The Changing Role of Government: The Reform of Public Services in Developing Countries (The Role of Government in Adjusting Econ)

The Changing Role of Government The Reform of Public Services in Developing Countries Richard Batley and George Larbi ...

Author: Richard Batley | George A. Larbi

30 downloads 1393 Views 692KB Size Report

This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!

Report copyright / DMCA form

DOWNLOAD PDF

The Changing Role of Government The Reform of Public Services in Developing Countries

Richard Batley and George Larbi

The Changing Role of Government

The Role of Government in Adjusting Economies General Editor: Professor Richard Batley, International Development Department, School of Public Policy, The University of Birmingham Over the last two decades there has been a strong emphasis on reducing the role of government and on reforming traditional public sector bureaucracies. The new conventional view has become that, where possible, services should not be provided directly by government but be contracted out or privatized. Where this is not possible, the predominant view has been that the public sector itself should change by setting up semi-autonomous agencies and by making public management more performance and customer-oriented. This series investigates the application of such reforms in Africa, Asia and Latin America. Underlying the enquiry is the question whether reforms, which were initially conceived in countries such as Britain and New Zealand are appropriate in other contexts. How much sense do they make where levels of public management capacity, market development, resources, political inclusiveness, legal effectiveness, political and economic stability are quite different? To investigate these issues, the series covers four service sectors selected to be representative of types of public sector activity – health care, urban water supply, agricultural marketing services and business development services. Titles include: Richard Batley and George Larbi THE CHANGING ROLE OF GOVERNMENT The Reform of Public Services in Developing Countries Mike Hubbard DEVELOPING AGRICULTURAL TRADE New Roles for Government in Poor Countries Paul Jackson BUSINESS DEVELOPMENT IN ASIA AND AFRICA The Role of Government Agencies Anne Mills, Sara Bennett and Steven Russell THE CHALLENGE OF HEALTH SECTOR REFORM What Must Government Do? Andrew Nickson and Richard Franceys TAPPING THE MARKET The Challenge of Institutional Reform in the Urban Water Sector The Role of Government in Adjusting Economies Series Standing Order ISBN 0–333–94618–9 (outside North America only) Please contact your bookseller or, in case of difﬁculty, write to us at the address below with your name and address, the title of the series and the ISBN quoted above. Customer Services Department, Macmillan Distribution Ltd, Houndmills, Basingstoke, Hampshire RG21 6XS, England

The Changing Role of Government The Reform of Public Services in Developing Countries Richard Batley and George Larbi University of Birmingham, UK

© Richard Batley and George Larbi 2004 All rights reserved. No reproduction, copy or transmission of this publication may be made without written permission. No paragraph of this publication may be reproduced, copied or transmitted save with written permission or in accordance with the provisions of the Copyright, Designs and Patents Act 1988, or under the terms of any licence permitting limited copying issued by the Copyright Licensing Agency, 90 Tottenham Court Road, London W1T 4LP. Any person who does any unauthorised act in relation to this publication may be liable to criminal prosecution and civil claims for damages. The authors have asserted their rights to be identiﬁed as the authors of this work in accordance with the Copyright, Designs and Patents Act 1988. First published 2004 by PALGRAVE MACMILLAN Houndmills, Basingstoke, Hampshire RG21 6XS and 175 Fifth Avenue, New York, N.Y. 10010 Companies and representatives throughout the world PALGRAVE MACMILLAN is the global academic imprint of the Palgrave Macmillan division of St. Martin’s Press, LLC and of Palgrave Macmillan Ltd. Macmillan® is a registered trademark in the United States, United Kingdom and other countries. Palgrave is a registered trademark in the European Union and other countries. ISBN 0–333–73617–6 This book is printed on paper suitable for recycling and made from fully managed and sustained forest sources. A catalogue record for this book is available from the British Library. Library of Congress Cataloging-in-Publication Data Batley, Richard. The changing role of government : the reform of public services in developing countries / Richard Batley and George Larbi. p. cm. – (The role of government in adjusting economies) Includes bibliographical references and index. ISBN 0–333–73617–6 (cloth) 1. Public administration–Developing countries. 2. Municipal service–Developing countries. I. Larbi, George A. II. Title. III. Series. JF60.B385 2004 351.172’4–dc22

10 13

9 12

8 11

7 10

2003069642

6 09

5 08

4 07

3 06

2 05

1 04

Printed and bound in Great Britain by Antony Rowe Ltd, Chippenham and Eastbourne

Contents List of Tables

viii

List of Figure and Boxes

ix

List of Acronyms and Abbreviations

x

Preface

xi

About the Authors 1

xiii

Changing Views of the Role of the Government Introduction The role of government in development The background to public service reform in developing countries Capacity to perform the ‘new’ roles of government An overview of the study Conclusion

7 15 20 28

2

Changing Approaches to Public Sector Management Introduction Theoretical basis of new approaches to public management Pragmatic rationales for public sector reforms From old public administration to new public management The era of new public management A review of new public management in practice Conclusion

31 31 31 38 39 40 44 52

3

The Politics of Service Reform Introduction Political stakes in the reform of public management Principals and agents in public service reform External bodies – agents or principals? The principal-agency characteristics of services The ‘principals’ – citizens and political leaders The ‘agents’ – public service administrators, professionals and workers Emerging constituencies for change Conclusion

54 54 56 58 58 62 65

v

1 1 2

71 76 78

vi Contents

4

Decentralizing Organizational Arrangements for Service Delivery Introduction Pre-reform organizational arrangements Types of organizational arrangements Reform of organizational arrangements Have reforms improved performance? Institutional and capacity issues Conclusion

81 81 81 85 88 98 102 104

5

The Experience of Charging for Public Services Introduction Explaining user charges Charging – what and how? Charging: performance and impact Conclusion

106 106 106 110 116 123

6

Working with Private Partners The case for privatizing management Organizational arrangements for service provision The application of public-private arrangements for service delivery Matching organizational arrangements to capacity Conclusion

125 125 128

7

The Experience of Contracting Introduction The use of contractual arrangements Assessment of the experience of contracting Capacity to contract Conclusion

144 144 145 158 167 177

8

Regulating and Enabling the Private Sector Introduction Forms of regulation De-regulation and support to market development Re-regulation under the new management Economic regulation of urban water supply The social regulation of health services Enablement of private providers – business support services A comparison of the experience of regulation and enabling

180 180 181 184 185 186 195 203 215

133 136 142

Contents vii

9

Conclusion Introduction Summary Performance and capacity constraints in speciﬁc reforms Implications for capacity development Conclusion

219 219 220 224 232 235

Notes

238

Bibliography

240

Index

253

List of Tables 1.1 1.2 1.3 2.1 3.1 3.2 3.3 4.1 5.1 6.1 6.2 7.1 7.2

7.3 7.4 8.1 8.2 8.3 8.4 8.5 9.1 9.2

Economic and social indicators in the study countries Human Development Index trends from HDR 2003 Sector characteristics and the motive for intervention Key elements of NPM by different authors Reform types analyzed by sector The public and bureaucratic arenas of response and resistance to reform Principal-agent relationships by service Summary of performance in urban water supply Cost recovery as a percentage of recurrent health expenditure From bureaucratic administration to charging, short-term contracts and the market Distribution of functions in different organizational arrangements Contract types and examples mentioned in the text Evidence on health policy outcomes: comparison of contracted-out service with directly-provided government service The feasibility of contract types Key constraints on governments’ capacity to contract Forms of regulation and enablement of the private sector Actors involved in regulation in Tamil Nadu, India Key regulatory statutes in Sri Lanka and Zimbabwe The relationship of business support services with government and the private sector Key constraints on governments’ capacity to regulate and enable private ﬁrms Summary of performance of organizational reforms Key constraints on governments’ capacity to perform new management roles

viii

12 13 25 42 55 57 64 101 112 125 130 145

159 166 169 182 198 199 207 217 225 227

List of Figure and Boxes Figure 1.1

A framework for analyzing capacity

19

Boxes 1.1 Market failure arguments 3.1 Line ministries have often had little inﬂuence on sector liberalization 4.1 Corporatization of the water sector in Ghana 5.1 Addressing equity and efﬁciency in tariff reforms: the case of Bulawayo 6.1 Public-private organizational arrangements 7.1 A service contract – an example from Santiago, Chile 7.2 Contracting for the services of private GPs in South Africa 7.3 A concession contract – an example from Buenos Aires, Argentina 8.1 Models of price regulation 8.2 Boundary problems in water regulation in Buenos Aires 8.3 Ineffectual regulation of private health providers in Zimbabwe 8.4 The Council for Scientiﬁc and Industrial Research, India 8.5 Zimtrade, the export promotion agency of Zimbabwe

ix

24 73 93 120 132 150 152 155 189 192 201 209 212

List of Acronyms and Abbreviations BOT CWE DfID DWR ESAP FCD GDP GNP GWC HDI IFIs IMF LBA MDG MTM NHS NWDB NPM OECD SAP UNDP USAID

