suggests that the Bank has been re- principles of OP 4.50 have remained relevant sponsive to that need: almost half of its projects to global sector trends and needs. However, have been in lower-income countries (including they need to be updated in two areas: the gov- 15 in Sub-Saharan Africa). Reflecting investors' in- ernment's (and the Bank's) role in fostering 28 PROJECT AND SECTOR OUTCOMES access by the poor (especially in rural areas) and gies has contributed to the fragmentation of the the implications of convergence, particularly Bank Group's approach to II in individual coun- of the Internet revolution (the next chapter tries, with critical decisions too often left to staff elaborates on these two issues). Most important, or consultants-or worse yet, not addressed at these general policy principles have not been all for lack of budget resources (as appears to translated into integrated sector operational be the case in many multisector projects). (For strategies, either at the global or Regional/coun- example, regarding what advice to give on sec- try level, until the very recent Eastern Europe tor structure, licensing regimes, approach to uni- and Central Asia Region strategy. The IFC's an- versal service, or interconnection pricing; how nual "strategies" have been more akin to de- to decide on the respective roles of the Bank tailed marketing plans and have lacked a and IFC; or how to establish linkages with well-articulated development agenda and strate- other sectors such as health and education.) gic framework. The absence of detailed strate- 29 r 1 * / " b "'irtl OUTSTANDING POLICY AND STRATEGIC ISSUES Rural and Universal Access T * 1he Bank's role in rural telecommunications and the linkages of the sector to the poverty-reduction agenda has been unclear. While the importance of telecommunications to the efficient functioning of the government and the economy is recognized, when it comes to allocating Bank resources for poverty reduction, the telecommunications sector has taken a back seat to health, education, and water in the Country Assistance Strat- egy (CAS) process. The linkages of informatics and telecommunications to the Bank's poverty-reduction mandate have not been sufficiently well re- searched or sold to the Regional teams. Rural components have generally been rare in Bank Group as to when rural access issues are telecommunications projects-only five proj- best addressed. Some have argued that it should ects before 1993 had such components, al- be a two-stage process, since there is little ev- though that situation may be changing (see idence that pure rural telephony is commer- box 4.1). One reason is the ambiguity of the cially viable on a stand-alone basis in most Bank Group's policy on the role of the public developing countries. In their view, initial ef- sector in fostering rural access, which has led forts should concentrate on sector liberalization staff to avoid promoting innovative policy op- and competition to improve the overall effi- tions or proposing Bank financial support (for ciency of the country's telecommunications example, to private-public partnerships) in that network and help reduce the build-out cost to area. This has left a gap that IFC has not been extend coverage to the rural areas. Only in a able to fill, apparently for lack of investor in- second stage, several years later when nu- terest (only two such IFC investments have merous private competitors are active, can a been made, and their performance has been less sector-wide universal access charge be levied than satisfactory). to subsidize expansion of rural coverage. This Besides the role of the public sector, there was the approach followed by Chile. Others has also been a lack of agreement within the have argued that there is too high a risk of rural 31 INFORMATION INFRASTRUCTURE Somerocng areqnon pates ruaralv conpsts,A refore ae$ mndtaInrineaMstrutuhre Issueswhponaeivt approved rank n t fro Bthe outset,faswas d~oneb ei-u m( osee Nonwofthlme Bakpoets aYppOrttove beore- pone to pepar listrtegy tor concluively WoLIpporect oneorthe n chotragin tend, fxoterlof tae newser. pjcs co.l and related hervie apporuach. lavatpdcmuute h ont e o tahlaese th poe access ntio nremotetctien Desa- The proj ThcteIsaespcMl ogieent man itsenentino edamples oatf p-vlpetq n t u rojectisn.nei,hsAcotc IsISSO WC iui catoions ines Afrnica aR etin Ainteria end Asiah$ wrigeaerenera rsobj cies ofiex moning frOMMtUenfcain orates at (sowmedliaeliist worrte kb in fonsev tHoevworW theisr Ian Ieinformai#ocrni nem*1ork to&faciliatesustanacels poorst ad lthetppoach$ outries, Tbe promect wlionl prox 4a d.n2 . the viTlae paimsto exprgamd wthl posftfr etomrkt wide p c at pears1#~O pdsroleisig, Theexerienlpscettoftk Red pyCie cces to r tra hegsponso; extensive-maic,Itrncet, hank area of tfficc