DIRECTIONS IN DEVELOPMENT Privatization in Africa | 17972 OLIVER CAMPBELL WHITE ANITA BHATIA 1 .~~~~~~~~~~~~~~~~~~~~~~~~~~~~4 * >^ # E ,,~~~~~~~~~~~~~~~.,.; ....... ..... . .... DIRECTIONS IN DEVELOPMENT Privatization in Africa ; Oliver Campbell White Anita Bhatia The World Bank Washington, D.C. © 1998 The International Bank for Reconstruction and Development / THE WORLD BANK 1818 H Street N.W. Washington, D.C. 20433 All rights reserved Manufactured in the United States of America First printing April 1998 The findings, interpretations, and conclusions expressed in this study are entirely those of the authors and should not be attributed in any manner to the World Bank, to its affiliated organizations, or to the members of its Board of Executive Directors or the countries they represent. Cover photographs by Ray Witlin/The World Bank. Oliver Campbell White is senior public enterprise specialist in the Private Sector Finance Unit of the World Bank's Africa Region. Anita Bhatia is an investment promotion specialist in the Investment Marketing Services Department of the Multilateral Investment Guarantee Agency. Library of Congress Cataloging-in-Publication Data Campbell White, Oliver, 1945- Privatization in Africa / Oliver Campbell White, Anita Bhatia. p. cm.-(Directions in development) Includes bibliographical references. ISBN 0-8213-3978-8 1. Privatization-Africa, Sub-Saharan. I. Bhatia, Anita, 1960- . II. Title. III. Series: Directions in development (Washington, D.C.) HD4338.C36 1997 97-22582 338.96-dc2l CIP Contents Preface vii Acknowledgements xi Acronyms and Abbreviations xiii Summary 1 1 What Is Privatization? 9 Privatization and Divestiture 9 Privatization and Divestiture Methods 10 Transactions 11 2 Why African Governments Have Privatized 21 The Stated Objectives of Privatization 21 Reducing the Fiscal Deficit: Is It Really the Key Objective? 25 Evidence That Nonstated Objectives Influence Privatization 26 The Real Incentives to Privatize 27 Broadening Ownership 31 Methods Used to Broaden Ownership 34 Voucher Schemes and Mass Privatization 39 3 How African Governments Have Privatized 42 Program Design and Preparation 42 Program Implementation 48 Program Management 53 The Legal Framework 65 4 The Record to Date 68 Sectoral Distribution 68 Value of Transactions 70 Privatization Methods Used 72 Ownership, Control, and Exit 76 5 The Impact of Privatization 78 Impact on Government Financial Flows 78 Enterprise-Level Performance 85 iii iv PRIVATIZATION IN AFRICA A Significant Reduction in the Number of State- Owned Enterprises 90 Employment 93 Severance Pay and End-of-Service Benefits 95 The Social Impact of Privatization 99 Foreign Investment 99 Capital Market Development 101 6 Assessing Privatization Programs in Africa 104 Measuring the Success of Privatization 104 Eight Indicators for Assessing Privatization Programs 106 Why Zambia Has the Most Successful Program 111 Dealing with Constraints on Privatization 114 7 The Role of Donors 121 The Effectiveness of Conditionalities 122 Lessons for Donors 123 8 How to Improve the Process 129 Demonstrate Commitment 129 Pay Greater Attention to Securing Consensus 129 Ensure Transparency 132 Invest More in Design and Preparation 132 Put Institutional Building Blocks in Place before Launching a Program 134 Do More to Broaden Ownership 135 Appendices 137 Notes 155 Bibliography 157 Tables 1.1 Privatization Methods and Their Outcomes 12 2.1 A Comparison of Stated Objectives 22 2.2 The Dominance of Noncompetitive Privatization Methods in Kenya 26 3.1 A Comparison of the Institutional Framework for Privatization in Ten Countries (1995) 55 4.1 Ten Countries Account for Most Privatization Activity in Africa 69 CONTENTS v 4.2 Numbers of Transactions Are Not Correlated with Values of Transactions 71 4.3 Total Value of Transactions Exceeded $50 Million in Only 12 Countries 71 4.4 Most Privatized Enterprises Are Worth Less Than $100,000 72 4.5 The Top 20 Transactions Accounted for More Than a Third of Total Value 73 4.6 Competitive Sale of Shares Is the Most Widely Used Method of Privatization 75 4.7 Governments Have Withdrawn from More Than Two-Thirds of Privatized Enterprises 76 5.1 Privatization Sales Values as a Percent of Government Revenue 81 5.2 Privatization Sales Values as a Percent of GDP 82 5.3 Total Cumulative Privatization Sales per Capita 82 5.4 Proceeds from Privatization Transactions Are Used in a Variety of Ways 83 5.5 Post-Privatization Investment Often Exceeds Sales Values of Enterprises 86 5.6 Share Prices in Nigeria Reflect Improved Enterprise Performance 87 5.7 Privatization Has Reduced the Number of State-Owned Enterprises by One-Third 93 5.8 Job Losses Following Privatization 94 5.9 Privatization Is Contributing to Capital Market Development (to End 1995) 102 6.1 Government Commitment to Privatization (to End 1995) 109 6.2 Case Study Programs Compared Using Eight Performance Indicators 113 6.3 Major Constraints on Privatization in Africa 115 Boxes 2.1 A Successful Sale with a Sting in the Tail 28 2.2 The Involvement of Indigenous Investors Is Sometimes Understated 33 2.3 Burkinabe Pharmacists and Ugandan Tea Growers Participate in Privatization 37 2.4 Zambia's Efforts to Broaden Ownership 40 3.1 Issues That Have Delayed Privatization in Zambia 44 3.2 Diagnostic Studies for Privatization: Re-Inventing a Wobbly Wheel? 45 3.3 More Efforts Are Needed to Mobilize Investors 51 3.4 Ghana Uses Outsourcing to Accelerate Privatization 53 3.5 How Weak Institutional Arrangements Hinder Privatization: The Case of Kenya 58 vi PRIVATIZATION IN AFRICA 3.6 The Single Biggest Constraint on the Zambian Program Was Interference by Holding Companies 59 3.7 Press Accusations of Lack of Transparency: The Case of Kenya 62 3.8 How Failure to Carry Out Due Diligence Affected Togo's Privatizatiori Program 63 5.1 The Many Meanings of Privatization Proceeds 80 5.2 Privatization Has Brought Changes in Management Practices 89 5.3 Changes in Labor Practices: Evidence from Benin 90 5.4 C6te d'Ivoire's Mixed Experience with the Power Sector 91 5.5 In Guinea, Improved Urban Water Supply but Fewer Gains Than Anticipated 92 5.6 Uncertainty about Severance Pay Fuels Distrust of Privatization 97 5.7 Disparity in End-of-Service Benefits Fuels Distrust of Privatization 98 6.1 Ghana: How Government Commitment Has Developed over Time 110 6.2 Senegal Typifies the Major Lessons of Experience 112 6.3 A Participatory Approach to Privatization in Cape Verde 117 7.1 Pressure to Privatize Can Be Counterproductive 124 Figures 2.1 Methods for Broadening Ownership 34 4.1 The Number of Privatization Transactions in Africa Is on the Rise 69 4.2 Privatization Transactions in Industry Outnumbered Those in Other Sectors during 1986-96 70 4.3 Privatization Methods Applied in Sub-Saharan Africa 74 4.4 Percentage of Transactions That Reduced Government Ownership to Zero, as of End 1996 77 5.1 Foreign Direct Investment Has Risen as a Result of Privatization 100 Preface As privatization progresses globally, comparisons between achieve- ments in Sub-Saharan Africa and the rest of the world sometimes give the impression that little privatization is taking place in the region. (Throughout the text, "Africa" refers to Sub-Saharan Africa.) To estab- lish the facts, and so confirm or dispel this view, the World Bank commissioned a study to determine what has been happening, what impact privatization has had, and what the key lessons are for deci- sionmakers, advisers, practitioners, and donors. This book presents the findings of that study. The study had three key components: the construction of a database of privatization transactions in Africa, a set of ten country case studies, and an analysis of the role and extent of World Bank and donor as- sistance in support of privatization. In the course of the study, we reviewed the available literature on privatization in Africa, both to draw on additional information and to compare our findings (see bibli- ography). Obtaining reliable information on privatization in Africa has been time consuming and difficult, principally because of weak monitoring of and reporting on the process but also because of confusion over the terms used and inconsistencies in the way in which information is presented. The transaction database was constructed using information from privatization agencies and task managers in the World Bank's Africa Region. During fieldwork and desk research, efforts were made to fill gaps in the data, to try to corroborate information on transactions, and to present data in a consistent format. Although still incomplete, the information from the database-summarized in this report-is the most comprehensive and reliable that is currently available.' The country case studies covered Benin, Burkina Faso, Ghana, Kenya, Madagascar, Mozambique, Nigeria, Togo, Uganda, and Zambia. In the case of Nigeria, the country case study covered only the privatiza- tion of federal public enterprises; the time available precluded inclusion of privatization in the states. The countries were selected mainly on the basis of geographic representation, inclusion of economies of differ- ent sizes and characteristics in which privatization is under way, varied experiences with the execution of privatization programs, and availabil- vii viii PRIVATIZATI ON IN AFRICA ity of data. The sample included countries that were early starters in privatization (Ghana and Togo) and countries in which privatization programs had only recently begun in earnest (Uganda and Zambia). The case studies sought to describe the design and implementation of the privatization programs, the institutions involved, the procedures followed, the problems encountered, the methods of privatization em- ployed, the transactions that took place, the proceeds involved and the impact of the programs. The primary sources of information for the case studies were privatization agencies and the relevant government ministries. Information provided by them was corroborated, to the extent possible, by field research using questionnaires and on-site inter- views at privatized enterprises. A key aim of the questionnaires and interviews was to establish the post-privatization performance of the enterprises concerned. Although information on enterprise perfor- mance was generally difficult to obtain, sufficient data were collected to draw a clear picture of what has been happening. Interviews were also conducted with successful and unsuccessful bidders and with bankers and other stakeholders. The analysis of the progress of privatization in Africa falls into two principal parts: findings and conclusions. In setting out the findings we first look at why African governments have privatized (chapter 2). We then examine how they have privatized (chapter 3), what has been achieved (chapter 4) and what the impact of privatization has been (chapter 5). Information for chapters 2 through 5 was drawn from the case studies, the transaction database, and other reports available to the Bank on privatization in countries not covered by the case studies. Our conclusions are set out in the final three chapters, covering an assessment of progress (chapter 6); the role of donors (chapter 7); and how to improve the process (chapter 8). The principal aim of this book is to contribute toward an understand- ing of and improvement in the process of privatization in Africa. With that in mind, we have written the book for several audiences. First, and foremost, are the stakeholders in African countries, who include senior government decisionmakers, civil servants, local community leaders, managers and employees of public enterprises, potential local investors, taxpayers, and consumers and users of public enterprise products and services. We argue that representatives of the full cross- section of stakeholders should have had a greater role in the formulation and implementation of privatization policy. We hope that this book, which provides stakeholders with a concise picture of privat- ization in Africa, will stimulate and contribute to further discussion on important issues, including assessment of the impact of privatization, PREFACE ix consensus building, broadening ownership, transparency, uses of privat- ization proceeds, debt overhangs, and low investor interest. The book is also aimed at implementing agencies, bilateral and multilateral finance and donor agencies, potential foreign investors, and commentators who have a keen interest in the experience of privat- ization in Africa. Acknowledgements The authors would like to thank the following for their contributions to the material and data gathered for the background country case studies and for comments on the synthesis draft: Michel Cramer, Rich- ard Thompson, Herve Dandois, and Charles Vellutini, all of Investisse- ment Developpement Conseil (IDC), Paris, and Raul Toro, Anne Castle, and Patrice Dufour of the World Bank. The study was carried out under the supervision of a Bankwide steering committee comprising Michel Wormser, John Nellis, Kevin Young,-Shyamadas Banerji, Syed Mah- mood, Douglas Webb, Eric Haythorne, and Jeffrey Katz. We thank the committee for its support of the project, and especially Michel Wormser for his encouragement. Many committee members, particularly John Nellis, Syed Mahmood, and Douglas Webb, provided helpful com- ments. Other useful comments were provided by lain Christie, William Steel, Gerardo Sicat, and Colin Xu. Several external reviewers provided useful comments as well; in particular we thank Roger Leeds, Elliot Berg, David Stafford, Mick Foster, and Chris Austin. Lucy Fye provided invaluable research assis- tance, particularly on the Africa privatization database; this book could not have been completed without her help. Cameron Khosrowshahi and Meg Garlinghouse also provided research assistance. Financial support for the study was provided by the World Bank and the Overseas Development Administration of the United Kingdom, which sponsored complementary work on "The Impact of Privatization on Labor in Africa." That study was undertaken by the consulting firm London Economics, and we thank Robert Laslett, Moazzam Malik, and Gil Yaron for their contribution to the work. Several task managers in the Bank helped by providing us with information for the privatization database. A special note of thanks is extended to all the people in the field who were interviewed for this study, particularly staff of privatization agencies and other government offices, managers of pri- vatized firms who agreed to be interviewed, and the myriad other stakeholders in the privatization process who were interviewed for this study. xi Acronyms and Abbreviations DFIs Development finance institutions EBO Employee buyout EU European Union FDI Foreign direct investment GDP Gross domestic product IDA International Development Association IFC International Finance Corporation IMF International Monetary Fund IPO Initial public offering KLM Royal Dutch Airlines MBO Management buyout PTF Privatization Trust Fund SAC Structural Adjustment Credit SAL Structural Adjustment Loan SMEs Small and medium-size enterprises SOEs State-owned enterprises TA Technical assistance UEMAO Union Economique et Monetaire d'Afrique Occidentale UNCTAD United Nations Commission for Trade and Development UNIDO United Nations Industrial Development Organization USAID United States Agency for International Development ZCCM Zambia Consolidated Copper Mines ZIMCO Zambia Industrial and Mining Corporation ZPA Zambia Privatization Agency xiii Summary This book presents the findings of a study designed to determine what privatization has taken place in Africa over the past ten years, what its impact has been, and the lessons for decisionmakers, advisers, practitioners and donors. The study involved reviewing the literature on privatization in Africa, developing a database of privatization trans- actions, undertaking ten country case studies to enable detailed infor- mation collection and analysis, and reviewing donor assistance in sup- port of the process. Governments in Africa have not always adopted privatization for the reasons stated. The process has been prompted in many cases by economic necessity and enabled by the political changes occurring across Africa. Although reduction of fiscal deficits is commonly cited as the main objective, the choice of enterprises for privatization suggests that the primary motivations for privatization have been the need for World Bank, International Monetary Fund (IMF), and donor financial support and the need to generate proceeds and divest some troubled state-owned enterprises while minimizing political fallout. Until a few years ago, there was doubt about many African govern- ments' commitment to privatization. That has changed; it is no longer a question of whether or not-or what-to privatize; it is how and when to privatize. Today, most governments are committed to the process, but they have to do more to demonstrate it. Commitment is essential, but so too is consensus, and in most coun- tries there has been a lack of consensus in favor of privatization. This has been the main factor inhibiting its pace. Here, too, the picture is changing: privatization is widespread and is regarded as inevitable, and consensus is growing as governments become more open and the public becomes more informed. Although many governments are taking steps to better inform and consult with all stakeholders, recogni- tion for the need for this drive has come late. Hence, when most programs commenced, the public was generally unaware of the extent of government ownership and of how poorly the public enterprise sector was performing. Generally, the public's-and, at times, even the government's-knowledge and perception of public enterprises has led 1 2 PRIVATIZATION IN AFRICA them to overestimate the enterprises' performance, condition, prospects and value. There is clearly a need to convince the public that privatization offers something for the vast majority, if not for all. To do that requires a substantial investment in public information campaigns, including ma- jor advertising campaigns so that people are aware of the proven bene- fits of privatization and the efforts being made to find ways for them to participate. Perhaps because of the desire not to play into the hands of political opponents at a time when democratization is under way, there has been a strong tendency in Africa to shy away from the difficult issues that privatization brings to the surface. As a result, there is a virtual absence of policies on retrenchment and related benefits, social safety nets, land ownership reform, the retirement of enterprise debts, and the regulation of privatized utilities. Debate and action on these issues would have been difficult but would have contributed to the process of building consensus and to smoother implementation of privatization programs once they were launched. They would also have led to a better understanding of the tradeoffs between the desire to broaden indigenous ownership and the need to encourage inward investment. As it is, these unresolved issues continue to cause delays that impede major transactions and deter investors. Other problems hindering pri- vatization include an overemphasis on enterprise valuation and weak efforts to mobilize investors. In order to respond to public distrust of privatization, governments also need to remedy the lack of publicly disclosed information on the impact of privatization and on the use of sales proceeds. A number of approaches have been adopted for planning and imple- menting privatization, including a variety of institutional models; nev- ertheless, many programs have been characterized by inadequate de- sign and preparation. Some institutional models have evolved in ways that have resulted in fragmented efforts and weak implementing agen- cies. Agencies generally suffer from a lack of sufficient legal authority and insufficient resources, on the one hand, and from government interference and delay, on the other. The case studies indicate that, if the process is to be efficient and transparent, a strong central agency should be established that is empowered, independent, and provided with adequate resources. Furthermore, the influence of state-owned holding companies should be neutralized at an early stage, since they invariably delay or obstruct the privatization process. Despite a myriad of constraints facing African governments and their implementing agencies, privatization is moving ahead throughout the continent. More and more countries have embraced privatization, SUMMARY 3 and the number of transactions has steadily increased. Around 2,700 transactions were reported by the end of 1996, with a combined sales value of some $2.9 billion. The indications are that 1997 and 1998 will show a significant increase in privatization activity and, with more utilities being privatized, the cumulative sales value of concluded deals is expected to be well over $6 billion. The picture across the continent is far from uniform, however. In some countries the pace is picking up, not only in terms of numbers but also in terms of the size and quality of the transactions. Elsewhere, privatization programs are just beginning. In several countries the process has slowed, and mounting discontent could conceivably delay it further or even stop it altogether. Many privatization methods have been used, but capitalization and voucher-based mass privatization schemes have not been among them. Although sales of shares through competitive tender is the principal method, formal liquidations and asset sales represent a significant pro- portion of transactions, reflecting the poor state of many state- owned enterprises. Significant numbers of transactions have been concluded through noncompetitive methods. Indeed, the lack of transparency in the privat- ization process is a common finding of the country case studies. How- ever, in most countries, greater public information and increasing involvement of the private sector in the process is helping to make the process more transparent. Broadening ownership is frequently cited as an objective of privatiza- tion, but, with a few exceptions, only minor efforts have been made in this direction. Public offerings and some directed group schemes have given more Africans the opportunity to invest, and at the enter- prise level, management buyouts and employee participation have extended ownership to a greater number of individuals. But more needs to be done to reach out to the majority of the population in every country that is poor, has no savings, and is therefore unable to participate in privatization under the methods currently employed. Some governments still retain equity interests in privatized firms, but most governments have exited completely from at least two-thirds of the companies privatized. Overall, between 1990 and 1995 the total number of public enterprises in Africa is estimated to have fallen by a third. Because until 1995 privatization throughout Africa focused primarily on small enterprises, its impact on fiscal deficits, economic efficiency, foreign investment, and employment has so far been small. This situa- tion illustrates the results of failing to privatize major enterprises, par- ticularly some of the biggest loss-makers. Transactions concluded dur- ing 1996 and 1997 are expected to show a much greater impact, with 4 PRIVATIZATION IN AFRICA significant foreign investment in utilities-both for the purchase of shares and concessions and for rehabilitation and expansion of the businesses themselves. As more major enterprises enter the process, the economic impact of privatization will become more obvious, but the process has already brought about tangible benefits. While a small number of privatized businesses have failed, the overall impact of privatization at the enter- prise level is encouraging, with clear evidence of improving per- formance. To improve efficiency, managements of privatized enterprises have inevitably sought to reduce unit labor costs. Available data indicates that privatization has resulted in some job losses. However, where major layoffs have occurred, they have generally taken place before, not after, privatization, as a result of the application of hard budget con- straints. Despite concerns about the social effects of privatization, African governments have done very little to monitor employment or to gauge the welfare consequences on the families of retrenched workers. For countries yet to embark on privatization, experience shows clearly that time and money spent on thorough program design and preparation are wise investments. The conditions under which privat- ization is to take place must be fully understood so that effective action can be taken to deal with known constraints. Although the emphasis should normally be placed on large loss-makers, enterprise selection criteria should relate directly to program objectives. At the same time, attention should be given to privatizing government-operated commer- cial services that may not have been legally incorporated as enterprises. Clear measures for gauging the success of privatization need to be established. Ideally, the primary indicator of privatization should be the decline in the ratio of assets employed by the public sector to assets employed by the commercial sector. Given the practical difficulty of measuring this ratio, we propose that employment be used as a proxy. The second measure of privatization should be the change in perfor- mance of privatized enterprises, and here our study disclosed a surpris- ing and disturbing finding. Despite concerns about consensus and the outcome of privatization, not one program in Africa has in place routine procedures for monitoring and evaluating post-privatization perfor- mance. In fact, so far as we are aware, our research represents the first major effort to collect data on post-privatization performance of enter- prises. In terms of adherence to its program and the results achieved, the most successful privatization program to date is that in Zambia. The key to the success of Zambia's program has been the government's SUMMARY 5 evident commitment. That commitment has been demonstrated by the following factors, which provide best-practice guidelines for other countries: * Sufficient resources invested in careful program design and prepa- ration * Full support of privatization by appropriate legislation * Active involvement of the private sector * Establishment of a well-financed, legally mandated agency that is free to undertake its mission as the sole privatization institution, with minimal political interference * A fully transparent process that has depoliticized the privatiza- tion process . Adequate support by donors, who cooperate fully and coordinate their assistance * A government and privatization agency that take decisive action to eliminate constraints, notably in addressing the weak capital market and the delaying maneuvers of holding companies * Major efforts to inform the public about the process and to encour- age participation. Given commitment, privatization can be achieved. To be of maxi- mum benefit, however, the process should be undertaken within an overall program of reforms, of which full liberalization is the most important. We note that privatization programs have not addressed the need to ensure competition and effective regulation of monopolies; this is a major policy weakness. Regulation is being examined as part of individual sector initiatives, but these efforts are uncoordinated, and implementation is being left to follow privatization instead of being put in place concurrently. All programs face constraints; and an important feature of a govern- ment's commitment is its readiness to deal effectively with them. By far the most important constraint, and one that has received insufficient attention from both governments and donors, is the lack of consensus on privatization across the continent. This has been at the root of delays in adopting privatization. Other important constraints include the scar- city of resources, the absence of social safety nets and problems associ- ated with employee end-of-service benefits (both of which fuel concerns about retrenchment), and underdeveloped capital markets. Although the level of donor support has been impressive, more could have been provided and it could have been better focused. In many countries, donor support needs to be better coordinated. A gov- ernment can help improve donor coordination by insisting that all 6 PRIVATIZATION IN AFRICA donor ideas and assistance connected with privatization be channeled through a single implementing agency, thus avoiding disparate and competing efforts. Donors should place less emphasis on privatization for privatization's sake and avoid playing the "numbers game." Assis- tance should be directed to supporting the privatization of those enter- prises whose transfer to private ownership or management control will have a measurable impact on the economy. To this end, donors should direct support toward securing commitment, mobilizing investors, pro- viding more assistance to agencies in using sector and financial special- ists for specific deals, and helping install and maintain monitoring procedures to track post-privatization enterprise performance and em- ployment. Donors should also consider ways to help kick-start pro- grams. At the beginning of a program, for example, donors could finance more hands-on technical assistance and provide bridging fi- nance to meet employee end-of-service benefits, thus facilitating the privatization of major loss-making enterprises. The book concludes with recommendations for governments on how to improve the privatization process. The key actions to be taken are to: * Demonstrate commitment by providing the framework within which the process can be carried out expeditiously. * Invest much more effort in securing consensus. Governments must provide more public information, allow more debate on privatiza- tion, and deal with labor issues up front. There should be open recognition that the majority of the population may be unable to participate as owners of privatized enterprises. To gain and main- tain the confidence of those who do not directly participate, infor- mation on the direct and indirect benefits realized by the privatiza- tion program should be disseminated. . Introduce post-privatization monitoring and reporting on the per- formance of privatized enterprises. * Make the process as transparent as possible; involving the private sector to the maximum extent possible will help achieve this. Transparency should be ensured not only in the privatization pro- cedures but also in providing information on how proceeds have been used. * Invest more effort in program design and preparation, with more emphasis on the major enterprises in order to achieve early visi- ble benefits. * Pay more attention to broadening ownership. This calls not only for capital market development in tandem with privatization. but also for imaginative methods that can extend ownership while SUMMARY 7 being consistent with the need for investment capital and im- proved corporate governance. A further review of progress in a couple of years will show the extent to which these lessons have been applied. Before then, however, introduction of routine monitoring of the larger privatized enterprises in Africa is recommended in order to track and measure the lasting impact of the process. Despite a slow start and a mixed record to date, privatization has now taken off, and its effect will soon be visible. The most encouraging sign of its impact across Africa is the widespread recognition of the need to redefine the role of government and to place commercial activity firmly in the hands of the private sector. 1 What Is Privatization? The use of terms related to privatization varies greatly. Therefore, it is useful to begin by first defining some of the more important terms as they are used in this book. Privatization and Divestiture Privatization is not a word or concept that has come easily to Africa. Until recently, the term was not used in countries with socialist legacies. The terms divestiture or state withdrawal were preferred because they more easily accommodated government withdrawal through the liqui- dation of uneconomic enterprises and were considered to sound politi- cally less radical than privatization.2 In most reports, discussion on privatization relates to governments selling some or all of their equity interests in specified enterprises. Often it is not clear what a government's role is in enterprises in which it continues to maintain a majority or minority equity interest. Nonetheless, transactions that have not resulted in the government yielding ownership control are often reported as privatizations. Thus, privatization is used to cover everything from the sale of a 1 percent minority equity stake in a business or a management contract for a state-owned enterprise to the outright sale of a multi-million-dollar, fully state-owned enterprise. At times, its meaning also embraces pri- vate participation in infrastructure with respect to new investment projects. We do not subscribe to this last extension of its meaning, and nowhere in our report do we use it as such. A second source of confusion is that the term divestiture is used as much as-and often interchangeably with-privatization. To some, divestiture is closely associated with asset sales and liquidations and is thus seen as an extension of privatization; others see divestiture as a subset of privatization. References to published literature on priva- tization do not clarify the meanings of and differences between priva- tization and divestiture. Indeed, the extent of the confusion over these terms is evident from the definitions that have appeared in some World Bank publications in recent years. 9 10 PRIVATIZATI ON IN AFRICA We define privatization as the transfer of operational control of an enterprise from the government to the private sector. Although opera- tional control can be placed in private hands through leases, concessions, or management contracts, control is most often secured by majority own- ership. Privatization thus includes any transaction that results in a gov- ernment ceding ownership control by decreasing its equity stake from 50 percent or more to less than 50 percent. This definition would therefore include equity dilutions (where government has moved from a majority position to holding a minority equity stake) and joint ventures, where a government holds no more than 50 percent of the equity and has ceded management control to the private shareholders. We know that many observers will argue about the level of equity needed to have effective ownership control-whether 30 percent, 50 percent, or whatever-and, of course, there may be situations in which, despite its minority equity position, a government retains undue influence over the running of a business. Generally, however, there is a perceived significance in a gov- ernment yielding its majority equity position to private shareholders, and 50 percent is the rational numerical threshold. We define divestiture to mean any transaction by which a government has transferred title in or sold some or all assets or shares in an enter- prise, regardless of any transfer of operational control. This definition of divestiture would thus embrace sales of minority government-held shares, transfers of shares or assets, public enterprise mergers, official liquidations, and asset sales made through means other than an official liquidation process. All divestiture transactions may be viewed as steps toward or as proxies for privatization. However, divestiture would not include equity dilutions, joint ventures, leases, concessions, and management contracts because the government, although it has ceded ownership control under these methods, has not divested (that is, sold or parted with, or transferred title to) shares or assets. Privatization and Divestiture Methods Up to the end of 1996, eighteen different methods of privatization had been used in Africa. Because they have different consequences on ownership and on outcomes, it is useful to group them by type of transaction: sales of shares; sales of assets; management/employee buyouts; transfers; equity dilutions; joint ventures; restitutions; liquida- tions; leases; concessions; and management contracts. Some types of transactions can be arrived at by different processes. For example, shares may be sold through public offerings, open auc- tions, sole sourcing, or competitive tender. The privatization methods (the full combination of transaction types and processes) applied in WHAT IS PFRIVATIZATION? 11 Africa are set out in table 1.1 together with notes on some important characteristics and on their outcomes. There are some important points to bear in mind when reading any report on privatization transactions. First, two or more methods may be applied concurrently for the same enterprise. Second, there may be several transactions relating to a single enterprise, but that enterprise can only be privatized once (unless a deal is reversed, for whatever reason). Third, although divestiture may take place, privatization will occur only if the transaction is with private investors or contractors; shares, assets, leases, concessions, or management contracts taken up by another public enterprise do not count. Fourth, certain methods of privatization (such as public offerings) offer a wider scope for broaden- ing ownership than restrictive methods such as sales of shares with preemptive rights or debt-equity swaps. Fifth, sales are often negotiated to include payment on deferred terms. Nonpayment of amounts due can result in a sale being aborted, the assets or shares reverting to government, and the privatization or divestiture process recommenc- ing, using the same or an alternative method. Transactions A transaction (or "deal"), whether it results in privatization or is simply a step toward eventual privatization, tends to be reported as completed when a deal has been reached. Sometimes, however, the approval process is incomplete or the final legal document lacks all of the necessary signatures or supporting documentation. For the pur- poses of this study, all reported transactions were included if they were known or thought to have been finalized in terms of final approval and completion of all necessary legal documentation; agreements that were negotiated but not yet approved were excluded. One exception was made to the requirement that the necessary supporting legal docu- mentation is required for a transaction to be considered complete: the presence of title to land. In many African countries there are problems of land title transfer; a transaction may be concluded in all other re- spects, but land title transfer (discussed in chapter 3) may not be com- pleted until several years later. Because of the different actual or implied definitions of privatization, many reports on the subject are unclear as to the effects of ownership changes arising from transactions. In addition, each country uses its own reporting format, so that comparison of activities between coun- tries is difficult and, without explanatory notes, are open to misinterpre- (Text continues on page 20.) Table 1.1 Privatization Methods and Their Outcomes Transaction type Notes Oiutcome 1. Sales of Shares Whatever process is adopted: * A sale may result in government remaining as the majority shareholder. Partial privatization * A transfer of ownership control takes place when government becomes a Majority privatization minority shareholder (usually judged to have occurred when government ownership is less than 50 percent). * Full privatization occurs when government disposes of all of the shares it Full privatization holds. A government may hold shares directly or indirectly through a state-owned holding company. In looking at the extent of privatization, government indirect ownership should be taken into account. Sales of shares refers to shares with voting rights. There are no reports of sales of nonvoting shares. Competitive sale Shares owned directly or indirectly by the government are offered for sale to private investors through competitive means-usually through open public tender, but occasionally through prequalification or shortlisting of suitable potential investors. On occasion, eligibility criteria are set-for example, to ensure the broadening of ownership or to secure a technically qualified core investor. There are no reports of shares sold through an open public auction. Public flotation Shares are offered to the general public through the stock market. Sometimes referred to as an initial public offering or simply a public offering. On occasion, eligibility criteria are set in order to broaden ownership or to avoid a concentration of ownership. May be linked with a private placement. Although there is a degree of competition when subscribing for shares offered through a public flotation, applicants are not competing on price. Hence, a public flotation is categorized as a separate sale method. Preemptive right sale Shares are offered for sale to existing shareholders in accordance with the rules Usually results in set out in the enterprise's charter. majority or full privatization Some misunderstanding can occur when a government sells equity interests in a company whose shareholders have preemptive rights because the public is unaware of either the existence of the rights or what those rights may mean in terms of the price to be paid to acquire shares from the outgoing shareholder. Preemptive right sales are noncompetitive sales. They are classified separately because of the legal right assigned to the existing private shareholders, which precludes competition (at least without their prior agreement). Noncompetitive sale A deal is negotiated for a sale of shares without using competitive means to select the new private shareholders. Frequently includes private placements and direct sales to private investors without previously advertising the opportunity for other potential investors to submit proposals. This type of sale is normally regarded as nontransparent. Widely used for any of the following possible reasons: (i) desire to grasp an opportunity to sell to an interested investor: (ii); an urgent need to initiate action on a company (either because of its distressed condition or its strategic importance); (iii) desire to avoid the opportunity cost of delay that a competitive procedure may involve; and (iv) corruption. 