Build-Operate-Transfer Cooperative Whole Establishment (Sri Lanka) Department for International Development (UK) Department for Water Resources (Zimbabwe) Economic Structural Adjustment Programme Food Commissioners Department (Sri Lanka) Gross Domestic Product Gross National Product Ghana Water Company Human Development Index International Financial Institutions International Monetary Fund Licensed Buying Agency (Ghana) Millennium Development Goals Market-type Mechanism National Health Service (UK) National Water and Drainage Board (Sri Lanka) New Public Management Organization for Economic Co-operation and Development Structural Adjustment Programme United Nations Development Programme United States Agency for International Development

x

Preface This is the ﬁnal volume of a series on the reform of service delivery in developing countries. The research on which the ﬁve volumes are based was funded by the economic and social research fund of the British Department for International Development (DfID), but the facts presented and the views expressed here are those of the authors and do not necessarily reﬂect the views of DfID. Each of the four previous volumes has examined a particular area of government activity: health care, urban water supply, business development, and services to support agricultural trade. A broad pattern of reform has affected these different sectors: liberalization, the introduction of private sector management approaches, charging for services, and new forms of working with the private sector. The leading question running across the ﬁve volumes is whether these approaches are appropriate to the context and capacities of developing countries. This ﬁnal volume draws together the strands, comparing the experience between these sectors in selected developing countries of Africa, South and Southeast Asia, and Latin America. The research programme involved research groups from ﬁve British universities: the International Development Department of the School of Public Policy, University of Birmingham; the Health Economics and Financing Programme, Health Policy Unit, London School of Hygiene and Tropical Medicine; the Overseas Development Group of the School of Development Studies, University of East Anglia; the Water, Engineering and Development Centre, Loughborough University of Technology; and the Department of City and Regional Planning, University of Wales. The overall research programme was coordinated by Richard Batley of the International Development Department, University of Birmingham. The health sector research was developed and organized by Sara Bennett, Anne Mills and Steven Russell, with support in the undertaking of country cases by: A. Asamoa-Basah, Nimal Attanayake, Siﬁso Chikandi, Charles Hongoro, Philemon Kwaramba, V. R Muraleedharan, Paul Smithson, Anuwat Supachutikul and Viroj Tangcharoensathien. The water sector research was led by Richard Franceys and Andrew Nickson, with support on country studies by Philip Amis, Frank Asiedu, Hartash Aturupane, Richard Batley, Srinivas Chary, Ngoni Mudege and xi

xii Preface

Kevin Sansom. The research on business development services was led by Paul Jackson and Richard Slater, with support in country studies by R. Basu, T. K. Bhaumik, Sunil Chandrasiri, Asoka Gunawardena, Ram Khanna and George Larbi. In the case of the agricultural sector, the research was led by Michael Hubbard and Marisol Smith with country studies by Frank Ellis, R. Kohli, D. Atse, S. Chikandi, G. Onumah, J. Rusike, Andrew Shepherd and C. Sukume. User surveys on the health and water sectors were undertaken by Carole Rakodi with the support of Steven Russell, Tudor Silva and Dorothy Mutizwa-Mangiza. Together this team produced 41 papers and case studies. This book was written jointly by Richard Batley and George Larbi, but it has depended greatly on the work undertaken by the whole team and particularly by the authors of the previous four volumes: – Sara Bennett, Anne Mills and Steven Russell, The Challenge of Health Sector Reform: What Must Governments Do? – Paul Jackson, Business Development in Asia and Africa: The Role of Government Agencies – Michael Hubbard, Developing Agricultural Trade: New Roles for Government in Poor Countries – Andrew Nickson and Richard Franceys, Tapping the Market: The Challenge of Institutional Reform in the Urban Water Sector We gratefully acknowledge their contribution to our thinking.

About the Authors Richard Batley is professor of development administration in the International Development Department of the School of Public Policy, the University of Birmingham. His research interests are in governance, service delivery and urban policy in India, Latin America and Africa. George Larbi is senior lecturer in public sector management in the International Development Department of the School of Public Policy, the University of Birmingham. His research interests are in governance (including corruption and ethics), new approaches to public sector management and service delivery. He has experience in Africa, the United Kingdom and Kazakhstan. He is the assistant editor of the journal Public Administration and Development.

xiii

1 Changing Views of the Role of the Government

Introduction Over the last two decades there has been emphasis, particularly in the English-speaking advanced countries, on reducing the role of government and on reforming public management by adopting aspects of private sector practice. The research programme on which this study is based was concerned with the fact that similar practices were being introduced in developing and transitional countries, often in association with economic adjustment. There has been considerable research on the difﬁcult process of adjustment but little on the process and outcomes of public management reform for improved service delivery. The research investigated the application of such reforms outside their countries of origin, in Africa, Asia and Latin America. Underlying the enquiry is the question whether approaches generated from a diagnosis of the ‘over-interventionist’ state in, say, the United Kingdom or New Zealand are appropriate responses in states where the levels of public management capacity, market development, resources, political inclusiveness, legal effectiveness, political and economic stability are quite different. Governments may be ill-equipped to adopt unfamiliar approaches to public service provision, where the institutional conditions on which the new management practices are premised may not be present. This chapter will describe the backdrop to these new management approaches in two respects. First, it will survey the broad shifts over the last 50 years in practical understanding of the role of government, particularly as this applies to developing countries. It will then focus in more detail on the underlying situation of the countries that faced radical reform of their public services, on the social, political and 1

2 The Changing Role of Government

managerial strains to which they were already subject. This will lead to an account of the questions that this volume will address and the approach that it will adopt. The following chapter will describe the new approaches to managing service delivery.

The role of government in development The history of the ﬁrst three quarters of the twentieth century can be seen as one where states held clear authority within their borders, and governments acquired growing functions which they tended to perform on their own. The latter part of the twentieth century and early twenty-ﬁrst century have seen the emergence of a more porous view of the nation state, and changed views of the role of government: it would perform fewer functions on its own and more in partnership with other actors. Associated with this change in public policy about what the state was to do was a consequential change in how it was to act. The extension of the state in developing countries The point of reference for the countries which became independent after the second world war was of an extensive state, whether in the form of the socialist or of the advanced capitalist nations. This was partly a matter of policy transfer from the more developed capitalist and socialist countries in their race to ﬁll the post-colonial ‘vacuum’. In the 1950s and 1960s, aid programmes and academic advisers propagated the idea of the state bureaucracy as the lead agent for the transition to what was then known as ‘modernization’ (Stone, 1965; Sifﬁn, 1976). Aid agencies favoured large-scale projects of industrial and agricultural development which, in their turn, required the guarantee of government involvement (Esman, 1988). On the side of the new political élites, the idea of state-led development was attractive: it apparently offered a way of satisfying popular expectations and demonstrating a concern with social justice and development. The case for state-led development was not only imitative but also built on a response to local circumstances. Where market institutions and indigenous entrepreneurs were weak, often only state enterprises were capable of investing or taking over foreign-owned plants. In the case of developing countries linked to the international economy through primary goods exports, the case was made for the state to take the lead in re-structuring the economy towards ‘inward directed’ industrial development on the basis of import substitution (Todaro, 1994).