a Prunraloa nogvrm netwlorgan tL0 riMsce n ovrmaiong somvie s;~l inrdiae 1ie snailb izaiesdvlopmnt aggrruravly arLsRrengthae installtono r vealedtheatdte iledsvie pa lsio g onextistn postale Each to sel siannetawhol rkl f r ietiocmnotpruflthrpLnghzLtPrLmr woma eletreprneircn networkn avergeop $2ga ande or$gLla halt illa empayp iess the onustrof tiveraiztottwll alove rniencan tsory traewicekso otattwill enhncalpa capista Inco ie paccs isse neovici erd beingy anddeedi theyifiLl HoitadstiAproacthe istitoInfrationpbliya( ares not putliin surlyt ofinanthe freorm aoegend Infratrcituriieevn Isses ce n d ita prighte fronteouset,oas whotsast done byperuoit (se Nione ofThe Banrpojectprmts aprovaed beftorePrt l)oxy3.1).oDevelopd ing icountr experienc is shp 1993 tadd ese broadisioissues,butainmanten- 3unic n in A ri L , s genl (somes isuppred by P,fo Dev). Howdevser,i theiran inormation networks to Ifaciliatsutain al saling-upu hsquprovdelusive soe rfarm (althoug reionaldevlopmen anIeonmiugowh rih rmthe ap oahtaken,a asdn by Ommeen Phe one, boo42 the Bn projects aimpoepadteprostal network appar promiseeoing) The tr experiene nfed Cien to9 rualdregios;exteno.d e-mailses Intrnet,i anden Othfc eruna lp) ocalnnoermna organ tourihsm , i hnformation senice; itroducte new izTionagressive lemay prsungteinsteamllaio of p-vale-addedn servies uing Icoexsting postaclpel an aoin networ gasofnoceter thproughouto Pru , tlcxm- eolectroic nerochtworks; developmlegand. ro egua- empihet>s thicnstfruca,tive roloernment canc tory framewokstatwll olj eo xenhance sostauniabion- playmi sproviintsed my1loe oney vand thediffcl-iy and stfrengthen institutisona capailityt staindl tis ofmoblizn lots p ofe fluinancin faro f}oregn capcityiona reevalpmnt agncie acndoaministra-th prvthe invesorsh whoe usuGallyn givne, prioriy to2 tioshe project prsoiepaclthes privtel setworkprt heavyfi coprthe aonds high-in ome usversmen c netrshp inmeo the t proisio andhanagemn oftanlz 32 OUTSTANDING POLICY AND STRATEGIC ISSUES Key Dimensions of a National Figure 4.1 Information Infrastructure Strategy . .... . ~~~~~~~~~~~~~.. . .... .... . . .. ... . ... .. ... Policy and legal issues Telecommunications infrastructure Applieations C-ovemance institutions * Network expansion * Access issues * Rule making processes * Market structure standards ^ Tele-medicine * Regulatory agencies * Transmission networks * Education and research * Impementing agencies * Switching * Distance learning * Financirg institutions and options * User equipment * E-govemance * E-commerce Telecommunications * Rural seNvices delivery * Competition policy - Citizen participation * Interconnection rules • Tariff regime * Accounting rates * Universal service Information policy Information technology industry * Taxation * Hardware * Intellectual property rights * Software * Content regulation * Export promotion * Privacy, encryption, and security * Incentives Source-Adapted from Nirupam Bajpai and Anupama Dokeniya. Information Technology-Led Growth Policies:A Case Study of TamilNadu. Research Paper, Harvard Institute for International Development, August 1999 information networks and systems and promotes for II in the development process. More re- private sector development at the user end of the cendy, the Comprehensive Development Frame- value chain of information services. The pro- work has attempted to treat development as an posed Information Infrastructure Sector Devel- integrated process. These changes in direction opment Loan in Morocco follows a similar and priorities for the Bank's development pro- approach. It supports the government in ex- gram all point to a potentially more influential tending its liberalization of the telecommunica- role for telecommunications and II. Although tions sector to the information technology (IT) these developments are more relevant to the fu- and postal sectors, including the financial serv- ture direction of Bank activities than to past ices provided through the postal network, rec- projects, they are useful for judging trends and ognizing the importance of strong synergies directions of the telecommunications and IT among the three sectors. programs. The Bank has gradually taken actions that rec- ognize the recent rapid changes in the infor- World Bank Group Coordination mation and communications technology (ICT) The location of most of the Bank's telecom- sector. IT activities have been evaluated twice in munications expertise in a central unit-the recent years, and most of the specialized IT staff Telecommunications and Informatics Division has been incorporated into the telecommunica- (EMTII-now the Global Information and Coin- tions division. infoDev has been established as munications Technology unit)-should, in theory, a catalyst to stimulate information infrastructure have fostered cross-fertilization