2. Sales of Assets The sale of assets is selected as the method of divestiture when an enterprise Sale of some principal has become dormant or is financially distressed and it is considered that an assets represents offer for sale of shares would fail to generate investor interest. A previous partial divestiture. attempt to sell the shares may or may not have occurred. Since creditors have certain legal rights, the sale of principal assets of a Full divestiture is financially insolvent company should not occur without the prior agreement achieved when the of the major creditors, including any secured creditors, and preparation of a enterprise is formally schedule of verifiable creditors to establish how sale proceeds should be used wound up, but for (Table continues on following page.) Table 1.1 (continued) Transaction type Notes Outcome to retire debt according to legal preference. Despite these requirements, some practical purposes, asset sales appear to have taken place regardless of the rights of the full privatization is creditors, presumably on the basis that much of the enterprise's debt was considered to have directly or indirectly owed to government. Caution is required in this occurred when all of connection, as claims from creditors might arise after the assets are sold. the principal assets Disregard for adherence to insolvency laws and procedures might render have been divested. government liable to those creditors for the full amount of outstanding debts. Subsequent legal claims could therefore be potentially more costly. Assets offered for sale normally comprise only the physical assets included in the enterprise's balance sheet as fixed assets or stocks, although on occasion they can include exploration and mining rights, copyrights, and patents. For enterprises with multiple facilities, the assets are sometimes sold in lots corresponding to the geographic locations where the enterprise had its operations. Strictly speaking, the assets are privatized but the enterprise is not. After the sale of the principal usable assets has been completed, some assets (which usually include debtors) as well as all of the liabilities still remain with the enterprise and the enterprise as a legal person continues to exist. A formal winding up process is therefore required. Competitive sale by Through press advertising and other means of communications, the assets of open tender an enterprise are offered for sale to private investors. Price is normally the determinant for selecting the successful bidder. Occasionally, especially for major production assets, technical competence and financial strength may be additional criteria for selection. Competitive sale by Through press advertising and other means, the public is notified of the date of public auction the public auction at which the assets are to be sold to the highest bidder. Noncompetitive sale Direct sale of assets to a private investor. This method has been widely used, with consequent concern among some stakeholders about transparency and fairness. 3. Management or Acquisition by management or by employees generally of the shares or Usually full Employee Buyouts principal assets of an enterprise. privatization Can be through either a competitive process (with or without the management or employee team being given preferred terms) or a noncompetitive process. There can be confusion as to how to categorize an MBO. We distinguish between a management/employee buyout (MBO), as defined above, and employee participation. Employee participation, although probably on concessionary terms, is the same as a sale of shares. It may occur as part of a negotiated sale of shares to private investors or alongside a public flotation when a small block of shares is reserved for employees. 4. Transfers Transfers can relate to assets or shares (although transfers of assets are rare). In both cases, they normally can be applied only when the existing public enterprise is wholly govemment owned. A transfer, as opposed to a sale, implies no financial consideration. A transfer of shares or assets from an enterprise that is not wholly owned by Ul the government often requires the agreement of the other shareholders and, in the case of a transfer of assets, agreement of the major creditors. Transfer of assets A transfer of assets might be considered appropriate when a public enterprise is inactive and therefore has unutilized assets while another enterprise is in need of such assets but lacks the funds to purchase them. The transfer enterprise undergoes partial divestiture. However, privatization would not have occurred since the transferee would almost certainly be a public enterprise, too. Privatization will occur only if the transferee is in the private sector. The rights of creditors also have to be carefully considered (see notes on sales of assets). (Table continues on following page.) Table 1.1 (continued) Transaction type Notes Outcome Transfer of shares A transfer of shares happens when a government cedes ownership to another institution without financial consideration. The reason for the transfer may be to warehouse the shares for later direct or indirect sale to the public, or so that the government can provide an additional revenue source to a government-sponsored institution which takes over full responsibility for portfolio management. For example, shares may be transferred to a local municipal authority or state pension fund which, as owner, would derive dividend income from the investment (or later realize cash by selling the shares). The warehousing of shares for subsequent sale to the public can be effected through a privatization trust or an investment fund. In the case of a privatization trust, shares are later sold to the general public through public offerings or to targeted. sections of the community. In the case of an investment fund, shares or units in the fund itself are offered to the public or to specific sections of the public. Normally, shares transferred to a trust (whether public or private) or to a privately managed investment fund are considered to represent a change of ownership (privatization) if the government ceases to exercise any ownership rights. A transfer of shares from an enterprise that is not wholly owned by the government would likely require the agreement of the other shareholders. 5. Equity Dilutions An equity dilution occurs when an enterprise's share capital is increased, with Partial or majority the new share capital subscribed by private shareholders. The equity held by privatization will the government, although unchanged and the same in absolute terms, occur, depending on becomes a smaller proportion of the total. the proportion of shares held by the government in the expanded equity. Debt-equity swap Government-held equity is diluted when the privately held debt of an enterprise is converted to equity This method may affect ownership and management of the enterprise, but there is no transfer of funds; additional capital is not necessarily introduced into the enterprise, and the government does not receive any proceeds. Hence, while this method is suitable where the sale of an enterprise's shares might be politically difficult, it would not be applicable as the sole privatization method if the enterprise required additional investment capital. An equity dilution through a debt-equity swap can be useful in reducing an enterprise's debt ratio and, when used, is often done in conjunction with another privatization method. New capital issue The share capital of an existing enterprise is increased by private shareholders subscribing new, additional capital to the enterprise. The government does not sell the shares it holds. The enterprise continues to operate as the same legal entity. An equity dilution through new capital can be useful when, for political reasons, a government does not want to be seen to be divesting any or all of its equity interest but nevertheless wants to privatize an enterprise in need of capital investment by introducing majority privately held capital. 6. Joint Ventures The term joint venture is sometimes used in a sense different from the classical 50:50 temporary business arrangement. In the context of privatization, a joint venture has come to mean a new company whose share capital is subscribed partly by the government-in the form of assets transferred from an existing but financially distressed public enterprise-and partly by private investors (typically in cash but sometimes also in kind). The existing public enterprise ceases operation and the new successor company takes over, usually without a break in business activities. Because a new legal person (the new company) is taking over, persons employed by the existing public enterprise, even if employed immediately by the successor company, will have to be paid their end-of-service benefit entitlement. (Table continues on following page.) Table 1.1 (continued) Transaction type Notes Outcome The existing public enterprise still has to be wound up through a formal liquidation process. This method is suitable for an enterprise that has some or all of the following characteristics: provides services that are economically important and need significant improvement; is politically and/or practically difficult to sell; and is in need of investment capital. 7. Restitution Restitution is a return of assets or shares to former owners from whom they Normally regarded as had been acquired by government through nationalization or confiscation full privatization. without adequate compensation. This is sometimes referred to as deconfiscation. A negotiated payment from the former owner to reacquire the assets or shares may accompany a restitution if significant investment has been made by the government since acquisition. 8. Liquidation The closure and winding up of an incorporated enterprise in accordance with Full divestiture when procedures under insolvency laws. the enterprise is wound up. Involves the appointment of a liquidator who realizes all assets and applies proceeds toward settling the enterprise's liabilities according to legal preference. Since liquidations are almost invariably applied to insolvent companies that have ceased productive activity, unsecured creditors are likely to receive little or nothing. Contingent liabilities, notably employee end-of-service benefits, become actual liabilities when the resolution to put the enterprise into liquidation is passed. 9. Lease In return for an agreed fee (rent), a private operator is given custody for a Temporary privatization specific period of time of some or all of the assets of a public enterprise to of the business takes employ them in a productive manner. Ownership of the assets remains with place and lasts as the enterprise, and ownership of the shares remains unchanged. long as the lease arrangement. Often applied to enterprises that have proved difficult to sell, either for political reasons or because of market conditions. Lease rents should accrue to the enterprise. Rents may accrue to the government if the enterprise is wholly government-owned and the government has under- taken to settle the enterprise's liabilities. 10. Concessions A concession is a contractual arrangement whereby, in return for a negotiated fee, Temporary privatization a selected private operator is awarded a license to provide specified services of the service takes over a certain period of time. Ownership of the principal assets remains with place and lasts for as the enterprise, and ownership of the shares remains unchanged. long as the conces- sion lasts. Concessions are normally awarded following a competitive tendering process. The concession agreement sets out the rights and obligations of the service pro- vider, including minimum service standards. This method is well suited to parts or all of an enterprise with some or all of the following characteristics: provides services that are economically important and need significant improvement; is large and usually enjoys a monopoly position; is politically and/or practically difficult to sell; and is in need of investment capital. 11. Management A management contract places a public enterprise under private management Temporary privatization Contracts for a specific period of time, for which the contractor is paid a fee. The fee may be based partly on performance. Ownership of assets remains with the enterprise, and ownership of the shares remains unchanged. The private manager has extensive autonomy, as set out in the contract. Often employed in situations where there is a need to turn around a company in readiness for eventual privatization. 20 PRIVATIZATION IN AFRICA tation. For example, readers often assume that the reported number of privatization transactions represents an equivalent reduction in the number of public enterprises, unaware that the "privatization" of some enterprises may have involved multiple transactions while some other transactions may have related not to public enterprises but to minor- ity investments. When looking at data on privatization, the reader is advised to bear in mind the features of reported numbers of transactions. First, several transactions occurring at different times may relate to the same enter- prise. Separate transactions for the same enterprise that occur simulta- neously-such as a sale to a new core investor combined with a sale of some shares to employees-are treated as one transaction. Nonsimul- taneous transactions may arise when an enterprise is being privatized gradually or when an enterprise is effectively being liquidated and its principal assets are being sold separately. Second, even though a deal may be finalized, legally the transaction is not concluded until all parties have fulfilled their obligations under the deal contract. This may take place several months-or even several years-after the deal is signed. For example, deals involving larger companies normally provide for new investors to carry out due dili- gence work once they have taken over control of the business. In isolated cases, due diligence work can result in a price adjustment or, in an extreme situation, in cancellation of the contract. Deals often provide for payments in installments, so that new investors do not secure legal title until the final payment has been made. In some cases, deals include an undertaking on the part of the new investors to in- troduce a minimum specified amount of additional capital into the business. Only when all such conditions are fulfilled is a transaction legally completed. 2 Why African Governments Have Privatized The choice of enterprises for priority privatization, the lack of trans- parency in some countries, the poor dissemination of information about the process and the outcome of transactions, and the failure to report on the actual use of proceeds suggest that governments in Africa have rarely adopted privatization for the reasons publicly stated. Yet to assess the success of a program, outcomes need to be measured against the objectives that governments had set for privatization. Objec- tives are usually articulated in policy statements. They guide imple- menters and inform the public about what a privatization program is designed to achieve and how the program will achieve its aims. Hence, as statements of what is desired to be achieved, they should provide the litmus test for judging the success of privatization. The Stated Objectives of Privatization Most governments have issued policy statements that set out the objectives of their privatization programs. The objectives for the case study countries and four other countries are shown in table 2.1. As the table shows, the stated objectives vary from one country to another. Objectives in every country are expressed in such general terms, and without related targets for achievement, that it is difficult, if not impossible, to judge objectively the impact and success of the privatiza- tion programs. For example, how should success in broadening owner- ship be measured? How can the contribution of privatization to private sector development be gauged? Comparison of Objectives In some countries, such as Zambia, the privatization program is self- standing and has its own set of objectives. Elsewhere (in, for example, 21 Table 2.1 A Comparison of Stated Objectives Reduce Access Raise Reduce Develop Broaden Increase adminis- Develop markets, revenue fiscal private owner- economic trative capital capital, & for Country burden sector ship efficiency burden market technology treasury Others Benin k P t Burkina Faso o o C6te d'Ivoire i. kw" Ghana ii' vS v S Kenya / / 0 Lesotho z. ,.' Nigeria Madagascar t t Mozambique i.' Senegal I-O Tanzania j..' 0 g. vS Togo a-' a-' Uganda a-' AO" ' a' Zambia v" v- a-' Source: Government policy statements. WHY AFRICAN GOVERNMENTS HAVE PRIVATIZED 23 Benin, Cameroon, Ghana, Kenya, Nigeria, and Tanzania) privatization has been included as part of a broader program of public enterprise reform, and objectives have been set collectively for commercialization, restructuring, and privatization. The Kenyan government's 1992 policy paper states five objectives of state enterprise reform, three of which apply to privatization: to enhance the role of the private sector, to reduce the fiscal burden of public enter- prises on the exchequer, and to broaden the base of ownership. The four objectives set out in Senegal's Privatization Law of August 1987 were to increase the autonomy and responsibility of management, to mobilize and direct public savings toward productive investment, to eliminate or substantially reduce public sector subsidies, and to help small savers gain access to capital markets. Only one of these objectives-the elimination or reduction of subsidies-is directly re- lated to privatization. Benin has not specified objectives for its privatization program; the objectives included for Benin in table 2.1 are inferred from those set out in the 1989 Policy Paper for the Structural Adjustment Program, in which the government drew up a list of potentially profitable sectors that would be attractive to private investors. Burkina Faso and Togo-and, indeed, most African countries-were motivated to adopt privatization by fiscal considerations. This is evi- denced by the inclusion of privatization in structural adjustment pro- grams and the correlation of privatization activity with the poor fiscal position of the countries concerned. Six years after initiating privatization, the government of Ghana is- sued a policy paper in 1993 in which the privatization program was stated to be "an ambitious attempt to unlock the economic potential of the country by permitting resources of people, money, and technology to be put to their best use and by increasing efficiency to achieve better living standards for all." The policy paper goes on to say that the program "is meant to reduce the size of the public sector and improve its perfor- mance by mobilizing private sector management and capital, thereby reducing the financial and managerial burden on Government," and that "privatization also promotes competition through which consumers get a better deal." These statements are fine rhetoric but do not provide yard- sticks against which progress on privatization can be measured. An interesting feature of table 2.1 is that the privatization objectives of countries in West Africa tend not to include references to ownership. This is in marked contrast to East Africa and southern Africa, where own- ership is a sensitive issue and the broadening of ownership is regarded as a key objective. Later in this chapter, we examine the objective of broad- ening ownership and look at what has been done to meet that objective. 24 PRIVATIZATION IN AFRICA Change in Objectives over Time Objectives can change over time. In Togo, some early transactions showed that privatization could galvanize entrepreneurial skills, stimu- late private investment, and offer scope for broadening private Togolese ownership. As a result, the objectives were expanded to include objec- tives other than fiscal deficit reduction. Francophone countries gener- ally led the privatization process in the mid-1980s, with privatization prompted by the need to stem government financial support for ineffi- cient and ailing enterprises. Emphasis was thus given to liquidations and, for potentially viable enterprises, to finding core investors with the requisite technical and management expertise who were willing to introduce investment capital. Only later did broadening of ownership emerge as an objective. Slow implementation and the difficulties involved in finding suitable local investors have recently led Ghana and Uganda to reduce the emphasis on broadening ownership and to concentrate on securing good deals with investors, regardless of origin, who have industry skills and access to investment capital and markets. Competition: The Overlooked Objective Except for Ghana, African countries have not cited encouragement of competition as a specific objective of privatization, although it may be inferred from the objectives of private sector development and in- creased economic efficiency. This is a crucial point because, in most cases, competition is more important than privatization per se. In Tanza- nia and Zambia, for example, competition, facilitated by the liberaliza- tion of crop marketing and road transport services, has brought about major benefits to growers and consumers before or without privatizing public enterprises in those sectors. Ghana also stands out as the first African country to license a second general telecommunications operator, and it is doing so at the same time as it is introducing a core investor into the existing utility. The government's policy is to stimulate investment and improve- ments in telecommunication services by introducing competition as early as possible. This is in contrast to other countries that pre- fer to privatize but continue with a telecommunications monopoly for some time in order to maximize the proceeds from privat- ization.3 WHY AFRICAN GOVERNMENTS HAVE PRIVATIZED 25 Conflicting Objectives Because stated objectives tend to be a set of desirable aims that appeal to a broad cross-section of stakeholders, objectives are not al- ways mutually consistent. When privatizing a business, whether by sale of assets or shares, the public expects the government to try to secure the best possible deal. However, price maximization may not necessarily be consistent with the long-term development and success of a privatized business. Also, broadening of ownership is often at variance with maximizing price and other benefits of privatization. Mobilization of investors with savings to invest (usually foreign invest- ors or wealthy minority groups) is inconsistent with the objective of broadening ownership within a population in which the majority have little or no savings. Reducing the Fiscal Deficit: Is It Really the Key Objective? Table 2.1 shows that stated objectives vary across countries but that all of the countries listed identified fiscal deficit reduction as a principal objective. Despite this, the evidence suggests that the desire to reduce fiscal deficits has not prompted privatization. If that had been the case, then priority would have been given to divesting the largest enterprises that are dependent on government support, principally the utilities and railways, which offer the greatest scope for improving economic efficiency. It is true that countries like the Republic of Congo and Gabon are today giving priority attention to the large loss-makers, but this is a recent development. At the end of 1995, except in Cote d'Ivoire and Guinea, not only were the larger enterprises in Africa not included in privatization programs, they were specifically excluded. Of some 57 power sector utilities in Africa, only 2 had been privatized (in Cote d'Ivoire and Guinea), and in two other cases (Gambia and Sao Tome & Principe) management contracts had been entered into. These four utilities employ some 6,000 people, compared with a total of some 150,000 employed in the power sector across Africa. Only Tanzania and Lesotho estimated pre- and post-privatization fiscal flows during program planning; but no country has monitored fiscal flows since privatization. This lack of monitoring renders it impos- sible for African governments to report to the public on the success (or otherwise) of their programs in meeting the stated primary objective. 26 PRIVATIZATION IN AFRICA Evidence That Nonstated Objectives Influence Privatization There is overwhelming evidence that factors other than the stated objectives have influenced privatization in Africa. First, if broadening ownership had been a principal objective, one would not have expected to see so many sales of investments in companies with existing private shareholders having preemptive rights, or sales on noncompetitive terms to foreign investors. For example, in Kenya, over 70 percent of privatization transactions up to the end of 1995 were preemptive rights sales (see table 2.2). Of 110 completed privatization transactions in Kenya between 1991 and 1995 (excluding the Kenya Airways' manage- ment contract), only 20 percent are known to have involved the use of one or more competitive methods; nearly 73 percent involved non- competitive methods. These transactions appear to have been given priority because they were believed to be easier to conclude and would therefore help meet World Bank and IMF loan conditionalities related to the number of privatization transactions to be completed. Second, if the goal of broadening ownership had been paramount, there would have been less emphasis on maximizing sale values and more emphasis on broadening ownership. In fact, considerable time and resources were devoted to valuations, and, according to privatiza- tion agencies and bidders, price was the single most important determi- nant in selecting investors in most cases. Selecting a bidder who oiffers the highest price renders it easier to convince observers that the process is transparent and thus provides political protection. However, it has Table 2.2 The Dominance of Noncompetitive Privatization Methods in Kenya Competitive methods 24 Sale of assets through competitive bidding 17 Sale of shares by public flotation 4 Equity dilution by public flotation 3 Sale of shares through competitive bidding 0 Noncompetitive methods 80 Sale of shares with preemptive rights 69 Direct sale of shares 5 Sale of assets through direct negotiation 5 Debt conversion I Unknown transaction methods 6 Sale of assets through unknown process 5 Unknown method 1 WHY AFRICAN GOVERNMENTS HAVE PRIVATIZED 27 often served effectively to exclude indigenous investors. The emphasis on securing the best possible financial deal has also provided plausible reasons for delaying transactions, especially when the obtainable price has been well below what was expected or hoped for. Third, although transparency has improved over time, the privatiza- tion process in many African countries continues to be insufficiently transparent. The cloud surrounding some transactions (in Ghana, Guinea, Kenya, and Tanzania, for example), especially those handled outside the privatization agencies (in Ghana, Kenya, and Tanzania), strongly suggests the presence of vested interests and rent seeking. Fourth, objectives of individual transactions are never stated. Some- times they may be deduced, but often they remain unclear. Take the case of Kenya Airways, the first major privatization deal in Kenya (see box 2.1). As a result of a successful management contract, the company was turned around and became profitable. What, then, was the reason for its subsequent privatization? Despite acclaim for stimulating the country's flagging privatization program, this deal, which may be con- sistent with other objectives of the program, does not appear to be fully consistent with the objective of enhancing the role of the private sector. This raises an interesting question. Is it preferable to have a privately- owned company operating in a protected market or to continue with an inefficiently run public enterprise? A protected enterprise like Kenya Airways with a core industry investor is probably less damaging than an uncontrolled public enterprise. Some may point to the turnaround achieved by the management contract prior to Kenya Airways' privat- ization, but it is by no means certain that that improvement would have been maintained, nor that it would have been achieved without the incentive of forthcoming privatization. The Real Incentives to Privatize Case study interviews suggest that the principal incentives for Afri- can governments to divest have been the following: * Political change * The need for World Bank, IMF, and donor financial assistance * The need to generate proceeds * The precarious state of some public enterprises * The need to maintain employment levels * At times, the need to satisfy vested interests. The demise of communism, coupled with the worldwide informa- 28 PRIVATIZATION IN AFRICA Box 2.1 A Successful Sale with a Sting in the Tail In January 1996, the government of Kenya signed a deal with KLM Royal Dutch Airlines that concluded the first phase of Kenya Airways' privatiza- tion. KLM purchased 26 percent of the government's existing shareholding in the airline for $26 million. At the same time, Kenya Airways and KLM signed a comprehensive agreement covering corporate governance and management and providing for KLM to deliver an additional $3 million in goods and services to Kenya Airways over the next two years. Following the signing ceremony, a campaign was launched to complete the privatization by selling shares to the Kenyan public through an offering on the Nairobi Stock Exchange and to Kenyan and international institutional investors through private placements. In May 1996, domestic and international public offerings of government- held shares resulted in the following ownership of Kenya Airways' equity: KLM, 26 percent; international institutional investors, 14 percent; Kenyan institutional investors, 14 percent; Kenyan public, 20 percent; Kenya Air- ways' employees, 3 percent; government of Kenya, 23 percent. A month after the agreement, in February 1996, the government unveiled its plan to sell 51 percent of its shareholding on the Nairobi Stock Exchange. The small portion of shares that were to have been made available to Kenya Airways' staff through an employee share ownership plan were instead to be offered to the public. For the time being residual shares will be held by the Government. The deal with KLM included several features that appealed to the Kenyan authorities. First, the price paid by KLM reflected a value for Kenya Airways that was more than three times its book value and at the top of the range indicated in the valuer's report. Second, the proceeds went directly to the Treasury. Third, KLM did not in the end require a management contract or unusual minority privileges, which means that Kenyan interests will control the company-a feature expected to help attract Kenyan investors. But there is a sting in the tail. In anticipation of privatization, the govern- ment confirmed that Kenya Airways will remain the designated national carrier and will retain all traffic rights, and that all charter operations will be banned from Nairobi airport for at least the next five years. This stifling of competition presumably might have influenced the price negotiated. It is at odds with Kenya's public enterprise reform and its privatization program, which is intended, among other things, to level the playing field and promote private sector development. Source: World Bank, 1996e. WHY AFRICAN GOVERNMENTS HAVE PRIVATIZED 29 tion revolution, has brought about profound political changes through- out Africa. This process of democratization has heightened the need for openness, accountability, and private sector development led by a reduction in state intervention in commercial activities, and, in so doing, has probably been the single most important stimulus for privatization. Obtaining World Bank and IMF Assistance The need for financial assistance has also been a great incentive for privatization and explains why the World Bank and IMF have played such significant roles in getting privatization off the ground in Africa. Case study interviews and internal World Bank records show that the need to satisfy Bank disbursement conditionalities has at times prompted action on transactions that would not otherwise have oc- curred as early. However, as is shown in chapter 7, the effect of condi- tionalities has not been as great as is generally thought. Donors have played, and continue to play, a crucial role in promoting privatization in Africa. Without donor pressure and support, it is doubt- ful that privatization would have progressed as far as it has in Africa. Donors have put pressure on governments to divest and have supported every program in the region. Assistance for pre-privatization studies, design, preparation, and implementation has been essential in moving the process forward. But there are lessons for donors, application of which would sharpen the effectiveness of their support (see chapter 7). Generating Proceeds Although governments have been reluctant to part with equity stakes in profitable enterprises that are attractive to investors for fear of being criticized for "selling the family silver," some sales of shares in such companies have taken place. In 1994, the government of Ghana sold part of its shares in Ashanti Goldfields, raising $316 million in the largest single flotation in Africa. That transaction and similar trans- actions appear to have been motivated by the need to generate cash to support immediate budgetary needs rather than to reduce budget- ary deficits. Divesting Troubled Public Enterprises Some large loss-making enterprises in Africa have been-and still are-in poor financial and physical shape and on the brink of collapse. 30 PRIVATIZATION IN AFRICA The need to secure investment in these businesses while avoiding further government financial support has persuaded governments to divest these enterprises, and governments want to avoid a loss of face by closing troubled public enterprises. These types of enterprises are unattractive to investors and account for the slowness in concluding such transactions. Maintaining Employment Since decisionmakers have seen pressure from a hard budget con- straint force public enterprises to eliminate surplus staff, retrenchment has become a major issue and employment retention an important short-term political objective. For some financially distressed public enterprises, privatization has offered the opportunity to exploit their untapped potential by introducing much-needed finances for invest- ment and working capital. In this way privatization has offered hope that existing employment levels may be maintained. Case study interviews revealed that most negotiations have involved discussions about retention of the existing workforce. Retaining the existing employees has been a condition of all of the deals in Mozam- bique and of some fairly major deals elsewhere, including, for example, the Lusaka brewery in Zambia. Satisfying Vested Interests Government and public enterprise employees are reluctant to relin- quish power and patronage as long as rents can be extracted. But privatization can sometimes be attractive in creating or even retaining opportunities for rent seeking. For example, where liberalization threat- ens the future of an inefficient public enterprise that has enjoyed a virtual monopoly, privatization may offer some partial protection. The incidence of noncompetitive methods of privatization where pre- emptive rights are not present and of unclear bidder selection processes leads observers to suspect that vested interests have come into play. Press articles in Ghana and Kenya are evidence of public concern about these types of transactions. Remaining Competitive Despite general apprehension in the public enterprise sector, most managers recognize that privatization is a worldwide phenomenon WHY AFRICAN GOVERNMENTS HAVE PRIVATIZED 31 and is on the rise in Africa. Some African public enterprises, particularly in the telecommunications and utilities sectors, are already keenly inter- ested in being privatized. For these enterprises, the incentive to privat- ize is linked not only to operational budgetary support; directors and managers of the major enterprises are becoming increasingly aware of the underinvestment in their businesses that forces them to operate at low efficiency and hampers their ability to compete internationally. Underinvestment has also rendered public enterprises unable to take advantage of local competitive advantages, such as low labor costs. Directors and managers of some enterprises (notably telecommunica- tions companies) are pushing for privatization and have taken steps to initiate the process. Broadening Ownership Many governments claim that they would like to use privatization to broaden ownership, but are faced with widespread poverty, underde- veloped capital markets, ethnic and tribal frictions, and a colonial leg- acy, all of which make broadening ownership difficult to achieve. De- spite these difficulties, the involvement of nationals in acquiring assets and shares through privatization indicates that some progress has been made toward meeting this objective. Why Broaden Ownership? By extending to more people the opportunity to participate in the ownership of businesses, the broadening of ownership satisfies national aspirations; in addition, it can help garner political acceptance of privat- ization. The unwarranted fear of so-called "economic neocolonialism" through privatization is a powerful force against foreign investment in some countries; particularly in countries with a legacy of socialism, public sentiment in favor of state ownership remains strong. At least in principle, privatization that results in private ownership by indigenous citizens is more palatable than outright sale to a foreign investor. The issue of ownership provokes heated discussion at any open debate on privatization. Since independence, governments and donors have invested heavily in public enterprises, and these institutions are rightly regarded as public assets. It is therefore understandable that there is a strong desire for sovereignty and for the benefits of privatiza- tion to be enjoyed by the people if privatization must take place at all. What adds to the difficulty is that the public is not informed of the extent of government ownership in commercial enterprises and how 32 PRIVATIZATION IN AFRICA these enterprises are performing. The public perception of the major public enterprises tends to overestimate actual performance and value. Governments have either themselves been ignorant of the facts-in which case they are guilty of poor governance-or have been reluctant for political reasons to inform their constituents of how poorly most public enterprises have performed. There is no doubt that senior politi- cians have been afraid to face up to the truth and to disclose the likelihood that privatization of many major enterprises will yield values much below what was originally invested. This is all the more embar- rassing when those enterprises have substantial debts guaranteed by their governments. Although broadening ownership is difficult, it is important if privat- ization is to gain the necessary political acceptance. Governments need to be able to show that some of the promises made in the earlier years of the privatization programs are being met. At the same time, as the process reaches across a wider spectrum of the public, broadening ownership can help to promote transparency. Broadening Ownership to Whom? Broadening ownership is a legitimate objective. But broadening own- ership to include whom? When the subject is discussed, resentment against relatively prosperous ethnic minority groups and foreign inves- tors comes to the surface. So, too, do political differences and perceived nepotism when groups identified with one or more tribes are believed to benefit from opportunities that are not open to all. Heated debate is not surprising when ownership facts are not disseminated (box 2.2) and when the large majority of the population is poor and unable to buy shares in firms being privatized. Despite the political need to meet the desire of indigenous people to broaden ownership, governments often prefer to involve a foreign strategic investor. Such investors bring with them new investments, proven managerial capabilities, new technology, and market access. Selling to a foreign investor is also often the only way to avoid accusa- tions of nepotism or tribalism. This has, on occasion, been a factor behind sales to foreigners in Ghana and Uganda. Attempts have been made to increase the number of indigenous persons with an ownership stake through privatization. To assist local investors, virtually all countries have negotiated deals with nationals for payment on deferred terms. Sales on deferred terms have been widely used in Ghana, Madagascar, Mozambique, and Uganda. But the news is not good. Credit administration has been weak, with privat- WHY AFRICAN GOVERNMENTS HAVE PRIVATIZED 33 Box 2.2 The Involvement of Indigenous Investors Is Sometimes Understated In some countries, there is a mistrust of foreign investors or ethnic minorities. Publicity given to sales to foreign investors has tended to confirm fears that privatization is placing national assets in nonindigenous hands. Sometimes, however, popularly held opinion about ownership is not based on the facts and understates the level of involvement of indigenous investors. In Kenya there has been a strong public feeling that public enterprises have been sold to foreigners or to Kenyan citizens of Asian origin. Yet, except for sales to existing shareholders with preemptive rights, the majority of transactions have in fact involved indigenous Kenyan investors. In Mozambique, an analysis of firms by nationality of owners, undertaken in late 1995, indicates that 92 percent of the transactions up until that time were sales to Mozambican nationals, and that many of the sales were made on an installment basis to enable the investors to purchase the firms. Source: World Bank, 1996e, 1995b. ization agency personnel too busy with current transactions to monitor and take action to collect payments due on past deals, and credit sales have proved problematic because many buyers have failed to meet the repayment schedules. Two factors have contributed to these defaults. First, in preparing their bids, local investors have not always given adequate consideration to the requirement for investment and working capital in addition to that needed for purchase, and often the same mistake has been made in evaluating those bids. Second, in some instances, the buyers did not have the financial resources they claimed to have and defaulted on their obligations. In Uganda, deals negotiated with nationals that provided for pay- ment on deferred terms were backed by bank guarantees. In most cases the buyers failed to meet their payment obligations and the privatiza- tion agency had to call on those guarantees. Although the problem of recovering the. amounts due has been transferred from the agency to the banks, the experience led the privatization agency to decide in early 1995 that single-payment sales should be given preference. When the banks concerned are state-owned and lack the capability to intervene to turn around financially distressed businesses, privatization can effec- tively become reversed. The experience with credit sales poses a dilemma. On the one hand, all-cash deals avoid both the risk of default and the burden of adminis- tering credit. On the other hand, a lack of credit that effectively excludes 34 PRIVATIZATION IN AFRICA a majority of local investors runs counter to the aim of broadening ownership. Privatization agencies would prefer not to have to adminis- ter credit, so the preferred solution would be for private sector develop- ment programs (running in tandem with privatization programs) to ensure that local investors have access to venture capital to support viable projects. Methods Used to Broaden Ownership Figure 2.1 shows the spectrum of methods that could be used to broaden ownership in Africa, ranging from methods that provide op- portunities for small groups of investors to those that provide opportu- nities for the entire population. Sales to Local Investors Local investors have participated as bidders for shares and assets, usually of smaller enterprises. Probably because of the publicity given to the larger deals that involve foreign investors, there is a widespread belief that local participation has been low. In fact, as the case studies show, local investors have been active. Management and Employee Buyouts Although management and employees of public enterprises would clearly benefit from a policy of broadening ownership, there have been Figure 2.1 Methods for Broadening Ownership Minimum Population Coverage Maximum MBOs EBOs Directed