Changing Views of the Role of the Government 3

Until the mid-1980s, the role of the state and employment in the public sector generally expanded faster in the advanced countries than they did among developing countries, except in the case of the oil-rich nations of the Middle East, Mexico, Venezuela and Nigeria. By the end of that period, the proportion of Gross Domestic Product (GDP) spent by central government in the rich Organization for Economic Co-operation and Development (OECD) countries was on average 47 per cent, compared with less than 25 per cent in all developing countries outside the Middle East (World Bank, 1997). On average, public sector employment (in government and public enterprises) occupied a higher proportion of the total population of afﬂuent (6.4%) than poorer (2.6%) countries. However, while in the rich countries more public sector workers were involved in the delivery of services, in developing countries there was a far higher proportion in public enterprises (Rowat, 1990). The ‘statist’ model was quickly subject to criticism. Until the full force of the neo-liberal perspective arose in the 1980s, this was not a matter of ideological opposition to the extension of the state as such. It was rather a question whether the states and public administrations of developing countries were structurally (not just technically) capable of acting as the agent of development. This raised issues of the nature of bureaucracy and the social composition of the state. An extensive critical literature will here be reduced to three themes. They overlap but derive from different theoretical positions and emerged from different regional contexts. The ﬁrst came from the early (1950s and 1960s) experience of operating aid programmes across a wide geographical spectrum but particularly in Africa, South and South-East Asia. Critics challenged the view that public administration could act as an agent of development. They argued that bureaucracies were biased to stability or only incremental change, were anti-developmental and suppressed entrepreneurial interests. Moreover, public sector agencies, which on the surface looked like rational bureaucracies, in practice often served particular interests (Hyden, 1983). The second theme has been applied particularly, but not only, in Sub-Saharan Africa. It refers to the ‘softness’ or ‘weakness’ (Migdal, 1988) of the state, in which the state apparatus maintains only a tenuous hold over society, lacking legitimacy and therefore the capacity to enforce policy. This is often associated with personal rule where leaders exercise their own interests through the ofﬁcial apparatus and patron-client networks, or resort to control through ‘hard’ militaristic régimes.

4 The Changing Role of Government

The third theme, which emerged from neo-marxism and dependency theory, saw the apparatus of the state (the bureaucracy and military) as being subordinated to non-national interests, particularly international capital. This approach was used to explain the rise of various forms of authoritarian rule in the 1970s – bureaucratic authoritarianism in Latin America (Collier, 1979), and the postcolonial ‘over-developed’ state in South Asia. Far from being an agent of development, the state was seen as an agent of underdevelopment or distorted development. By the end of the 1970s, with international recession, growing national debt in Latin America and Africa and widespread military régimes, there was deep pessimism about the scope for development and scepticism about the state’s role. Two ways forward became evident in the 1980s – the example of the newly industrializing countries of East Asia and the neo-liberal model developed originally in the United Kingdom and the United States. The East Asian ‘newly industrializing countries’ demonstrated, contrary to left radical thinking of the 1970s, that it was possible for poor nations to develop in a sustained and inclusive way whilst engaging with the international economy. Contrary to Right liberal thinking of the 1980s, they also showed that governments could play a positive role in achieving development. State administrations had authority and capacity to give direction to market development, respond ﬂexibly to private sector needs, develop technology and human resources, while controlling labour. Wade (1990) describes this approach as one of a ‘governed market’ rather than either a free market or a command economy. While these arrangements were adequate for the management of relatively closed economies, they could not cope with the global liberalization of ﬁnancial and capital markets in the 1990s (Flynn, 1999). Destabilizing capital ﬂows overwhelmed established institutions and contributed to the East Asian ﬁnancial crisis of 1998 (Watkins, 1998; Stiglitz, 1998). The ‘Asian model’ seemed now to be a less convincing alternative to the neo-liberal consensus. Neo-liberalism and economic adjustment The immediate roots of the neo-liberal revival lay in the economic crisis which followed the massive rise in oil prices in the early 1970s. The United Kingdom and USA, with histories of low investment and high trade deﬁcits, were particularly badly affected by economic recession. The United Kingdom, in effect, experienced the ﬁrst ‘structural adjustment’ programme when, in 1976, it negotiated a loan from the

Changing Views of the Role of the Government 5

International Monetary Fund. In return it had had to accept public expenditure cuts, divestiture of public enterprises, a ﬂoating exchange rate and restraints on money supply. These became the principal elements of the structural adjustment programmes that were later applied globally. Many countries, particularly in Latin America but also central Europe and Africa, were slow to adjust to the new economic order and got themselves into deep difﬁculties of debt and inﬂation. They eventually had to resort to the International Monetary Fund and World Bank for loans. These two international ﬁnancial institutions became the main propagators of the ‘Washington consensus’, a panoply of precepts to do with the liberalization, privatization and stabilization of economies, and the reduction of the role and scale of the public sector. Earlier criticisms of the state’s role in development had questioned whether the conditions existed for ‘third world’ states to operate effectively. In the absence of these conditions, the tendency was for state agents to become self-serving, patrimonial and linked to powerful, often foreign, interests. The new liberal view held a deeper scepticism about the capacity of state administration to play a developmental role under any circumstances (Friedman, 1962). The bias was now against state expansion and for the market: the motto was ‘public provision, the exception not the rule’ (World Bank, 1996). Some broad critical points can be made about the experience of adjusting the role of the state in developing countries (Mosley, Harrigan and Toye, 1995; Turner and Hulme, 1997). First is the question of ‘ownership’ – who conceives, drives and supports reform? Advanced countries, such as the United Kingdom, New Zealand and Australia, were able to mould the pace and nature of their responses and did so with the compliance of their electorates. Developing and post-socialist transitional countries, on the other hand, have usually made their adjustment under strong external pressure from the international ﬁnancial institutions and major donor countries. Their governments have often not been the chief architects of programmes, understood their implications or gained an electoral mandate for change, as the ex-vice president of the World Bank has himself argued (Stiglitz, 1998). Second is the question of uniformity of application. The main tenets of state withdrawal and adjustment, ﬁrst developed to adjust to ﬁscal crisis in advanced welfare states, were applied to countries with very different circumstances whose only common feature was debt: postsocialist central and east European states, post-authoritarian Latin

6 The Changing Role of Government

American states, weak states in Africa and bureaucratic states in South Asia. Variations in programmes have resulted less from design and more from failures to honour loan conditions (Mosley, Harrigan and Toye, 1995). Following from this, the third point is that adjustment has often been implemented in an unbalanced way, leading to deeper crisis and poverty at least in the short term. Quick economic policy changes affecting consumption and the distribution of income – ﬂoating exchange rates, freeing trade, cutting public expenditure and subsidies – have often been made. Structural changes affecting production – reforming state administration, privatizing, reforming the ﬁnancial sector, building agricultural and industrial markets, providing basic services and infrastructure – have often stalled or taken longer (Mosley, Harrigan and Toye, 1995; Turner and Hulme, 1997; World Bank, 1997, p. 13). The pain was felt before the (uncertain) gain. The adjustment of adjustment Partly in response to the experience of liberalization, during the 1990s the international reform agenda was modiﬁed in various ways which are important to this study to include new focuses on poverty reduction as well as efﬁciency, and state capacity-building as well as market development. First, the goals of reform were extended from freeing market forces and making economies efﬁcient to directly addressing problems of poverty. This was initially in the shape of add-on policies to provide social safety nets and employment to the people adversely affected by economic liberalization. By the end of the 1990s, in the case at least of the most highly indebted countries, the commitment to the poverty reduction objective was at the core of reform. The support of donors and governments was to be coordinated through agreed ‘international development targets’ and ‘comprehensive development frameworks’ whose main mechanism was country-speciﬁc ‘poverty reduction strategies’. In principle these are nationally owned but they have to be approved by donors and international ﬁnancial institutions (Panos Institute, 2002). Second, the early 1990s saw recognition that market development and poverty reduction depended on effective states: ‘State-dominated development has failed, but so will stateless development’ (World Bank, 1997, p. 25). The earlier stages of structural adjustment had addressed the state essentially as something to minimize by cutting back the responsibilities, expense and size of the public sector. The