of experiences (II) activities. World Development Report 1998: from different countries. However, it has also KnowledgeforDevelopment identifies a new role meant that the telecommunications sector has 33 INFORMATION INFRASTRUCTURE often been given a lower priority by country di- monopoly as an important component of the rectors and, consequently, in CASs. Coordination reform process that the Bank loans have been with the country directors has been handled aimed at, this objective has often been part of through informal relationships between the more general multisector privatization projects, country/Regional unit and the sector staff at often administered by Private Sector Development EMTII. Where these relationships have been good, units. Perpetuation of exclusivity has occasion- the Bank's presence in the telecommunications ally resulted, sometimes because of insufficient sector of that country has been significant and the inputs by telecommunications experts in project country has benefited from sound technical ad- management. This happened in Armenia, for in- vice (such as in Latin America and Caribbean stance, where the government awarded 15-year countries, Sri Lanka, and Indonesia). Where co- exclusivity to the privatized monopoly. ordination between the country/Regional unit IFC's recent support for pre-privatization proj- and the sectoral unit was limited, sectoral experts ects marked a departure from the traditional have had less opportunity to be extensively in- Bank and IFC respective fields of intervention and volved in the process of sector reform. has presented new challenges for coordination. In IFC, coordination between the Regions As of end-FY99, IFC had approved four pre- and the Telecommunications Division was also privatization projects, three of which have been an issue until FY99 when the Regional depart- committed (Hungary, Latvia, and Macedonia). ments were given responsibility for setting sec- Two of these operations have been evaluated in toral priorities within the countries. While it is this report, and both of them achieved a satis- too early to evaluate the outcome of this arrange- factory-or-better development outcome. The first ment, it clearly allows for more coordinated IFC pre-privatization investment was structured IFC support in the telecommunications sector. in FY94, in close coordination with Bank staff, However, competition among industry depart- to help complete the financing of an expansion ments for the attention of Regional investment project partly funded by a Bank loan. IFC's key departments is high. There may be a bias among objectives were to help launch the first major Regional departments toward more familiar sec- telecommunications privatization in Eastern Eu- tors that may not include telecommunications, rope and to give confidence to potential strate- especially given the rapid change in technology gic investors. These objectives were met based and scope of this sector. on the amount of interest shown by world-class Organizational issues have been particularly telecommunications operators and the privati- prevalent in the informatics sector. Most no- zation revenues realized by the government. tably, the sector has lacked institutionwide lead- But investing in a pre-privatization project is ership and focus. And the absence of a clearly seen by some as having the potential to create a recognized and funded informatics thematic financial incentive for IFC to lobby for extending group has further contributed to the Bank's frag- exclusivity while it is in a strong position to do mented approach to IT-specific issues. so-before its public partner is actually privatized. Fragmentation in the administration of Bank- The review, however, did not find any evidence managed trust funds and internal pricing distor- to support such perception. In fact, in all four tions for advisory and nonlending services (such preprivatization projects supported by IFC, the as between economic and sector work, trust- basic parameters (including exclusivity and the at- funded technical assistance, and Corporate Fi- tendant build-out obligations of the project) of the nance Services Department [CFS] services) have license have been structured prior to IFC's in- often resulted in poor coordination of Bank volvement. In one operation, the exclusivity pe- Group advice to countries. Privatization of state- riod is being renegotiated in response to increased owned monopolies has been particularly con- private sector interest following the early suc- tentious due to the differing perspectives on this cess of the project and to align with current ex- issue within the Bank Group. While most Bank clusivity periods being granted in other markets. projects mention privatization of the dominant To avoid the perceived conflict of interest in a