group Public Mass ownership flotations privatization Local Employee Unit trusts investors share Capitalization participation Investment trusts Note: In addition, privatization trusts can be used to warehouse shares subsequently sold through any of the above methods. WHY AFRICAN GOVERNMENTS HAVE PRIVATIZED 35 surprisingly few management buyouts (MBOs) or employee buyouts (EBOs). They have been tried only inBurkina Faso, Madagascar, Nige- ria, and Zambia. In Zambia, where five MBO transactions were con- cluded in 1995, the government was eager to encourage employee participation and identified nearly 30 public enterprises that were con- sidered suitable candidates for MBOs or EBOs. As elsewhere, however, examination by the privatization agency revealed that the past perfor- mance of the individual enterprises raised doubts about the manage- ment capability of incumbent employees. In several cases, unrealistic business plans relied too heavily on government or bank support, and inadequate allowance was made for much-needed investment and working capital. The only enterprise privatized through a management buyout in Nigeria is now in a financial and legal deadlock, with plum- meting sales and defaults on large bank loans. Although the experience with MBOs and EBOs has not been encour- aging, their political acceptability and relative ease of implementation will continue to make them an attractive approach to broadening own- ership. To improve their success, privatization agencies must make available detailed guidelines for the submission and evaluation of MBOs, and resources should be provided to train enterprise employees. Employee Share Participation MBOs and EBOs tend to be feasible only for small firms or service companies in which capital investment is minimal. Nevertheless, gov- ernments are often keen to encourage employee equity participation in medium-to-large companies alongside new investors. In Benin such participation was mandated by the 1992 Privatization Law, which lays the principle for the creation in every denationalized firm of a savings fund (plan d'epargne d'entreprise) that is to be used to allow employees to purchase shares in their company or any other company. The scheme is unclear on when and how employees will participate, and the pro- gram has not been implemented. In Mozambique, 20 percent of the value of each privatized enterprise at the time of its sale was set aside pending the establishment of a stock ownership program for employ- ees, who will be able to acquire shares on concessionary credit terms. In at least eight privatization deals in C6te d'Ivoire, between 1 and 5 percent of shares have been reserved for employees. Elsewhere, bidders have been encouraged to include employee equity participation, but no general policy has been formulated. Although employee share participation can be politically attractive, its low incidence reflects the difficulties it poses for policymakers and 36 PRIVATIZATION IN AFRICA implementers. First, it cannot be applied across the board to all deals but can take place only when shares, not assets, are sold and when no major changes in the structure and size of the enterprise are foreseen. It is thus available as an option in only a limited number of cases; and employees in firms that do not offer employee share participation may feel that they are being treated unfairly. Second, by introducing additional potential shareholders, employee share participation ex- tends deal negotiations and reduces the price that outside investors will pay. Directed Group Ownership Directed group ownership, which aims to broaden ownership by offering equity at discounted prices or on deferred terms to people involved in a particular sector, has been tried in a few instances. It provides the opportunity for people with a long-term interest in a sector-but with limited or no capital-to participate in privatization (box 2.3). Public Flotations The infancy of most capital markets in Africa represents a serious constraint on privatization and has given added urgency to the need to improve banking systems, make venture capital available, and establish markets for trading financial instruments. In this context, privatization is making an important and lasting contribution to private sector devel- opment. Although close linkage between privatization and capital mar- ket development was an explicit objective of many privatization pro- grams, only Nigeria has mandated that all public enterprises be sold through the public offering of shares. Even so, the Nigerian federal government has approved a few exceptions for private placements or sales of assets. Interestingly, regional quotas have been applied to the public flotations in Nigeria in order to achieve an equitable geographic distribution of shares. Thirty-five of the 49 federal public enterprises sold to date in Nigeria have been through public offerings. In Ghana, shares in Ashanti Goldfields were offered concurrently in Accra and London. In Kenya, shares in the National Bank of Kenya were allocated on a regional basis to ensure equitable distribution. Public flotations are the easiest way to reach as many people as possible. Where they have occurred, they have given many Africans their first opportunity to become investors. Their use is, of course, restricted to those countries that have stock exchanges. In West Africa, WHY AFRICAN GOVERNMENTS HAVE PRIVATIZED 37 Box 2.3 Burkinabe Pharmacists and Ugandan Tea Growers Participate in Privatization When the pharmaceutical importing company SONAPHARM was put up for sale in Burkina Faso, the highest bidder was a foreign investor. At the bid selection phase, this investor was asked to form a consortium with an association of Burkinabe pharmacists that had submitted a lower bid. Although this condition led to long negotiations that delayed the transaction, a satisfactory conclusion was reached, with both the foreign investor and local investors becoming owners of the enterprise. The Uganda Tea Growers Corporation (UTGC), a parastatal, owned four operational tea factories that had been incorporated as limited liability com- panies. Having considered the options for privatizing those factories, the government decided to adopt a proposal that would involve the farmers themselves. Consultants and lawyers were engaged to examine and work out the modalities. The first hurdle to overcome was the value of the factories, each of which was valued at about $3.5 million-a price considered too high for farmers to afford. It was therefore decided to base the sale price on the value of the factory shell (by omitting land and other values), which decreased the value of each factory to $.5 million. Donors agreed to pay off the loans attaching to all of the factories. In order to get the privatization process started and managed properly, a neutral bank, the Development Finance Company of Uganda (DFCU), was appointed to act as trustee and manager of the process. UTGC transferred its shares to DFCU acting as trustee so that, technically, the tea factories have been privatized. Tea growers are acquiring ownership through a scheme of registration and purchase at the time of sale of green tea to the factories. Each farmer initially pays 5,000 Ugandan shillings to participate. Share allocations are based on the minimum subscription and thereafter on production, up to a limit. Individual farmers can acquire only up to a maximum of 5 percent of voting shares. Farmers are pleased with the scheme and feel that they are involved in the privatization process. It was the first example in Africa of farmers being directly involved. Each tea factory has a board of directors comprising six elected shareholders' (farmers') representatives and a representative of the institutional investor. UTGC has formed a new company, Agri-Industrial Management Agency Ltd., which offers management services to the factories. It currently has a three-year management contract, with experts funded by the European Union. Buoyed by the success to date, the government is considering adopt- ing the same approach for the dairy sector. Source: World Bank, 1996c, 1996j. 38 PRIVATIZATION IN AFRICA the planned regional stock exchange will extend this form of investment opportunity to citizens of all the Union Economique et Monetaire d'Afrique Occidentale (UEMAO) countries. Although public flotations are designed to encourage small invest- ors, in practice some flotations involve companies in which there is already a core investor. Governments often want a core investor in the larger enterprises before shares are floated to ensure that a sound management team and good corporate governance are in place, and sometimes to improve operational performance prior to a flotation. Without a core investor, corporate governance is a potential problem. Unscrupulous enterprise managers can take advantage of a widely dispersed ownership. In Nigeria this risk was recognized, and, by promoting the establishment of shareholders' associations, the privat- tization program sought both to encourage the active participation of small investors and to safeguard their interests. Some public flotations pose risks. Shares in the National Bank of Kenya and the Cooperative and Rural Development Bank in Tanzania have been offered to the public despite the poor track records of both banks. Although these flotations may have reached a large number of investors in the short term, the long-term impact of selling shares in risky enterprises may outweigh the benefits, since the public may as- sume that the government is implicitly guaranteeing the value of the shares and expect the government to intervene if the value plumrnets. Such failures would set back privatization. Already there are examples where falls in share prices have affected the confidence of small local investors. In Ghana, the price of shares in Ashanti Goldfields in early 1997 was about two-thirds of that paid at the time of the initial public offering in 1994. This has currently diminished the appetite of small Ghanaian investors for initial public offerings (IPOs). In Zambia, the temporary, unexpected fall in the mar- ket value of shares in Chilanga Cement affected the confidence of Zambian investors and this delayed the sale of other shares held by the Zambia Privatization Trust. It is proving difficult to convince new small investors that they should not expect to buy shares and then make a quick financial gain through an early sale; the culture of a long- term investment will take time to develop. Given the stage of stock market development in Africa, most govern- ments have adopted a policy of retaining an equity interest in some medium-sized businesses (such as breweries and cement factories) with the intention of floating shares at a later date (deferred public offers) or of transferring these shares to investment funds that will either resell the shares or float their own shares. Deferred public offerings are planned in Kenya, Tanzania, and Uganda. Malawi has set up a WHY AFRICAN GOVERNMENTS HAVE PRIVATIZED 39 unit trust investment fund for this purpose, and Ghana is considering establishing an investment fund or a privatization trust along the lines of the Zambian model. The problem is knowing when a government is serious about acting in a temporary trusteeship capacity and when such an arrangement is being used as a means by which the government will keep its foothold in commercial enterprises. To avoid any doubt, the transfer of government equity to an independent trust with a clear role and finite life is preferred. This is the case in Zambia, where the Privatization Trust Fund (PTF) was established. Financial Intermediaries African countries have been slow to use financial intermediation for privatization, although some countries have tried to do so. In Malawi, a unit trust was established. In Zambia, the PTF was established to warehouse shares in the larger privatization companies. The PTF holds up to 30 percent of shares in companies the trustees consider to be worthwhile investments and later sells the issues to the public through public offerings (box 2.4). Pension funds can be an indirect way for people who may not be able to buy shares to benefit from privatization. Data on the extent of institutional investor investment suggests that, although privatization has provided an opportunity for state-organized funds to earn higher yield investments, there are potential drawbacks.4 Where the govern- ment continues to control a fund, it can be argued that shares purchased by that fund do not represent privatization. This was the case in Ghana, where the Social Security and National Insurance Trust (SSNIT) pur- chased shares in privatized enterprises. Voucher Schemes and Mass Privatization The problem with all the methods for broadening ownership that have been tried in Africa is that they directly reach only the minority of citizens who have savings or are employees in enterprises in which equity participation is available. Of course, some people benefit from improved utility and transport services, but still a significant proportion of the population, which is very poor, does not directly gain from or observe any indirect benefits of privatization. Given the importance attached to broadening ownership, it is surprising that more effort has not been put into developing a mass privatization model that could be used widely in Africa. Although the implementation of mass privatization through a 40 PRIVATIZATION IN AFRICA Box 2.4 Zambia's Efforts to Broaden Ownership To develop the country's capital market, the establishment of a stock ex- change and a Privatization Trust Fund (PTF) were planned alongside the privatization program. The Securities Act, enacted in December 1993, estab- lished the Securities and Exchange Commission (SEC) to regulate all partici- pants in the stock market with the prime purpose of protecting investors. The Lusaka Stock Exchange (LSE) began operations in January 1994 and was formally opened for business on February 21, 1994, after only 12 months of concerted effort on the part of the government, the private financial sector, and the World Bank project team. The LSE has been constituted as a private nonprofit company owned by the members of the exchange, with each member having a single share. The LSE has a small full-time staff led by a general manager. Zambia's principal mechanism for broadening ownership is the PTF, set up in 1994 as a vehicle for warehousing and then selling to the public minority shareholdings in certain major, newly privatized enterprises once the majority of shares have been sold by the Zambia Privatization Agency (ZPA) to core private investors. The PTF's objectives are to enable the widest number of Zambian citizens to participate in the privatization process, to protect the value of the minority shareholdings until they are sold, and to ensure transparency. The PTF gives priority to individual Zambian investors who wish to acquire a small number of shares. If, however, less than favor- able market conditions prevail, it will sell shares to financial intermediaries who are investing on behalf of Zambian citizens. To avoid flooding the market with public offerings, the PTF coordinates its efforts with the ZPA, the LSE, and institutional investors. A board of five trustees, four of whom are drawn from the private sector, oversees the operations and financial position of the PTF. The PTF is managed on a daily basis by a management company which is remunerated by a fixed monthly fee plus a performance fee determined on the basis of the company's ability to sell shares. The PTF was established for a period of only five years. If at the end of that period all shares transferred to the PTF have been sold, the PTF will be liquidated. If some shares remain unsold, the PTF will be converted into a unit trust. The ZPA has identified 14 companies as prospective candidates for even- tual residual government shares to be transferred to the PTF. By the end of 1996, the ZPA had transferred the residual government minority sharehold- ings in four companies. Twenty-seven percent of the shares in the first, the Chilanga Cement Company, were all offered in May 1995 and were fully subscribed. Seven percent of the shares in the second, Zambia Sugar Plc, were offered for sale in August 1996, leaving a further 23 percent in that company to be sold by the PTE Source: World Bank, 1996k. WHY AFRICAN GOVERNMENTS HAVE PRIVATIZED 41 voucher scheme would be constrained by logistical difficulties, the low level of interest in this mechanism does not appear to stem entirely from logistical concerns. Although communications are poor, voter registration difficult, and illiteracy widespread, most African countries already have active private sectors and commercial banking, so the reluctance to use a voucher-based scheme is perplexing. Improving communications and voter registration procedures that would be needed for a voucher-based scheme would be enormously beneficial to the economy and the process of democratization in addition to priva- tization. Several reasons account for the reluctance to examine mass privatiza- tion further. First, financial and technical resources are scarce. The considerable resources required for planning and implementing a mass voucher-based scheme are beyond what most African countries can muster. As it is, government and donor commitments to privatization are much lower in Africa than elsewhere in the world, and programs for the development of mass privatization like the schemes used in Eastern Europe and the former U.S.S.R. have not been sponsored in Africa. Second, mass privatization would not only be costly, it would raise no capital. Third, unless all of the utilities were included, the market value of the privatized firms might be embarrassingly low. When com- panies with shareholders with preemptive rights are excluded, the earnings potential of the remaining public enterprises is often very low. The cost of a mass privatization scheme thus might well outweigh any benefits assigned to the public. Fourth, in some countries it is possible that potential ethnic and political frictions have added to the reluctance to adopt a voucher scheme. Despite these drawbacks, governments in Africa may have to recon- sider some form of voucher scheme if the general public is to be con- vinced that privatization will benefit them. Perhaps, as an alternative to voucher schemes, more effort should be directed toward identifying targeted groups to whom share discounts or credit could be offered. Without an attempt to broaden ownership on a larger scale than has been achieved so far, privatization will continue to be perceived as benefitting principally those who already have savings. 3 How African Governments Have Privatized Privatization is an intensely political process involving players from the executive, legislative, and judicial branches of government, as well as private sector representatives and institutions. It can also in- volve significant social adjustments. For privatization to be successful, political leadership at the highest levels must demonstrate commitment to the process, and a structured process for executing and coordinating privatization transactions must be in place. Strong institutions and well- prepared programs must be established and legal constraints removed before the privatization program begins. This chapter presents the find- ings of our review of institutional arrangements for privatization in Africa and highlights the key program design, implementation, and management issues. It also reviews the legal framework in which pri- vatization has taken place in African countries. Program Design and Preparation Privatization involves considerable detailed preparatory work, both at the general and at the enterprise levels. In addition to establishing the privatization agency and providing it with resources, this work involves defining the scope of the program; selecting enterprises to be divested and examining the specific issues in each privatization; collecting data and examining the condition, performance, and pros- pects of public enterprises, particularly those enterprises selected as priorities for privatization; and preparing detailed operating policies and procedures. Poor design and preparation lead to delays and low investor inter- est. The case studies revealed that many privatization programs have suffered from poor design, inadequate preparation, and weak or frag- mented implementation. This is clearly evidenced by the long average time to complete transactions, by the number of incomplete transac- tions, many of which have been under way for several years, and 42 HOW AFRICAN GOVERNMENTS HAVE PRIVATIZED 43 apparent low transparency.5 The delays in initiating and processing transactions may well have deterred investors. Why have program design and preparation been so weak? First, the programs have, by and large, not been home grown. Because of a lack of consensus on and commitment to privatization, governments have been slow to take the initiative, and the World Bank and other external agencies have often initiated much of the design work. Second, institu- tional arrangements are weak in Africa. Third, the skills needed to oversee privatization transactions are scarce. Fourth, governmerits in Africa have underestimated the factors that are likely to constrain the process and have been very slow to remove those constraints. Even when design and preparation have been done carefully, prob- lems have arisen. Even Zambia's program-in our view the most suc- cessful privatization program in Africa-has suffered from constraints that have delayed transactions (box 3.1). It is therefore not surprising that in other countries, where design has been less thorough, implemen- tation has been slow and difficult. Initial conditions have not been adequately understood. Ideally, the first step in the design of a privatization program should be to establish the initial conditions by taking stock of the public enterprise sector, the environment in which privatization is to take place, and the constraints that may be encountered. Once a government is confident that all of its direct and indirect commercial investments have been identified, the next step should be to determine how those investments are performing and what factors affect their performance. Study of these factors identifies the changes that are necessary for privatization to proceed smoothly in conjunction with other necessary reforms. In practice, many African countries have proceeded with privatiza- tion without adequate knowledge and understanding of the initial conditions. In some cases, the extent to which the enterprises were underperforming was not known, either because of lack of sufficient information on the nature and level of subsidies or because of weak corporate governance. Some countries, including Cameroon, the Re- public of Congo, Gabon, Lesotho, Mauritania, Tanzania, Uganda, and Zambia, undertook detailed preparatory work to establish the financial and technical performances and the position of enterprises as a precur- sor to developing their programs; other countries did not. For example, in Benin, Burkina Faso, Ghana, and Togo diagnostic studies on the methods for privatization were undertaken only after the programs had been initiated and enterprises had been selected for privatization, but before transactions were undertaken; some countries, including 44 PRIVATIZATION IN AFRICA Box 3.1 Issues That Have Delayed Privatization in Zambia The constraints that delayed transactions in Zambia during 1992-94 in- cluded the following: Claims of former owners Negotiation and preparation of legal Damaged, stolen, and vandalized documentation relating to sales assets on deferred terms Death of selected bidder Delays by Ministry of Finance Negotiations with shareholders Outstanding order to cease operations with preemptive rights Failure by minister of finance to sign Delays in obtaining funds for agreement because of EBO asset valuations representation Delays in Zambia Industrial and Inability of selected bidder to obtain Mining Corporation (ZIMCO) finance signing sales documentation Change of basis of negotiation by Rehabilitation of enterprise successful bidder Inability of highest bidder to Lack of title to land or site provide evidence of financial of operations capability to conclude deal Temporary withdrawal of donor Leakage of commercial assistance information by ZIMCO Uncertainty regarding enterprise Missing, spoilt, or expired legal liabilities documents, including operating Unclear terms of bids licenses, title deeds, leases, Unpaid sales tax share certificates, shareholders' Poor offer from shareholder with agreements, memoranda and preemptive rights articles of association, and Contested valuations by ZIMCO directors' service agreements Withdrawal by top bidders Lawsuit by disgruntled MBO team Source: World Bank, 1996k. Guinea and Kenya, also subsequently undertook studies of indebted- ness and public sector flows. Some practitioners question the usefulness of detailed studies of individual enterprises when preparing a privatization program. Studies take time and are costly. When the time comes to begin to privatize many of the enterprises, those studies can be years out of date, necessi- tating further reviews. Hence "diagnostic" studies are preferred. But diagnostic studies have varied in depth and quality, as is to be expected when there is little time to identify and assess the range of factors that can affect a business in its specific environment. Therefore it is at times difficult to reach convincing conclusions. What is surprising is that, HOW AFRICAN GOVERNMENTS HAVE PRIVATIZED 45 after years of public sector reform and now more than ten years of privatization in Africa, there is still no standard format for undertaking diagnostic studies (box 3.2). Case studies show that privatization programs have most often been based on lists of public enterprises, which in turn were based on out- of-date or anecdotal information. They have thus been embarked on with insufficient information on the condition and performance of those enterprises-and even less understanding of what outcome to expect. This voyage into the unknown has had several consequences. Con- Box 3.2 Diagnostic Studies for Privatization: Re-Inventing a Wobbly Wheel? Either as a precursor to developing a privatization program or in preparation for divesting selected enterprises, diagnostic studies of medium and large public enterprises have been undertaken to determine or verify their owner- ship, condition, and operating environment. Ownership: This study component involves examining the laws under which an enterprise was incorporated and is governed, tracing the ownership history and structure to identify private shareholders with preemptive rights or former owners with potential claims, and identifying any legal constraints that might hinder privatization and the steps needed to remove such imped- iments. Condition: This involves examining the financial and operational perfor- mance and condition of an enterprise, determining what will be on offer, and identifying issues that must be addressed prior to an offer for sale (for example, major restructuring and investment requirements, the potential liability for employee end-of-service benefits, the likely effects of removing price controls or subsidies, evidence of title to key assets, and issues that have led to audit qualifications). Operating Environment: This involves studying the markets in which an enterprise operates, the likely effects of policy changes, notably liberaliza- tion, and linkages with other sectors undergoing change. The depth and quality of diagnostic studies have varied according to the time and budgets available. Most studies have not included a review of financial flows between enterprises and government or identified and quantified direct, indirect, and implicit subsidies enjoyed by the enterprises. These issues have tended to be examined as separate exercises, if at all, and often only after a privatization program has commenced. Surprisingly, despite their widespread use, there is no standard format for diagnostic studies; so the imperfect wheel continues to be re-invented. 46 PFRIVATIZATION IN AFRICA straints have surfaced during implementation that should have been identified and dealt with during design and preparation. Delays in privatization have occurred because some policy issues were not identi- fied early on or, when they were, they remained unresolved. These issues include far-reaching concerns such as ownership, settlement of enterprise liabilities (including what is due to employees), and land reform, but they have also been sector specific (for example, competition policy and national security) and enterprise specific (for example, the existence of an undertaking in a bilateral grant agreement that restricted privatization). As a result, transactions have dragged on for yeaLrs be- cause of unresolved issues, and those delays have sent the wrong signals to investors. Poor design has also led to confusion in reporting on privatization. Information on completed transactions is collected in varying degrees of detail so that reports, even from the same agency, are prepared in various formats that use the terms privatization and proceeds differently. Most programs have been designed without attention to systemati- cally monitoring and assessing the impact of the programs. None of the case study countries has a mechanism for monitoring the impact of the process or the performance of privatized enterprises. Deci- sionmakers have thus had little idea of the expected or actual outcomes of their programs. Only Lesotho and Tanzania have attempted to rnodel the outcome of their programs in terms of proceeds and fiscal impact, but even they do not yet have mechanisms in place for monitoring post-privatization performance. Classification as "strategic" has delayed privatization of key enter- prises. There has been a common tendency to categorize enterprises as "strategic" or "nonstrategic," although these terms have never been satisfactorily defined. Almost without exception, utilities, which usu- ally enjoy monopoly positions, were categorized as strategic. As a consequence, most countries have, until recently, either specifically excluded utilities from their privatization programs or deferred action on them to an unspecified date in the future. Nonutility monopolies, such as state-owned banks, development institutions (including some holding companies), and defense-related industries, have also been classified as strategic. This type of categorization has meant that programs have focused on smaller, relatively insignificant enterprises rather than on larger, economically more important enterprises, including those that are the biggest fiscal drains on the economies. Because a case-by-case approach has been employed, privatization of these small enterprises has been just as difficult as privatization of the few larger companies that have HOW AFRICAN GOVERNMENTS HAVE PRIVATIZED 47 been privatized. Recently, auctions have been used for very small enter- prises in Uganda; but elsewhere the case-by-case approach continues to be used. Selection of enterprises for privatization. The case studies showed that enterprises have generally been selected on the basis of the following criteria: size (smaller enterprises, which are less controversial and which are less likely to involve labor issues), performance (get rid of loss-makers), and the presence of an existing private shareholder (thought to make privatization easier). Program priorities have also been influenced by approaches from interested investors and by the governments' need to raise money from privatization proceeds. Most programs have begun by listing public enterprises by groups (or tranches), beginning with the smaller, less controversial enterprises. Ghana, Kenya, Senegal, and Zambia published lists of the first groups for privatization; Lesotho, and Tanzania described the factors that will influence privatization decisions but remained flexible as to the identi- ties of the actual enterprises. Some countries tried at first to tackle a few big loss-makers (the State Fishing Corporation in Ghana and Southern Paper Mills in Tanzania are typical examples). These enterprises, however, proved difficult to privatize because of the difficulty of attracting investors to a poorly performing business and the political problems associated with the acceptance of very low market values, particularly when there are large unsettled debts and no consensus in favor of privatization. In many cases, liquidation-which would have curtailed further losses, max- imized proceeds, and put assets back into productive use sooner- would have been more efficient than attempting other methods of privatization. Governments have hesitated to take early action to divest loss-makers, however, because of the uncertainty as to what to do when liquidation proceeds are well below outstanding debts and, more importantly, because of the immediate impact on jobs and the negative publicity that that would attract. Government services have been overlooked as candidates for pri- vatization. In addition to public enterprises established as corporate legal entities, many state-owned commercial activities in Africa are undertaken through unincorporated entities. Government departments own and operate vehicle workshops, road and building construction and maintenance equipment, printing plants, and many other forms of commercial enterprises such as the abattoir and the flour mills in Lesotho. Governments also operate services such as catering, cleaning, 48 PRIVATIZATION IN AFRICA maintenance, and transport services that could easily be privatized by contracting out. Very few programs address the privatization opportunities that these services afford. These activities could easily be privatized through em- ployee buyouts at very little cost. The issues of ownership, redundancy, valuation, and debt would be far less complicated in these cases, com- pared with many other privatization transactions. These privatizations could also help build consensus while at the same time assisting civil service reform. Regulation has been ignored. In not one country with a privatiza- tion program has there been an effort to develop a regulatory frame- work as an integral component of that program. Financial sector regula- tion is being addressed under reforms for that sector, and now that utilities are being privatized the need for regulation in those sectors is recognized. But efforts to develop regulation in different sectors have not been coordinated and have not been linked to what is happening and planned in the privatization programs. There is pressure to proceed to privatize and to deal with regulation afterward, and this can already be seen in key sectors such as energy and telecommunications. In the absence of the requisite experience, detailed information, resources, and enforceable legislation, it is difficult to see how African countries can quickly develop the types of sophisticated regulatory institutions and mechanisms that exist in developed economies. Despite awareness of this weakness, little has been done to address it. Just as surprising has been the lack of attention to dealing with the absence of antitrust legislation in most African countries. Privatization and competition are helping to improve the availability and quality of products and services, but African consumers are not necessarily get- ting a good deal. For example, it is rumored that cartels are forming and operating in the brewing and cement industries in some regions. Program Implementation Well-managed implementation requires detailed operating policies and procedures covering the routine parts of the program: the selection of enterprises and privatization methods, due diligence, the safe- guarding of assets prior to divestiture, recruitment of consultants, bid evaluation and negotiation processes, contract preparation, determina- tion and payment of severance pay and employee end-of-service bene- fits, safety net arrangements, settlement of outstanding debt, reporting, and public information. The case studies showed that operating policies normally deal ade- HOW AFRICAN GOVERNMENTS HAVE PRIVATIZED 49 quately with the selection of enterprises and privatization methods, the bidding and evaluation process, and the negotiation and approval procedures, and there is usually a policy on transparency even if its practical application is not described or always followed. In most cases detailed operational procedures have usually been developed as the programs have been launched and during the initial period thereafter. There is, however, much reinventing of the wheel; despite the wealth of experience, no set of standard procedures representing best practices has been established. A good example is the bid selection process, which is the subject of ongoing debate in many countries, including Madagascar and Uganda. Distillation of the experience with various transactional aspects of privatization would benefit programs that are flagging or have stalled (including those in Guinea, Kenya, Madagascar, and Senegal) and countries that have yet to embark fully on the process (including Botswana, Ethiopia, Sierre Leone, Somalia, South Africa, Swaziland, and Zimbabwe). Some important operating policies have not been established. While programs tend to be strong on transactional procedures, other operating procedures tend to be inadequate. In all of the countries reviewed, the absence of policies means that operating procedures fail to cover retrenchment (level and payment of severance pay and end- of-service benefits, and-compensation, if any, for loss of housing and other benefits); land ownership, title, and use; and retirement of enter- prise debts. During case study interviews in Zambia these omissions were said to be deliberate; they were viewed as providing flexibility and avoiding the creation of rigid arrangements that might be difficult to undo at a later stage. Such flexibility suits the government and implementers but does not satisfy labor, creditors, and investors, who need assurances if they are to support privatization fully. It is particu- larly important to resolve the issue of debt settlement. Under current procedures it is often unclear whether privatization proceeds have been or will be applied to settle the relevant enterprises' outstanding and contingent liabilities, and how the excess debt of those enterprises will be handled if proceeds are insufficient to meet the total debt. Use of proceeds is not transparent. In virtually every country the use of proceeds is not reported. This issue is expected to receive increasing attention as governments and their agencies come under more pressure to demonstrate transparency through better public infor- mation. We return to this topic in chapter 5. Emphasis on enterprise valuation is excessive and unnecessary. During preparation, much emphasis has been given to enterprise valua- 50 PRIVATIZATION IN AFRICA tion because of the importance attached to privatization agencies being seen to have secured the best possible price. Considerable effort and money have been spent on valuations. Where the deal negotiated has been within the range indicated and accepted by the privatization agency, valuations are perceived as having been helpful. In many cases, however, privatization of loss-making enterprises has been delayed because an "acceptable" valuation has not been obtained. In cases in which a professional valuation has provided a market range of values that is well below the book value of the assets, a second (and even a third) valuation has been commissioned. Privatization agencies are still finding it difficult to accept-let alone to convince the public-that the market should determine the values of enterprises up for sale, and they continue to place too much emphasis on the depreciated replace- ment cost and net book values of assets, which often bear little resem- blance to the market value. Mobilization of potential investors is weak. A thin capital market has constrained privatization in Africa. Despite clear recognition of this fact, investor mobilization has remained one of the weakest aspects of preparation. The need for capital market development in tandem with privatization has been recognized but, in terms of implementation, it has been weak (box 3.3). Only in Zambia was the opening of a share warehousing facility (the PTF) and a stock exchange planned and implemented alongside the privatization program. There are plans to open a regional stock market in Abidjan for the UEMAO countries, but even if the market opens by 1998, it will be ten years overdue for many of the Francophone countries. Public flotations have mobilized investors in Ghana, Kenya, Nigeria, and Zambia. Except in Zambia, public flotations have generally been combined not with the entry of core private investors, but rather with a retention of government shares. Only Ghana has floated shares out- side the country. For selected large enterprises, other countries should consider floating shares internationally. Predictably, the major single investors in privatized African enter- prises are foreign. In some countries, including Ghana, Kenya, and Zambia, sales to existing foreign shareholders with preemptive rights have been among some of the more important transactions. Attracting new investors has been difficult. Up to the end of 1995, a small number of new investors had arrived (South African investors have been partic- ularly active). Many countries find that foreign investors tend to be those who were already present before privatization began. That said, the picture is changing as the larger enterprises come on stream. In HOW AFRICAN GOVERNMENTS HAVE PRIVATIZED 51 Box 3.3 More Efforts Are Needed to Mobilize Investors Efforts to mobilize investors in Africa have been weak, and only four of the ten countries surveyed-Benin, Mozambique, Togo, and Zambia-have used predominantly competitive methods of privatization. Zambia estab- lished the Lusaka Stock Exchange and the PTF to promote share ownership, and the Zambia Privatization Agency is also making a big effort to encourage foreign investors. Guinea received investor search assistance financed by donors for its first series of industrial privatizations. Madagascar called on the services of a merchant bank for the privatization of a state-owned bank, and Uganda is using merchant banks to assist in the privatization of its telecommunications sector and a large bank. Ghana has included privatiza- tion in its general foreign investment promotion program, but it took the privatization agency eight years to produce its first pamphlet on privatiza- tion, and many of the early deals were noncompetitive direct sales. Nigeria's marketing efforts were aimed at tapping small local investors proportion- ately among the different states, and public flotations were designed accord- ingly; opportunities for attracting larger investors were therefore limited. In 1992 Kenya and Mozambique adopted policies that required international advertising and competitive bidding on a prequalification basis. Kenya failed to observe these policies, and Mozambique has privatized mostly very small enterprises, to which these policies do not apply. Burkina Faso intended to advertise internationally but was constrained by a lack of funds. Only Benin is known to have regularly advertised its enterprises outside Africa, although Uganda and Zambia are now using the Internet to advertise enterprises for sale. Source: World Bank country case studies. Ghana, the recent privatizations in the banking and telecommunica- tions sectors have introduced new foreign (as well as local) investors. Although an uncertain and uninviting investment climate may at least partly account for the difficulty in attracting new investors, it does not explain why so few agencies have used investor search facilities and why only one agency (in Zambia) maintains a dedicated publicity and marketing function. The case studies showed that, except for the occasional use of merchant banks and the International Finance Corpo- ration (IFC), which has advised on transactions in Gabon, Ghana, Kenya, Tanzania, Uganda, and Zimbabwe, the search for qualified investors has been restricted to a few advertisements, often only in the local press. Privatization agencies often neglect to establish the profile of investors likely to be attracted by an enterprise and capable of meeting its needs. 