Changing Views of the Role of the Government 7

new agenda recognized that, while there may be too much state intrusion in the economy, there was also often too little government capacity to make policy, perform basic administrative functions, work with private partners, and ensure the provision of infrastructure and public services. There was ‘too much state’ and ‘too little state’ at the same time (Larbi, 1999). ‘Only after a decade of experimentation with reducing government did economic reformers become explicit about the importance of strengthening government’ (Grindle, 1997, p. 4). Failures in structural adjustment contributed to a renewed concern with the capacity of states. Brinkerhoff (1994) describes the ﬁrst round of structural adjustment as having been put in place by technocratic élites, while the second round gave more attention to the transformation of the ‘hard to change institutions which structure political and economic life’ (World Bank, 2000). These institutions include clear property rights, the rule of law, ﬁnancial systems, an active civil society, accountable government, and efﬁcient and effective public administration (Burki and Perry, 1998). While softening the earlier anti-statism of structural adjustment, this new focus paradoxically implied an extension of the liberal agenda into government itself. Whereas the ﬁrst wave of reform had been about reversing government growth, freeing markets from regulation and privatizing state-owned enterprises, the more recent waves have also been about ‘liberalizing’ government itself, including the management of public services. This book is about the experience of transforming governments to work with the private and non-governmental sectors and of introducing market-style reforms into public management. It considers the performance of such reforms in terms of their effects in bringing about greater efﬁciency, increasing public service access for the poor and stimulating market development. It particularly addresses the questions whether governments have adequate capacity and whether the necessary institutional conditions are present to enable these styles of operation to work.

The background to public service reform in developing countries The contribution of this book The world-wide trend to the redeﬁnition of public and private roles, and reform of government administration on neo-liberal lines has continued for 20 years, but there has been little systematic and comparative evaluation of the process of reform or of its outcomes. Most of

8 The Changing Role of Government

what exists is speciﬁc to particular service sectors or is based on individual country case studies (see Manning, 2001 for a survey). This study embraces several service sectors and countries, selected to maximize their representativeness and the possibility of comparison. It draws on a series of global sector surveys and country case studies undertaken between 1996 and 2000, and on four books which have already resulted from the research on particular sectors: Mills, Bennett and Russell (2001), Jackson (2002), Hubbard (2003), and Nickson and Franceys (2003). The research focused on four core countries, Zimbabwe, Ghana, Sri Lanka and India, but it also made reference to other countries of Latin America, Africa and South and Southeast Asia. The reasons for this choice and an account of the research approach are described later in this chapter. This section goes on ﬁrst to describe the circumstances and context in which developing countries, and particularly those that are the focus of this study, confronted the demand to reform their public services. The institutional background to reform The period from the late 1980s until the present has been one of attempts at extraordinarily radical public sector reform. They challenged that previous great period of radical reform in the 1950s and 1960s in South Asia and Africa (or the 1930s to 1950s in Europe and Latin America), which had established the responsibility of the state for social services and economic development. Developing countries shared in a practically world-wide convention in favour of direct public ownership or state management as the preferred model of intervention. Post-colonial governments added redistributive and nationbuilding intentions whose interventionism was often enhanced by commitments to reversing colonial inequalities, to state socialism and national planning. Economically, this model of direct state intervention was fragile, being even less ﬁscally sustainable in the face of economic crisis than in the advanced countries. However, as Chapter 3 will show, institutionally it presented strong barriers to change, which have had particular force in the poorer developing countries. Although they never achieved the same level of inclusiveness of beneﬁts, the statist model was more deeply ingrained in their power structures. On it was constructed a deep commitment to the responsibility of the state that carried its own constituency of interests in maintaining interventionism: politicians and bureaucrats with patronage opportunities, pro-

Changing Views of the Role of the Government 9

fessional staff with standards to protect, urban residents enjoying subsidized prices, services and employment, and industrialists and farmers with guaranteed but also controlled prices. In the poorer countries with weaker market systems, power and privilege existed in function of state action, and the relief of poverty depended on access to state redistributed wealth. To challenge the statist model therefore risked challenging the foundations of the state and its legitimacy (Sandbrook, 1993). Even in the more mundane sense of the political management of reform, changes which proposed the undoing and restructuring of public service organizations, especially where these delivered basic living requirements (health, water, price subsidies), are bound to be complex and long term. They present costs and uncertainties to politicians and administrators, and only rather abstract advantages to anybody (Grindle and Thomas, 1991; Grindle, 2003; Nelson, 2000). The circumstances under which reform was to take place compounded the difﬁculty. Crisis and delayed reform The initiating impulse to public sector reform in developing countries (as elsewhere) has been the economic crisis that became transparent in the early 1980s and which, for many, has since then persisted. This has been not only the impulse for but also the difﬁcult context of reform, particularly in Sub-Saharan Africa and Latin America. Economic reform in most developing countries was delayed and therefore ultimately more traumatic than it might have been. The oil crisis of the early 1970s, falling commodity prices and recession in the West led to increasing balance of payments difﬁculties not only for the advanced countries but also for those countries which were dependent on them for export markets. African and Latin American (and central European) governments generally avoided voluntarily adjusting to the new economic realities; instead they tried to maintain the scale of public sector employment and expenditure by borrowing in the late 1970s and 1980s to cover tax shortfalls and debt. In most countries of Africa as in Latin America, but not in South or East Asia, government expenditure therefore became a growing proportion of shrinking national wealth (Cornia et al., 1987; Sandbrook, 1993; Bulmer-Thomas, 1992). Among the countries that were the core subjects of this study, Sri Lanka and Ghana responded relatively quickly to ﬁnancial crisis and the inability to maintain public expenditure. Indeed, Sri Lanka can be

10 The Changing Role of Government

said to have embarked on its own economic reform – a self-motivated forerunner of the IMF/World Bank stabilization and structural adjustment package. Beginning in 1977, and then more fully with the support of the international ﬁnancial institutions from 1987, Sri Lanka launched a programme to liberalize the economy, encourage foreign investment and privatize state enterprises. In 1983, Ghana embarked on an Economic Recovery Programme funded by the World Bank and IMF to stabilize the macro-economy by reducing public sector control of exchange and interest rates and, in principle (though very uncertainly in practice), to reverse previous policies of public ownership and unsustainable public expenditure. The later adjusters were India and Zimbabwe. India adopted an IMF supported ‘economic reform package’ in 1991 after a macro-economic crisis, triggered by the ﬁrst Gulf War, which included spiralling ﬁscal and current account deﬁcits and a collapse of foreign reserves. The focus of its reforms was on macro-economic management and the liberalization of regulations on trade and industry, rather than on the reform of public administration itself. The Government of Zimbabwe introduced a civil service reform programme in 1987, and entered discussions with the International Monetary Fund and the World Bank which led to a ﬁveyear Economic Structural Adjustment Programme (ESAP) early in 1991. In addition to an opening up of the economy through the liberalization and deregulation of trade and investment, the programme included macroeconomic stabilization, privatization of state-owned enterprises, and a social element to protect those affected by transitional hardships. The degree to which these reform programmes were implemented differs between countries. In particular, though Ghana embarked early on the path of adjustment, it has made slow progress and has had to renegotiate terms with the international institutions. Zimbabwe has continuously faced the suspension of IMF programmes due to its inability to control budget deﬁcits and public borrowing. Another variation was in the application of liberalizing reforms in different service sectors, as we will see in Chapter 3. The general experience has been that reforms in the social sectors (health, education, and even water) have been cautious and introduced much later than economic reforms. However, for all sectors of government, structural adjustment in the 1980s and early 1990s was an important background of constraint on public expenditure. The re-distribution of poverty Whether it was due to pre-existing failures of economic policy or to the effects of structural adjustment, Cornia et al. (1987) demonstrated a