dual 34 OUTSTANDING POLICY AND STRATEGIC ISSUES role as advisor and financier, IFC does not par- also provide some breathing room for the gov- ticipate in projects where CFS is, or has been, an emient to implement drastic tariff rebalancing advisor to the govemment on privatization of the measures before full competition sets in. Indeed, state telecommunications company. As in non-pre- most IFC projects with expired exclusivity are privatization projects, IFC has accessed Technical now operating against key global players in a Assistance Trust Funds to provide project prepa- competitive environment. Evidence reviewed in ration assistance to pre-privatization projects. All this study appears to confirm that short (that is, IFC pre-privatization projects have been under- no more than 4 or 5 years) initial exclusivity pe- taken in coordination with the Bank. IFC did not riods for fixed telephony operations can be ben- proceed with the pre-privatization proposals in in- eficial in countries with very limited access to stances where the Bank and IFC have disagreed foreign private capital (for example, Peru in 1993). on the rationale for IFC's pre-privatization in- At the same time, exclusive licenses (including de volvement. facto restrictions on new entrants) are becoming While Bank and IFC telecommunications staff rare for the wireless and international segments agree on the opening up of the sector to compe- under current business environments. tition, there have been differences of opinion on IFC staff have generally argued that the state- speed and timing. IFC staff have generally sup- owned telecommunications operators should be ported the private sector's view that limited ex- privatized before liberalization to offer a level clusivity is important in attracting huge financial playing field for new entrants (Annex B). How- commitments from international operators to meet ever, this is rarely the case and, with appropri- the build-out requirements in pre-privatization ate risk mitigation measures, IFC has supported and in enhancing the viability of pioneering proj- private telecommunications operators in under- ects (such as mobile phones) in countries with less- taking projects in such markets with a less-than- developed regulatory frameworks. This view has desired enabling environment. For its part, Bank not been unanimously shared in the Bank, how- staff have usually argued that privatization of ever, since others see it as in conflict with the state-owned telecommunications operators is Bank's objectives of establishing competitive not a prerequisite for liberalization. regimes in its projects as quickly as markets will Both Bank and IFC staff agree that coordina- pursue. IFC's position is that limited exclusivity to tion between the Bank and IFC has been infor- the extent deemed necessary by private investors mal-with project task managers (on the Bank to commit to build-out requirements is not in con- side) and investment officers (on the IFC side) flict with the Bank's objectives. Rather, it lays the consulting and providing input to their respec- groundwork for a sustainable and healthy open tive projects (in some countries, Bank and IFC competition. Successful projects have a strong staff have consulted each other in advising the demonstration effect in attracting new players in regulators on international best practice). The the country once the limited exclusivity has ex- merger of Bank and IFC staff into a joint prac- pired. Failed projects could provide a negative tice group should resolve these coordination demonstration effect, especially in cases where the issues, but may exacerbate potential conflicting- global strategic partners walked away from their interest challenges when Bank Group staff wish projects because of a difficult enabling environ- to be involved in both policy advice and trans- ment. Furthermore, a limited exclusivity period can actions in the same country. 35 - I) * CONCLUSIONS AND RECOMMENDATIONS Successful Projects but Variable Sector Impact T the trend documented in this review is similar to the one identified in the 1993 OED telecommunications review: projects have generally suc- ceeded when measured against their physical and financial objectives, but evidence remains inconclusive that they have had any broad impact on developing countries as a whole in terms of ultimate outcomes. Most Bank and IFC investments were processed and delivered appropriately and achieved their physical targets and better-than-average returns. Indeed, there are some impressive success stories. But in the aggregate, (declining) Bank Group- financed investments have made only a limited contribution to the expan- sion and improvement of developing countries' national telecommunications systems, nor is there evidence that they have significantly