52 PRIVATIZATION IN AFRICA Part of the problem lies in the lack of experience with private invest- ment markets. That lack of experience has not been compensated for by external expertise because privatization agencies lack sufficient re- sources to hire outside experts. Seemingly weak efforts to mobilize investors may also reflect the inherent conflict between the aim of maximizing sale prices, which means selling to those who have or can raise capital, and the objective of broadening ownership among a population in which the majority have no savings. Agencies may have shied away from publicly generating investor interest because what makes good press for attracting investors, particularly foreign invest- ors, may not be politically appealing to the indigenous majority. But if privatization is to attract investment and extract optimum deals, program design and preparation must allow for more concerted efforts to make local and foreign investors aware of the opportunities afforded by privatization. Private sector service providers facilitate transactions. The pri- vate sector has provided services to facilitate transactions. Most valua- tions have been undertaken by private sector specialists. In Nigeria, the privatization body (the TCPC Secretariat) had sufficient autonomy to recruit its own staff from the banking sector and from private sector companies. For each privatization transaction, private sector special- ized expertise was used. Early on, most services, except for property valuations, tended to be provided by foreign consultants. In Ghana, Nigeria, Tanzania, and Uganda, for example, subcontractors have been used principally for asset and business valuations, but that has changed. The privatization process has provided an important opportunity to develop local profes- sional capacity and local firms, and individual consultants are increas- ingly being used. For example, the privatization agency in Zambia uses resources from outside to form negotiating teams, and in Ghana the agency started in 1996 to use subcontractors to manage selected transac- tions (box 3.4). There are sound reasons for using subcontractors to fulfill some privatization functions. Subcontractors can alleviate the problem of staffing shortages in the agencies and can help add transparency to the process and deflect interference in the privatization process away from the central agency. In some cases the use of subcontractors has speeded up transactions. Credibility with investors can also be en- hanced by the use of subcontractors, who can help the privatization agency conclude the best possible deal. The World Bank is encouraging agencies to use subcontractors both to increase the pace of privatization and to develop local professional capacity. To make the best use of HOW AFRICAN GOVERNMENTS HAVE PRIVATIZED 53 Box 3.4 Ghana Uses Outsourcing to Accelerate Privatization After eight years of slow progress on privatization, the government of Ghana decided to accelerate the process. To do this, policies for the use of subcontractors were approved at the cabinet level, and private firms were required to register with the Divestiture Implementation Committee (DIC). Subsequently, registered firms have been shortlisted and invited to bid for selected individual assignments, an arrangement known as "outsourcing." Firms are remunerated on the basis of a basic fixed fee and a success fee paid on completion, which is based on the total sale price negotiated. By the end of 1996 the divestiture of some 15 enterprises had been outsourced. Although it is too early to report on the final outcome of out- sourcing, the DIC believes that this arrangement is leading to more innova- tive ways of packaging the businesses to be sold and that it will result in quicker divestiture and higher sales prices because the contracted firms have the incentive to conclude deals quickly and to maximize sale proceeds. Source: World Bank, 1996d. subcontractors, a privatization agency must be able to manage their work and they must be given sufficient freedom to carry out their tasks. Local banks have been largely inactive in the privatization process. This is surprising given that the enterprises to be privatized are often among their major debtors, and that the banks need to develop their own debt work-out capacity and stimulate private sector initiative and investment. Program Management The study's review of institutional arrangements (both in the case study countries and some additional countries) shows that the key issues affecting program management are the ability of the privatization agencies to function independently, the level of resources made avail- able to the agencies, the lack of a centralized focus for privatization, and the role of holding companies. But there is another important factor that is commonly overlooked. While privatization may be a political process in terms of policy deter- mination (what and when to sell, how, and to whom), it is a business process when it comes to implementation. Getting the best deal calls for knowledge and experience of how to sell, how to identify and contact potential local and international investors, how to evaluate offers, how to package a deal, and how to negotiate. Too often, program 54 PRIVATIZATION IN AFRICA management is in the hands of bureaucrats and advisers who are well versed in policy issues and procedures but who do not know how to access and deal with the business world. Typically, institutional arrangements for privatization include an implementing body that reports to a secretariat or committee. Within this broad model there is considerable variation. In both Benin and Burkina Faso, the work is split between a committee, which oversees transactions, and an implementing secretariat, which implements the transactions. In two countries, ministries are in charge of privatization: the Ministry for Industry and State Enterprises in Togo and a special Ministry for State and Finance (Privatization) in Uganda. In Nigeria, Ghana, Kenya, and Tanzania the privatization agencies report to the head of state, and government approval is required for all transactions. In only one of the countries reviewed (Mozambique) is a distinction made between large and small enterprises, for which separate institu- tional arrangements exist. Large enterprises are handled by UTRE, a technical unit that oversees the implementing unit for privatization; small- and medium-sized enterprises are handled by individual parent ministries through special departments set up in each ministry. Table 3.1 outlines the institutional framework in the ten case study countries. It highlights the differences not only in institutional arrange- ments but also in the level of technical resources and legal powers of the principal implementing agencies. Weak agencies suffer from lack of empowerment. The level of authority, accountability, and capacity, the degree of political indepen- dence, and the legal powers accorded to privatization agencies vary considerably across countries. Most agencies recommend the method of privatization of a particular enterprise, review bids, and select the winning bid. However, final approval for transactions is often a govern- ment prerogative, and this lack of final authority has undermined the reputation of agencies and caused delays in finalizing transactions. In Benin, for example, all transactions are subject to the approval of the Council of Ministers. In Burkina Faso, a committee of ministers ap- proves all transactions, although the privatization agency has nominal authority that empowers it to carry out some transactions on the basis of a prepared list. In Kenya, the privatization unit must obtain clearance for transactions from a committee composed mainly of government personnel and headed by the minister of finance. In Ghana and Tanza- nia, the agencies can plan, initiate, and negotiate privatization transac- tions but are still dependent on the government for formal approval. In Nigeria, the agency reports directly to the president, and strict limits are placed on its portfolio, which is concentrated on federal, not state, Table 3.1 A Comparison of the Institutional Framework for Privatization in Ten Countries (1995) Country Lead institution(s) Powers Constraints Other institutions involved Benin Commission Technique * Commission hires Liquidation process is too * Council of Ministers (which de Denationalization consulting firms to carry long (3-9 years) must approve every and Cellule du Project out technical studies privatization transaction); d'Assistance Aux * Cell executes privatization * Tutelle ministries Enterprises (6 staff) transactions Burkina Faso Privatization * Reviews bids for valuation * Tutelle ministries Commission (a * Selects winning bids committee of * Analyses offers ministers) and * Authorizes Permanent Permanent Secretariat Secretariat to initiate (4 professionals) negotiations * Signs deeds of sale Ghana Divestiture * DIC Secretariat plans and Committee lacks authority * State Enterprises Commission Implementation implements selected to execute transactions; * Ministry of Finance Committee (DIC) and privatization transactions can recommend but not * Sector ministries Permanent Secretariat approve transactions * Ghana Stock Exchange (10 professional staff that require government and advisers) approval Sector ministries * Sector ministries oversee major transactions Non-Performing Assets * Trust liquidates and sells Recovery Trust assets, negotiates, resched- ules or sells debt to third par- ties, and participates in the privatization of tradable en- terprises within its portfolio Kenya Parastatal Reform * ESTU selects enterprises ESTU has no legal status * Ministry of Finance Programme Committee for PRPC approval and does not deal with * Holding companies (DFIs) (PRPC) and its * PRPC recommends method all transactions * Stock Exchange Commission Executive Secretariat of privatization * Line ministries Technical Unit (ESTU) (3 professional staff) (Table continues on following page.) Table 3.1 (continued) Country Lead institution(s) Powers Constraints Other institutions involved Madagascar Commission * Formulates legal Interference from * Office of the President Independante de framework Government and * Tutelle ministries Privatisation (CIP); vested interest Direction Generale pour * DGGP participated in the groups and DGGP's la Privatisation initial privatization lack of a full (DGGP) transactions until 1993 mandate resulted when the program was in ad-hoc suspended privatizations outside DGGP's purview Mozambique For large enterprises: * CIRE advises Prime Interministerial Minister on decisions Committee for based on UTRE's work Restructuring of * CEP prequalifies bidders, Enterprises (CIRE) recommends and chaired by the PM evaluates bids, selects with Minister of winners and negotiates Finance as deputy; deals Privatisation * UTRE reviews CEP's Executive decisions, analyzes each Commission (CEP) enterprise and its Technical Unit, UTRE For small and medium enterprises: departments within individual parent ministries; technical work is handled by units; mostly in GREI (Ministry of Agricul- ture and Fisheries) Nigeria Technical Committee on * Most decisions other than Encountered some * Securities & Exchange Privatisation and valuations for public resistance from line Commission Commercialisation offerings made by TCPC ministries * Private sector technical advisors (TCPC), subsequently Restricted to federal * Federal sector ministries succeeded by the enterprises * State governments Bureau of State Enter- prises, reporting di- rectly to the President Togo Ministry for Industry * Clear mandate for * Ministry of Finance and State Enterprises privatization and/or * Ministry of Planning & Industry u-i improving state * Secretary of Public Enterprises 9*1 enterprises Uganda Privatization (8 * Handles all matters No legal status-cannot * Line ministries professional staff) relating to privatization sue or be sued under the Minister of Holding companies State and Finance (Privatization) Zambia Zambia Privatization * Executes privatization Holding companies (until * Cabinet (approves privatization Agency (ZPA) (12 transactions dissolved) method) board members, of * Monitors redundancies * Private firms, who assist in which only 3 are from * Approves disposal and valuations, bid evaluations and government, 32 lease of assets in all SOEs negotiations professional staff and to be privatized * Privatization Trust 6 full-time advisers) Fund * Stock Exchange 58 PRIVATIZATION IN AFRICA public enterprises. In Madagascar, the agency retains powers that in- clude acting as a liquidator. In Kenya, weak institutional arrangements have hindered privatization (box 3.5). Only in Zambia does the privatization agency have the authority to conclude deals. In fact, the agency's mandate extends beyond privatiza- tion transactions and includes the safeguarding of public enterprise assets and the monitoring of the social impact of privatization. Another notable feature of the Zambian agency is that it has a board of 12 directors, 9 of whom are from the private sector, thus providing a high degree of independence from the government. Why are so many governments reluctant to empower their privatiza- tion agencies? It is often argued that privatization is a political process and that therefore a political entity has to approve transactions, but the example of Zambia shows that this argument does not hold true. What is true is that the political backdrop varies from one country to another and that those variations may lead to differences in approval processes. For example, government ministries typically have different agendas for privatization than the central privatization agency, not least because line ministries, like holding companies, view privatization as a curtailment of their authority and influence. Full consensus within the government is therefore rare. Ultimately, the public wants to be Box 3.5 How Weak Institutional Arrangements Hinder Privatization: The Case of Kenya Kenya's privatization program was designed to be coordinated centrally by a privatization implementation agency reporting to a govemment-appointed committee that would oversee both parastatal reform and privatization. In reality, the agency, which lacks legal status and authority, has played a much less important role than envisaged and at end 1995 had handled less than one-half of all privatization transactions. Public flotations have been man- aged by enterprises themselves. Receiverships and liquidations, which have constituted a large part of the privatization program to date, have been supervised by holding companies. Some privatizations in the agricultural sector have been managed by the Ministry of Agriculture and holding companies. Even transactions handled by the agency, which were vetted by the oversight committee, have been resisted in many cases by other stakeholders, such as the holding companies, sector ministries, and enter- prise managers. Organizational changes made during the first half of 1995, which entailed moving the agency to the Ministry of Finance, did not resolve some of the shortcomings in the organizational structure. Source: World Bank, 1996e. HOW AFRICAN GOVERNMENTS HAVE PRIVATIZED 59 Box 3.6 The Single Biggest Constraint on the Zambian Program Was Interference by Holding Companies The most significant constraint on privatization in Zambia was, until its closure in March 1995, the Zambia Industrial and Mining Corporation (ZIMCO), whose portfolio included some 130 public enterprises ranging from the single largest commercial concern in the country to small govern- ment-owned service shops. The management of this powerful holding com- pany did its utmost to frustrate privatization transactions and to perpetuate ZIMCO's empire and influence, principally by deliberately withholding information and legal documents, transferring funds from enterprises (be- fore and after deals had been negotiated), and stirring public animosity toward privatization. Because ZIMCO was the legal holder of shares in its portfolio, senior ZIMCO personnel needed to sign the transfer certificates before shares could be transferred. ZIMCO thus had ample scope to delay transactions that its personnel viewed as the selling off of its portfolio and, ultimately, the dismantling of the holding company. The government took a brave and necessary step in deciding to dissolve ZIMCO. The experience of ZIMCO's interference and its negative impact on the process offers a clear lesson for successful privatization: where a large holding company exists, remove its influence by dissolving it and transferring the shares temporarily either to the privatization agency or to some form of trust. There is no doubt that in Zambia's case, transactions in the early years of the privatization program would have been concluded sooner and public opinion less adversely influenced had it been possible to remove ZIMCO's influence earlier. Source: World Bank, 1996k. assured that fair and rational decisions are made, and there may be instances in which only a senior person in government can provide that assurance. State-owned holding companies obstruct privatization. Reluctant to lose their empires, state-owned holding companies have obstructed and delayed privatization transactions. To minimize this obstruction, governments have tried to involve the holding companies in the process (Tanzania), sometimes by giving them the responsibility to privatize (Kenya and Zambia). These efforts have been unsuccessful and have merely played into the hands of the obstructionists. As the Zambian experience shows (box 3.6), it is better to restructure and privatize or dissolve holding companies as early as possible in the privatization program before their resistance to the process can significantly delay the process (as it did in Kenya and Tanzania, as well). 60 PRIVATIZATION IN AFRICA Fragmented institutional arrangements. Fragmented institutional arrangements for privatization are common in Africa. Privatization agencies in Ghana, Madagascar, Nigeria, and Togo were set up or given a legal mandate after privatization transactions had begun. In Madagascar, the early privatization deals were handled by ad hoc committees that were empowered by decree to privatize. Although a central agency for privatization was set up when transactions became frequent, ad hoc privatization deals continued outside the cerntral agency, which undermined its authority and caused the frequent stall- ing of transactions it had been handling. In most countries surveyed, some privatization transactions have taken place outside the purview of the central agency. In some cases, such transactions took place because certain sectors, such as mining and natural resources, had been specifically exempted from the privat- ization program or reserved to the state for special disposition (Burkina Faso, Ghana). In other cases, some transactions had taken place prior to the formation of the agency or the inception of the program (Kenya). In some instances (Tanzania) it is unclear why certain transactions were not handled by the privatization agency. Transactions handled outside the central agency have typically been either very large (as in the case of Africa's largest transaction, the Ashanti Goldfields) or were for enterprises not in the manufacturing and trading sectors. The privatizations of tea factories in Kenya and Uganda, for example, were handled by the parastatals in charge of the tea factories; banking sector transactions in Burkina Faso and Nigeria were handled outside the purview of the agency. Privatization in Nige- ria has proceeded on two tracks, one for enterprises handled at the federal level and another for enterprises being privatized by the states. One of the side effects of the fragmented institutional structures for privatization has been the lack of centralized data on privatization. In Ghana, Kenya, Madagascar, Mozambique, Nigeria, and Uganda, it has been difficult to establish with precision the total number of privatiza- tion transactions that have taken place. As a consequence, the lists and summaries of transactions compiled for the case studies and the database may be incomplete. Even where a multitrack approach to implementation is preferred, a central focus should be maintained. Centralized coordination-not nec- essarily control-of privatization can have many benefits, including a greater assurance of policy consistency; the pooling of expertise and full employment of experience; easier data collection, monitoring, and dis- semination of information; and greater confidence that transparent pro- cedures have been followed at all stages. Centralizing privatization can also serve to augment the authority of the central privatization agency. HOW AFRICAN GOVERNMENTS HAVE PFRIVATIZED 61 Resources forprivatization are inadequate. The two mostcommon problems faced by privatization agencies are the lack of adequate re- sources and the lack of legal authority. Many agencies are severely under- staffed and suffer from serious shortages of qualified personnel and other budget resources (Burkina Faso, Ghana, and Kenya). Although under- staffing should have been expected in countries that suffer from human resource capacity constraints, it was not fully taken into account when privatization programs were developed. Where the World Bank and do- nors have made financial or human resources available to the agencies, assistance has sometimes been slow to arrive. In other cases, including Kenya, countries have been slow to use the credits and assistance avail- able to them. In some quarters this is interpreted as a sign of the low importance that the governments have attached to privatization. In only a few cases (Benin, Nigeria, and Zambia) is the privatization agency geared up with a full complement of staff and advisers to a level that is consistent with the volume and complexity of privatization transactions being handled. The lack of financial and human capital resources has had a marked effect on the pace of privatization. Agencies with insufficient resources have not been able to process transactions at the rate envisaged during program design. As a consequence, targets for privatization that were agreed on by the countries, the World Bank, and the IMF have often not been met. To meet these targets, some countries have concluded insignificant deals in order to increase the reported number of com- pleted transactions. Some agencies have begun to address their resource constraints by the use of subcontractors (Ghana). Legal authority is lacking. Experience shows clearly that for a privatization agency to be able to make timely decisions and to con- clude deals expeditiously, it must be vested with the appropriate legal authority to do so and must be headed by a person of sufficiently high status and reputation. Some agencies lack the necessary legal authority to implement the privatization program; Kenya and Ghana are cases in point. The prob- lems arising from a lack of legal authority to negotiate and conclude transactions expose them to obstruction from other stakeholders, such as enterprises, holding companies, and line ministries. In many cases (Burkina Faso, Ghana, and Kenya) the agency has been required to defer to the government (the President, the Cabinet, or the Council of Ministers) on many issues that considerably delayed the privatization process. Even where vetting or privatization decisions are not a formal requirement, the privatization programs have encountered resistance 62 PRIVATIZATION IN AFRICA Box 3.7 Press Accusations of Lack of Transparency: The Case of Kenya In January 1995, 41 opposition members of parliament in Kenya published a press statement complaining about the "severely chaotic, blatantly messy, grossly irregular and shamelessly fraudulent manner and the secrecy in which the Privatisation Programme has been and is being conducted." They called for the immediate removal of the head of the privatization agency, an investigation of the agency's activities, and the reversal of a privatization transaction. The opposition members correctly pointed out that the agency had no legal status and that "Parliament and the Kenyan public are entitled to know the amount of money so far realized from privatization sales and where that money has been deposited." On the same day a press article stated that "the privatisation programme has opened up perhaps the most lucrative means of looting and legitimizing the plunder of public investments as well as transferring state wealth to some members of the ... community." It went on to say that "profitable parastatals, which have taken years of nurturing at enormous public cost, are disposed of for paltry sums in pre-arranged deals." This hyperbolic attack clearly reveals the lack of public relations on the part of Kenya's privatization agency, the lack of awareness about the condition of the para- statal sector, and the lack of understanding of the presence and rights of minority shareholders with preemptive rights. In March 1995, the head of the privatization agency was removed. Since then, the govemment has revamped its privatization program, and greater attention has been given to public relations. Source: World Bank, 1996e. from sector ministries and the ministries of finance, which have, on occasion, withheld information on the enterprises in their portfolios. Transactions lack transparency. Lack of transparency is a key prob- lem in many privatization agencies, and the problem is often com- pounded by poor public relations (box 3.7). In Madagascar, the lack of a legal framework for privatization re- sulted in a host of vested interest groups manipulating the process for their own ends and rendered the program nontransparent and suspect in the eyes of the public. The exercise of preemptive rights by existing shareholders in Kenya was misunderstood by the media and the public, and the government was slow to explain the facts and to allay concerns. In Ghana, the lack of information on certain procedures and transac- tions fueled criticism by the press, and the government's failure to provide an immediate response increased public suspicion of the pro- HOW AFRICAN GOVERNMENTS HAVE PRIVATIZED 63 gram. In Mozambique, the public perception was that most public enterprises had been sold to foreigners, although in fact 93 percent of the buyers were nationals. Information on the program in Burkina Faso was communicated to the public through statements in newspapers and through public speeches by ministers, but the public at large felt left out of the process. In Benin, the government failed to explain its privatization policy to the public, and the 1992 Privatization Law was openly criticized. Al- though this may have been more a problem of poor communication than lack of transparency, it played into the hands of political opponents. In Togo, the lack of a legal framework and the lengthy negotiations to con- clude transactions rendered the process nontransparent in the eyes of the public, especially when the sale prices of enterprises were well below book values. The eventual outcome of one deal seriously undermined the business community's confidence in privatization (box 3.8). In eight of the ten country case studies-the exceptions are Nigeria and Zambia-transparency was cited as a major issue of concern to parti- cipants and the general public. In view of the lack of consensus on privat- Box 3.8 How Failure to Carry Out Due Diligence Affected Togo's Privatization Program An American businessman set up a company, STS, in Togo in 1984 that leased the facilities of a state-owned steel plant (Societe Nationale de Siderur- gie). The businessman negotiated the lease with the government and, thanks to favorable incentives as well as aggressive marketing, managed to make STS a profitable showcase that did much to promote the international reputa- tion of Togo as highly conducive to investment. For the first few years STS paid its annual lease rent and seemed to be performing very well. After two years, one-third of the shares in STS were sold to Togolese investors, and the company began to diversify into new products and new markets (in Benin and C6te d'Ivoire). As a visible marketing innovation, STS used a floating market, including a small boat that plied West African ports selling r-bars. Behind the scenes, the businessman gradually milked the company and covered his traces with creative accounting techniques. On departing the country in September 1991, he left behind a debt-ridden shell with liabilities of 3 billion CFA francs. Suppliers, bankers, the government, and Togolese shareholders had no hope of recovering sums due to them and the company ended up in bankruptcy. This regrettable episode seriously weakened the public's confidence in privatization. Source: World Bank, 1996i. 64 PRIVATIZATION IN AFRICA ization, transparency and accountability are of paramount importance in addressing public concern and building confidence. In many African countries there is a perception that the privatization process is influ- enced, if not directly determined, by vested political and financial inter- ests. If a government wants to convince the public that this is not the case, it must ensure that its privatization agency follows transparent proce- dures and keeps the public informed. If not, it is not surprising if the public assumes that vested interests have come into play. Transparency can do more than simply reduce allegations about vested interests and secure public support. It may help in attracting more and better investors through a more credible process, obtaining better deals, yielding higher proceeds, reducing legal complications and costs, and avoiding reversions. However, transparency is costly; it demands time and resources that few implementing agencies have. Transparency requires written enforceable procedures against which actions can be judged. These may include, among others, the adoption of competitive methods of privatization (as far as is possible), provision of adequate notice, publication of appropriate information before and after deals are concluded, public availability of nonproprietary informa- tion, and the use of objective, quantitative criteria for selecting bidders. It also requires that claims of nontransparency be dealt with expedi- tiously and fairly through an appropriate appeal procedure. Most coun- tries lack such procedures; some lack the requisite judicial support for transparency (legislation in Ghana, for example, specifically prevents legal action against the privatization agency; Kenya's agency lacks any legal status). The Zambian Privatization Agency (ZPA) is exemplary for the atten- tion it has paid to ensuring public accountability and transparency. The ZPA has been given wide-ranging powers to recommend and execute privatization transactions. Following the removal of the principal state- owned holding company, the ZPA controls the information flow from each public enterprise, has approval power over public enterprises' capi- tal investment programs, and can direct enterprises as to actions they need to take prior to privatization: Because its financial statements are audited and semiannual reports are submitted to the minster of finance (the latter being public documents available to the general public), the ZPA is subject to independent public scrutiny, which has enhanced its reputation for transparency. The ZPA has also undertaken measures de- signed to win the public's confidence. To reach a large percentage of the population, it has published information on privatization in eight local languages. It also publishes monthly status reports on enterprises that are being and have been privatized, the privatization methods em- ployed, the identity of all bidders, the terms offered and negotiated, and HOW AFRICAN GOVERNMENTS HAVE PRIVATIZED 65 other information on each transaction. These measures provide a model for agencies in other parts of the continent. The Legal Framework The legal framework can directly support privatization through en- actment of laws covering the processes for preparing for privatization and transacting deals. It can indirectly support privatization by making the investment environment more conducive by affording private firms access to credit and land, enforcing contracts, and allowing the private sector to compete on an equal basis with the residual state sector. This section is concerned primarily with the direct effect of the legal framework. Typically, the authority for a program emanates from a privatization law, which defines the key principles on which the program will be based and the institutional arrangements for implementing the pro- gram. Other laws provide for the legal steps that are necessary before privatization can commence (such as corporatization of state-owned enterprises), and stipulate how much foreign ownership in an enter- prise is permitted and who holds legal title to the privatized assets. With the exception of Kenya, all of the countries in the sample had a privatization law or decree governing the privatization activity and, in some cases, creating the privatization agency charged with imple- menting privatization transactions. In one instance (Burkina Faso) par- liamentary approval of the entire privatization program was required before the program could be launched. In other cases legal sanction of the privatization process was obtained after the fact. In Ghana, for instance, the law granting authority to the Divestiture Implementation Committee to move ahead with privatization was not passed until six years after the start of the privatization program and had retroactive effect. Similarly, in Togo, privatization had proceeded for several years without a specific law; only in 1994 was an ordinance passed specifying the acceptable methods of privatization, employee participation incen- tives, and the membership of the privatization commission. In yet other cases, the legislature has issued omnibus laws authorizing the privatization program (Zambia) or laws listing enterprises to be privat- ized (Burkina Faso). In Tanzania an amendment to the State Corpora- tions Act was passed to facilitate privatization. Often, parliamentary approval is required for transactions involving enterprises established by discrete acts of parliament (as in the case of statutory corporations in the Anglophone countries) or for transactions in a particular sector. In Ghana and Burkina Faso, legislative approval 66 PRIVATIZATION IN AFRICA is required for transactions in certain core sectors such as mining and natural resources. In Kenya, although parliamentary approval was not required for the overall privatization program, it was a necessary condi- tion for the privatization of Kenya Airways. In Burkina Faso, legislative approval was required on a case-by-case basis for certain significant transactions. Evidence shows that having a privatization law in place prior to the onset of the program facilitates the privatization process. Enactment of a law can help reduce public criticism and prevent an institutional void. Experience clearly shows that such a law, when it accords the requisite authority to the privatization agency, gives investors confidence and re- duces the risk of transactions being stalled or derailed. The most compre- hensive privatization legislation is Zambia's Privatization Act of 1992, which created an agency with an extensive mandate and far-reaching powers. Although the effect of this law on the subsequent success of pri- vatization is impossible to gauge, it is noteworthy that Zambia's program is the most successful program studied (see chapter 6). A timely privatization law that explains the objectives and principles of the program can be a powerful tool for a public relations campaign and can deflect public criticism of the program. In Benin, there was no privatization law at the time of the sale of the brewery, La B6ninoise, leaving the government open to criticism. Although a privatization law was subsequently passed, public opinion about privatization had already been negatively influenced, the most recent manifestation being the elections that resulted in success for the party opposed to priva- tization. A major legal constraint hindering privatization transactions has been the uncertainty in many countries over land rights and title to land. First, there are problems due to outdated property laws that do not recognize private interests in land or that prevent foreigners from owning land or securing long leases. Second, disputes arise over owner- ship where nationalization occurred in the past and former owners either were not adequately compensated or consider that they have a preferential right to reacquire property that is to be privatized. Third, records in companies and official registries are not always up-to-date or readily available. In Ghana and Uganda, uncertainty and disputes over title to land have hindered some transactions. Virtually no sales contracts under the Ghanaian privatization program have included documentation to support the legal transfer of title to land. In Kenya, a presidential exemption is required to acquire agricultural land, and transfers are subject to the discretion of provincial land boards. In Zambia, the Land Act that enables foreign investors arriving through the Investment Centre to purchase land was only recently passed. HOW AFRICAN GOVERNMENTS HAVE PRIVATIZED 67 Although an increasing number of African countries are now offering incentives to foreigners for investment, in many countries the business environment for foreign investment (encompassing the macroeconomic situation, the banking system, and the availability and quality of utility services, as well as the legal regime and judicial enforcement) is still a deterrent to potential investors. This has seriously constrained Africa's ability to attract core investors for enterprises in dire need of infusions of capital, technology, and management expertise. In Nigeria, until mid-1995 foreign participation in certain activities was limited by the Nigerian Enterprises Promotion Decree (1989), which barred foreign participation in 40 activities, including trade, transport, and the media, and the Industrial Policy of Nigeria (1989), which limited foreign equity participation to 40 percent in banking and insurance companies. The Exchange Control Act of 1962 required strictly regulated foreign ex- change transactions and approvals for foreign equity investments and repatriation. The decree and act were repealed in mid-1995, but the industrial policy remains in effect. In Benin, the 1992 Privatization Law limits foreign investment in a privatized public enterprise to no more than 65 percent. Although the law allows for certain exceptions by the government in particular circum- stances, it is nonetheless a deterrent to foreign investors because 33 per- cent ownership of the shares can constitute a "freezing minority" under Benin's company law. Thus, even if a core investor were to take a 65 percent stake in an enterprise, it would not necessarily suffice to gain control over the management of the company. In Togo, foreign invest- ment may be limited on a case-by-case basis, and foreigners are not al- lowed to buy land and buildings unless authorized by the government.6 In Burkina Faso, ordinances limit foreign investment to 65 percent of the shares of any enterprise, and to 49 percent in "vital or priority" sectors.7 In both Ghana and Madagascar, the law does not permit foreigners, and therefore foreign controlled companies, to own freehold title to land, but foreigners can enter into long-term leases of up to 50 years.8 Laws and judicial proceedings relating to liquidations have caused delays in privatization transactions in some countries. In Togo, liquida- tions initiated in 1990 were still in process in 1995. Delays also were reported in Burkina Faso, Ghana, Kenya, Madagascar, and Tanzania. Apart from liquidations, however, few privatization transactions have ended up in the courts, and overcrowded African court systems have not been major impediments to privatization. 4 The Record to Date privatization in Africa is on the rise. Although many public enter- prises remain in government portfolios, acceptance of the need to reduce the size of the public enterprise sector has grown. By the end of 1996, just under 2,700 privatization transactions had taken place in Sub-Saharan Africa, with a total sales value of about $2.7 billion (see figure 4.1). Up to 1996, emphasis was primarily on divesting small and medium-size enterprises. 1997 and 1998 may not witness an increase in activity in terms of transaction numbers. However, with attention focusing more on the larger enterprises, including several privatizations in the telecommunications sector, we expect a significant increase in the value of the deals concluded. Since data for many countries are not yet available for 1997, we record and comment here on the activity to the end of 1996. At the beginning of 1990, about a dozen countries in Africa had undertaken some form of privatization; by 1993 that number had dou- bled and by end 1996 all but five countries had divested some public enterprises. Appendix A provides a breakdown of transactions by year for each of the 48 countries in the region. In terms of the number of transactions, most of the privatization activity has been concentrated in a few countries. Of the 2,689 reported privatization transactions that took place between 1988 and the end of 1996, 1,891 (just over two-thirds) were concentrated in ten countries: Mozambique, Angola, Ghana, Zambia, Kenya, Tanzania, Guinea, Mad- agascar, Nigeria, and Uganda (see table 4.1). Sectoral Distribution Privatization has occurred most often in the manufacturing and industrial sectors which accounts for about 30 percent of all reported transactions for which a sectoral breakdown was available (figure 4.2). Enterprises involved in agricultural production, marketing, and pro- cessing accounted for 23 percent and transactions in service industries for 22 percent, but transactions involving financial institutions ac- counted for less than 5 percent of all transactions. 