Changing Views of the Role of the Government 11

general increase in levels of poverty in most countries of Latin America and Africa during the 1980s. Moser et al. (1993) showed that the impact of structural adjustment on prices, wages and employment structures also had the effect of redistributing the incidence of poverty. The urban poor and ‘border-line poor’ were particularly affected by the removal of subsidies in government services and controls on food prices. A ‘newly poor’ category was created by the shift of employ-ment out of public administration, urban commercial services and protected industries into tradeable export goods. This generated underemployment and drastically lower wages in the previously privileged urban centres. However, the experience of South Asia (at least India and Sri Lanka) and South-East Asia was dramatically different with accelerating growth and a declining incidence of poverty, in the same period. In 1996, when the research was initiated, the core countries selected for this study had a similar level of national wealth (GNP) per capita of between $420 and $580 per head per annum. By the year 2000, the relative position of the four countries was less in step, with Ghana at $350 and Sri Lanka at $870 per head. If income is adjusted for ‘purchasing power parity’, India and Zimbabwe were at similar income levels, with Ghana lower and Sri Lanka much higher (Table 1.1). There has been a marked historical change in the relative positions of the countries. The African countries have seen a decline or stagnation in income since the 1970s while the South Asian countries have experienced economic growth. Most strikingly positive is the case of Sri Lanka which has not only achieved sustained economic growth but also, though still a poor country, good human development indicators (UNDP, 2002). By the late 1990s Sri Lanka exceeded the average for middle-income countries on two of these indicators (life expectancy and adult literacy), and was near their average on education enrollment rates. By contrast, Zimbabwe saw a decline in its overall performance: real per capita income is now lower than in the 1970s and it is one of only ﬁve countries whose human development index rating is now lower than in 1980. Ghana has made a slow recovery from negative economic growth since 1985; its human development indicators remain low and mainly below the average for low-income countries. India, which had started the 1970s with the weakest position on most indicators, achieved accelerating growth from 1980 and an improvement of its human development status. However, by the late 1990s it still had life expectancy, literacy and education levels that were only about average for low-income countries, and almost half its population lived on less than $1 per day.

12 The Changing Role of Government Table 1.1

Economic and social indicators in the study countries

Indicator

Ghana

Zimbabwe

India

Sri Lanka

Average for lowincome countries*

GNP per capita $ at 1995 prices (i)

350

480

460

870

420

GNP per capita $ PPP at 1995 prices** (i)

1940

2590

2432

3470

1990

% of population below national poverty line*** (i)

31.4 (1992)

25.5 (1991)

35.0 (1994)

25.0 (1996)

–

% of population with less than 1$ per day at 1985 prices (i)

38.8 (1998)

36.0 (1991)

44.2 (1997)

6.6 (1995)

–

Govt. consumption as % GDP (i)

10

23

11

9

11

Life expectancy at birth in years (ii)

57

43

63

73

60

% Adult literacy rate (ii)

72

88

57

92

62

% Education enrollment**** (ii) (1999 ﬁgures)

42

65

55

70

51

Rank on the Human Development Index of 174 countries (ii)

129

128

124

89

–

* Low income countries are those where per capita income is below $760 p.a. ** Purchasing power parity (PPP) represents GNP in ‘international dollars’ so as to provide a better comparison of average income between countries. *** Deﬁnition of poverty line is country speciﬁc based on local consumption basket and costs. **** Calculated by dividing the number of children enrolled in each level of schooling (primary, secondary and tertiary) by the number of children in that age group. Sources: (i) World Bank World Development Indicators, 2001. 2000 data unless otherwise stated. (ii) UNDP Human Development Report, 2002. 2000 data unless otherwise stated.

Table 1.2 indicates the changing status of the four core countries in terms of the UNDP’s human development indicators. Taking into account per capita income levels, life expectancy, literacy and education enrollment, countries are classiﬁed from 0 to 1 (the maximum)

Changing Views of the Role of the Government 13 Table 1.2

Ghana Zimbabwe India Sri Lanka

Human Development Index trends from HDR 2003 1975

1980

1985

1990

1995

2001

0.444 0.544 0.416 0.609

0.474 0.570 0.443 0.644

0.487 0.626 0.481 0.670

0.515 0.614 0.519 0.692

0.537 0.567 0.553 0.715

0.567 0.496 0.590 0.730

Source: UNDP Human Development Report, 2003.

in a composite human development index (HDI). All the countries but one see an improvement over the period 1975–2001. By 2001, Zimbabwe falls into the low human development category. Declining public services The rate of growth of social expenditure (particularly health and education) declined almost everywhere during the 1980s and up until the mid-1990s, particularly in ‘intensely adjusting’ countries (Cornia et al., 1987). Actual reductions in expenditure per head were more common in Latin America than in Asia or Africa: however, Ghana undertook one of the most severe cuts, as in Asia did Sri Lanka. India saw an incremental decline from an already low base (Moser et al., 1993; Jalan and Subbarao, 1995). The proportion of India’s economy devoted to public expenditure on health (0.6 per cent in 1998) is one of the lowest among low income countries and the percentage allocated to education declined in the 1990s to 3.2 per cent in 1997 (UNDP, 2000). Zimbabwe maintained a relatively high proportion of government spending on social expenditure through the 1980s and into the late 1990s (3.1 per cent and 7.1 per cent on health and education), but in real terms, with the decline of the economy, the value of the expenditure fell. Internationally, where cuts did take place, they often particularly affected basic services (including preventive and primary health care and primary education) and infrastructure: To meet their interest obligations, countries mired in debt squeezed programs in education, health, and infrastructure as often as – or more than – they cut low-priority programs, bloated civil service rolls and money-losing enterprises. (World Bank, 1997, p. 24) The widespread decline in investment in and maintenance of infrastructure had adverse effects not only on households but also on enterprises and particularly small ﬁrms (ILO, 1995; Stren and White, 1989;

14 The Changing Role of Government

Moser et al., 1993; Sandbrook, 1993). Public employment and public subsidies were often maintained for political reasons, while expenditure on investment and the maintenance of public services and infrastructure was allowed to slip. International shocks thus helped to enhance the visibility of public sector inefﬁciency and ineffectiveness, even before the stabilization and adjustment programmes of the International Monetary Fund and the World Bank were introduced. Cuts and demoralization in public administration Programmes for the reduction of employment in public administration were among the most tangible aspects of the ﬁrst phase of structural adjustment. Over-stafﬁng after a long period of growth (Abernethy, 1988) was one of the World Bank’s main criticisms of public administration, and staff reductions were seen as an important way of controlling public expenditure (World Bank, 1983 and 1989). However, cutting employment represented a considerable political challenge, undermining government’s role as employer of last resort and threatening its constituency of support (Sandbrook, 1993). This was a particular problem in those democratic regimes, such as India and Sri Lanka, where public sector unions represent a considerable block of electoral support, and especially in countries with strong ethnic differences (Malaysia, Sri Lanka, South Africa, Kenya) where public sector employment offers the possibility of balancing advantage (Esman, 1999). Effectiveness in restraining civil service expansion has been greatest in the more ideologically committed countries of Latin America (such as Chile) and in the more donor-dependent countries of Africa (Moser et al., 1993, Sandbrook, 1993; World Bank, 1991). Cuts in civil service staff were largest in certain African countries – in Uganda by almost half and in Ghana by 40 per cent since 1987 (Larbi, 1999; Nunberg and Nellis, 1995). Zimbabwe reduced its government staff by 12 per cent under the civil service reform programme from 1991 (Makumbe, 1997). However, staff cuts have often not led to the expected budget savings. Redundant staff had to be compensated and often the better staff were lost. A study of civil service reform in 15 countries showed that ‘wage bill reduction and salary decompression had been achieved in fewer than half the cases’ and that sometimes staff cuts were later reversed. Wages have therefore often declined in the face of ﬁscal pressures (World Bank, 1997). It is easier for governments to allow real wages to erode than to shed jobs. Public sector wages fell below subsistence level in, for example, Sudan, Ghana, Uganda and Guinea. The

Changing Views of the Role of the Government 15

result was widespread demoralization, corruption and moonlighting (Sandbrook, 1993; Amis, 1992; Hilderbrand and Grindle, 1995). The public administration that was expected to carry out the reforms was itself demoralized by the decline in real salaries and by threats to job security: ‘government agencies were expected to co-operate in diminishing or dismantling their own power’ (Hirschmann, 1993). The climate of change has often been one of suspicion and resistance, unmatched by support from any clear constituency for change. Because government and the public sector are simultaneously the target and agent of reform, the reform process has been compared to ‘rebuilding the boat in the open sea’ (Elster, 1993).