contributed to over- all development or to the alleviation of poverty. The magnitude of the problem of building na- nition of the broader objectives in designing tional telecommunications systems in develop- "new agenda" projects and assessing their ef- ing countries and the limited effectiveness of fectiveness has not been standard Bank and IFC loans for system expansion were key factors in practice. While the evaluation of most IFC proj- the shift in Bank Group priorities to institutional ects has shown demonstrable development im- reform and promotion of private sector invest- pacts, development objectives have not been well ment. This was justified not by failures of past articulated at entry or monitored at supervision. Bank projects to achieve physical and financial IFC, then, faces a risk parallel with the Bank of objectives, but rather by gradual recognition achieving well-documented financial success that the program was not helping to meet broader but being less clear on development impacts, es- Bank Group objectives. This shift in priorities is pecially since not all projects are evaluated under justified not only by Bank experience but also the Investment Assessment Report/Expanded by external evidence. However, explicit recog- Project Supervision Report system. Unless such 37 INFORMATION INFRASTRUCTURE practices change, the Bank and IFC may continue of guidelines that set best-practice targets and the pattern of successful projects with limited identify bounds on the range of unacceptable in- demonstrable impact on the broader objectives. stitutional reform conditions (for example, terms for exclusive monopoly licenses). Such guidelines An Enhanced Focus on Institutional would not only have improved the Bank Group's Reforms effectiveness, they would have preempted the In one respect the Bank's role as a catalyst for in- Bank's critics who have viewed some Bank ac- stitutional reform has been impressive, as many tivities as primarily in the interest of foreign cap- countries have launched major structural reform ital rather than domestic development. programs in association with Bank loans. More- Effective institutional reform also requires spe- over, as major institutional reform generally takes cial attention to building human capital, partic- considerable time to be fully implemented, the ularly knowledge and understanding to create long-term effects cannot be expected to be evi- constituencies for reform and competence for im- dent in the short term. But the effectiveness of long- plementing the new institutional structure beyond term institutional reform will be heavily influenced the initial transition phase. Most Bank activities by the impact of initial reforms in early adopters. for building human capital have been small and The Bank's policy on sector reform has spec- brief, directed more toward fixing a current prob- ified an integrated package of privatization (for lem by calling on external consultants than on additional capital and skill), competition (for building human capital to make the new insti- market efficiency), and regulation (for fair com- tutional structures promote long-term develop- petition, reasonable prices, and universal serv- ment objectives (El Salvador is an exception; see ice development). The integration of the three box 5.1). In part this has been the result of an elements is important not only because they re- excessive reliance on multisector loans, which inforce one another and provide synergy bene- have proved inadequate to pursue long-term fits, but also because the adoption of one element sector goals. The Bank's more recent emphasis without the otlhers risks achieving little, or even on stand-alone telecom/information technology creating greater difficulties. Moreover, some technical assistance projects with a focus on forms of privatization, competition, and regula- long-term institutional strengthening (Chad, Mali, tion are more likely to promote development Mauritania) should help address this pitfall. than others. But the Bank Group's implementa- In many countries the new structures are not yet tion of this policy has not been entirely consis- well grounded; in some, they are precarious; and tent for a variety of reasons. Most notable is the in some, they appear to be failing. If the reforms divergence of views between the Bank and IFC initiated by Bank projects are to be effective over on the priorities of the different policy elements. the longer term, the Bank will need to mount a sig- In practice, there seem to have been no bounds nificantly expanded program for building the on how the policy principles of OP 4.50 could human capital necessary to implement them. This be interpreted and implemented, in the absence will require assistance over longer periods and 3 8 g ¢ - iafE af ebru ey hc sbn neue i,>.^