68 THE RECORD TO DATE 69 Figure 4.1 The Number of Privatization Transactions in Africa Is on the Rise 500- 450- 199 400 99-92 19 19 95 19 0 5- 3_ 00 .2 w250 __200 o6 150- E 100- 50- 0 1990 1991 1992 1993 1994 1995 1996 Note: Privatization transactions prior to 1990, which totaled 334, are not included. Source: Privatization agencies, government ministries, and World Bank data. Table 4.1 Ten Countries Account for Most Privatization Activity in Africa Number of transactions Country (as of end 1996) Mozambique 548 Angola 331 Ghana 191 Zambia 191 Kenya 152 Tanzania 123 Guinea 114 Madagascar 84 Nigeria (federal government only) 81 Uganda 76 Total 1,891 Source: Privatization agencies, government ministries, and World Bank data. 70 PRIVATIZATION IN AFRICA Figure 4.2 Privatization Transactions in Industry Outnumbered Those in Other Sectors during 1986-96 Mining Energy, telecom.. and utilities Transport Financial Trade Services, tourism, and real estate Agriculture, agroindustry, fisheries Manufacturing and industry o 150 200 300 400 So0 600 700 800 Number of privatization transactions in Africa, 1986-96 Note: Excludes 291 transactions for which a sectoral breakdown was not available. Source: Privatization agencies, government ministries, and World Bank data. Value of Transactions Looking at the numbers of transactions alone as an indicator of the extent of privatization is misleading because of the typically small size and low values of the enterprises divested. For the ten-country sample that provided most of the data for this report, a numerical count would suggest that Mozambique, with 548 transactions, is the leader in privat- ization (table 4.2). Most transactions in Mozambique, however, were of small retail establishments, many employing fewer than 100 employees. Similarly, the total number of transactions in Kenya was affected by the privatization of many minority equity stakes in tea factories. As table 4.2 shows for the ten country case studies, there is no correlation between the number of transactions and the sale values: 191 transactions in Ghana yielded more than $400 million, whereas 548 transactions in Mozambique yielded only about $179 million. In the period 1988 to 1996, only 12 countries reported transactions totaling more than $50 million and only 9 countries (South Africa, Ghana, Nigeria, Kenya, Mozambique, Zambia, C6te d'Ivoire, Uganda, and Tanzania) had sales totaling more than $100 million (table 4.3). The value of transactions has been highest in South Africa, where privatization of two major enterprises yielded $762 million. Only 128 of the total of 1,259 transactions for which sales value data (excluding liquidations) were available grossed more than $3 million. THE RECORD TO DATE 71 Table 4.2 Numbers of Transactions Are Not Correlated with Values of Transactions Number of transactions, Total sales value Average sales value Country 1988-96 (US$ million) (US$ million) Ghana 191 417.3 2.2 Nigeria 81 206.9 2.6 Mozambique 548 179.7 0.3 Zambia 191 146.3 0.8 Uganda 76 136.9 1.8 Benin 46 63.5 1.4 Kenya 152 184.6 1.2 Togo 45 35.9 0.8 Madagascar 84 18.8 0.2 Burkina Faso 16 9.6 0.6 Total, 10 countries 1,430 1,399.5 1.0 Other Sub-Saharan 1,259 1,304.6 1.0 African countries Total, Sub-Saharan 2,689 2,704.1 1.0 Africa Source: Privatization agencies, government ministries, and World Bank data. Table 4.3 Total Value of Transactions Exceeded $50 Million in Only 12 Countries Total number of Total sales values Country transactions (US$ million) South Africa 2 762 Ghana 191 417 Nigeria 81 207 Kenya 152 185 Mozambique 548 180 Zambia 191 146 C6te d'Ivoire 46 123 Uganda 76 137 Tanzania 123 126 Senegal 48 65 Benin 46 64 Malawi 44 56 Total 1,548 2,468 Source: Privatization agencies, government ministries, and World Bank data. 72 PRIVATIZATION IN AFRICA Most transactions involved small to medium-size firms. Of the 1,259 transactions, as many as 472 (37 percent) involved sales values of less than $100,000. Table 4.4 shows the sales value distribution of privatiza- tion transactions in all African countries to end 1996. A few transactions have accounted for most of the privatization sales values. The top 20 transactions (see table 4.5) totaled $949.53 million. Of that total of $949.53 million, transactions in South Africa and Ghana together accounted for $445 million, or almost 47 percent of the total. Privatization Methods Used As privatization has progressed in Africa, so has familiarity with different methods of privatization. The use of a wide range of different methods is now a well established feature of privatization efforts. A 1994 report (Berg 1994) found that liquidations were the most com- monly used form of privatization up to that time, followed by sales as going concerns. By December 1996, 16 methods of privatization had been employed in Africa, with 32 percent of all transactions (for which information on the method used was available) involving the sale of shares by competitive tender. Figure 4.3 summarizes the privatization methods applied in Africa, and an analysis by country is set out in appendix B.9 Table 4.6 summarizes the methods employed by year. Not all sales of shares have been open and transparent. In An- gola, Ghana, and Senegal there are many examples of shares being Table 4.4 Most Privatized Enterprises Are Worth Less Than $100,000 Share of Sales value of Share of Total sales total value transactions Number of transactions value (US$ million) (US$ million) transactions (percent) (US$ million) in percent More than 3 128 5 2,267 83.7 1-3 142 5 245 9.1 0.5<1 117 4 83 3.1 0.1 <0.5 400 15 97 3.6 Less than 0.1 472 18 13 0.5 Value not known, 1,430 53 and/or liquidations Total 2,689 100 2,704 100.0 Source: Privatization agencies, government ministries, and World Bank data. Table 4.5 The Top 20 Transactions Accounted for More Than a Third of Total Value Government ownership Date (percent) Total sales of Before After value Country Name of enterprise sale Sector sale sale Method of sale (US$ million) Ghana Ashanti Goldfield Corporation 1994 Mining 55 30 Public flotation 316.00 South Africa Radio Stations 1996 Telecom 100 0 Assets (competitive) 129.00 Kenya Kenya Airways 1996 Transport 100 20 Shares (competitive) 74.07 Tanzania Tanzania Cigarette Company 1995 Agroindustry 100 0 Shares (competitive) 55.00 Senegal SOCOCIM 1991 Industry 97 0 Shares (competitive) 39.20 Zambia Zambia Sugar Plc 1995 Agroindustry 100 0 Shares (direct) 36.80 Nigeria Festac 77 Hotel Plc, Lagos 1992 Service 100 0 Shares (direct) 30.53 Benin La Beninoise 1992 Industry 100 0 Assets (competitive) 28.37 ] Uganda Nile Hotel Complex 1995 Service 100 41.4 Joint venture 26.90 QJ Nigeria Tourist Co. of Nigeria 1992 Service 100 0 Shares (direct) 25.10 Tanzania Tanzania Breweries 1993 Industry 50 50 Shares (competitive) 22.50 Cote d'lvoire Societe des Caoutchoucs de 1995 Agroindustry 95 35 Public flotation 22.21 Grand-Bereby (SOGB) Uganda Hima Cement Factory 1994 Industry 100 0 Assets (competitive) 20.50 Mozambique Cimentos de Mozambique 1994 Industry 100 49 Shares (competitive) 20.00 Cape Verde Cabo Verde Telecom 1995 Telecom 100 60 Shares (competitive) 20.00 Kenya National Bank of Kenya 1996 Financial 63 23 Public flotation 18.18 Ethiopia Coca Cola Bottling Plant 1995 Industry 100 0 Assets (competitive) 16.60 Nigeria Nigeria Hotels Plc. 1992 Service 47.46 0 Shares (direct) 16.40 Benin Societe Nationale des Ciments 1991 Industry 100 0 Assets (competitive) 16.39 (SONACI) Ghana Achimoto Brewery Co. Ltd. 1991 Industry 100 0 Public flotation 15.78 Total 949.53 Note: Excludes transactions prior to 1990. Source: Privatization agencies, government ministries, and World Bank data. 74 PRIVATIZATION IN AFRICA Figure 4.3 Privatization Methods Applied in Sub-Saharan Africa Open auction 2 Equity dilution 2 Debt/equity swaps 7 Direct sales of assests 27 Joint ventures 27 Transfers 3t Restitutions 39 Management or employee buyouts Management contracts 47 Public flotations 71 Preemptive rights sales 76 Leases and concessions 92 Direct sales of shares 0O Competitve sales of 404 assests Liquidations 514 Shares sold by compebtive 7 tender 0 100 200 300 400 500 600 700 800 900 Number of transactions Note: Excludes transactions for which data were not available. Source: Privatization agencies, government ministries, and World Bank data. sold directly. In Kenya, shares were sold to existing shareholders with preemptive rights. Some asset sales have also been handled on a non- competitive basis, which has led to accusations of lack of transparency. In many countries, including Ghana and Kenya, it is difficult to ascer- tain how many bidders participated in competitive processes and what selection criteria were applied. Although the number of enterprises being privatized through indi- rect methods such as lease arrangements and management contracts is on the rise, their use is not yet widespread. Because such indirect mechanisms are most often employed in the privatization of utilities, the low usage of this method of privatization underscores the ex- tent to which utilities in Africa remain in the public sector. A signifi- cant increase in direct and indirect means of privatization of utilities is expected in 1998 and subsequent years, including combinations of sales to minority core investors who are also given management control. Methods that broaden ownership, such as management employee buyouts, have not been commonly used in Africa. Despite the expressed THE RECORD TO DATE 75 Table 4.6 Competitive Sale of Shares Is the Most Widely Used Method of Privatization (number of transactions) Prior Year to not Privatization method 1990 1990 1991 1992 1993 1994 1995 1996 known Total Sale of shares Public flotation 15 8 4 12 9 12 5 6 71 Competitive tender 127 49 70 85 85 162 154 131 12 875 Open auction 2 2 Direct sales 10 15 12 17 8 19 16 11 108 Preemptive rights 1 10 5 56 4 76 Sale of assets Liquidation 137 47 33 62 21 59 98 54 3 514 Competitive tender 21 27 20 33 15 68 70 150 2 404 Direct sale 2 8 3 5 6 2 1 27 Other Management or employee buyout 1 1 1 4 5 16 15 44 Equity dilutions 1 1 2 Transfer to trustee 3 2 3 12 6 26 Transfer without remuneration 2 2 1 5 Debt-equity swap 1 2 3 1 7 Joint venture 1 1 4 1 2 9 7 2 27 Restitution 1 1 8 2 7 13 7 39 Lease 11 8 14 12 20 14 7 6 92 Management contract 3 5 8 1 4 11 12 3 47 Method not specified 1 5 7 2 3 22 283 323 Total 334 171 163 256 195 367 474 428 301 2,689 Source: Privatization agencies, government ministries, and World Bank data. desire by many governments to broaden ownership, only 44 manage- ment/employee buyouts, representing less than 1 percent of all transac- tions, have taken place. Other methods used include restitution and trusteeship arrangements. In some cases, notably in Ghana, Madagasar, and Zambia, and to a lesser degree in Burkina Faso, Tanzania, and Nigeria, some enterprises have reverted to private ownership through a process of restitution. Trustee arrangements have been used where governments want to demonstrate their commitment to transferring ownership to domestic private investors. The trusts are set up to ware- 76 PRIVATIZATION IN AFRICA house shares temporarily until they are sold through public flotation. This was the mode used in Zambia. In Ghana and Tanzania the govern- ments have preferred to retain a block of shares for later sale to the public. Ownership, Control, and Exit Table 4.7 summarizes the changes in government ownership of enter- prises resulting from all reported privatization transactions on which such information was available. The summary shows that prior to privatization, governments had owned the majority equity in over 80 percent of the enterprises which were subsequently the subject of privatization transactions. Following privatization, the governments retained no ownership in at least 72 percent of all privatized enterprises and had maintained a majority ownership in only 6 percent; and 10 percent of transactions resulted in minority government ownership. In some cases, shares remain in government ownership with a view to subsequent public flotation once the capital market develops. For this purpose, shares in some enterprises in Zambia are held by the privatization trust. In other countries the rationale for government retention is unclear-for example, the governments in Ghana and Ke- nya continue as partial owners of some hotels. Some 8 percent of transactions involved sales of minority-held interests and did not con- stitute privatization in the sense of a transfer of ownership control. Ownership changes usually provide a good picture of the degree to Table 4.7 Governments Have Withdrawn from More Than Two-Thirds of Privatized Enterprises Government Government ownership Percentage ownership Percentage Level of government before of total after of total equity ownership privatization transactions privatization transactions Majority ownership 2,198 81.7 164 6.1 (50% or more) Minority ownership 232 8.6 278 10.3 (less than 50%) Ownership not 259 9.6 294 10.9 known Zero ownership - 1,953 72.6 Total 2,689 100 2,689 100 Note: Numbers may not total 100 because of rounding to the nearest decimal place. Source: Privatization agencies, government ministries, and World Bank data. THE RECORD TO DATE 77 Figure 4.4 Percentage of Transactions That Reduced Government Ownership to Zero, as of End 1996 Zambia Togo Benin Nigeria Uganda Ghana Mozambique Madagascar Kenya Burkina Faso 0 10 20 30 40 50 60 70 80 90 100 Source: Privatization agencies, government ministries, and World Bank data. which the government is willing to relinquish its involvement in a particular sector. Figure 4.4 shows, for the ten case study countries, the extent to which governments have been willing to relinquish control to the private sector. In Zambia, Togo, and Benin the government was willing to reduce its ownership share to zero in more than 85 percent of the transactions reviewed. In Zambia the total percentage of firms in which the government exited fully was an impressive 98 percent. At the other end of the spectrum is Burkina Faso, where, in most cases, the government did not divest fully, preferring instead to retain an equity interest in more than 60 percent of the enterprises reviewed. Appendix C provides an analysis of the ownership changes in the ten countries studied. 5 The Impact of Privatization Despite an impressive level of privatization activity across Africa, D the economic and financial impact of privatization appear so far to have been minimal. We should emphasize that the data and analysis presented in this chapter provide a picture of the impact of privatization only up to the end of 1996. When data for 1997 become available, we expect to see a significantly increased impact as more major enterprises are privatized. We note, for example, that at least six African telecom- munication enterprises went to the market in 1996 and that privatiza- tion of Zambia Consolidated Copper Mines, one of the largest enter- prises in Africa, is expected to be completed by 1998. On the basis of available data, this chapter examines the impact of privatization from several different perspectives: the impact on govern- ment financial flows, the changes in enterprise-level performance, the level of foreign investment, and employment. (The impact on broaden- ing of ownership was discussed in chapter 2.) Information on enter- prise-level performance changes was gathered from firm-level inter- views and privatization agencies during fieldwork in 1995 for the ten country case studies. In addition, data on employment were subse- quently gathered in Benin, Ghana, and Zambia. Impact on Government Financial Flows In theory, privatization should improve a government's financial flows by raising one-off revenue from the sales of assets and shares, by reducing the need for operating subsidies and investment capital (which become the responsibility of the new owners or managers), and by increasing tax revenues as a result of improved enterprise per- formance. At the time of this writing, it is not possible to assess the impact of privatization on government financial flows, for several reasons. First, information on relevant financial flows prior to privatization is unavail- able. Second, in several countries the amounts of cash actually received by governments from the sale of assets and shares have not been 78 THE IMPACT OF PRIVATIZATION 79 reported. Third, some-indeed, we believe a high proportion-of the proceeds from privatization have been used by holding companies and privatization agencies to meet enterprise severance payments, end-of- service benefits and debts, and their own current expenditure. Not only is there a dearth of data on financial flows, there is also a general misunderstanding about enterprise debts. Many public enter- prises carry excess debt, much of it on-lent by or guaranteed by their respective governments or owed to state-owned banks. In the case of an insolvent enterprise of little economic importance, the assets of that enterprise are sold (that is, the entity is wound up), and the government receives the sale value but remains with the full debt liability. In the case of an important enterprise, a government normally prefers to keep the business going and therefore assumes responsibility for excess debt. In both cases the government would, without privatization, have been liable for the excess debt in any event. Hence, the only effects privatiza- tion has on government financial flows are those listed in the previous paragraph."0 However, when an investor buys an enterprise with part or all of its debt, which the government would otherwise have had to meet, the debt relief that the government gains through privatization is reflected in a lower price paid by the investor. Determining how much revenue governments have raised through privatization is difficult. Few privatization agencies maintain system- atic records of proceeds, and deals invariably involve payments in installments, which are difficult to track. Installment payments are particularly common in larger deals, where final payments are contin- gent on completion of due diligence work (and occasionally on addi- tional contract conditions), and for smaller deals where deferred terms have been negotiated with local investors. In Burkina Faso, one-third of the total sale price remains unpaid because of credit purchases. As box 5.1 shows, the term proceeds is used in vastly different ways. In some cases, proceeds refer to the cash received by the government. In other cases, proceeds include amounts of capital committed by the investor for rehabilitation and expansion of the privatized enterprise. Cross-country comparisons of reported proceeds can thus be mis- leading. To ensure consistency across countries and to avoid the definitional problems raised by using proceeds figures, we use the sale values of transactions to gauge the financial impact of privatization transactions. Although the use of sale values overstates the amounts that govern- ments actually received, it enables us to make more meaningful cross- country comparisons. Because of their size, or because a hard budget constraint was intro- duced, most enterprises divested to the end of 1995 were not a drain 80 PRIVATIZATION IN AFRICA Box 5.1 The Many Meanings of Privatization Proceeds There is considerable laxity in the use of the term proceeds. Research for the case studies revealed that the term is used in various, sometimes confusing, ways by the different privatization agencies, which use the term to mean any one of the following: 1. Amounts paid to the government for the purchase of assets or shares, excluding amounts paid to government-owned holding companies and parent state-owned enterprises for sales of assets or shares 2. Amounts paid to the government for the purchase of assets or shares and amounts paid to government-owned holding companies and par- ent state-owned enterprises for sales of assets or shares 3. Amounts as described in (1) or (2), plus the amount of capital commit- ted by the investor for rehabilitation and expansion in the enter- prises concerned 4. Amounts as described in (3), plus the amount of enterprise liabilities assumed by the investor. The broader definitions are misleading. An investor may pay little or nothing to the government but may instead invest substantially in a com- pany by way of a capital investment that makes cash available to the busi- ness. If the broader definition is used, the impression is mistakenly given that the government or government-owned entities have received all of the funds. Lack of systematic recording and reporting also contributes to the problem of disaggregating the gross values of transactions from the net cash proceeds received by governments or other public shareholders. on government financial resources (at least in the period immediately prior to privatization). While their privatization may have resulted in a one-time injection of cash and, in some cases, may have yielded some increases in tax revenues through improved results, such flows will not have had a significant effect compared with the fiscal deficits caused by the large loss-making parastatals needing government support. Comparisons of transaction sale values with deficits are therefore not informative. To give an idea of the potential financial impact of privatization (prior to consideration of how proceeds have been used), it is more useful to look at reported total sales values as a percentage of govern- ment revenue (excluding grants) and GDP over the period from the inception of the privatization program to end 1996, with some excep- tions as noted below. When looked at in terms of government revenue, privatization has been of most significance in Zambia, Mozambique THE IMPACT OF PRIVATIZATION 81 and Uganda (table 5.1). In terms of GDP, Mozambique, Zambia, Cape Verde, and Ghana remain in the top five countries (table 5.2). It should be noted that the Ashanti Goldfields transactions contrib- uted substantially to Ghana's ratings. Were they excluded, privatization sales in Ghana would have been equivalent to only 1.3 percent of government revenue and 1.7 percent of GDP over the period of the program to the end of 1996. The above comparisons show that privatization has so far had no significant financial impact in most African countries. This is further underlined when total sales values are related to population (table 5.3). One country, Cape Verde, scores highly; but given that privatization has been underway for some years, the cumulative average of only US$5 per capita across Africa is very low indeed. Table 5.1 Privatization Sales Values as a Percent of Government Revenue Average annual sales value as percent of average annual government revenue over period of privatization program Zambia 6.8 Mozambique 6.4 Uganda 6.3 Cape Verde 5.8 Ghana 5.5 Tanzania 4.3 Ethiopia 3.3 Benin 3.0 Malawi 2.2 Kenya 1.9 Mali 1.7 Gambia, The 1.2 C6te d'Ivoire 1.1 Togo 1.1 Notes: Government revenue figures used for this table excluded grants. Countries with below I percent were Senegal, South Africa, Nigeria, Madagascar, Guinea-Bissau, Burkina Faso, Burundi, Zimbabwe, Guinea, Niger, and Comoros. No privatization sales value data were available for other countries. For some listed countries, data on privatization sales values were incomplete. For each country, the percent relates to the period from the year of the first reported complete privatization transaction to end 1996, except for Benin (1986-95), Ethiopia (1995), The Gambia (1987-94), and Mali (1988-92). For The Gambia, 1991 government revenue figure assumed for 1992-94. For Malawi, 1995 government revenue figure as- sumed for 1996. Source: World Bank data. 82 PRIVATIZATION IN AFRICA Table 5.2 Privatization Sales Values as a Percent of GDP Total sales values to end 1996 as percent of: Anntual average GDP over period Country of privatization GDP for 1996 Mozambique 12.8 10.5 Ghana 7.0 6.6 Cape Verde 6.3 5.7 Zambia 4.5 4.2 Benin 3.7 2.9 Gambia, The 3.4 2.7 Malawi 3.3 3.8 Tanzania 3.2 3.0 Uganda 3.2 2.3 Togo 2.9 2.5 Kenya 2.3 2.0 Notes: Countries not shown in the table either had total reported sales values equivalent to less than 2 percent or sales value data were not available to make the comparison. [For The Gambia, Malawi, and Tanzania GDP figures are for 1995.1 For some listed countries, data on privatization sales values were incomplete. For each country, the percent relates to the period from the year of the first reported completed privatization transaction to end 1996, except for Benin (1986-95), Ethiopia (1995), The Gambia (1987-94). Source: World Bank data. Table 5.3 Total Cumulative Privatization Sales per Capita Country US$ Cape Verde 60.3 Ghana 25.4 Zambia 19.7 South Africa 19.2 Benin 12.5 Mozambique 11.9 C6te d'Tvoire 10.6 Gambia, The 9.9 Togo 9.2 Senegal 8.2 Source: World Bank data. THE IMPACT OF PRIVATIZATION 83 Not All Sales Proceeds Go to the Government Monies from sales do not always accrue to the government. De- pending on the method of privatization, proceeds may remain at the firm level. Holding companies tend to retain proceeds from the sale of their investments. In Kenya, for example, until the sale of 26 percent of Kenya Airways to KLM in early 1996, less than one-half of the total reported proceeds from privatization had gone to the government; most proceeds accrued to government-controlled holding companies (the legal shareholders selling their shares) and enterprises that issued new shares (a profitable supermarket and a financial institution were privatized through share dilutions). In some cases of restitution of assets to former owners, transactions included compensation for the earlier nationalization of the enterprise. Uses of Sales Proceeds Data on uses of sales proceeds have so far been difficult to obtain, but information collected during fieldwork shows that they are used for a wide variety of purposes (table 5.4). In Benin and Zambia, the use of privatization proceeds is mandated by law. In Zambia, proceeds are deposited in the Privatization Revenue Table 5.4 Proceeds from Privatization Transactions Are Used in a Variety of Ways Employees' Enterprise Privatization end-of- non- agency and Government Investment service government transaction central in public Country benefits creditors costs budget enterprises Benin v' Burkina Faso vK Ghana P o / Kenya - - / Madagascar Mozambique Nigeria Togo Uganda / vfv Zambia o P PO Source: Government policy statements, privatization agencies, and World Bank data. 84 PRIVATIZATION IN AFRICA Account under the control of the minister of finance. The Privatization Act permits the funds to be used for a variety of purposes, including expenditures in connection with the privatization process, funding of retrenchment costs, investments in companies to be privatized, and general government expenditures. Proceeds from sales of shares trans- ferred to the Privatization Trust Fund are to be remitted to the govern- ment along with any dividends received by the fund before the sale. In Kenya, proceeds from the sale of investments held by the development finance institutions (DFIs) are to be held at the Central Bank in special accounts, one for each institution, with the Treasury as countersigna- tory. Proceeds are to be used for enterprise debt repayment, DFI debt repayment, staff redundancy payments, and new investments by DFIs. In Uganda, proceeds in the Divestiture Account had been plowed back into the parastatal sector but a much tighter control over the use of proceeds was introduced in 1995 so that management and uses of proceeds are now in accordance with the legislation of Public Enterprise Reform and Divestiture Statute No. 9 (1993). Some countries, including Tanzania and Zambia, have general poli- cies on the use of privatization proceeds that give prominence to pay- ment of severance or end-of-service entitlements for retrenched em- ployees and to settlement of enterprise debts. In Nigeria, the government has stated that a substantial portion of privatization pro- ceeds will be used to support investment programs of major public enterprises. As a result, 50 percent of net proceeds has been allocated to ailing public enterprises, such as Nigerian Railways. In the latter case, taking a one-time gain and putting it into a perennial loser appears to be tantamount to frittering away a benefit of privatization. It should be noted that the privatization agency recommended that 10 percent of the proceeds be used to assist ailing enterprises and the majority of the re- maining proceeds be used to finance other public sector activities."1 In countries where the use of proceeds is neither mandated by law nor subject to a published government policy, proceeds are generally applied against privatization costs. In Ghana, proceeds are handled by the privatization agency, which uses them to pay enterprise debts and retrenchment costs. In Mozambique, proceeds are first used to pay debts, retrenchment costs, and costs related to privatization; any re- maining proceeds are put in a fund to promote entrepreneurship. Given the lack of information on this subject, it is not surprising that there is already public concern about who decides how proceeds are used and exactly how cash proceeds from privatization have been applied. THE IMPACT OF PRIVATIZATION 85 Enterprise-Level Performance Available data indicate that firms are indeed performing better after privatization. A few caveats about the data are in order, however. Information on firm performance was obtained from mail question- naires, on-site visits, and interviews; but no attempt was made to adjust for the effects of economic liberalization measures that often accompanied or preceded privatization. Some of the firms interviewed were reluctant to reveal financial information. Even when firms were willing to divulge information on post-privatization performance, they were often unable to provide reliable pre-privatization data for com- parison. The problem posed by the reluctance of firms to reveal financial and operating data was compounded by the virtual absence of post- privatization monitoring by the privatization agencies. In Benin, firms are required to submit financial statements to the privatization agency for five years after privatization, but this requirement has not been enforced. Although agencies in Tanzania, Uganda, and Zambia intend to commence monitoring shortly, only the privatization agency in Nige- ria was able to provide some information on the post-privatization performance of firms. Moreover, at the time of the fieldwork for the case studies, the recentness of privatization transactions (less than 40 percent of the 1,430 transactions covered in the case studies took place prior to 1994) meant that in many cases it was too early to gauge the lasting effect of privatization on enterprise performance. No information is available on businesses that are using assets ac- quired through formal liquidations or other forms of asset sales, all of which represent small transactions. Information on the post-privatiza- tion performance of enterprises privatized through share sales and other methods is scant. However, some of the larger companies were forthcoming with information, and here the news is good: the case studies reveal that the transfer of ownership control to private investors has provided the confidence for investment in many of the privatized businesses, some of which had long been starved of investment. Post-privatization investment, notably in enterprises purchased by foreign investors, tends to be greater than the amount paid to purchase the enterprise. This commitment to invest in the enterprise after it has been privatized has become a major criterion in competitive bidding for the selection of investors in major industrial and agricultural enter- prises. Although systematic data on post-privatization investment are lacking, interviews with firms suggest that the investment commit- 86 PRIVATIZATION IN AFRICA ments are being kept. Table 5.5 gives some examples from Benin, Mo- zambique, and Uganda. In many cases investment in privatized companies has resulted in increased capacity utilization, expanded capacity, the introduction of new technology, product diversification, and expanded markets. These changes-coupled, in some cases, with new and better management- have contributed to increased turnover, reduced unit operating costs, Table 5.5 Post-Privatization Investment Often Exceeds Sales Values of Enterprises Transaction Additional sales value commitment Country Privatized ($ millions) ($ millions) Ratio and year enterprise Investor (1) (2) (2/1) Benin 1990 Manucia Rothman's 5.4 5.5 1.0 International (UK) 1992 La Beninoise Castel BGT (France) 25.7 7.6 0.3 1991 Sonaci Scancem (Norway) 15.1 3.5 0.2 Mozambique 1995 Cimentos de Cimpor, SA (Portugal) 20.0 76.4 3.8 Mozambique 1995 Companhua Namib Mills (Pty) Ltd, 7.3 6.4 0.9 Industria (South Africa) Matola 1995 Fabrica de Indol BV 5.0 4.2 0.8 Cerjas da (Netherlands) Beira 1995 Fabrica Cade Indol BV 9.0 4.2 0.5 Cervejas (Netherlands) Uganda 1992 East African International 0.7 1.5 2.1 Distilleries Distillers & Vintners Ltd. Ltd. (UK) 1992 Lake Victoria Local investors 3.9 8.0 2.0 Bottling Co Ltd. 1992 Shell (Uganda) Shell Petroleum 7.25 13.0 1.8 Ltd. (Netherlands) 1993 Agriculture Commonwealth 9.9 11.8 1.2 Enterprises Development Ltd. Corporation/Finlay (UK) 1994 Hima Cement BV Rawals Group of 20.5 40.0 2.0 Industries (India) 1995 Nile Hotel Tahar Fourati et al 19.0 26.9 1.4 (Tunisia) a. Not the sale value but the value of the assets contributed by the government to the new joint venture. Souirce: World Bank country case studies. THE IMPACT OF PRIVATIZATION 87 and improved financial performance. Not surprisingly, many of the managers interviewed cited the removal of government influence in the management of the firm as one of the positive effects of privatization. For firms privatized through public flotations, the evidence suggests that privatization has had a significant positive impact on the firms' financial performance. This is particularly true in Nigeria, where large nonfinancial enterprises such as National Oil, African Petroleum, and AIICO have reported robust profits since privatization. The share prices of privatized firms in Nigeria (table 5.6) have been stable or have increased steadily, indicating that financial performance is acceptable to investors. In Kenya, profits of the Housing Finance Company, privat- ized in 1992, rose by more than 100 percent between 1992 and 1993. Anecdotal evidence suggests that privatization has enabled firms to diversify product lines and upgrade and rehabilitate facilities. In Uganda, Shell International reported that privatization allowed the firm to concentrate on core activities and to contract out noncore activities such as canteen services and gardening. Investment far exceeded the re- quirement under the privatization deal to invest $10 million over three years: in just two years $13 million was invested in new and rehabilitated filling stations. In Madagascar, a firm engaged in the production of furni- ture discontinued its activities and began importing furniture and other home furnishings. In Benin, the Societe Nationale d'Equipement, privat- ized in 1988, diversified and now owns a supermarket and a library. In Ghana, West African Mills rehabilitated and modernized its plant and Table 5.6 Share Prices in Nigeria Reflect Improved Enterprise Performance (share price in Kobo, as of October 1995) Name of enterprise Price at Lowest offer, Highest offer, and date of sale privatization October 1995 October 1995 Aba Textile, 1990 75 358 485 African Petroleum, 1989 190 596 1870 AIICO, 1989 165 350 565 Ashaka Cement, 1990 120 363 1015 Benue Cement, 1990 90 213 310 Cement Co of Northern 100 230 331 Nigeria, 1992 Crusader, 1989 110 120 255 Impresit, 1992 70 82 122 National Oil, 1989 200 725 1890 Sun Insurance, 1989 125 122 177 Unipetrol, 1991 200 475 2104 Source: World Bank, 1996h. 88 PRIVATIZATION IN AFRICA machinery and now exports food products. The Tema Steel Company, which had been closed down prior to privatization, has undergone a major rehabilitation program. The Western Veneer and Lumber Com- pany, which has been only partially privatized, is successfully marketing secondary species and has extended its processing activities. Management changes and new investments associated with priva- tization have resulted in improvements in performance. In Ghana, the acquisition of a majority stake by Lever International in Unilever Ghana Limited, which increased its shareholding from 45 to 70 percent, brought about significant changes in the company. The increased shareholding gave the investor the necessary flexibility on matters relating to strategic direction, capital investment, and mergers and acquisitions, and significant capital investment was made, enabling the company to reduce unit costs. The company has reported a 50 percent increase in production without increasing energy consumption. SCANCEM, the majority owner of the Ghana Cement Company, re- ported that additional investment after privatization has led to signifi- cantly increased production output and increased profits. The Tropical Glass Factory, privatized in 1990, is now one of the leading producers of beer bottles in West Africa. Privatization has also brought about changes in management and labor practices that have led to improved production and financial performance (boxes 5.2 and 5.3). Some privatized firms have shown a dramatic increase in profitability. In Kenya, two firms that were in receivership reported increases in profits after privatization as well as increases in employment. Also in Kenya, the Uchumi supermarket chain saw sales triple between 1992 and 1994; reported profits increased 107 percent between 1992 and 1993 and 42 percent between 1993 and 1994. Mount Kenya Textile Mills, which had accumulated large losses in the period 1987 to 1992, managed to break even in 1993 following privatization (though the firm continues to enjoy a government subsidy through exemption from government sales tax.) Since privatization, the Premier Bags Company has modernized its facili- ties, recruited additional employees, and shown a modest increase in profits. At Unilever Ghana, turnover nearly doubled between 1990 and 1993, and gross profit rose almost fourfold; even discounting for infla- tion, the results are impressive. In Benin, profits at SONACI, privatized in 1992, have risen and production has increased by 50 percent. Not all firms surveyed did well. The small sample of firms surveyed indicates that privatization has resulted in the closure of several of the privatized firms. Examples include a Togolese firm, privatized in 1987 (one of the earliest privatizations in the study sample), which was unable to hold its own against domestic and foreign competition and THE IMPACT OF PRIVATIZATION 89 Box 5.2 Privatization Has Brought Changes in Management Practices Privatization has brought about changes in corporate culture in many of the sampled companies. CIMBENIN, the cement company in Benin, privat- ized in 1991, is a case in point. At the time of privatization, the new investor, SCANCEM, brought in three appointees to take over key managerial posi- tions. Substantial privatization related changes in management practices include: * Quicker decisions at all management levels; the managing director is no longer required to consult with the Ministry of Industry for decisions as simple as ordering spare parts. * Recruitment and dismissal of personnel are no longer subject to politi- cal interference. * Efficient stock management and budget monitoring systems have been introduced in each department. * A regular schedule for the maintenance of equipment and machinery, now in place, has eliminated production stoppages. Source: London Economics, 1996. was put into receivership in 1993. Causes of failure included overesti- mation of the market, insufficient working capital, and difficulty in raising additional capital. Even some firms with monopoly or dominant positions have failed to improve performance. Import liberalization has put competitive pressures on some privatized firms, notably in the textile sector. Sobetex, a Beninese textile manufacturer privatized in 1990, initially increased employment by 11 percent but today cannot compete with, cheaper Nigerian textiles. Closure of enterprises after privatization is not necessarily undesir- able. If an enterprise is inherently nonviable-and this only becomes clear when other concurrent policy reforms remove distortions- privatization, by facilitating closure, eliminates the economic cost of continuing with a nonviable business. Privatization has occurred at a time of major economic reforms, including trade liberalization and removal of some monopoly rights. For privatized enterprises it is sometimes difficult to disaggregate the effects of these reforms from the effect of privatization. In addition, the devaluation of the CFA franc in 1994, combined, in some cases, with political upheavals, makes changes in enterprise performance resulting from privatization even more difficult to disaggregate. A major cottonseed oil company in Togo, privatized in 1987, was ad- 90 PRIVATIZATION IN AFRICA Box 5.3 Changes in Labor Practices: Evidence from Benin A sample of eight firms in Benin showed significant post-privatization improvements in labor practices and conditions, and better and more fre- quent training opportunities. * In SONAEC, a car servicing company, specialized training in automo- tive engineering and maintenance was introduced after privatization. * In Societe Beninoise de Textiles, a flexible workday schedule was adopted and an employee-of-the-month bonus scheme was introduced. * Rothmans, which bought Societe B6ninoise de Tabac et Allumettes, introduced a training exchange program. * In the brewery, SOBEBRA, working conditions, including hygiene, are significantly better after privatization. * In FACS, the airport catering services company, an intensive profes- sional training program was initiated to enable staff to meet the trade's high performance requirements. There are now job descriptions for each post, and theft has virtually been eliminated. Source: London Economics, 1996. versely affected by the devaluation because its debt was denominated in foreign currency; it has subsequently gone under. In Benin, La Beninoise, the brewery privatized in 1992, reported large losses as a result of debt denominated in dollars, an inability to pay for imported goods, and a pricing policy hamstrung by government price control. In Burkina Faso, firms reported being adversely affected by increased import duties and the higher cost of raw materials. In Togo, firms were adversely influenced by political turmoil in the 1992-93 period. The power sector in C6te d'Ivoire and the urban water supply in Guinea offer important examples of the few major public utilities whose management has been privatized through lease or concession arrange- ments. Although operational performance has improved in both cases, lack of clarity with regard to some of the contractual obligations or their enforcement has restricted the gains that were expected (boxes 5.4 and 5.5). A Significant Reduction in the Number of State-Owned Enterprises Across Africa, as well as in the ten case study countries, the number of state-owned enterprises (SOEs) had by end 1995 fallen by more than THE IMPACT OF PRIVATIZATION 91 Box 5.4 Cote d'Ivoire's Mixed Experience with the Power Sector In March 1990, the government invited two French companies, the Bouygues Group (already involved in water distribution in Abidjan) and EDF (Electricite de France), to take over the operation of the power sector in the country. A concession contract was signed on October 25,1990. On Novem- ber 1, the newly formed Compagnie Ivoirienne d'Electricite (CIE) took over operation of the generation, transmission, and distribution system, while EECI, the former public concession holder, retained responsibility for the sector's public infrastructure and for overseeing the application of the con- cession. This arrangement thus separated operations from the ownership and investment functions. In April 1994, a National Electricity Fund (FNEE) was established to ensure balanced financial management of power sector resources. The FNEE services EECI's debt, makes payments for maintenance and extension work, and oversees the regular payment by CIE of its fees to the govemment. The FNEE, which is under the technical oversight of the Ministry of Energy and the financial oversight of the Ministry of Finance, also plays a part in electricity rate setting. The EECI's role is now to oversee the CIE concession and serve as an engineering consultant. In this role, it plans future invest- ment, acts as works supervisor on behalf of the government, and is responsi- ble for major equipment overhaul and sector development projects. CIE imports and exports power and is responsible for customer services, daily operations, and upkeep of facilities and infrastructure. Two major advantages were claimed for the Ivorian model. First, the government retains ownership and responsibility for infrastructure. Second, daily operational control rests with an operator whose mission is to improve the quality of service and commercial management. In the first year, the new firm succeeded in earning an operating profit of US$2.5 million, and the quality of service did increase considerably, particularly in terms of reliability of supply. But the arrangement is not without its problems. There is a lack of clarity regarding responsibility for major maintenance and investment in the distribution system and for tariff setting. Actions to launch major rehabil- itation programs are EECI's, not CIE's, responsibility, so that failure on EECI's part can hamper CIE's daily functioning. The CIE does not control any investment, including connection of new users. There are also funda- mental issues relating to market structure, institutional capacity and the regulatory framework. These issues, which are fundamental to the efficient running of the sector, are being addressed through ongoing World Bank op- erations. Source: World Bank, 19961. 