Capacity to perform the ‘new’ roles of government The above background indicates a difﬁcult basis on which governments must construct reform programmes for improving the management of public services. Government was being slimmed down and, at the same time, new roles and relationships with non-government service providers were being deﬁned. The following sections will outline the core concern of this book: why governments face questions of capacity to undertake these new roles and how their capacity might be assessed. The new roles of government: why is capacity in question? The broad thrust of reform is to redeﬁne the roles of government and then to inject a more entrepreneurial or managerial (versus bureaucratic and administrative) style into the performance of those roles. The underlying shift has been described as one of government moving from ‘a concern to do, towards a concern to ensure that things are done’ by collaboration with other actors (Kaul, 1997). Following this principle, responsibility for arranging for a service to be delivered is separated from the actual production of the service (Savas, 1987, 2000). According to this argument, governments could, in principle, usually achieve their objectives without getting involved in the direct production or delivery of services. Several writers (Koldarie, 1986; Wunsch, 1991; Ostrom et al., 1993) distinguish between the responsibility for ‘provision’, which might be governments’ concern, and ‘production’ which might be done by private, public agency or com-munity actors. This book adopts a different terminology to distinguish between – ‘direct provision’ which is the act of physically producing (constructing, creating, maintaining) and delivering a service; and

16 The Changing Role of Government

– ‘indirect provision’ which is the business of ensuring that a service is available; this might involve decisions about policy and standards of service, about organizational arrangements, coordinating, ﬁnancing, enabling and regulating producers. Rather than a reduction of government, instead there is a redeﬁnition of its roles. Governments always performed ‘indirect roles’ but these have become more important with recent public policy reforms. Tasks such as analyzing policy options, setting standards and monitoring their enforcement, raising and allocating ﬁnance, managing budgets or informing citizens have always been part of government, but they take on clearer importance and become more complex where government is working with networks of independent direct providers. Some indirect roles, such as contracting, regulating and creating incentives for private producers have assumed vastly greater signiﬁcance in many countries, and can effectively be regarded as new for many. Even where there is no engagement with private or community producers, governments have had to ﬁnd new ways of managing within the public sector, where decentralized government or decentralized management have become the main direct providers. The case for the shift has been that many governments have proven to have weak capacity to deliver basic services, and that they should therefore ﬁnd new ways of providing them jointly with markets and civil society (World Bank, 1997 and 2003). The question is whether new approaches imply new demands on the capacity of governments. While it may sometimes be possible wholly to withdraw from manufacturing (for example, the complete privatization or divestiture of steel production), governments can rarely simply wash their hands of all responsibility for a public service. Reforming governments may shed certain direct roles – for example, constructing social housing, running agricultural markets or managing urban water supplies. But they take on other indirect ones – for example, giving incentives for housing for poor people, providing market information to producers, and regulating water companies. This redistribution of roles is defended on grounds of efﬁciency, but there are important questions whether the ‘transaction costs’ of bringing about new forms of management may outweigh the efﬁciency gains. The theory underlying these questions will be described in the next chapter. If we represent government as involving the following main categories of activity, then all of them may be affected by new manage-

Changing Views of the Role of the Government 17

ment reforms, but the key ‘new’ roles that will be examined in this book are numbers 3, 4, 5 and 6: Government’s indirect roles 1. Policy and rule-making: analyzing policy options, identifying the need for intervention in markets, advising ministers, setting policy frameworks, monitoring effects. 2. Enforcing and upholding the law. 3. Regulating markets: maintaining competitive conditions or ensuring efﬁciency in the absence of competition. 4. Enabling providers: facilitating provision by the market, households and groups, ﬁnancing and supporting producers, providing information to producers and consumers. 5. Contracting independent providers: designing, managing and monitoring contracts so as to ensure that policy objectives are achieved. Government’s direct roles 6. Managing service delivery through decentralized management and market mechanisms within government. 7. Direct provision of services by government administration. With regard to these roles, the key capacity questions that will be addressed throughout the following chapters are these: • Are the appropriate skills and organization available for the exercise of the activity? • Does the exercise of the role require collaboration and coordination going beyond the particular organization? • Are the mechanisms of accountability of the agent to the principal clear, and can the principal measure and therefore control the performance of the agent? (The conceptualization of principal and agent will be described in the next chapter). • Is the institutional environment supportive of the roles and organizational arrangements? Concepts of capacity The assessment of capacity to perform the ‘new roles of government’ requires an approach that goes well beyond the traditional concern with the quality of human resources. The previous section emphasized the importance of the wider institutional context within which reform was taking place. Human resources are constrained or enabled by the

18 The Changing Role of Government

organizations and networks of organizations within which they operate and by the institutions which surround them. The importance of institutions in shaping the way people are motivated and interact is recognized in the theory that underlies the new models of public service management described in the next chapter. The concept of capacity adopted here reﬂects these different dimensions and is close to the approach presented by Hilderbrand and Grindle (1995) who deﬁne capacity as ‘the ability to perform appropriate functions effectively, efﬁciently and sustainably’. ‘Capacity’ is the factor that explains human performance within a given organizational framework and institutional environment. One way of seeing the factors that inﬂuence government capacity is as a series of concentric circles going from those that are internal to speciﬁc organizations to those that are more external: from interorganizational relationships, to the surrounding managerial and political practices, to the inﬂuence of wider social, political and legal conditions. Figure 1.1 sets this out diagrammatically. In this approach there are three levels of analysis (Batley, 1997; Hilderbrand and Grindle, 1995; Larbi, 1998a). First, among the aspects of capacity that are internal to the organization are: • human resources, skills and incentive systems; • organizational and administrative arrangements including decision making and control structures, procedures, authority relations, the distribution of functions and responsibilities, management style and leadership; • the organization’s equipment, capital and ﬁnancial resources. Second, there are the arrangements through which bodies coordinate within the network of organizations that together perform a task: are their roles, the ﬂow of decisions and ﬁnance clear or contradictory? Third, the approach considers the wider institutional environment and how this affects (enables or constrains) the mobilization and deployment of personal and organizational capacity. This includes: • the more immediate institutional context of the apparatus of government, such as civil service rules and regulations, and the way political control is exercised over public administration; and • the broader context: the state of the macro-economy, the strength of civil society and the private sector, the stability of the political

Changing Views of the Role of the Government 19 LEGAL/ADMIN. FRAMEWORK & CULTURE

• • • •

rules and regulations prevailing admin. cultures, values, practices meaning systems and behaviour patterns public sector reward & incentive framework

EXTERNAL TASK

ENVIRONMENT NETWORK

ORGANIZATIONAL CAPACITY • human resources and incentives

SOCIAL CONDITIONS • societal values • strength of civil society • degree of literacy • level of satisfaction with current conditions

• organizational structures and relationships

• management and leadership capital and financial resources

• information resources TASK

EXTERNAL

MACROECONOMIC CONDITIONS • stability of the economy • govt. financial position • strength of private sector • skill levels

NETWORK

ENVIRONMENT

POLITICAL/GOVERNANCE CONDITIONS

• • • • •

political commitment and support nature of political decision-making systems of appointment to leadership posts respect for rule of law/political values political stability/regime legitimacy

Figure 1.1 A framework for analyzing capacity Source: Larbi, 1998a:164.

system, technological change, the legal framework and its enforcement. This approach emphasizes the importance of ‘ﬁt’ between the levels. Skills development or organizational reforms may often be introduced in a particular agency but fail because they ignore the operational context in which people and organizations have to operate – for example, the civil service rules, political pressures, public expectations and economic situation. The question of ﬁt raises the possibility that agencies and governments need to adapt their organizational arrangements and skills

20 The Changing Role of Government

to changing institutional environments, for example as technologies, public demands and economic conditions change. Capacity requirements are therefore not static but continuously change with task and context. One of the major issues for this study is whether governments have been able to show this sort of ‘adaptive capacity’. Lastly, the approach also emphasizes that capacities are speciﬁc to task as well as to context. The different organizational elements of capacity and the institutional conditions come into play in different country contexts, sectors, organizational arrangements and for different types of governmental activity. This argument can be turned around in the other direction. Governments should only take on tasks and organizational reforms which they have capacity to perform: ‘ﬁrst focus the state’s activities to match its capability’ (World Bank, 1997, p. iii).