92 PRIVATIZATION IN AFRICA Box 5.5 In Guinea, Improved Urban Water Supply but Fewer Gains Than Anticipated Until the late 1980s, Guinea had one of the least developed urban water supply sectors in West Africa. Less than 40 percent of urban dwellers had access to piped water through either connections or standpipes. Where connections existed, service was often interrupted, and water treatment inadequate. In 1989, the government entered into a lease arrangement for private sector oper- ation of water services in the capital city and 16 other towns. Two organizations are central to the lease arrangement: the state-owned national water authority, Societe Nationale des Eaux de Guinee (SONEG), and a water management company (Societe d'Exploitation des Eaux de Guinee, SEEG). SONEG owns the water supply facilities in the cities and towns covered by the lease and is responsible for new investments and for tariff setting. SEEG, jointly owned by the state and a foreign private consortium, operates and maintains facilities and bills users. At the start of the lease the consumer tariff was raised from US$0.12 to US$0.25 per cubic meter, but this was still too low to cover operating and debt service costs. So, in the initial years of the lease, the difference between tariff revenues and costs was funded by an IDA credit. During the contract's first five years there was a big increase in the population that had access to safe water, from about 15 percent in 1989 to 52 percent in 1994. This was achieved through investments in new supply capacity (external to the lease), combined with rehabilitation and mainte- nance. Connections increased from 12,000 to 30,000, and metering increased from about 5 to 95 percent of all connections. During the same period SEEG's revenues rose almost tenfold. Using a lease arrangement, rather than a full-fledged concession or asset sale, meant that the government did not attempt to sell a major strategic enterprise and that private investors were not required to commit to major capital investment. Using an IDA credit to smooth the differences in tariff revenues and costs meant that the operating business could function on a quasi-commercial basis from the beginning. The expected benefits were twofold: early and lasting gains in the availability of services and the effi- ciency of service delivery and, in the medium term, an improvement in the business environment that would make private sector investment and risk taking in Guinea more attractive. Despite these gains, there are concerns about performance. Over the past few years the water supply system has not improved and expanded as fast as had been hoped, and unaccounted-for water remains high, at 47 percent. Also, the relationship between SONEG and SEEG has not been smooth, with risk sharing between the parties proving difficult to implement and enforce. Source: Brook Cowen, 1996, pp 89-92. THE IMPACT OF PRIVATIZATION 93 one-third as a result of privatization (table 5.7). Of course, this decline does not reflect the size and relative importance of the enterprises privatized, nor does it reflect the different stages of the privatization programs. Nonetheless, it does represent a decisive step toward govern- ment withdrawal from commercial activities. Employment While it seems that no one doubts that privatization will bring about improvements in efficiency, there is widespread concern that it will have a negative effect on employment. The country case studies show that governments have taken rather different policy approaches to employment in privatization, reflecting the political capacity to absorb the consequences of retrenchment. In Zambia, where the government depends heavily on an urban electorate, liberalization and privatization have been controversial. Accordingly, the government required the privatization agency to try to negotiate employment retention condi- tions in privatization transactions and also legislated improved em- ployee end-of-service benefits. In Ghana, where the government de- pends more on a rural electorate that has benefited from liberalization of Table 5.7 Privatization Has Reduced the Number of State-Owned Enterprises by One-Third Estimated Estimated number of number of SOEs SOEs Percentage Country 1990 1995 reduction Benin 60 25 58 Burkina Faso 71 57 20 Ghana 329 277 31 Kenya 255 146 43 Madagascar 184 114 38 Mozambique 1,200 690 43 Nigeria 485 404 17 Togo 50 28 44 Uganda 159 137 14 Zambia 160 92 35 Total, 10 countries 2,953 1,920 35 All other countries of Sub-Saharan Africa 3,116 2,138 31 Total Sub-Saharan Africa 6,069 4,058 33 Source: World Bank data. 94 PRIVATIZATION IN AFRICA agricultural markets, the government has set few employment-related conditions on privatization. Despite the concern about possible job losses, the case studies showed that African governments have done very little to track the effect of privatization on employment. This reinforces the need to im- prove the gathering and dissemination of employment data generally throughout the continent. In view of the dearth of employment data, a special review was undertaken for this study on the impact of privatization on labor in Benin, Ghana, and Zambia."2 Data collected during that review and from the Burkina Faso and Togo case studies are set out in appendix D. Table 5.8, which summarizes the firm-level data in appendix D, shows that, overall, employment has declined by an average of almost 15 percent across the five countries. The limited available data do not tell the whole story. In some cases downsizing took place prior to privatization. With the application of a hard budget constraint, the day of reckoning on the financial side has come to many public enterprises in Africa, and this has led to downsizing regardless of privatization. Layoffs, however, have not occurred in some enterprises because bidders were required to maintain existing employment levels or, at the very least, were encouraged to do so by selection criteria that favored bidders prepared to retain existing Table 5.8 job Losses Following Privatization Number of Employ- privatized Employment ment, Net Country firms in at first quarter change in Percentage and period sample privatization 1996 employment change Benin, 8 1,872 1,197 -675 -36.0 1988-95 Burkina Faso, 10 895 901 +6 +0.1 1991-95 Ghana, 7 5,363 4,431 -932 -17.3 1990-95 Togo, 1984-95 19 2,882 2,338 -544 -18.8 Zambia, 10 6,150 5,733 -417 -6.7 1993-96 Total 54 17,162 14,600 -2,562 -14.9 Note: This table is a compilation of data from a representative sample of firms surveyed in each of the listed countries. See appendix D for details on individual firms. Source: Data on Benin, Ghana, and Zambia are from London Economics, 1996. Data for Burkina Faso and Togo are from World Bank, 1996c and World Bank, 1996i, respectively. THE IMPACT OF PRIVATIZATION 95 employees. In Mozambique, one of the criteria for evaluating bids is the bidder's commitment to maintaining the labor force at the existing level, although, in practice, the commitment has not always been kept. In Burkina Faso, employment conditions are included in the sales con- tracts, and buyers are requested to maintain employment at particular levels. Both Benin and Zambia have used employment guarantees in an attempt to contain post-privatization retrenchment. These guarantees have not always been easy to enforce. In Zambia, several companies retrenched workers despite the guarantee. By contrast, companies in Benin wanting to downsize their workforce have been prevented from doing so by a "five-year no lay-off" clause standard in privatization con- tracts. Moreover, the figures in table 5.4 are the net total changes in employ- ment in the sample firms. In some enterprises job losses have been higher, while other enterprises have increased their workforces. In some cases employment rose because the company was not operational before privatization, and an increase in the workforce was necessary before production could recommence. In other cases increased produc- tion has led to increased employment. What is clear is that governments have delayed privatizing enter- prises where a high social cost has been expected. More than 80 percent of the enterprises divested in the ten case study countries had fewer than 500 employees. Many of the largest employers, with sizable work- forces that will need to be downsized, remain in government portfolios. In Benin, only 4 enterprises of the 12 enterprises with more than 500 employees were or are being privatized. In Uganda, divested parasta- tals account for less than 5 percent of the total public enterprise la- bor force. Although privatization has not resulted in large layoffs, the general fear of privatization persists. A survey undertaken in Zambia in 1992 found that a majority of those interviewed opposed privatization pri- marily because of fear of job losses. The negative publicity surrounding the liquidation of several major public enterprises in late 1994 and early 1995 convinced the public that privatization is synonymous with job losses. Press coverage on the employment effects of privatization has also negatively influenced public opinion in Kenya. Severance Pay and End-of-Service Benefits When employees are retrenched, the compensation package can comprise severance pay, end-of-service benefits, or both. Severance pay is payable only to employees who lose their jobs as a result of the 96 PRIVATIZATION IN AFRICA downsizing of their employer's workforce. End-of-service benefits are lump-sum benefits payable to employees on retirement. Some retrench- ments may be treated as early retirement. Typically, there are statutory legal minimum levels set for both end- of-service benefits and severance pay. However, in many countries these levels are considered to be too low, especially where the levels do not take into account inflation or where employment packages include a high level of nonmonetary benefits. Hence collective bar- gaining agreements may specify more generous levels. Where collective agreements do not exist or benefits are still regarded as low, the amount of severance pay is often a matter of negotiation between the employees and whoever is handling the labor affairs of the divested business. The case studies revealed that no country has a uniform policy on retrenchment. As a result, some workers are entitled to the end-of-service benefits or severance packages determined earlier through collective bargaining agreements, others have to negotiate case by case when an asset sale or downsizing occurs, while yet others receive only the legally mandated benefits, which are often very low. Moreover, governments, notably in some Francophone countries, have at times been unable to afford the packages that have been negotiated with unions. The means by which an enterprise is privatized also affects the amount and timing of these types of benefits. When an enterprise is privatized through a sale of shares, there is no break in the continuity of employment. While the contingent liabilities remain with the business, there is no immediate outlay of cash required to meet severance pay or end-of-service benefits. When an enterprise is privatized through a sale of assets, its revenue-earning capacity is removed, so that all employees have to be dismissed. They then become immediately enti- tled to severance pay, whether or not they are reemployed in the same capacities by the buyer of the assets. Two additional issues may also arise: retrenched workers may be entitled to claim immediate payment of part of their end-of-service benefits, and funds may be insufficient to meet immediate or future end-of-service benefits. Hence a sale of assets-which typically is resorted to because the enterprise is insolvent and cannot be sold as a going concern-frequently results in the govern- ment having to meet the liabilities for severance pay and end-of-service benefits. This has been an issue in Benin and Ghana, where the govern- ments had to meet costly severance awards that had to be paid to many workers whose legal employer had changed (by virtue of the sale of assets) but who had not, in effect, lost their jobs. It is difficult to see how this could have been avoided, but the cost to the taxpayer would have been less if the retrenchment process and entitlements necessitated by asset sales could have been redesigned. THE IMPACT OF PRIVATIZATION 97 Despite several years of experience with the problems arising from unclear severance pay entitlement, many countries continue to leave the matter unresolved, and this has an adverse impact on their privatization programs (box 5.6). There is a need for countries to define more clearly the framework for determining and paying severance pay and end-of- service benefits so as to reduce inequities, minimize disputes, smooth industrial relations, and reduce public suspicion of privatization. In Ghana, the level of end-of-service benefits has been a constant Box 5.6 Uncertainty about Severance Pay Fuels Distrust of Privatization Disputes over severance pay have arisen between government and unions mainly where companies were privatized through asset sales rather than as going concerns. In these asset sale cases, there has been a wide variation in the severance awards. In Zambia, no legislative framework has been created to help determine severance awards. In Benin and Ghana, minimum severance awards have been defined, but superior settlements have been awarded in many cases. In Ghana, for example, the level of awards seems to depend on the entitlement to severance pay had the company continued as a going concern, the funds raised from privatization, the ability of negotiators to extract a good deal from the government, and the willingness of the government to be generous in particular cases. In some cases, the government has made ex gratia severance awards some time after the completion of a privatiza- tion transaction. Although the negotiation process for severance pay has been smoother between companies and workers where firms have been privatized as going concerns, there is again a wide range of experiences. In Zambia, the lack of a defined legislative framework has thrust the responsibility for negotiating severance pay onto the Zambia Privatization Agency (ZPA), which is not well equipped for this role. In at least one case it appears that retrenched workers may have had relatively poor settlements imposed on them by new owners. In all three countries this negotiating process has reduced the transpar- ency of severance awards and has thereby contributed to the perception that awards are made on an arbitrary basis. As a consequence, workers and their families are uncertain as to how retrenchment might affect them. This heightened uncertainty simply serves to reinforce popular distrust of the privatization programs. This is particularly the case for nonunionized and less educated workers, who are the most vulnerable. Source: London Economics, 1996. 98 PRIVATIZATION IN AFRICA problem and was a major cause of delay in the privatization program between 1987 and 1990. During that period, as a response to the collapse of the social security system, end-of-service benefits were maintained at extremely high levels-typically eight months of final salary for every completed year of service. In 1990 a cap was put on these benefits, both to limit the extent of the government's liability and to make the benefits uniform. In practice, however, benefits continued to be negotiated on an individual enterprise basis, and the government has now accepted full responsibility for their settlement. In Benin, the amounts of severance pay due to employees have been decided by the courts, and in at least one case (Sobetex) payments to laid- off employees were set at the equivalent of ten years of wages. In Zambia, employees who accepted voluntary redundancy packages reportedly re- ceived payments in line with the conditions of service prevailing in the company at the time of termination, although there are examples of wide discrepancies in redundancy or end-of-service payments which can fuel the public's distrust of privatization (box 5.7). In a sample of 229 re- trenched workers, roughly 10 percent left voluntarily under priva- tizations in which a retrenchment package was offered, and about 77 percent received some cash end-of-service benefits, the median value of which was equivalent to 40 months' salary (London Economics 1996). Box 5.7 Disparity in End-of-Service Benefits Fuels Distrust of Privatization The United Bus Company of Zambia was insolvent and was forced into liquidation. Under the applicable bankruptcy law employees were entitled to 200 kwacha (equivalent to $0.30) as preferred creditors for the legal minimum end-of-service benefit; for any additional amount, they ranked as ordinary creditors in accordance with their contract terms of employment. In contrast, a cleaning worker who left Zambia Telecommunications after working there for only two years reportedly received a payment of 2 million kwacha ($3,000). The legal minimum end-of-service benefit has recently been raised to the equivalent of three months' salary. In practice, however, most public enterprise terms of service are reported to provide for approximately 4.5 times annual salary. Such egregious disparities between the legal and the contractual minimums for end-of-service benefits can create opposition to privatization and delay transactions, as governments and firms struggle to establish and minimize the extent of their contingent liabilities. Source: World Bank, 1996k. THE IMPACT OF PRIVATIZATION 99 In Uganda, many companies have their own rules on termination pay. These rules vary considerably, with some companies including standard allowances received with pay as a basis for calculating termi- nal benefits and others ignoring allowances but providing formulas that compensate for them. In Burkina Faso, the labor code is flexible with respect to severance pay, and tribunals have approved generous redundancy packages of up to six years of wages. The Social Impact of Privatization Mitigating the social impact of privatization has not been a part of the government's agenda in most countries. Doing so may become increasingly necessary as larger enterprises are prepared for privatiza- tion. Only a few attempts to design and implement redeployment schemes have been made (in Benin, Cape Verde, Guinea, Madagascar, Senegal, and Zambia). In anticipation of redundancies, training in job search and business development has been offered to more than 5,000 people in Zambia. In Benin, two training schemes-a project to develop microenterprises and one to develop SMEs-were set up, but their impact has been minimal, with less than 1 percent of retrenched staff benefiting from these programs."3 At the instigation of donors, the requirements for social safety nets and redeployment training and assis- tance were studied in Kenya and Tanzania, but no action has been taken by the governments to introduce social safety nets. The review of the impact of privatization on labor in Benin, Ghana, and Zambia confirmed that retrenchment has had an adverse effect on household welfare; retrenched workers are more likely to be poorer than before privatization. However, they are still much less likely to be below the poverty line of the general population (London Economics 1996). Not surprisingly, the review also found that the scale of severance payments and their use is a significant determinant of the incidence of poverty. This again highlights the need for governments to have a clear policy on severance pay. Foreign Investment Privatization is attracting foreign direct investment (FDI), although the amount to the end of 1995 was small relative to the total FDI (figure 5.1). As figure 5.1 shows, there was only a modest rise in the FDI from privatization in the period to 1993. In 1994, privatization-related FDI reached almost $500 million. In all enterprises surveyed as part of the 100 PRIVATIZATION IN AFRICA Figure 5.1 Foreign Direct Investment Has Risen as a Result of Privatization 3.000 25600 2,4 2,187 2.000 ~,1,500 1,000 1.022 5000 167.6 19S9 1990 1991 1992 1993 1994 199& -Total net FDI, Sub-Saharan Africa ~Total net FDI, ten case study countries -.-Total value of FDI from privatization, ten countries a. Estimated. Source: World Bank data. country case studies, foreign investors have introduced capital into the enterprises in addition to the capital required to acquire assets or shares. Although the increase in transactions is resulting in an increase in foreign investment, privatization-related flows as a percentage of total FDI flows are still low compared with other regions of the world. A 1995 (United Nations 1995) study found that privatization-related flows represented less than 5 percent of all foreign investment in the region, compared with 43 percent in the transition economies of Eastern Europe and the CIS, and 15 percent in Latin America and the Caribbean. What explains the relatively low levels of foreign investment? For- eign participation in privatization has been deterred by the small size of the privatized enterprises, the lack of transparency of transactions, and the preference for local investors. Investor promotion efforts have also been weak. In most cases privatization programs were not designed to attract foreign investors, and often there has been little coordination between the privatization agency and the investment promotion agency. Foreign investment has played a negligible role in the privatization THE IMPACT OF PRIVATIZATION 101 of small enterprises but an important role in the case of larger enter- prises. More than 50 percent of the large transactions in the case study sample involved foreign investors, typically those with a history of investment in the countries concerned. Among the case study coun- tries Ghana has attracted the largest amount of privatization-related foreign investment ($339.5 million, principally from the Ashanti Gold- fields flotations), Mozambique attracted $152 million of investment in the past five years, and Uganda attracted $137.8 million (see appen- dix E). We expect data for 1997 to show a dramatic increase in FDI. Privatiza- tion is now extending to major enterprises in the financial, energy, telecommunications, and transport sectors. The sale of equity in six telecommunications companies in West Africa (Cape Verde, Congo, C6te d'Ivoire, Ghana, Guinea, and Senegal), which went to the market during 1996, has reportedly raised more than $500 million so far; more- over, the new investors are committed to additional investments. For example, the new core investor in Ghana Telecommunications is com- mitted to installing an additional 225,000 lines within the next few years, and that alone could mean a further $350 million-$400 million in FDI for Ghana. In Zambia the privatization of ZCCM is expected to be finalized in 1997; privatization proceeds and new investments in the mining sector are expected to exceed $1 billion. Capital Market Development Privatization has highlighted the need for capital market develop- ment in Africa. In so doing, it has provided the strongest impetus to date for such development. Government sales of shares by public offerings have accounted for the increase in the number of quoted companies on existing stock exchanges. Although the amounts raised through privatization issues remain low relative to total market capitalization, they are likely to increase. As table 5.9 shows, almost 12 percent of the companies listed on the Nairobi Stock Exchange were privatization flotations, as were one-third of the companies listed in Abidjan. The need to be able to trade shares has led to the opening of a stock exchange in Zambia, and Tanzania and Uganda are expected to follow suit shortly. In West Africa there are active plans to establish a regional stock exchange in Abidjan that will service the UEMAO countries. In the Gambia, where shares were sold in CFAO, the National Trading Corporation, and Standard Chartered Bank, shares have been sold to the public despite the absence of a stock exchange and a secondary Table 5.9 Privatization Is Contributing to Capital Market Development (to End 1995) Numnber of Amount Amount Market privatization raised from raised from capitalization, flotations Number of Number of privatization other flotations end 1995 Country and sales other flotations listed companies ($ millions) ($ millions) ($ millions) Kenya 7 18 56 20 23 1925 C) Nigeria 35 4 182 112 628 1350 Ghana 10 29 17 387 154 2005 C6te d'Ivoire 10 0 30 66 n/a 428 Zambia 1 0 10 9 n/a 442 Note: Up to the end of 1995, privatization flotations had not taken place in the other African countries with stock exchanges (Botswana, Mauritius, Namibia, South Africa, and Swaziland). Soturce: IFC (1996) and World Bank data. THE IMPACT OF PRIVATIZATION 103 market. Public sales also occurred in Tanzania, where shares were sold by the Cooperative and Rural Development Bank, and in Togo. Trading in privatized shares on the secondary market appears to be an increasingly important source of financial sector activity, especially in Nigeria, and privatization programs are becoming an important source of fees for financial institutions, in the form of advisers' fees, accountancy, and legal and underwriting costs. 6 Assessing Privatization Programs in Africa W e have seen that privatization has been adopted across Africa but that, despite a fairly large number of transactions, its eco- nomic and financial impact has been minimal, albeit generally positive. How should this record be judged? Since privatization objectives are expressed in general terms, with no measurable goals stipulated, the stated objectives do not readily provide benchmarks against which the success of individual programs can be measured. Moreover, the economies and their public enterprise sectors have different characteris- tics, and different programs have different privatization objectives. For example, although reducing the fiscal deficit is important and measurable in countries that suffer from deficits, not all countries have fiscal deficits related to public enterprise performance. Some better standard measure needs to be established to compare relative progress and success across countries. Measuring the Success of Privatization This report has noted that the number and value of completed trans- actions do not provide a full picture of privatization and can even provide a misleading impression. More appropriate indicators of suc- cess are required. Which measures should be used to assess and com- pare the success of privatization in Africa is a question that could be usefully debated among the countries themselves. To stimulate that discussion, we suggest that success could be related to the diminution in the size of the public enterprise sector, to the extent to which private sector investment has been mobilized, and to post-privatization perfor- mance (which reveals the extent to which benefits are sustained). The Size of the Public Enterprise Sector Privatization should lead to a reduction in the size of the public enterprise sector. The simplest way to measure this reduction would 104 ASSESSING PRIVATIZATION PROGRAMS IN AFRICA 105 be to compare the number of public enterprises privatized with the initial stock of enterprises. This method, however, does not take into account the relative importance of the enterprises. More informative measures would be the total value of assets vested in the enterprises that have been transferred from government control to private hands, expressed as a percentage of the total value of assets in public enter- prises at the beginning of the process, and the total value of assets remaining in public enterprises, expressed as a percentage of the total value of assets employed in all commercial businesses. These measures require a clear definition of what constitutes the public sector, data on all private and public enterprises, and a consistent basis for determining asset values. Relevant data on public enterprises are available in some Francophone countries, but much work would be needed elsewhere to compile this information. Data on the private sector might be obtainable from company returns to tax authorities, but even then extracting and aggregating the information would be an onerous exercise. For the foreseeable future, it would be prudent to assume that complete data for these measures will not be available. Employment as a Proxy In the absence of data for measuring the size of the public enterprise sector directly, employment can be used as a proxy for both intra- and intercountry comparisons. Although some enterprises are capital- intensive and others labor-intensive, employment gives some idea of the relative importance of the formal private and public sectors. The extent of privatization could be measured by looking at the number of employees who moved from the public to the private sector, ex- pressed as a percentage of total employment in the public enterprise sector prior to privatization and as a percentage of total formal employ- ment. Despite the importance attached to the potential effects of pri- vatization on labor, few privatization agencies track employment. Pro- cedures should be established to track enterprise employment using existing agencies, such as social security and tax authorities, which already require and collect employment data. Private Sector Investment Changes in private sector investment resulting from privatization reflect confidence in a country and in its reform program. For a complete picture, private sector investment should include the initial investment made to acquire shares and assets of public enterprises and any addi- 106 PRIVATIZATION IN AFRICA tional capital invested in those enterprises for expansion, rehabilitation, or working capital requirements. Year-to-year changes in foreign direct investment (related specifically to enterprises undergoing privatization and to those that have been privatized) provide clear indications of success. Measurement of local investment would require that the amount of incremental investment be determined, which is likely to be very difficult. Enterprise Post-Privatization Performance Privatization is expected to result in increased efficiency in privatized enterprises as a result of new investment, new technology, and im- proved corporate governance. No country has monitored post-privat- ization performance to observe to what extent such improvements are being realized. This is an area in which the World Bank and donors can-and should-assist. The universal absence of monitoring systems provides an opportunity to design and introduce standard monitoring and reporting procedures across Africa. Post-privatization performance can be judged by the rates of return achieved and by financial flows, both compared against performance prior to privatization. Eight Indicators for Assessing Privatization Programs Eight general indicators (including some that approximate the ap- proaches described above) have been used to provide some indication of which case study countries are doing well and why. Where relevant data are available, they have been used as measurements in arriving at a rating for a given indicator; otherwise, a subjective judgement has been made on the basis of our analysis of the material gathered for our study. The first indicator, the extent of privatization, reflects the number of privatization transactions (both completed deals and those in progress) and the sizes of the enterprises involved. This indicator is easier to establish in some Francophone countries, where public enterprise asset values are reported and aggregated. The second, third, fourth, and fifth indicators, financial impact on government, broadening ownership, foreign direct investment, and post- privatization enterprise performance, measure the impact of privatization. All of them relate directly to some of the key objectives of many pro- grams. For the purposes of table 6.2, financial impact is assessed by reference to table 5.1. Countries receive high ratings when the average annual total of privatization transaction sale values has exceeded 5 ASSESSING PRIVATIZATION PROGRAMS IN AFRICA 107 percent of the government's average annual revenue budget (excluding grants) over the same period. Countries receive medium ratings for achieving between 2.5 and 5 percent, and receive low ratings for achiev- ing less than 2.5 percent. Broadening ownership is a purely subjective judgment based on efforts made to encourage local investors (see the section on broadening own- ership in chapter 2), including measures to develop the capital market alongside the privatization process. Foreign direct investment (FDI) through privatization is measured by comparing the total equity purchased through privatization transac- tions and post-privatization capital investment with the average annual FDI prior to the start of a privatization program (see appendix E). If the average FDI through privatization has been more than 50 percent higher than the pre-privatization average FDI, the country receives a high rating; countries with incremental investments of 25-50 percent receive a medium rating, and countries with incremental investments of less than 25 percent receive a low rating. Since earlier reported figures for FDI may be of doubtful accuracy or unavailable, the average FDI through privatization may also be compared with the reported average FDI during the privatization period. The ratings for post-privatization enterprise performance are based on the limited data that were collected for the country case studies, with additional weight given to the performance of the larger enterprises. The sixth indicator, program design and management, measures how well a program was conceived, planned, and executed. Program design and management can be assessed by looking at the existence of a detailed privatization blueprint; statements of policies on important issues (such as enterprise selection and priorities, selection of privatiza- tion methods, broadening ownership, capital market development, debt retirement, redundancy, and end-of-service benefits); the level and quality of work in establishing the performance and condition of public enterprises; organization (which ideally should cover such functions as pre-privatization studies and enterprise preparation, legal advice and documentation, social impact research, investor mobilization, and public relations); efforts to mobilize investors and, where applicable, to meet the objective of broadening ownership; and the experiences of bidders (both successful and unsuccessful) in dealing with the privat- ization agency. A judgment must make allowance for the environment in which the program has had to operate. For example, the operating environments in Burkina Faso (where there is no consensus on privat- ization and virtually no capital market) and Mozambique (where the economy has been devastated by war and capacity constraints are 108 PRIVATIZATION IN AFRICA severe) have been far less conducive to privatization than in Kenya or Zambia. The degree of transparency, the seventh indicator, is apparent from the existence and strict application of clear written policies and proce- dures for bid prequalification, bidding, evaluation, and negotiation; the choice of competitive methods of privatization (except where pre- emptive rights are a constraining factor); and the publication of results so that the public is informed of who the bidders were, what selection criteria were used, and what the outcome of the negotiations was. The eighth and most important indicator is commitment. There is no doubt that government commitment drives the privatization process. This was borne out by the country case studies, where lack of commit- ment emerged as a major issue, and it is a recurring theme in virtually every publication on the subject of privatization in Africa. The level of government commitment to privatization in each of the case study countries can be gauged using the following indicators: 1. Publication of government's privatization policies and program 2. Legislative support (including, for example, laws pertaining to bankruptcy, land, investment) and judicial support as evidenced in a well functioning court system 3. A privatization agency empowered with the legal authority nec- essary to carry out the job 4. A privatization agency that has adequate resources to carry out the job 5. Visible executive support from the head of state and senior gov- ernment ministers 6. Positive action by the government and privatization agency to eliminate constraints, such as holding companies 7. Willingness by the government to divest fully, as evidenced by the use of full privatization methods, except in those cases where a minority equity stake in large companies has been specifically and publicly earmarked for later privatization through the open market 8. Steps taken to make the investment environment conducive to privatization 9. The level of dialogue between the government, stakeholders, and the private sector in general 10. Public opinion regarding government commitment. Indicators I and 9 are used to gauge both transparency and commit- ment. Under transparency, they refer to evidence of the extent to which the government has informed the public and ensured the application Table 6.1 Govemment Commitment to Privatization (to End 1995) Adequacy of Policies Legal Legal resources Positive Willing- Capital Dialogue Public and support status of for Visible action to ness market with all opinion of program for privatization privatization executive counter to divest develop- stake- government Country published privatization agency agency support constraints fully ment holders commitment Benin Medium High Medium Medium High Low High Low Low High Burkina Faso Low High Low Low Low Low Low Low Low Low Ghana Low Medium Low Low Medium Low High Medium Medium Medium Kenya Medium Medium Low Low Low Low Low Low Low Low Madagascar Low Low Low Medium Medium Low Low Low Low Low Mozambique Medium High High High High Medium Medium Low Medium High Nigeriaa High High High High Medium Medium Low Medium Low High Togo Medium Low Low Low Medium Low Low Low Low Low Uganda Low High Low Medium High Medium Low Low Low Medium Zambia High High High High High High High High High High a. Based on the period when Nigeria's privatization program was active. 110 PRIVATIZATION IN AFRICA of transparent procedures. Under commitment, they refer to the govern- ment's determination to inform the public of the benefits of privatiza- tion and to mobilize investor interest while maintaining maximum transparency consistent with commercial discretion. Table 6.1 sets out the commitment indicators and includes the general ratings based on the country case studies and other available material. The table indicates that the commitment of the Government of Zambia is strong and that commitment in Burkina Faso, Kenya, and Madagascar is weak. Commitment may, of course, vary over time. Take, for example, the program in Ghana, where commitment has developed slowly but surely over the past ten years (box 6.1). This appears to be the trend in many countries. Senegal is another example. Low commitment led to a slow program with little impact. After five years the program came to a halt for several years, but it has recently resumed with much greater govern- ment commitment. Senegal's revitalized program has drawn on the Box 6.1 Ghana: How Government Commitment Has Developed over Time Privatization in Ghana has been slow, and commitment too has been slow to develop-but it has developed. Looking at the activity in the three-year period 1987-89, one would almost certainly draw the conclusion that the government was not committed to privatization. Program performance was very low in terms of what was being achieved (no privatizations) and lack of transparency. There was some movement in the next three years, 1990-92, when the first privatizations occurred. But with low transparency-and few resources for privatization, commitment was apparently low. The picture changes when we look at 1993-95. There were signs of real progress in terms of the number of completed transactions; the govemment announced its intention to accelerate its divestiture program; and policies and procedures for outsourcing were drawn up. Commitment was evident, but the privatization process was still not regarded as transparent, and many major public enterprises were still not approved for privatization. In 1996 the picture changed again. Outsourcing of many privatization transactions began. New procedures were adopted to provide greater trans- parency, including the dissemination of information to the public. Most convincing of all, the privatization of the telecommunications company was undertaken; this deal was concluded in 1997. Source: World Bank, 1996d. ASSESSING PRIVATIZATION PROGRAMS IN AFRICA 1ll lessons of the country's earlier experience that bear out many of the lessons highlighted in this report (box 6.2). Why Zambia Has the Most Successful Program The indicators described above allow us to compare the privatization programs of the case study countries (table 6.2). For this purpose, we have tried to rate performance using the indicators over the period from the beginning of each program up to the end of 1996. Table 6.2 clearly reveals that Zambia has the most successful pro- gram. Ranking of the other nine case study countries will, of course, depend on the relative weights given to the different indicators, but the assessments will also vary if different time periods are selected. Generally, the pace of privatization picked up in the second half of 1996, reflecting a growing commitment and consensus in many countries. The Zambia case study identified eight factors that contributed to the success of the privatization program: 1. Sufficient resources were invested in careful program design and preparation. 2. The process is supported by appropriate legislation (the Privatiza- tion Act). 3. The private sector is taking a leading role in the process. 4. The ZPA is the sole institution dealing with privatization, it has the legal mandate to execute its function, it has sufficient re- sources, and it is able to undertake its work with minimum politi- cal interference. 5. The process is transparent.'4 6. The program is well supported by donors, who have coordinated their assistance. 7. The government has taken decisive action to reduce constraints on privatization, notably by addressing the weak capital market and eliminating the influence of holding companies. 8. Although there is more to be done in this area, important steps have been taken to inform the public about the program and to encourage Zambian participation in the process, primarily through management buyouts and the Privatization Trust Fund. All of these factors add up to a clear demonstration of the govern- ment's commitment to privatization. That commitment was present at the beginning (it was in the party's election manifesto) and has been demonstrated by the government's total withdrawal from ownership of privatized enterprises. 112 PRIVATIZATION IN AFRICA Box 6.2 Senegal Typifies the Major Lessons of Experience The privatization program in Senegal, which started in 1986, has had a checkered history. Progress has varied over four time periods (with pri- vatization occurring mostly in the second and fourth periods): 1986-88, when little happened, principally because of low govemment commitment; 1988-90, when a new head of the privatization unit pushed the program forward; 1991-94 when, following a further change in program leadership, the privatization process stagnated; and from 1995, when the minister of finance took over responsibility for the program, to the present time, with much clearer evidence of government commitment. Senegal's experience with privatization epitomizes many of the features and problems of privatization programs in Africa: * There was a lack of leadership through the first and third phases, and only recently has the program had a "champion" who holds a senior position and who has the requisite communication skills and repu- tation. * More professional and financial resources were needed in the design, preparation, and implementation of the program. * The program was launched at a time when there was no consensus in favor of privatization. As a result, throughout the first three phases there was considerable resistance to the process from politicians, civil servants, and employees of public enterprises. * Enterprises were classified into "strategic" and "nonstrategic." Until the policy change in 1995, this somewhat arbitrary distinction resulted in most of the financially important enterprises being excluded from privatization. * Privatizing the smaller enterprises did not prove easy and did not contribute to developing public consensus. * There was insufficient involvement of all stakeholders. Only now does Senegal's program emphasize public participation and private sector partnership. * There was no publicly announced policy or reporting on the use of privatization proceeds. The total of about $65 million generated up to end 1996 appears to have gone directly to meet current government ex- penditures. * Commitment is what counts. Until recently, the privatization process did not have the necessary government commitment to tackle con- straints and push the process forward. Table 6.2 Case Study Programs Compared Using Eight Performance Indicators Enterprise Financial Foreign post- Program Extent of impact on Broadening direct privatization design and Trans- Government Country privatization government ownership investment performance management parency commitment Benin Medium Medium Low High Low Medium High Medium Burkina Faso Low Low Low Medium Medium Low Low Low Ghana Medium High Low Medium High Low Low Medium Kenya Low Low Medium Low Medium Low Low Low Madagascar High Low Low High Medium Low Low Low Mozambique High High Medium Low High Medium Medium High Nigeria Low Low High Low Medium High High Medium Togo Low Low Low Low Low Low Low Medium Uganda Low High Medium Medium Medium Low Medium Medium Zambia High High High Medium Medium High High High 114 PRIVATIZATION IN AFRICA Dealing with Constraints on Privatization An important feature of Zambia's program has been the govern- ment's readiness to deal effectively with constraints on privatization. Programs in every country have encountered constraints with varying degrees of intensity. The conclusions drawn from table 6.2 must there- fore be tempered by the constraints that individual countries have had to face. Table 6.3 lists the major constraints that have been identified, summarizes their causes and impact, and lists the countries in which the incidence of their impact has been reported. The remainder of this chapter looks briefly at some of the major constraints on privatization in Africa and compares experiences using Zambia as the benchmark where applicable. Lack of Consensus A lack of consensus is a recurring theme in all of the case studies and has been at the root of delays in adopting privatization and in dealing with other constraints. Without consensus, senior government officials cannot be expected to demonstrate commitment to privatization; politi- cians may understand the potential benefits of privatization but will tread the path of privatization warily if their constituents oppose the concept. In most African countries there has not been a broad consensus in favor of privatization. The notable exceptions are Cape Verde (box 6.3) and Nigeria. The general lack of consensus in Africa is hardly surprising. The public has not been informed of or has not understood the need for reform and the potential benefits to them of privatization. There has been no attempt to link poverty alleviation with the expected outcomes of privatization, and the public rightly questions where the money from privatization goes. Except in Nigeria, Tanzania, Zambia, and, more recently, Uganda, insufficient information has been made available to the public on pri- vatization activities and outcomes. The public information efforts cur- rently underway are welcome, but more needs to be done. To change public opinion to accept the notion that privatization is good requires a big advertising effort to present facts in a readily assimilable form: Ad- vertising should present information on the successful outcomes of pri- vatization, while the difficulties should not be ignored; the public should be shown how the difficulties are being dealt with. Zambia is the only country whose government was elected with a manifesto that included widespread privatization. This gave the newly Table 6.3 Major Constraints on Privatization in Africa Constraint Causes Effects Country examples Lack of consensus * Lack of information * Weak government commitment Everywhere except Cape * Lack of political will * Slow process Verde and Nigeria * Ideological beliefs * Reluctance to sell profitable * Vested interests enterprises Political uncertainty * Historical setting * Tardiness Benin, Ghana, Madagascar, * Democratization * Investor uncertainty Nigeria, Togo * Forthcoming elections Inadequate management * Lack of commitment * Poor design and inadequate Malawi, Guinea, Kenya, and implementation * Weak institutional and human preparation Cameroon, Niger capacity resources capacity * Lack of transparency * Fragmentation * Distrust of valuation methods * Long delays and incomplete m1 transactions Legal constraints * Old legislation * Insufficient authority given to See chapter 3 (see chapter 3) * Lack of commitment agency * Weak judicial system * Slow process Lack of ownership of the * Weak government commitment * Perception of program as driven Burkina Faso, Cote d'Ivoire, program * Institutional jealousies and by external agencies Guinea, Kenya, Malawi government interference * Lack of consensus * Lack of involvement of nationals * Tardiness and indigenous private sector * Donor driven Poor data on private * Weak governance * Initial conditions not clearly Ghana, Kenya, Mozambique, enterprise performance * Poor accounting and reporting understood Uganda and position * Too much emphasis on valuations * Liquidations executed too late (Table continues on following page.) Table 6.3 (continued) Constraint Causes Effects Country examples Preponderance of * Unwise investments * Difficulty attracting genuine Benin, Cameroon, Ghana, financially distressed * Poor management investors Mozambique, Tanzania, enterprises * Insufficient capital * Little to sell as a result of asset Togo, Uganda Adverse operating environment stripping * Slow process * Low proceeds * Difficulty meeting employee end-of-service benefits * Debt overhang Fear of unemployment * Lack of understanding of initial * Reluctance to privatize major Burkina Faso, Ghana, Kenya, conditions loss-makers Tanzania * Lack of understanding of privatization opportunities * Lack of social safety net Absence of a social safety * Shortage of resources * Fear of redundancy Kenya, Tanzania net * Wider implications beyond * Delay in privatizing inefficient privatization enterprises Contingent employee end- * Overmanning * Delay in dealing with SOEs with Ghana, Kenya of-service benefits * Inequity in benefits overmanning problem Underdeveloped capital * Socialist legacy * Low response from local All (in varying degrees) market * Weak financial sector investors * Low savings Low savings * Poverty * Participation restricted to the All countries but especially better off those with low GDP per * Difficulty gaining consensus capita * Weak capital market Unfavorable foreign * Unattractive investment code * Low investor interest Benin, Burkina Faso, Togo investment environment * Political uncertainty e Bureaucracy * Corruption and security problems Holding companies * Fear of dismantling * Delays and obstruction Ghana, Tanzania, Zambia * Vested interests * Diversionary tactics Vested interests * Desire to retain privileges/rents * Tardiness and interference Kenya, Madagascar, * Noncompetitive transactions Tanzania, Uganda ASSESSING PRIVATIZATION PROGRAMS IN AFRICA 117 Box 6.3 A Participatory Approach to Privatization in Cape Verde The government of Cape Verde successfully mobilized public support for reforming the public enterprise sector and implementing a privatization program. To do so, it developed a well-conceived, aggressive, and sustained television and radio information campaign entitled "Let's Modemize Cape Verde." The campaign highlighted opportunities for higher growth led by the private sector. Once the privatization program got under way, public anxiety rose re- garding the impact of the program on an already difficult unemployment situation. In response, the government, with the support of the World Bank, organized a series of eye-opening visits to countries such as Mauritius, which had gone through similar economic reform and liberalization processes. Participation in these sensitization programs was targeted at opinion leaders, such as journalists, union leaders, local elected officials, local business repre- sentatives, and key civil servants. During these trips, meetings and discus- sions were held with labor leaders, local entrepreneurs, and foreign invest- ors, managers of free-trade zones, and government officials in such ministries as justice, labor, industry, tourism, economy, and planning. These visits proved instrumental in convincing a large segment of the public of the potential of the reforms and in rallying support for the necessary steps to complete them. Finally, the privatization agency itself has tried at every step to ensure that the needed liquidations, restructuring, and privatizations have been carried out with a minimum of labor unrest and social disruption. This has been achieved through its participatory approach to the process, in particular its constant efforts to involve the people most affected (workers and manag- ers) in the discussion of privatization options; its creativity in designing socially acceptable solutions, mainly through the use of management or employee buyouts; and its proactive provision of support to the newly privatized enterprises in identifying potential domestic and foreign sources of technical assistance and financing. Source: World Bank, 1996a. elected government a mandate to proceed with its program. Even in Zambia, however, public opinion opposed privatization: a poll con- ducted in early 1992, shortly after the election, showed that 70 percent of the population did not favor privatization. Although the transac- tions completed since 1992 have begun to convince the Zambian public of the benefits of privatization, there undoubtedly remains a large section of the population that regards privatization as a threat to employ- 118 PRIVATIZATION IN AFRICA ment and sovereignty rather than as an opportunity for development and growth. In most countries the governments have tried to convince the public that private sector development will be the engine of growth. Attention has been drawn to desirable popular outcomes, particularly the broad- ening of ownership. The public has not been kept informed of the extent of inefficiency in the public enterprise sector; the major restructuring required for potentially viable businesses with negative net worths; the need to close nonviable businesses; and the need to introduce new capital investment, new technology, and improved corporate gover- nance. The public remains wary because of the history of widespread corruption and the slow emergence of democracy. Given this natural distrust, it is critical that more attention be paid to building consensus through public education and debate. Scarcity of Resources The case studies confirm that most programs have been severely restricted by the scarcity of resources. Privatization units have had to deal simultaneously with many transactions that have been more complex than designers and decisionmakers envisaged. Only the Zam- bian government set up its privatization agency with sufficient profes- sional and support staff and adequate facilities. Absence of a Social Safety Net Redundancy is a particularly sensitive issue in Africa because public enterprises tend to provide a range of nonmonetary benefits in addition to pay. These typically include transport, home leave travel, medical treatment, and housing but can extend to a wide range of other benefits, including education and clothing. Despite the concern about retrench- ment and despite some safety net studies (conducted in Kenya and Tanzania), there have been no concerted efforts to introduce national safety nets. The lack of action has probably contributed to a hardening of attitudes toward retrenchment, which in turn has delayed the pri- vatization of major enterprises and restricted the use of certain methods of privatization. Although Zambia did not introduce a social safety net, its privatization agency was set up with a department to examine the potential and actual social impact of each privatization transaction. ASSESSING PRIVATIZATION PROGRAMS IN AFRICA 119 Payment of Amounts Due to Employees Payment of amounts due to employees (salary arrears, severance pay, and end-of-service benefits) has been a constraint in some coun- tries. Ghana (1987-92), Cameroon (in the transport sector), Tanzania, Uganda, and Zambia are examples of countries where this has been an issue. The problem has three dimensions. First, insolvent enterprises are unable to meet this liability. Employees of debt-ridden public enter- prises are often owed back salaries and fear that they may never receive their end-of-service benefits. Second, some enterprises have generous packages written into negotiated collective agreements, while others are bound to pay only the legal minimum-usually an amount estab- lished long ago that is negligible in current terms. There are cases (in Uganda for example) where public enterprises slated for privatization have renegotiated severance provisions on overly generous terms. Third, redundancy pay and end-of-service benefits are normally calcu- lated on the basis of pay and do not take account of nonmonetary benefits. Governments in Ghana, Uganda, and Zambia have taken steps to settle end-of-service benefits, but these payments represent only the legal minimum based on basic pay and do not take into account the loss of nonmonetary benefits. Elsewhere (in Benin, Burkina Faso, and Togo) the legal position of redundant employees is uncertain. The contingent liability for employees' end-of-service benefits is reported to have delayed the privatization of insolvent enterprises, notably in Burkina Faso (where 30 percent of all railway staff may be laid off), Ghana, and Tanzania. If these payments were really the cause of these delays, the delays might have been reduced had program planning identified sources of funding to meet the liabilities, which could have been repaid from subsequent privatization proceeds. More likely, the delays had to do with the potential political repercussions of large-scale redundancies. Depending on how privatization proceeds are used, payment of employees' end-of-service benefits should not be a problem after the first year or two of a program. Enterprise Debts The heavy indebtedness of financially distressed enterprises has frequently been cited as a constraint on privatization because of uncer- tainty as to whether or not governments should retire the debts of public enterprises, and, if so, how this should be done. Certainly some transactions, such as the sale of the Morogoro Shoe Company in Tanza- nia, were delayed because governments were unsure about how to 120 PRIVATIZATION IN AFRICA proceed with privatizing heavily indebted companies. But the problem of debt exists with or without privatization, and it should not prevent privatization when that is what the government wants. For insolvent companies, the legal options available are to sell assets (through a liquidation process initiated by the creditors) or to relieve the enterprises by having the government assume direct responsibility for part of or all of the debt. Other methods have been used which, strictly speaking, are illegal unless the required majority of creditors has given its approval. One is simply to let a distressed company wither away (which too often allows management and employees to strip the assets while other creditors remain unpaid). Another method is to sell or trans- fer the assets to another company. There are many examples of privatiza- tion agencies selling the assets of public enterprises without going through a legal process of securing the agreement of the creditors and assuring those creditors that proceeds will be distributed among them. The problem of enterprise debt is mainly political: how to explain the buildup of excess debt and why and how the government has to deal with it. In some instances (of which there are many in Uganda) there have been doubts about the amounts of debt. In the end the governments are having to bite the bullet. Insolvent enterprises with very poor prospects have been put into liquidation; others with some ready, usable assets (such as the State Fishing Corporation in Ghana) have been sold piecemeal, while enterprises with prospects have been sold without debt. Underdeveloped Capital Markets The weakness of capital markets in Africa has contributed to the slowness of most programs. The absence of financial instruments such as unit trusts and a secondary market for trading restricts the methods of privatization and the ability of small savers to invest in companies. Privatization agencies are acutely aware of the importance of capital market development but are unable to rectify the problem, which is beyond their purview. Only in Zambia was capital market development planned and implemented alongside the privatization program. In all other countries-except Cote d'Ivoire, Kenya, and Nigeria, which al- ready had established stock exchanges-capital market development has lagged behind privatization efforts, particularly in countries with very weak financial sectors. 7 The Role of Donors It is unlikely that there would have been much privatization without donors. But not everyone agrees; some say that there might have been more and better privatization without donors because budget constraints would have exerted a more effective pressure. Nonetheless, there is no doubt that donors have played a crucial part in promoting privatization by putting pressure on governments to divest and sup- porting programs throughout Africa. Even in Nigeria, which chose to implement privatization without external assistance, the World Bank served as an adviser during the planning phase. Donor pressure to privatize has raised the issue of ownership of programs: with few exceptions, programs throughout Africa are perceived by the public to be, not home-grown programs, but World Bank-IMF driven and supported by the donor community. Donor support has taken several forms. The principal channel has been through structural adjustment programs that included a privatiza- tion component, for which World Bank loans and credits have been cofinanced by other agencies. These programs have generally (but not always) been followed by World Bank -loans and credits specifically to support privatization, including credits for technical assistance to implementing agencies. These have been complemented by World Bank loans for various sectors and by donor grants and credits for sector studies and technical assistance to privatization units. The World Bank has been closely involved in the design of many programs and throughout Africa has provided useful guidance on legal issues and privatization experience. By the end of February 1996, the World Bank had made available $450.5 million through 27 technical assistance credits to support privatization programs in Africa, less than one-half of which ($173.3 million) had been disbursed at that time (appendix F). Bank structural adjustment loans and credits for reform programs including privatization totaled $2,582.5 million by the end of 1995, of which $2,262 million had been disbursed. Other bilateral and multilateral agencies, such as the African Devel- opment Bank and the European Development Fund, have actively 121 122 PRIVATIZATION IN AFRICA promoted and supported privatization by pledging a total of $3,174.9 million in grants and loans (appendix G). Data on actual aggregate support from the donor community to individual countries in support of privatization are incomplete and extremely sketchy. Most donor agencies do not keep systematic records of their support for privatiza- tion or privatization related activities, often because it cuts across sectors. The Effectiveness of Conditionalities To encourage reform, the World Bank has negotiated with client governments for enterprises to be prepared for sale or to be brought to the point of sale (through, for example, public advertisements offering assets or shares for sale). As a matter of policy, the World Bank does not set conditions that require the actual sale of enterprises by specific dates. Thirty-four countries in Africa have World Bank loans or credits for programs or projects that include a privatization component. Loan preparation has involved extensive discussions between World Bank teams and government representatives. Pressure to achieve specified outcomes is sometimes reflected in conditionalities attached to the legal agreement for a loan. In some cases, credit conditions require a certain number of public enterprises to be privatized before a tranche is re- leased. The study reviewed the incidence of World Bank conditionalities on privatization to determine whether there was any direct correlation between such conditionalities and the pace of privatization. Appendix F lists all World Bank-funded programs and projects in Africa between 1988 and 1995 that supported privatization or included a component relating to privatization. Privatization conditionalities are attached to some three-quarters of World Bank loans or credits. Many of the credit agreements do not specify enterprises or set numerical privatization targets. Where enterprises are named or a number of enterprises for privatization has been specified, privatization has rarely occurred within the timeframe envisaged. Indeed, setting numerical targets has sometimes had adverse consequences, as is discussed below. Bank conditionalities may not have forced the process or the pace of privatization to the extent imagined. Although conditionalities may have contributed to stimulating the process, World Bank influence on privatization has more likely stemmed from persuasive discussion during loan preparation, appraisal, and supervision rather than through the loan conditionalities themselves. A comparison of the peak of activity in countries which have recorded most transactions with the timing of World Bank credit agreement conditionalities yields no THE ROLE OF DONORS 123 clear picture on the effect of those conditionalities. Case study inter- viewees did, however, suggest that the pace of privatization would have been slower without the conditionalities. Lessons for Donors The level of donor support has been impressive, although it could have been greater and better focused. This section identifies lessons that could be usefully applied to future privatization programs. Understand Initial Conditions Better Donors have exerted pressure to privatize without sufficient infor- mation. Anxious to see speedy action and results, donors have too often paid insufficient attention to the fact that the initial conditions are not generally well understood. Hence, donors have not done as much as they could to assist in developing a consensus in favor of privatization. Government officials feel that they have been pressed to adopt privatization on faith, and that donors have at times underesti- mated the constraints, particularly the capacity constraints, the diffi- culty of attracting the right types of investors, and the political sensitiv- ity surrounding ownership, potential unemployment, valuations, and undischarged debt. In the case study countries this view was expressed in Burkina Faso, Ghana, Kenya, Uganda, and Zambia and has been reported in Tanzania. Place Less Emphasis on the Number of Transactions Even though privatization targets have not always been set as condi- tionalities attached to loans, some governments and their agencies still feel that there has been-and continues to be-an emphasis, particu- larly by the World Bank and the IMF, on meeting numerical targets for privatization. The tendency to press for transaction numbers has arisen because of the perceived slowness of the programs and insuffi- cient information on the progress of privatization. Because of pressure to meet numerical privatization targets, attention has focused on completing many small transactions that have had minimal economic impact (box 7.1). Because of the focus on meeting a specified number of transactions, privatization agencies have played the "numbers game." This has meant that progress reports on privatization have at times included: 124 PRIVATIZATION IN AFRICA Box 7.1 Pressure to Privatize Can Be Counterproductive In 1993 the pressure on the privatization agency in Tanzania to achieve privatization targets agreed on with the World Bank led to the diversion of personnel resources to small transactions. Privatizing these small enterprises proved to be as complex and as difficult as for some larger businesses and diverted resources away from the main program. It is arguable, although not provable, that at the end of 1993 the pressure to meet the privatization targets resulted in fewer-rather than more-completed deals in Tanzania; it certainly diverted resources away from preparatory work on privatizing larger enterprises. transactions that have not in fact been completed; the winding up of companies that only ever existed on paper; insignificant transactions (such as the sale of a 1 percent equity interest by a government-owned development bank in Kenya); and other transactions that are classified as privatizations but that began or were even completed before the privatization programs commenced. Recognize the Complexity of Privatization Donors need to have a fuller appreciation of the complexity of privat- ization. Donors have spurred African governments into privatization without understanding the constraints or the resources and time needed to overcome them. Although constraints were identified, little was done to eliminate them. For example, the lack of consensus in most countries should have prompted greater efforts on the part of governments, with donor support, to engage in debate and to gain public confidence before programs got under way. Instead, public relations was ignored. Ghana, for example, is only now planning its first major public information campaign ten years after it embarked on privatization; by the time the campaign is under way more than one-half of the 195 public enterprises already approved for divestiture will have been privatized.'5 Provide More Hands-On Support Early While urging transparency and speed, donors have not taken into account the many detailed procedures involved and the resources those procedures require. Critical of the slowness of these programs, donors themselves have been slow to provide the type of assistance needed. Although donor projects can take several years from conception to THE ROLE OF DONORS 125 approval, African governments, already desperately short of the neces- sary technical skills, have been expected to launch complex programs quickly. The meager resources made available to privatization agencies in countries such as Ghana, Kenya, Tanzania, and Uganda have limited the volume of transactions they could handle. Assistance is most needed early in the process. The experience of the programs in Mozambique, Tanzania, and Zambia shows that early hands-on support to develop programs and to prepare for and transact deals is critical to kick-start the process. Early support should be com- plemented by on-the-job training and study tours so as to develop local capacity as quickly as possible. Provide Early Assistance to Meet End-of-Service Benefits Early action on loss-making enterprises has been slow, not simply because of the governments' apprehension about the public reaction to retrenchment but also because of a lack of funds to meet employees' severance pay or end-of-service benefits. Once a program is well under way, privatization proceeds flow in to meet these costs; until that time, financial support may be needed. It is debatable whether action to divest large loss-makers would have been taken earlier had donors provided financial support to meet these benefits. In this context, it should be noted that until recently, under the terms of IDA lending, countries were not able to use World Bank credits to meet these costs. World Bank policy has changed, and credits may now be applied to cover severance pay under certain specified conditions. Pay More Attention to Monitoring and Evaluation Donors must pay much more attention to the monitoring and evalua- tion of enterprises following privatization. Their attention has focused instead on pre-privatization studies, preparation for privatization, and early closure of deals-that is, on the process and the completion of the process. Monitoring and evaluation have largely been ignored; hence the paucity of data on-and the difficulty of judging-the prog- ress and impact of privatization to date. But the blame rests as much, if not more, with governments who themselves have done little to tackle information gaps. Africa is litteTed with various information systems and technical assistance projects, some very slowly or never completed, that have yielded little lasting effect. In the past this may have been due to a government's desire not to know. Whether or not that has been the case, today the information 126 PRIVATIZATION IN AFRICA revolution offers Africa a chance to catch up with the rest of the world. And privatization is but one example of how information can be used to help the process and gauge its impact. Demonstrate the Benefits of Privatization Most governments have pushed ahead with privatization in order to secure World Bank, IMF, and donor support (see chapter 2). This reinforces the importance of more support for monitoring, evaluating, and reporting on the impact of privatization in order to demonstrate to all stakeholders the benefits of the process. Donors could support efforts to advertise privatization, both to publicize its benefits and to encourage people to actively participate. In this way donors could make an important contribution to assist governments in building consensus. Improve Coordination Among Donors Poor donor coordination is not a new problem in Africa. As a prime example, UNCTAD, UNIDO, and USAID separately sponsored studies of privatization in Africa at the same time as this book. The self-interest of different donors has, on occasion, affected the level of support and advice to governments. It is not uncommon for donors to want to take the lead in providing privatization advice and assistance and, where another donor is already in "pole" position, to withhold support (as happened in Uganda) and direct it elsewhere. The meager resources that are generally made available to African privatization agencies are evidence enough of the lack of coordination among donors. In most countries-Zambia is an exception-the privatization agency is likely to receive assistance only from the World Bank or one bilateral agency, or both. More assistance is required, and that calls for greater coordi- nation. Zambia has received a high level of donor support, and cooperation and coordination among donors has been good. This is apparent in the strong support given to the privatization program, the monthly meetings with the privatization agency attended by all donors, and the many transactions that have come to completion since mid-1995. Elsewhere, coordination specifically for supporting privatization is less apparent. Governments and donors could do more to improve coordination and the quality of advice and assistance for privatization. Donors frequently operate in isolation, and governments find them- selves inundated with separate donor missions on privatization, all THE ROLE OF DONORS 127 offering advice on priorities and methods, as well as assistance, but assistance is often slow or fails to materialize. Where support for privatization is forthcoming, donors sometimes choose to direct assistance to enterprises and sectors outside of and uncoordinated with the program managed by the privatization agency. Without close coordination, it is difficult to plan privatizations to take account of the absorptive capacity of the capital market, to achieve more equitable distribution of participation opportunities, and to use fully the privatization knowledge and experience already accumulated in the countries concerned. As a result, uncoordinated initiatives have unwittingly dissipated privatization efforts. Donors that have sup- ported particular sectors or enterprises have also lobbied governments to give preference to investors from their own countries, confirming suspicions that bilateral assistance for privatization is to some extent determined by commercial self-interest. Provide More Assistance to Mobilize Investors Investor mobilization stands out as an area in which donors could have done much more to assist African countries, both internally and externally. Internally, more support is needed in two areas. The first is development of the capital market, so that tradable financial instru- ments become available early on to provide real opportunities for broadening ownership. The second is in the area of management or employee buyouts, where there is a lack of low cost professional advice to public enterprise managers to enable them to formulate and present convincing proposals for raising capital and for bidding when their enterprises are offered for sale. Externally, donors can do more to assist with raising the interest of potential investors in African privatization programs. This could be done by supporting the sales promotion efforts of privatization agen- cies, including greater international media coverage, organizing Afri- can investment promotion conferences in their own countries (with emphasis on positive examples and forthcoming investment opportuni- ties), and funding investment search and deal-making consultancies. Sharpen the Focus on Who Benefits If donors are serious about poverty alleviation, they must sharpen their focus on who will benefit from privatization. Although they state their support for broadening ownership, the fact remains that the large majority of the population in every country in Africa is unable to 128 PRIVATIZATION IN AFRICA participate. This has been a dilemma for donors. On the one hand, to spur private sector development and eliminate inefficiencies, they have wanted to support activities to quickly privatize public enter- prises. On the other hand, early action, without careful attention to public information and participation, has meant that opportunities have been open mostly to the few who have savings and to those able to benefit from the patronage which some methods of privatization offer. Relatively few poor people have become or will become share- holders through employee participation and directed schemes. Donors, among others, are of the view that consumers are benefiting from privatization and that private sector development through privatiza- tion will benefit the entire population in the medium to long term. What is not clear, however, is who the consumers are and to what extent they are benefiting. To build and maintain consensus, the majority of the people, who are poor, will need to be convinced that they are indeed sharing in the benefits. Hence we argue for more support to monitor the impact of privatization and to impart that information to the public. For a few countries, where time and circumstances permit, there may be a case for supporting an initiative to develop a mass privatiza- tion scheme, but for most of Africa it is too late. In most cases where privatization is well under way, we believe that donors should encour- age African governments to ensure that at least part of privatization proceeds is applied in ways that will demonstrably help the poorest segments of the population. 