An overview of the study The research focused on four service sectors, urban water supply, health care, agricultural marketing and business development, in four core countries, Zimbabwe, Ghana, Sri Lanka and India. However, it also made reference to other sectors and countries. The country focus The core countries were selected for study on the following main criteria: 1. South Asia and Sub-Saharan Africa: These regions account for most of the world’s poor population. Two countries were selected from each region so as to maintain clearer possibilities of comparison. 2. Low income: The focus is on the situation of low-income countries and their capacity to adopt new approaches to service provision. However, it was important to select countries that were not so poor or weak that no signiﬁcant services could be provided. 3. Anglophone and Anglo-colonial heritage: This focus allowed comparison within a range of countries with some common experience and similar administrative traditions. They present scope for extending generalization of ﬁndings to similar countries. However, a range of non-anglophone countries was also studied for comparative reference. 4. Continuity of public administration: India and Zimbabwe, relative to their respective continents, have traditions of strong conventional

Changing Views of the Role of the Government 21

public services. Ghana and Sri Lanka have experienced breakdown or considerable disruption and reconstruction of their governmental systems; Zimbabwe’s deepening state crisis occurred after the research was undertaken. India and Ghana, but most enduringly the former, represent the classic cases of interventionist states which the neo-liberal adjustment programme seeks to reform. 5. Macro-economic changes and liberalization: Two of the core countries (Sri Lanka and Ghana) embarked on structural adjustment in the early 1980s: Sri Lanka was a self-motivated fore-runner of the World Bank/IMF package, while Ghana is often cited as an exemplary case in Africa. Zimbabwe and India delayed structural reform into the 1990s and have been more hesitant about reforming public administration. The duration of adjustment and the willingness with which it was engaged appeared to be possibly inﬂuential factors in the success of reform. In these core countries, all four service sectors were studied. A further set of ‘reference countries’ was selected for study of speciﬁc sectors. This was in order to extend the comparison to known interesting cases and to redress the focus on poor countries with a British administrative tradition in South Asia and Africa. These included: • Health care: Thailand; • Agricultural marketing: Kenya; • Urban water supply: South Africa, Bolivia, Argentina and a number of other countries of Latin America and Francophone West Africa. However, even with a quite carefully selected group of core countries and a wide range of reference countries, there is a need to be cautious about both comparison and generalization. Institutions and administrative traditions vary between and even within core countries, in spite of certain similarities of tradition and recent experience of adjustment and reform. The sector focus The research explored how the role of government and the approach to the management of service delivery were changing in the following main respects across the four sectors: • Health care: decentralizing hospital management, charging user fees, contracting services from the private sector; regulating and enabling the private sector;

22 The Changing Role of Government

• Urban water supply: corporatizing public water utilities, private sector participation and contracting, regulating and enabling the private sector; • Business development: replacing direct intervention through public ownership with the provision of support services to industry. The textiles industry was selected as the focus of the study; • Agricultural marketing: privatizing food supply management, reforming public marketing agencies, contracting out marketing services, developing quality assurance and market information systems. At ﬁrst glance, not much seems to unite these spheres of activity. However, in all of them there was some shift either to adopt new internal management practices or to create new forms of relationship with non-governmental provider organizations. Each of the sectors is also important in its impact on the livelihoods of poor people. Health care and drinking water are core elements of any welfare concern and raise basic issues of distributive equity for consumers. The operation of agricultural markets raises fundamental issues of survival for both producers and consumers; adjustment in this sector has a direct impact in both urban and rural areas. Support for industry is of primary concern to producers but also to the prospects for national development and mass employment. Nevertheless, the four service sectors were also selected precisely because they span a broad range of types of government activity and because they are different in important respects that affect their ‘manageability’. The ﬁrst set of differences may affect the capacity of citizens and policy-makers to assert their control over the performance of service providers (these considerations are explained more fully in Chapter 3): (i)

Organization and deﬁnition of user groups: Health care and drinking water supply are essentially consumer services which apply to the general public, while agricultural marketing and business development are oriented to production and apply to speciﬁc producer groups; (ii) Choice and exit opportunities: The sectors are differentiated between those where users have market choice about whether to use services at all (agricultural and business support), one where they may have some choice about which public or private service provider to use (health care), and one where they may have little or no choice (piped water);

Changing Views of the Role of the Government 23

(iii) Measurability of performance: The ultimate outputs of the four sectors are to a different degree quantiﬁable, tangible and measurable as opposed to qualitative, intangible and difﬁcult to measure. Water supply is most clearly in the ﬁrst category; health care, business and agricultural support services most clearly in the latter category. The second set of differences is about the economic characteristics of each of the service sectors. None of the selected service sectors is a public good: the market could in principle, provide all of them. However, each sector is exposed to a different set of possibilities of market failure that inﬂuences the case for and perhaps the nature of government intervention. The concept of ‘market failure’ is described more fully in Chapter 2, but Box 1.1 in this chapter sets out some of the main reasons why market provision may fail and governments may have to intervene. An illustrative case for intervention in each sector is made below, but recognizing that the case may vary for particular aspects of sectors and in particular country contexts. For example, health care consists of many sub-components (ancillary and clinical services, hospital management etc.) and the extent of market failure will depend on circumstances: ….in a country where there are very strong professional ethics and a highly educated population, asymmetric information between health care provider and patient will not be as problematic as elsewhere (ceteris paribus) and the need for regulation may be more limited. (Mills, Bennett and Russell) Beginning with the weakest case for intervention and moving in a rough order towards stronger cases for government intervention, the following are the main arguments (summarized in Table 1.3). Support by government for industrial development historically has been a major area of government intervention often through public ownership. This has particularly applied to textiles and garment manufacturing which has been seen as the classical ‘lift-off’ industry, offering prospects of national development and employment. If there are market failure arguments for intervention, they relate to informational asymmetry (inequality) between producers at a national and international level and perhaps between national producers and large international purchasers. Governments may play a part in providing information and support to enable the insertion of businesses into the national and global economy (Jackson, 2002).

24 The Changing Role of Government

Box 1.1

Market failure arguments

Public goods: Some goods and services are described as by their nature ‘public’ in the sense that private ﬁrms on their own would simply not provide them either adequately or at all. These are goods for which there is no means of charging consumers (because non-payers are not ‘excludable’) and whose consumption does not reduce their availability to others (there is no ‘rivalry’). This would occur, for example, in the case of police services, street-lighting or street-sweeping where the beneﬁts are collective. It is impossible to measure how much any individual has consumed, and the service is not used up by consumption. Externalities: • the wider society would beneﬁt from extension of a service even though the direct consumers may be unwilling or unable to pay for it: an example of such ‘positive externalities’ would be vaccination; • producers can pass on costs to the wider society, allowing them to keep their costs to the direct consumers artiﬁcially low: an example of these ‘negative externalities’ would be industrial pollution. Monopoly: The nature of a good or service or the scale of the initial investment required prohibit competition – e.g. water supply systems. Scale and risk: The necessary investments are so large scale or the returns so risky that private ﬁrms are not prepared to undertake them – e.g. major underground infrastructure. Imperfect information: Consumers or producers have too little knowledge to make informed choices: this sort of situation may arise in the case of professional services (such as health advice) where clients may not be able reasonably to assess the value of options. Equity or ‘merit’ goods: There are in practice basic commitments in any society to the view that everybody should have access to certain goods and services, regardless of their ability and willingness to pay the market price. ‘Government needs to act to provide merit goods because individuals are not necessarily the best judges of what is in their own or the public interest’ (Walsh, 1995, p. 10). Education and health typically come into this category.