8 How to Improve the Process The more than ten years of experience across Africa has provided clear lessons on how to improve the process of privatization. Those lessons are summarized in this chapter. They require changes to im- prove the process and its outcome, and the changes require commit- ment. Today, most governments are committed to privatization, but they have to do much more to demonstrate it. Demonstrate Commitment Commitment can best be demonstrated by taking the following steps: * Have a "champion"-a government official at senior ministerial level who can, and can be seen to, drive the process. * Establish a central focus for privatization-an agency that has a clear legal mandate to carry out its functions. * Appoint a person of the right caliber to head up the agency, and provide the agency with adequate resources for the job. e Ensure full legislative support. * Involve the private sector to the maximum extent feasible (privat- ize the privatization process). * Identify and take actions to remove or reduce constraints. * Exit fully from ownership of enterprises. Commitment is essential, but so too is consensus. After all, it makes political sense for a government to proceed warily when it is not con- vinced that the public is in favor of privatization. If the commitment is there, the first step is to build consensus. Pay Greater Attention to Securing Consensus Countries embarking on privatization, and those with flagging pro- grams, should pay greater attention to securing consensus. The evi- 129 130 PRIVATIZATION IN AFRICA dence suggests that privatization in African countries has mostly been adopted as a matter of economic necessity, with strong pressure from the World Bank, the IMF, and donors. Senior persons in government may have been convinced of the benefits of privatization, but there has not always been a consensus throughout government and certainly not among the general public. It is understandable, therefore, that governments have hesitated to deal with constraints and that progress has been slow. But much more could have been done to convince stakeholders of the desirability and feasibility of privatization. Provide Public Information and Encourage Debate Because building consensus involves presenting facts and arguments to convince a majority of the population, governments must provide public information and encourage debate. People need to understand why privatization is important, what benefits it can bring, and the context in which privatization will take place. An early step should be to gather and disseminate information about the public enterprise sec- tor, the economy generally, and the possible options and tradeoffs. This calls for effective use of the media through carefully designed advertising and public debate. Representatives of all stakeholders should be encouraged and permitted to participate in discussions on privatization (its objectives, options, and priorities) and the alternatives. Informed and constructive debate requires full disclosure of informa- tion, and that is a test of commitment. Disclosure can be unpalatable for a government when there are major gaps in data and poor results, reflecting poor decisions and management in the past. It also requires that more be done to clarify objectives (both at the program and the enterprise levels) and, later, to report progress toward achieving those objectives."6 Employment and ownership have been and still are at the center of public concerns about privatization. These issues need to be debated openly so that the tradeoffs become known and better understood. Many public enterprise sector employees can be expected to feel threat- ened by the prospect of privatization. An important way in which to alleviate their concerns is to inform them of the experience of successful privatization cases so that they can learn how opportunities afforded by privatization have been realized elsewhere and thereby become motivated to prepare and push for change. Despite the benefits of privatization, doubts about and opposition to the process will arise as opportunists seek political gain. Although in some quarters there may be a hankering for a failed ideology, the princi- pal reasons for opposition are usually ignorance and misunderstanding. HOW TO IMPROVE THE PROCESS 131 Many stakeholders, including trade unions, have until recently been in- sufficiently informed and involved in the process and feel that they are mere bystanders. In view of the perceived effects on employment, trade unions everywhere are understandably wary of privatization. In the Francophone countries they are becoming more voluble in their opposi- tion-justifiably so in view of the dearth of information and dialogue. The lack of consensus in Africa may be overstated. After all, it is easier for doubters and dissenters to raise concerns about an untried process than to convince the public of a new concept that is remote from their experience. In some countries the large majority of the popu- lation may be ambivalent about privatization. If that is the case, it is incumbent on a government committed to the process to present the case clearly and unequivocally through a professionally managed pub- lic information program and debate. Once a program is under way, the public needs to be kept informed not only of the facts about particular transactions but also later about the impact of those deals on the enterprises, their stakeholders, and the economy at large. Information needs to be communicated in ways audiences will un- derstand, and this demands carefully designed and executed informa- tion campaigns. Particularly sensitive handling is needed for certain issues, including the need for nonviable businesses to be liquidated, the potential effects on labor and the social costs of privatization, the position and rights of existing shareholders with preemptive rights, the need for changes in land and other laws to encourage private investment, the need to attract foreign investors, the often wide gap between book and market values, and ways in which the poorer sections of the population can participate or will otherwise benefit. During field research we enquired about efforts to inform the public. A frequent response was that the resources, both technical and financial, were not available and that an information campaign could slow the process down. We do not share that view. In relation to the total cost of a privatization program and the benefits a well executed campaign can offer, the cost is insignificant; and although dissemination and debate may absorb time early on, the consensus and investor interest they build can later help accelerate the process. Today the debate is no longer about whether to privatize but how and when. Deal with Labor Issues Up Front Privatization programs should deal with labor issues up front. The case studies show how the potential impact on labor of divesting large loss-making or inefficient enterprises has delayed action on the very 132 PRIVATIZATION IN AFRICA enterprises whose privatization would yield the greatest benefits. Gov- ernments and their agencies have tended to avoid open discussion of labor redundancy. Difficulties over the level and payment of employee severance pay and end-of-service benefits and the absence of a social safety net have caused governments to be embarrassingly silent. But the issues will not go away; the longer they are not addressed, the greater the concern of workers, and the more difficult it is to build consensus. Labor representatives should be party to developing mecha- nisms to deal with these issues and be shown how privatization offers real opportunities for sustained employment through the introduction of capital into businesses that need investment. Ensure Transparency Transparency is essential for building and maintaining consensus. As more stakeholders enter the debate and greater press coverage is given to privatization, greater transparency will be demanded. It is preferable that transparency be achieved by privatization agencies ac- cording to published procedures and by issuing routine public reports rather than as a reaction to criticism. One area where transparency is conspicuously absent is in the publi- cation of financial accounts of privatization programs which disclose, among other things, the use of privatization proceeds. So far, only the privatization agency in Tanzania has prepared and published financial accounts, although Ghana is expected to do so shortly A government risks adverse speculation and publicity if the public is not informed about the use of privatization proceeds. There should be a clear policy on how proceeds will be managed and utilized. Progress reports on privatization should include a section setting out how much of the proceeds have been applied to meet transaction costs, program management, end-of-service benefits and debt retirement for specified enterprises, general debt servicing by the government, and other spe- cific government expenditure items, and how much is held in trust and for what purposes it is earmarked. Invest More in Design and Preparation An important lesson that countries which do not yet have a program in place can learn from other countries' experience is to invest more in program design and preparation. This means investing more in the critical early stages of the program where, as we have seen, the level HOW TO IMPROVE THE PROCESS 133 of resources made available is an early indication of government and donor commitment. Clearly, there is a need to avoid "re-inventing the wheel." Access to information about the wealth of experience across Africa and the rest of the world should be available to all privatization agencies. In this area the World Bank is now playing an increasingly important role in collecting and disseminating information and providing best practice guidelines. Key areas in which more effort in design and preparation will con- tribute to a successful program include the following: * Establishing the initial conditions and the anticipated impact of privatization. This work facilitates the processing of transactions and allows impact to be measured. * Determining objectives at the program and enterprise levels and setting priorities. * Examining and designing strategies for identifying and attracting core investors. * Drawing up detailed policies and procedures, which should be published to facilitate the processing of transactions and to en- sure transparency. * Giving priority to enterprises whose privatization will have the greatest impact. i Involving the private sector in the privatization process. * Establishing how the success of privatization will be measured. * Assessing the potential social impact of privatization and ensuring that clear policies are set and mechanisms are put in place for the payment of severance awards and end-of-service benefits. - Monitoring employment changes. * Monitoring post-privatization performance of enterprises. * Mustering and coordinating donor assistance. * Putting the institutional building blocks in place before launching a program. Policy issues affecting sectors and individual enterprises need to be identified as early as possible and efforts must be made to examine options, make decisions, and draw up detailed policies which enable the implementation agency to move forward quickly and consistently. Take, for example, the question of divestiture priorities. If economic efficiency and fiscal considerations are paramount, attention should be focused on large enterprises and on the enterprises incurring the biggest losses. The program should then be designed to have an early and evident impact. Less attention should be paid to small, insignificant 134 PRIVATIZATION IN AFRICA enterprises, privatization of which could be handled through auctions or other competitive methods. Ownership is another example. Al- though there will be a general policy on ownership, except where a number of enterprises with very similar characteristics can be aggre- gated, there will be a need to review and decide on the most appropriate ownership structure for each individual enterprise. Put Institutional Building Blocks in Place before Launching a Program Delays, uncertainty and poor publicity for privatization are deter- rents to investors. Hence it is important to put the building blocks for privatization in place before launching a program. The three most important building blocks are the enactment of appropriate legislation; the granting of sufficient authority, independence, and resources to the privatization agency; and the removal of the obstructionist influences of holding companies and parent ministries. Enact Appropriate Legislation Although it may give rise to lengthy debate, experience has shown that enactment of one all-embracing piece of legislation will facilitate privatization. Appropriate legislation allows deals to be approved and finalized more quickly, and that is crucial to attracting potential in- vestors. Give the Privatization Agency the Necessary Powers, Independence, and Resources The experience of the different institutional models that have been tried in Africa indicates that the best arrangement is to establish a central privatization agency with legal status and sufficient authority to initiate, negotiate, and conclude transactions. Establishment of such an agency does not mean that one agency must handle every transaction (some privatizations can be delegated), but it does mean that one agency should oversee the program and report to the main legislative body and the public on the progress of the program. The agency should, to the extent possible, be independent of the government, perhaps along the lines of the Zambian model. This helps depoliticize and speed up the process, indicates a strong commitment to transparency, and sends an encouraging signal to investors. HOW TO IMPROVE THE PROCESS 135 To be truly effective, the agency must have access to adequate re- sources and approach privatization in a commercial manner. This means being able to avail itself of the best advice, to use the media effectively, and to contract out the work on individual privatization transactions to experienced professional deal-makers. The agency should be headed by a person regarded by the public as having integ- rity, political astuteness, and commercial acumen. The agency head should be supported by adequate financial, human, and logistical re- sources. Eliminate the Influence of Vested Ownership Interests Privatization can be accelerated if steps are taken to eliminate the influence of institutions with vested interests, in particular state-owned holding companies (banks, industrial conglomerates and sector holding companies) and line ministries, which do not want to lose the privileges afforded by their exercise of ownership rights in public enterprises. When a sufficiently competitive environment has been established, holding companies themselves can be privatized. This would transfer their entire investment portfolio to the private sector, reducing the number of public enterprises to be privatized individually. When this course of action is not appropriate, the holding company's shareholding in subsidiaries that are to be privatized should be transferred to the privatization agency or an interim trust. Address the Need for Competition and Regulation We have highlighted the absence of antitrust legislation and the slow efforts to develop effective regulation. These weaknesses must be addressed if African countries are to be able to promote real competition and ensure a fair deal for consumers. Regulation will almost certainly have to be developed in stages. Work to introduce an appropriate regulatory framework for each country should have commenced sev- eral years ago. Given the commitment by governments to private sector development and privatization, there is no excuse for not giving high priority to this work now. Do More to Broaden Ownership More steps should be taken to broaden ownership, using all the methods available, with preference given to those methods that reach 136 PRIVATIZATION IN AFRICA out to more people. For some large public enterprises this may call for open discussion about the capital needs of those businesses, the risks involved, and the tradeoffs between price optimization and broadening ownership. Efforts should also be directed toward attracting available savings for sound investment opportunities and toward developing capital markets. Develop the Capital Market in Tandem with Privatization Capital market development should take place in tandem with priva- tization. To attract indigenous private entrepreneurs, venture capital funds need to be established and made accessible. To attract savers generally through public offerings, a secondary market needs to be in place, in the form of either a national or a regional stock market. Appendices A Summary of Divestiture Transactions and Sales Values in Sub- Saharan Africa, as of End 1996 138 B Divestiture Methods Used in Sub-Saharan Africa, 1988-96 140 C Ownership Changes in Firms in Sub-Saharan Africa: Case Study Countries, 1988-96 142 D Examples of Employment Changes in Firms after Privatization 1987-95 144 E Foreign Direct Investment and Privatization in Sub-Saharan Africa, 1989-95 148 F Summary of World Bank Lending Operations Supporting Privatization in Sub-Saharan Africa, to End 1995 150 G International Donor Support Available for Privatization Programs in Sub-Saharan Africa, 1988-95 152 137 Appendix A Summary of Divestiture Transactions and Sales Values in Sub-Saharan Africa, as of End 1996 Transactions completed Total Total Average Year no. sale sale Prior to not trans- value value Country 1990 1990 1991 1992 1993 1994 1995 1996 known actions (US$m) (US$m) Angola - - - - - - 56 275 331 25.1 0.1 Benin 17 3 8 3 5 3 5 - 2 46 63.5 1.4 Botswana 0 0 0 0 0 0 - - 1 1 - Burkina Faso 0 0 2 1 5 5 3 - 16 9.6 0.6 Burundi 4 0 2 12 7 1 - 13 39 10.8 0.3 Cameroon 0 0 8 0 0 5 3 23 39 Cape Verde 0 0 0 0 1 10 10 9 30 24.1 0.8 + Central African Rep. 17 4 5 0 0 0 0 - 26 - o: Chad 0 0 0 0 0 12 12 10 34 - Comoros 0 0 0 0 4 - - - 4 0.2 0.1 Congo, Dem. Rep. of 16 0 0 0 0 0 3 3 22 - Congo, Rep. of 0 0 0 2 0 0 44 14 60 C6te d'lvoire 0 1 3 4 3 8 18 9 46 123.1 2.7 Djibouti 0 0 0 0 0 0 0 - 1 1 0.0 Equatorial Guinea 0 0 0 0 2 - - - 1 3 0.2 0.1 Eritrea 0 0 0 0 0 0 0 - 0 0.0 Ethiopia 0 0 0 0 0 0 9 - 9 31.3 Gabon 0 0 0 0 0 0 0 25 25 - Gambia, The 8 7 4 5 3 3 0 - 30 9.9 0.3 Ghana 4 33 13 14 9 83 24 12 191 417.3 2.2 Guinea 76 8 2 6 16 2 4 - 114 8.9 0.1 Guinea-Bissau 2 3 3 7 1 3 11 - 30 1.3 0.04 Kenya 0 0 0 27 21 16 66 22 152 184.6 1.2 Lesotho 0 0 0 0 0 0 8 - 8 - Liberia 0 0 0 0 0 0 0 - 0 0.0 Madagascar 14 11 10 21 1 3 24 - 84 18.8 0.2 Malawi 35 - - - - - - 9 44 55.9 1.3 Mali 2 4 18 30 - - - - 54 31.8 0.6 Mauritania 13 1 - 15 - - 3 15 47 1.2 0.03 Mauritius 0 0 0 0 0 0 0 - 0 0.0 Mozambique 65 40 53 62 42 136 112 38 548 179.7 0.3 Namibia 0 0 0 0 0 0 - - 1 1 - Niger 20 4 2 1 1 1 2 - 31 3.47 Nigeria 15 36 8 14 8 0 0 - 81 206.9 2.6 Rwanda - - - - - - - - - - Sao Tome & Principe 0 5 0 2 1 0 1 - 9 - Senegal 12 7 17 - - - - 9 3 48 65.0 1.4 Seychelles 0 0 0 0 0 0 - - 1 1 - Sierre Leone 0 0 0 0 6 1 1 - 1 9 - Somalia 0 0 0 0 0 0 0 - 0 0.0 South Africa 2 0 1 0 0 0 0 1 4 761.7 190.4 Sudan 0 0 0 12 17 3 - - 32 - Swaziland 0 0 0 0 0 0 0 - 0 0.0 Tanzania 0 0 0 13 33 16 27 34 123 125.5 1.0 Togo 12 5 3 2 0 3 1 19 45 35.9 0.8 Uganda 0 0 1 3 3 29 25 15 76 136.9 1.8 Zambia 0 0 0 0 6 20 58 107 191 146.3 0.8 Zimbabwe 0 0 0 0 0 4 0 - 4 25 6.3 Total 334 171 163 256 195 367 474 428 301 2,689 2,704.1 1.0 - Not available. Source: Data from government privatization agencies and ministries and the World Bank. 140 PRIVATIZATION IN AFRICA Appendix B Divestiture Methods Used in Sub-Saharan Africa, 1988-96 Sale of shares Sale of assets Direct Competitive Direct Pre- Competitive sales tender of sale Open emption Public sales of of Country shares' of shares' auction rights flotations Liquidations assets assets Angola - - 56 - Benin 1 2 - - 23 11 4 Botswana - - - - - Burkina Faso 9 3 - - 2 Burundi 11 3 - - 13 4 - Cameroon 19 - 16 - Cape Verde 9 7 - 6 2 Central African Rep. - - - - 27 - - Chad 20 - - - 12 Comoros - Congo, Dem. Rep. of I - - - 15 Congo, Rep.of - - - - 53 Cbte d'lvoire 12 11 - 10 - 5 1 Djibouti - - - - - - - Equatorial Guinea - - Eritrea - - Ethiopia - - - - - 9 Gabon - - Gambia, The 13 7 - - 3 1 Ghana 16 21 - 10 49 68 8 Guinea 42 - - - 67 1 - Guinea-Bissau 8 2 8 2 - Kenya 12 5 73 12 25 18 5 Lesotho 3 1 - - 4 - - Liberia - - - - - - - Madagascar 6 12 - - 29 15 6 Malawi 22 5 - I - 13 - Mali 14 1 - - 22 10 - Mauritania 21 - - - 15 5 - Mauritius - - - - - - - Mozambique 495 - - - - 5 - Namibia - - - - - - - Niger 13 - - - 17 - - Nigeria 5 5 - 36 2 29 - Rwanda - - Sao Torn & Principe 1 - - 6 - - Senegal 24 - - - 20 1 - Seychelles - Sierre Leone - - - - 1 Somalia - South Africa 3 - - - - 1 Sudan 7 - - - - 15 Swaziland - - Tanzania 38 4 - - 32 15 2 Togo - 7 - - 17 15 1 Uganda 12 2 2 - 16 28 - Zambia 37 7 3 1 15 75 - Zimbabwe 2 2 - - - - - Totals 875 108 2 76 71 514 404 27 -Not available. a. Includes tenders. APPENDICES 141 Other methods Manage- Debt- Joint ment or Manage- Transfers Transfers Method equity Equity ven- employee ment to without Resti- not swaps dilutions Leases tures buyouits contracts trustees payment tution' known Total - - - - - - 275 331 - 2 - - 2 1 - - 46 _ - _ _ I I - 16 - - - 1 7 - - - 39 - - - - - - 4 39 - - - 6 - 30 - - - - - 26 - - - - 2 - - - 34 - - - - - - 4 4 - - - - 2 5 - - 60 _ 3 1 - I _ _ 2 46 _ _ _ _ _ _ _ ~ ~ ~~~~~~~~~~~~~~~~~I I _ I _ _ _ _ - ~ ~ ~~~~~~~~~~~~~~~~3 3 - - - - - - 25 25 - 5 - I - - - 30 - 4 9 - I - 5 - 191 - -2 - 2 - - - 114 2 5 - 7 - - - 30 - 2- - 2 - - - 152 _ _ _ _ _ - - - 8 - 2 1 1 - 3 - 8 1 84 - - - - - 3 - - 44 1 - - 1 3 - 2 54 - - - 2 3 - 1 47 - 37 10 - I - - - 548 1 I - - - - - - 31 - - - 1 - - 2 - - 81 _ 1 1 - - - 9 2-- - - -- 48 -~~~~ _ - 1 -1 - 5 - - 3 - - - 9 -~ ~ _ _ _ - - 4 5 1 - - 4 - - 32 - 19 3 2 1 - 3 2 2 123 - 3 - - 2 - - - 45 1 1 2 1 - 5 6 76 - - - - 6 - - - 22 3 - 31 - 1 18 - 191 _-_ _ _ _ - - - 4 7 2 92 27 44 47 26 5 39 323 2,689 b. Includes private placement. c. Includes transfers of assets and shares. Source: Goverment privatization agencies and ministries and World Bank data. 142 PRIVATIZATION IN AFRICA Appendix C Ownership Changes in Firms in Sub-Saharan Africa: Case Study Countries, 1988-96 Government ownership Majority in SOEs 100% (>1=50%) Minority Not known Total Benin Before privatization 4 40 2 0 46 After privatization Majority (>/=50%) 4 4 Minority 1 1 N.K. 0 Zero 39 2 41 Total 4 40 2 0 46 Burkina Faso Before privatization 1 9 5 1 16 After privatization Majority (>/=50%) 1 Minority 1 7 3 10 N.K. 1 1 Zero 2 2 4 Total 1 9 5 1 16 Ghana Before privatization 151 17 23 0 191 After privatization Majority (>/=50%) 9 9 Minority 13 8 3 24 N.K. 0 Zero 129 9 20 158 Total 151 17 23 0 191 Kenya Before privatization 47 16 89 0 152 After privatization Majority (>/=50%) Minority 3 5 45 53 N.K. 0 Zero 44 11 44 99 Total 47 16 89 0 152 Madagascar Before privatization 76 8 0 0 84 After privatization Majority (>/=50%) 2 2 4 Minority 14 1 15 N.K. Zero 60 5 65 Total 76 8 0 0 84 Mozambique Before privatization 548 0 0 0 548 After privatization Majority (>/=50%) 42 42 APPENDICES 143 Government ownership Majority in SOEs 100% (>1=50%) Minority Not known Total Mozambique (continued) Minority 69 69 N.K. 0 Zero 437 437 Total 548 0 0 0 548 Nigeria Before privatization 34 22 24 1 81 After privatization Majority (>/=50%) 2 2 Minority 6 1 7 N.K. 1 1 Zero 32 16 23 71 Total 34 22 24 1 81 Togo Before privatization 34 4 7 0 45 After privatization Majority (>/=50%) 4 4 Minority N.K. Zero 30 4 7 41 Total 34 4 7 0 45 Uganda Before privatization 69 6 1 0 76 After privatization Majority (>/=50%) 4 4 Minority 5 2 7 N.K. Zero 60 4 1 65 Total 69 6 1 0 76 Zambia Before privatization 182 7 2 0 191 After Privatization Majority (>/=50%) 0 Minority 2 2 4 N.K. 0 Zero 180 5 2 187 Total 182 7 2 0 191 Totals Before privatization 1146 129 153 2 1430 After privatization Majority (>/=50%) 67 2 0 69 Minority 106 32 52 190 N.K. 0 0 0 2 2 Zero 973 95 101 1169 Total 1146 129 153 2 1430 Appendix D Examples of Employment Changes in Firms after Privatization 1987-95 Employment Employment, Country, name of enterprise, at 1st quarter Jobs and date of sale Sector privatization 1996 discontinued New recruits Benin SONAEC, 1988 Industry 123 126 n.a. 3 SOBETA, 1990 industry 145 116 29 ma, CIMBENIN, 1991 Industry 168 162 6 n.a. SOBETEX, 1991 Industry 273 138 135 n.a, g SOBEBRA, 1991 Industry 1,110 613 497 n.a, FACS, 1994 Service 50 29 21 n.a. Drift Benin, 1995 Petroleum 0 8 0 8 SONAR, 1995 Financial 3 5 0 2 Total 1,872 1,197 688 13 Burkina Faso SBCP, 1991 Industry 43 53 11 15 GMB, 1993 Industry 175 140 52 13 SBMC, 1993 Industry 84 89 23 28 SIFA, 1993 Industry 200 179 22 1 SOBCA, 1993 Financial 47 56 0 9 SONAR, 1993 Financial 82 80 3 1 Zama Publicite, 1993 Service 97 68 51 20 CIMAT, 1994 Industry 11 120 n.a. 109 SONAPHARM, 1994 Trade 52 51 1 n.a. FLEX-FASO, 1995 Industry 104 65 39 n.a. Total 895 901 202 196 Ghana Golden Tulip Hotel, 1990 Service 116 306 0 190 Lever Brother Ghana Ltd, 1990 Industry 790 916 0 126 ABC Brewery, 1994 Industry 718 350 368 0 West African Mills, 1993 Industry 600 446 154 0 Paper Conversions, 1994 Industry 289 143 146 0 Social Security Bank, 1995 Financial 2,000 1,500 500 0 GAFCO, 1995 Industry 850 770 80 10 Total 5,363 4,431 1248 326 Togo SNS, 1984 Industry 350 221 129 n.a. ITP, 1986 Industry 101 79 22 n.a. SORPLAIT (FAN MILK), 1986 Industry 40 72 n.a. 32 COMPLEXE SUCRIER DANIE Agro/industry 0 550 0 550 (SINTO), 1987 (Table continues on following page.) Appendix D (continued) Employment Employment, Country, name of enterprise, at 1st quarter Jobs and date of sale Sector privatization 1996 discontinued New recruits IOTO (NIOTO), 1987 Industry 169 131 38 n.a. SODETO, 1987 Industry 78 55 23 n.a. SOTOMA, 1987 Industry 144 25 119 n.a. SBHT, 1988 Trade 50 35 15 n.a. STB, 1989 Industry 159 145 14 n.a. ITT & TOGOTEX (TOGOTEX), 600 1990 Industry 1255 655 n.a. SOTEXMA, 1990 Agriculture 173 n.a. n.a. n.a. STALPECHE, 1990 Agribusiness 30 n.a. n.a. n.a. STH, 1990 Petroleum 164 87 77 n.a. LE MOTEUR (now AMINA- TOGO), 1991 Industry 0 160 0 160 SOTOPROMER, 1991 Agribusiness 7 n.a. n.a. n.a. TOGOROUTE (now ARI LA GAZELLE), 1991 Transport 68 50 18 n.a. HOTEL MIRAMAR (now KRIMASI), 1992 Service 0 30 0 30 SOTOTOLES, 1992 Industry 64 60 4 n.a. OTODI (now POINTE Industry 30 38 n.a. 8 D'IVOIRE), 1994 Total 2,882 2,338 1,114 780 Zambia AFE, 1993 Agriculture 225 90 135 n.a. Auto Care, 1993 Service 160 37 123 n.a. Chilanga Cement, 1994 Industry 751 700 51 n.a. Coolwell Systems, 1994 Industry 28 15 13 n.a. Eagle Travel Ltd, 1993 Service 105 87 18 n.a. Monarch Zambia Ltd, 1995 Industry 167 147 20 n.a. v Munkumpa Ipumbu Farm, 1995 Agriculture 287 235 52 n.a. National Breweries, 1995 Industry 560 560 0 n.a. Zambia Sugar, 1995 Agroindustry 3,765 3,765 0 n.a. Total 6,048 5,636 412 n.a. Note: The listed companies are a representative sample of privatized enterprises. n.a. = data not available. Source: Burkina Faso and Togo data are from World Bank country case studies. Benin, Ghana, and Zambia data are from London Economics, 1996. Appendix E Foreign Direct Investment and Privatization in Sub-Saharan Africa, 1989-95 (millions of U.S. dollars) 1995 1989 1990 1991 1992 1993 1994 (estimate) Total Total net FDI SSA' 2,540.6 861.8 1,810.0 1,482.5 1,751.4 2,989.4 2,137.9 13,573.6 Total net FDI from ten countriesb 2,155.2 903.7 847.2 1,050.4 1,587.4 2,315.5 1,021.7 9,881.2 Total value of FDI from privatization (ten countries)c 12.6 43.1 58.4 65.5 20.9 466.5 167.6 834.6 Share in total FDI 0% 5% 3% 4% 1% 16% 8% 6% Benin ~ Total FDI 1.3 1.8 13.0 7.0 10.0 5.0 1.0 39.1 m FDI from privatization - 12.0 24.1 35.9 - - - 71.9 Burkina Faso Total FDI - - - - - 3.0 3.4 6.4 FDI from privatization - - 2.0 - 2.9 - 4.9 Ghana Total FDI 15.0 14.8 20.0 22.1 125.0 233.0 230.0 659.9 FDI from privatization - 8.4 - - 316.0 15.0 339.5 Kenya Total FDI 62.2 57.1 18.8 6.4 1.5 9.0 22.0 177.0 FDI from privatization - - 3.9 - - - 3.9 Madagascar Total FDI 12.8 22.4 13.7 21.1 15.4 5.7 9.7 100.8 FDI from privatization - 13.3 - - - - 13.3 Mozambique Total FDI 3.4 9.2 22.5 25.3 30.0 33.0 36.0 159.4 FDI from privatization 12.0 12.0 12.0 58.0 58.0 152.0 Nigeria Total FDI 1,882.3 587.9 712.4 896.6 1,345.4 1,959.0 650.0 8,033.6 FDI from privatization - - - - - - - Togo Total FDI 16.6 7.7 11.6 18.9 1.8 6.3 2.0 64.8 FDI from privatization 0.6 19.1 0.6 - - - - 20.3 Uganda Total FDI (2.0) - 1.0 3.0 3.4 4.5 5.0 14.9 FDI from privatization - - 23.7 20.9 61.4 31.8 137.8 , Zambia 't Total FDI 163.6 202.8 34.3 50.0 55.0 60.0 66.0 631.7 FDI from privatization - - - - 28.2 62.8 91.0 a. Net in the sense of net of affiliated company transactions but not net of aggregate outflows. b. The Africa Live Data Base of the World Bank. c. Ten country case studies. Source: World Bank data. Appendix F Summary of World Bank Lending Operations Supporting Privatization in Sub-Saharan Africa, to End 1995 Number of credits Millions of U.S. dollars Project Project Total Total Total Program Program Program Program and and program program project credits credits credits credits sector sector credits, and sector Total approved, disbursed, approved, disbursed, credits credits Total Country SAL/SAC credits, TA credits credits SAL/SAC SAL/SAC TA TA approved disbursed credits Angola 0 1 2 3 0.0 0.0 23.0 5.0 78.7 18.7 101.7 Benin 3 0 2 5 140.0 121.2 0.0 0.0 22.1 8.4 162.1 Botswana 0 0 0 0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Burkina Faso 1 0 1 2 80.0 85.4 0.0 0.0 7.0 1.4 87.0 g Burundi 2 0 3 5 120.0 99.6 0.0 0.0 63.3 25.2 183.3 Cape Verde 0 2 0 2 0.0 0.0 12.3 5.4 0.0 0.0 12.3 Cameroon 2 1 1 4 225.0 228.0 10.2 0.0 34.6 13.8 269.8 Central African Rep. 3 0 2 5 98.2 96.8 0.0 0.0 77.0 58.7 175.2 Chad 0 0 1 1 0.0 0.0 0.0 0.0 18.6 13.1 18.6 Comoros 1 0 1 2 8.0 7.6 0.0 0.0 5.1 0.5 13.1 Congo, Dem. Rep. of 0 0 0 0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Congo, Rep. of 2 1 1 4 170.0 170.0 15.2 11.7 9.0 0.0 194.2 C6te d'Ivoire 1 1 1 3 17.0 10.9 15.0 3.8 80.0 80.0 112.0 Djibouti 0 0 0 0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Equatorial Guinea 0 0 0 0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Eritrea 0 0 0 0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Ethiopia 1 0 1 2 250.0 178.7 0.0 0.0 120.0 0.0 370.0 Gabon 1 1 0 2 50.0 50.0 5.0 4.8 0.0 0.0 55.0 Gambia, The 1 0 1 2 23.0 24.0 0.0 0.0 10.0 3.1 33.0 Ghana 1 1 4 6 120.0 119.8 10.5 9.4 232.5 144.8 363.0 Guinea 1 1 4 6 65.0 65.0 7.3 3.2 113.6 39.5 185.9 Guinea-Bissau 1 0 0 1 23.4 23.4 0.0 0.0 0.0 0.0 23.4 Kenya 0 2 2 4 0.0 0.0 29.3 8.4 121.4 100.6 150.7 Lesotho 0 1 0 1 0.0 0.0 11.0 1.1 0.0 0.0 11.0 Liberia 0 0 0 0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Madagascar 1 1 4 6 125.0 76.8 22.0 19.0 108.0 18.9 255.0 Malawi 0 1 2 3 0.0 0.0 22.6 1.9 33.8 5.6 56.4 Mali 0 0 3 3 0.0 0.0 0.0 0.0 91.8 36.2 91.8 Mauritania 0 1 3 4 0.0 0.0 10.0 9.9 105.0 101.8 105.0 Mauritius 0 0 1 1 0.0 0.0 0.0 0.0 30.5 0.0 30.5 Mozambique 1 1 3 5 180.0 143.0 74.3 12.7 247.4 41.3 501.7 Namibia 0 0 0 0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Niger 0 1 0 1 0.0 0.0 5.5 5.3 0.0 0.0 5.5 Nigeria 0 1 4 5 0.0 0.0 20.0 4.2 359.6 214.5 379.6 Rwanda 0 1 3 4 0.0 0.0 4.4 2.5 78.8 19.1 83.2 Sao Tome & Principe 1 0 1 2 9.8 6.4 0.0 0.0 9.8 4.6 19.6 Senegal 1 1 1 3 80.0 59.4 12.5 0.0 46.0 31.5 138.5 Sechelles 0 0 0 0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Sierre Leone 2 1 1 4 93.1 78.1 10.0 7.0 45.0 23.9 148.1 Somalia 0 0 0 0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 South Africa 0 0 0 0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Sudan 0 0 1 1 0.0 0.0 0.0 0.0 85.0 83.0 85.0 Swaziland 0 0 0 0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 Tanzania 0 3 7 10 0.0 0.0 84.4 25.2 894.3 531.4 978.7 Togo 2 0 1 3 100.0 83.6 0.0 0.0 11.5 2.2 111.5 Uganda 3 1 5 9 270.0 200.4 15.0 11.7 210.9 22.2 495.9 Zambia 1 2 5 8 210.0 208.9 31.0 21.4 664.0 535.6 664.0 Zimbabwe 1 0 0 1 125.0 125.0 0.0 0.0 0.0 0.0 125.0 TOTAL 34 27 72 133 2582.5 2262.0 450.5 173.3 4014.3 2179.5 6796.3 Source: Government privatization agencies and ministries and World Bank data. Appendix G International Donor Support Available for Privatization Programs in Sub-Saharan Africa, 1988-95 (millions of U.S. dollars) Donor agency 1988 1989 1990 1991 1992 1993 1994 1995a Total UNSPECIFIEDb (loans and grants) 101.2 1.8 147.3 62.8 - 147.3 241.5 - 701.9 Multilateral donor agencies Loans AfDB/ADF 151.6 39.0 54.1 40.0 - 60.4 83.4 - 428.5 EIB/EDF - 1.9 - - - 31.9 15.0 - 48.8 IFAD - 10.8 - 4.0 14.8 IMF - 15.8 2.6 11.5 - - - - 29.9 OPEC - 4.5 4.0 - - - - - 8.5 World Bank 1,029.7 836.9 716.2 813.3 1,286.9 672.6 848.5 592.2 6,796.3 Subtotal 1,181.3 908.9 776.9 868.8 1,286.9 764.9 946.9 592.2 7,326.8 Grants EEC/EC 17.3 11.3 34.4 79.9 - 83.4 97.9 0.5 324.7 UNDPC - 12.1 8.4 7.2 0.4 11.8 3.6 - 43.5 World Bank-managed Trust Funds 1.7 8.7 9.6 4.1 3.7 1.3 13.6 1.0 43.7 Subtotal 19.0 32.1 52.4 91.2 4.1 96.5 115.1 1.5 411.9 Bilateral donor agencies Loans Germany (KfW) - - - 20.0 18.6 - 25.8 50.0 114.5 Japan - - 1.8 2.9 - 1.1 1.0 3.8 10.6 Saudi Arabia-Saudi Fund 5.9 - - - - - - - 5.9 Subtotal 5.9 - 1.8 22.9 18.6 1.1 26.8 53.8 131.0 Grants Belgium 14.3 0.1 14.4 Canada (CIDA) - - - 13.0 - - 13.0 Denmark (DANIDA) - - 4.2 - - 15.2 4.0 - 23.4 Finland (FINNIDA) 17.9 17.9 France (CCCE/FAC) 23.9 1.6 47.6 61.1 - 5.7 - 2.2 142.1 Germany (KfW/GTZ) 6.0 20.4 110.2 55.7 102.7 71.9 38.8 59.2 464.8 Japan (JICA/OECF) 50.2 - 66.9 7.6 - 8.8 1.0 3.8 138.4 Netherlands - 7.3 45.4 4.5 - 0.5 - - 57.7 Norway (NORAD) - - 11.5 - 0.3 51.6 0.5 0.6 64.5 Saudi Arabia 4.9 4.9 Sweden (SIDA) 17.5 45.5 63.0 Switzerland (Swiss Fund) 4.0 4.8 15.7 12.0 - 7.2 - - 43.7 United Kingdom (ODA) - - 20.0 1.3 18.1 15.8 2.0 3.3 60.6 United States of America (USAID) 54.7 50.7 42.0 47.4 39.4 11.7 36.1 9.2 291.2 Subtotal 143.7 84.8 399.0 202.6 174.9 188.5 127.9 78.2 1,399.6 Total donor support 1,451.1 1,027.6 1,377.4 1,248.3 1,484.5 1,198.3 1,458.2 725.7 9,971.2 Total donor support excluding World Bank 421.4 190.7 661.2 435.0 197.6 525.7 609.7 133.5 3,174.9 Note: The data includes company financing to IDA for SAL/SAC, ERL/C and sectoral projects that support privatization activities. a. The information for 1995 is incomplete. b. Because the amount could not be broken down by donor agency, the amounts under individual donor agencies support may be understated. c. World Bank-managed trust funds included. Source: World Bank Lending Operations Database (LOD) and donor agencies. Notes 1. The database continues to be maintained, and information from it is now included in World Bank publications such as African Development Indicators and Global Development Finance (formerly World Debt Tables). 2. In addition to privatization, the term disengagement de I'Etat (state with- drawal) is used in some Francophone countries. 3. For example, C6te d'Ivoire has recently concluded a privatization deal for its telecommunications company. Under the contract deal, the company will have a monopoly for seven years. 4. Until recently, many funds were limited to investing in government bonds and property development. 5. Information from privatization agencies suggests that the average time from initiation to completion of a transaction is between 18 and 24 months. This is a long time in view of the preponderance of small enterprises and the fact that in some countries the major privatizations, especially public flotations, have been handled outside the central agency. The privatization of some enterprises has extended over many years, either because of half- hearted divestiture efforts or because negotiated deals could not be finalized and the process had to be repeated. 6. This law has been modified by Law 80-25, which stipulates that the bar against foreign ownership of land and buildings does not apply to the guarantees of foreign financial institutions. 7. Ordinance No. 75/49 (October 1975) is included in the appendix to Law 35/94. The latter forms the basis of the current privatization program. Although the ordinance has not been applied in connection with the pri- vatization program, its provisions on State nationalization and the limits on foreign investment conflict with the intent of the privatization law. 8. In Madagascar, government decrees banning all foreign ownership of land were passed in 1978 and 1982. In 1992, the government created an intermin- isterial committee to consider foreigners' requests to own land. The commit- tee was to base its decisions on socioeconomic, environmental, and geo- graphic factors, as well as the morality of investors. The committee never met and was eventually suspended. In 1995, the government enacted a law that clarified the situation. Foreigners can now lease land (bail emphyteo- tique) for a period of up to 50 years. 9. The large number of privatizations in Mozambique may distort the picture: all sales transactions are reported to have been sales of shares, when they were more likely to have been sales of assets. 10. Some observers consider that the net amount, after any retirement of enter- prise debt, is what counts. In our view, this does not represent the full 155 156 PRIVATIZATION IN AFRICA financial impact since, without privatization, a government would in most cases have to bear the enterprise debt, either because the debt is directly guaranteed by the government or because it is held by other government- owned enterprises. 11. The agency recommends that 15 percent of these funds be used to expand the power supply in selected state capitals and some newly created govern- ment areas, 5 percent to expand facilities of the River Basin Development Authority, 5 percent to the Nigeria Television Authority and Federal Radio Corporation of Nigeria, and 10 percent of the proceeds for a student educa- tion fund. 12. Fieldwork for the study was carried out by London Economics for the World Bank during February-May 1996. 13. Few persons availed themselves of the schemes. According to one estimate, a total of 111 former public enterprise employees benefited (World Bank 1996b). 14. Zambia was the one country case study in which not one interview revealed any concern about the transparency of the process; detractors based their opposition to the program on the perceived effect on employment, not on the way the process was handled. 15. A $2.46 million public information component is included in the current World Bank-supported Public Enterprise and Privatization Technical Assis- tance project. 16. There should be a few simple and nonconflicting objectives which are translated into quantifiable targets that can later be used to gauge the success of program outcomes. Bibliography Balassa, Bela.1987. "Public Enterprise in Developing Countries: Issu of Privat- ization." Development Research Department, Economics and Re. arch Staff. Washington, D.C.: World Bank. Bellanger, Serge. 1995. "Fin de Siecle: The Privatization of Banks around the World." The Bankers Magazine. Berg, Elliot. 1994. "Privatization in Sub-Saharan Africa: Results, Prospects and New Approaches." Consultant report. Development Alternatives Inc., Bethesda, Md. Berg, Elliot, M. Graham, D. Hogan, and C. Ludvik. 1994. "Privatization and Development Project: Midterm Evaluation." Consultant report. Develop- ment Alternatives Inc., Bethesda, Md. Bhatia, Anita. 1995. "World Bank Support for Privatization in Africa." Africa Region Technical Department. Washington, D.C. Brook Cowen, Penelope J. 1996. "The Guinea Water Lease-Five Years On: Lessons in Private Sector Participation." Public Policy for the Private Sector, Special Edition, June, pp 89-92. Callaghy, Thomas, M., and E. J. Wilson III. 1988. "Africa: Policy, Reality or Ritual?" In Raymond Vernon, ed., The Promise of Privatization. New York: Council on Foreign Relations. Campbell White, Oliver. 1993. "Parastatal Reform Programme: Initiating Re- port." Consultant report on the Tanzanian Privatization Program. Merlin International, London. . 1994. "Parastatal Reform Programme Implementation: End of Mis- sion Report." Consultant report on the Tanzania Program. Merlin Interna- tional, London. Candoy-Sekse, Rebecca, with Anne Ruiz Palmer. 1988. Techniques of Privatization of State-Owned-Enterprises. Vol. 3: Inventory of Country Experience and Reference Materials. World Bank Technical Paper 90. Washington, D.C. Center for Economic Growth, U.S. Agency for International Development. 1994. "Privatization in Africa: Lessons and Opportunities." Consultant report. Price Waterhouse, Washington, D.C. Dia, Mamadou. 1992. "The African Experience with Privatization." Division Notes. World Bank, Africa Technical Department, Institutional Development and Management Division, Washington, D.C. Ejangue, Theodore. 1995. "Selling the Privatization Program." Paper presented at World Bank seminar on The Challenge of Privatization in Africa. Private Sector Development Division. Washington, D.C. Galal, Ahmed. 1990. "Does Divestiture Matter? A Framework for Learning from Experience." Policy Research Working Paper 475. World Bank, Policy Research Department, Washington, D.C. 157 158 PRIVATIZATION IN AFRICA Galal, Ahmed, Leroy Jones, Pankaj Tandon, and Ingo Vogelsang. 1994. Welfare Consequences of Selling Public Enterprises: An Empirical Analysis. New York: Oxford University Press. Guislan, Pierre. 1997. The Privatization Challenge: A Strategic, Legal, and Institu- tional Analysis of International Experience. Regional and Sectoral Study. Wash- ington, D.C.: World Bank. IFC (International Finance Corporation). 1995. Privatization: Principles and Prac- tice. IFC Lessons of Experience Series 1. Washington, D.C. . 1996. Emerging Stock Markets: Factbook 1995. Washington, D.C. Kikeri, Sunita, John Nellis, and Mary Shirley 1992. Privatization: The Lessons of Experience. Washington, D.C.: World Bank. Lewis, Peter. 1995. "Banking Privatization in Nigeria: Background, Context and Prospects." Consultant report. American University, Washington, D.C. London Economics. 1996. "The Impact of Privatisation on Labour in Africa." Report for the World Bank. Lopez-de-Silanes, Florecio. 1994. "Determinants of Privatization Prices. (Mex- ico)." Harvard University, Cambridge, Mass. Morin, Pierre. 1993. "Senegal: Towards the Next Stages in PE Reform and Privatization," World Bank, West Africa Department, Industry and Energy Division, Washington, D.C. Nankani, Helen. 1988. Techniques of Privatization of State-Owned-Enterprises. Vol. 2: Selected Country Case Studies. World Bank Technical Paper 89. Wash- ington, D.C. Orr, Daniel, and Thomas S. Ulen. 1993. "The Role of Trust and Law in Privatiza- tion." Quarterly Review of Economics and Finance 33 (Special issues). Sader, Frank. 1993. "Privatization and Foreign Direct Investment in the Devel- oping World, 1988-92." Policy Research Working Paper 1202. World Bank, International Economics Department, Washington, D.C. . 1994. "Privatizing Public Enterprises and Foreign Investment in Developing Countries, 1988-93." Foreign Investment Advisory Service Oc- casional Paper 5. World Bank, Washington, D.C. Shaikh, Hafeez, and Maziar Minovi. 1995. "Management Contracts: A Review of International Experience." CFS Discussion-Paper 108. World Bank, Cofi- nancing and Financial Advisory Services (Privatization Group), Washing- ton, D.C. Swanson, Daniel. 1993. "Public Enterprise Sector Reform: Case Studies for Cameroon, Mauritania, Malawi, Cote d'Ivoire, Niger, Mozambique, and Burundi." Background papers for World Bank, Adjustment in Africa: Reforms, Results, and the Road Ahead. New York: Oxford University Press, 1994. Swanson, Daniel, and Teffera Wolde-Semait. 1989. Africa's Public Enrerprise Sector and Evidence of Reforms. World Bank Technical Paper 95. Washington, D.C. Tanzania, Presidential Parastatal Sector Reform Commission. 1993a. "Parastatal Reform Programme: Initiating Report on Major Issues for Privatization and Restructuring." Dar es Salaam. . 1993b. "Parastatal Privatization and Reform: Master Plan." Dar es Salaam. . 1994. "Annual Report and Accounts for the Year ended June 30, 1994." Dar es Salaam. BIBLIOGRAPHY 159 Triche, Thelma A. 1990. "Private Participation in the Delivery of Guinea's Water Supply Services." Policy Research Working Paper 477. World Bank, Infrastructure and Urban Development Department, Washington, D.C. United Nations. 1995. World Investment Report 1995, Transnational Corporations and Competitiveness. New York. UNCTAD (United Nations Conference on Trade and Development). 1994. Final Report on the Ad Hoc Working Group on Comparative Experiences with Privatiza- tion to the Trade and Development Board. Geneva. .1996. Privatization in Africa: An Overview. Geneva. Vuylsteke, Charles. 1988. Techniques of Privatization of State-Owned-Enterprises. Vol. 1: Methods and Implementation. World Bank Technical Paper No. 88. Washington, D.C. World Bank. 1993a. "Regional Study on Public Enterprise Reform and Privatiza- tion in Africa." Africa Technical Department, Washington, D.C. . 1993b. "World Bank Assistance to Privatization in Developing Countries." Operations Evaluation Department, Washington, D.C. . 1995a. Bureaucrats in Business: The Economics and Politics of Govern- ment Ownership. New York: Oxford University Press. . 1995b. "A Continent in Transition: Sub-Saharan Africa in the Mid- 1990s." Africa Region, Washington, D.C. . 1996a. Private Sector Development in Low-Income Countries. Develop- ment in Practice. Washington, D.C. =.1996b. "Privatization in Benin." Africa Technical Department, Wash- ington, D.C. . 1996c. "Privatization in Burkina Faso." Africa Technical Depart- ment, Washington, D.C. . 1996d. "Privatization in Ghana." Africa Technical Department, Washington, D.C. . 1996e. "Privatization in Kenya, 1992-95." Africa Technical Depart- ment, Washington, D.C. . 1996f. "Privatization in Madagascar." Africa Technical Department, Washington, D.C. . 1996g. 'Privatization in Mozambique." Africa Technical Depart- ment, Washington, D.C. . 1996h. "Privatization in Nigeria." Africa Technical Department, Washington, D.C. . 1996i. "Privatization in Togo." Africa Technical Department, Wash- ington, D.C. . 1996j. "Privatization in Uganda." Africa Technical Department, Washington, D.C. . 1996k. "Privatization in Zambia, 1993-95." Africa Technical Depart- ment, Washington, D.C. . 19961. "Symposium on Power Sector Reform and Efficiency Im- provement in Sub-Saharan Africa." Report no. 182/96. Industry and En- ergy Department. . 1996m. World Debt Tables. Washington, D.C. . Various years. African Development Indicators. Washington, D.C. Directions in Development Begun in 1994, this series contains short essays, written for a general audience, often to summarize published or forthcoming books or to highlight current development issues. Africa's Management in the 1990s and Beyond: Reconciling Indigenous and Transplanted Institutions Are Financial Sector Weaknesses Undermining the East Asian Miracle? Building Human Capital for Better Lives Class Action: Improving School Performance in the Developing World through Better Health and Nutrition Counting the Full Cost: Parental and Community Financing of Education in East Asia Decentralization of Education: Community Financing Decentralization of Education: Demand-Side Financing Decentralization of Education: Legal Issues Decentralization of Education: Teacher Management Deep Crises and Reform: What Have We Learned? Early Child Development: Investing in the Future Everyone's Miracle: Revisiting Poverty and Inequality in East Asia Financing Health Care in Sub-Saharan Africa through User Fees and Insurance Gender and Law-Eastern Africa Speaks: Conference Organized by The World Bank and The Economic Commission for Africa Global Capital Supply and Demand: Is There Enough to Go Around? Implementing Projects for the Poor: What Has Been Learned? Improving Early Childhood Development: An Integrated Program for the Philippines (with a separate supplement) (continued on the following page) Directions in Development (continued) India's Family Welfare Program: Moving to a Reproductive and Child Health Approach (with a separate supplement) Investing in People: The World Bank in Action Livable Cities in the 21st Century Managing Commodity Booms - and Busts Meeting the Infrastructure Challenge in Latin America and the Caribbean Monitoring the Learning Outcomes of Education Systems MIGA: The First Five Years and Future Challenges Nurturing Development: Aid and Cooperation in Today's Changing World Nutrition in Zimbabwe: An Update Poverty Reduction in South Asia: Promoting Participation of the Poor Private and Public Initiatives: Working Together for Health and Education Private Sector Participation in Water Supply and Sanitation in Latin America Reversing the Spiral: The Population, Agriculture, and Environment Nexus in Sub-Saharan Africa (with a separate supplement) A Strategy for Managing Water in the Middle East and North Africa Safety Net Programs and Poverty Reduction: Lessons from Cross-Country Experience Taxing Bads by Taxing Goods: Pollution Control with Presumptive Charges Toward Sustainable Management of Water Resources Trade Performance and Policy in the New Independent States The Transition from War to Peace in Sub-Saharan Africa Unshackling the Private Sector: A Latin American Story The Uruguay Round: Widening and Deepening the World Trading System Water Markets in the Americas THE WORLD BANK 1818 H Street, N.W. Washington, D.C. 20433, U.S.A. i'E'l.lHONI: (202) 477-1234 FACS11MILE: (202) 477-6391 '1'1FI.t.X: MC.I 64145 WORIL)BANIK MCI 248423 WORLDBANEK WVORILD) 'I[)1WEB'FB: http://wwwv.worldbank.org/ E-NAII.: booksO@worldbank.org 1 3 9 78 9 780821 339787 COVER DESIGN: THE MAGAZINE GROUP ISBN 0-8213-3978-E