Agricultural marketing historically has been another area of major government intervention in what is by nature a private good. Governments have often owned or regulated food supplies, ﬁxed prices to producers and managed production. Market failure analysis might suggest more speciﬁc interventions to manage supply uncertainties (particularly sensitive in the case of food), and to overcome problems of monopolized transport, poor market information, inadequate ﬁnance and insurance markets. (Hubbard, 2003) Urban water supply, if piped, is not a public good. It is possible to exclude non-payers from beneﬁting and to charge for levels of consumption, so provision can be made through the market. However,

low high

high

high

Health – preventive – curative

Agricultural Marketing – food crops

– export crops

– city well – piped

low/ medium medium high

Rivalry

Production/Distribution characteristics

high

high

low high

low high

low

low

low

medium medium

low high

high

low/ medium medium

high low

low high

high

low/ medium low

high medium

medium high

high

high

low

high high

low low

low

Excludability Monopoly Economies Externalities Information tendency of scale asymmetry

Consumption characteristics

Sector characteristics and the motive for intervention

Water – stand pipe

Sector

Table 1.3

1 monopoly 2 health externalities 3 equity/ merit good

household and national food security

1 food security 2 market information 3 market structures

equity 1 public good considerations (preventive) and human 2 asymmetric capital information development 3 externalities 4 equity/merit

access basic necessary water among current and future populations

Social welfare Motive for considerations government intervention

25

Rivalry

Production/Distribution characteristics

high

low

medium

low

high

Excludability Monopoly Economies Externalities Information tendency of scale asymmetry

Consumption characteristics

Sector characteristics and the motive for intervention – continued

Textiles – export high manufacture

Sector

Table 1.3

national development and employment

1 national development 2 asymmetric information of local versus global producers

Social welfare Motive for considerations government intervention

26

Changing Views of the Role of the Government 27

water supply is subject to other forms of market failure which, historically, have been used to justify public ownership. It is the classical example of a natural monopoly, resulting from high investment costs and enormous economies of scale, which make it extremely difﬁcult for alternative suppliers to compete. The case for some form of public intervention is reinforced by water’s positive (health and productivity) and negative (pollution) externalities, and by its status as a merit good for basic consumption necessities. (Nickson and Franceys, 2003) Health care has also been an important area of government intervention, often through systems of publicly owned health services and hospitals. Preventive (environmental) health services do often have public goods characteristics, since it is impossible to exclude nonpayers from the beneﬁts. In the case of personal health care, which is the focus of this study, the main arguments for government involvement are to do with medicine’s high ‘external’ or social beneﬁts (avoiding the transmission of disease, reaping the beneﬁts of health for labour productivity). There is also the case for intervention to compensate for consumer ignorance of or incapacity to judge between medical treatments. Moreover, the need for health care is irregular and unpredictable, making it difﬁcult for consumers to afford or insure for it; there is a case for tax-based cost-sharing. There are also strong equity or merit goods considerations for providing health care to consumers who cannot exercise private choice, due to their poverty, childhood or mental incapacity. (Mills, Bennett and Russell, 2001) The research framework The research on which this book is based followed a broadly similar framework of analysis across the countries and sectors. The objective was to understand whether the types of reform indicated in this chapter were taking place, for what reasons and under what pressures, and whether such approaches are appropriate to developing countries. Sector and country case studies were undertaken following a systematic sequence of steps: • Analysis of the technical, economic characteristics of the sector in the particular country context – were there particular market failure ‘justiﬁcations’ for public intervention on technical grounds? • Description of the current organizational arrangements for service provision, covering direct and indirect provider roles, identifying

28 The Changing Role of Government

•

•

•

•

the roles (or forms of intervention) which public agencies had in practice adopted. Comparison of the actual intervention with the technical (theoretical) case for intervention so as to raise questions about whether market failures were being addressed and why the public sector was doing what it was doing. Analysis of the history which had generated these organizational arrangements and the (stakeholder) forces which either sustained them or pressed for change, so as to identify the real, as opposed to technical, motives for the status quo and for reform. Secondary source evaluation of the performance of the sector nationally, in terms of its outputs and outcomes, focusing on their allocative efﬁciency and effectiveness (coverage, quality, equity), and the relation of outputs and outcomes to policy objectives. Evaluation of the performance of speciﬁc organizational arrangements for service provision, in terms of (i) their outputs, focusing on their productive efﬁciency and effectiveness, and (ii) the organizational performance by government or public agencies of their roles – how far were these performed? how well were operational functions carried out? Assessment of the capacity of government to undertake indirect provider roles. This aimed for an explanation of performance in terms of (i) internal organizational capacity (human resources, structures, capital) and (ii) inter-organizational relations, and (iii) the extra-organizational or institutional conditions which impacted on the mobilization of internal capacity.

In addition to the analysis of the management of the four sectors, consumer surveys were undertaken in each of the four core countries (Mutizwa-Mangiza, 1997; Rakodi, 1996, 1998 and 2001; Silva, Russell and Rakodi, 1997). Principally through focus group discussions, these explored users’ opinion of the performance of health and water services and of the reforms.

Conclusion This chapter has identiﬁed the territory of the book locating public service reform in broader historical changes in the role of government in development. Changes in approaches to public management are associated with the neo-liberal movement of the 1980s and 1990s. This broad approach persists, even though it is now asserted with less simplicity and recognizes the need to take into account local circum-

Changing Views of the Role of the Government 29

stances and more directly to address poverty. The question that this book addresses is whether the approach is appropriate to the circumstances and capacities of the developing countries to which it has been exported. Reforms in public management are being undertaken in more difﬁcult circumstances than those experienced by reformers in more advanced countries. This is particularly – but not only – the case of the African countries where, paradoxically, reform proposals have often been most radical. Structural adjustment and public sector reform have often been delayed until the point that the ﬁscal crisis is deep and public resources exhausted. They have therefore often taken place in the context of already rapidly declining public services from which it is difﬁcult to recover. Moreover, the reforms themselves have often generated a ﬁrst impact of increased stress and poverty for those sections of the population that had had access to services and employment. A problem of the public sector, unlike private ﬁrms competing in the market, is that there are weak internal pressures for adaptive change to local institutional conditions. It is easy enough for public ofﬁcials and politicians to go on doing what they always did, because they are subject only to very diffuse demands from their electorates rather than from the very immediate requirements of competitive survival. Nevertheless, public organizations are being driven to adopt a model of service delivery where they assume an ‘indirect role’, managing, regulating or enabling other providers – whether these are privately, community or publicly owned. The question is whether this changed, and in some respects reduced, role of government is actually a less demanding one. The hypothesis of this book is that indirect management is more demanding on government capacity, and that current reforms therefore present new dangers of government failure. The following chapters are essentially about this question of government capacity to handle new roles and approaches. Chapter 2 describes the conceptual origins and the main elements of the new management reforms that have had such wide international impact. It uses this as a framework for locating the service reforms described in this book. Chapter 3 goes on to look at the political dynamics of change, identifying the leaders, the supporters and the opposition across the service sectors. The underlying question is whether reforms are driven and accepted locally. The following chapters examine the experience of implementation. Chapters 4 and 5 analyze reforms that have ‘imported’ market-type reforms into the public sector: decentralizing management and making managers and

30 The Changing Role of Government

agencies directly responsible for their budgets and service outcomes, and getting users to pay charges for services. Chapter 6 examines the logic of ‘exporting’ responsibility for service delivery through privatization, the purchase of the services of contractors and the regulation of non-government providers. Chapter 7 analyzes the experience of contracting, sector by sector, identifying the forms of contract that have been adopted and found by governments to be more or less feasible. Chapter 8 investigates even more arms-length public sector roles than contracting: ﬁrst, the regulation of private service deliverers in the water and health sectors; and second, the support (enablement) of the market operations of private ﬁrms. The principal focus of Chapters 4–8 is to assess the capacity of governments to take on these new modes of operation and to establish which seem to be more appropriate in the various service sectors. Chapter 9 concludes by drawing together the analysis, considering how our understanding of governmental capacity might inﬂuence future approaches to service reform.

2 Changing Approaches to Public Sector Management

Introduction Chapter 1 traced the changing perspectives on the role of government in development. It was noted that crisis in the welfare and developmental states in the 1970s and 1980s called into question the post-war consensus on the active role of the state in the economy and led to the ascendancy of neo-liberal economic policies from the 1980s. It was not just the welfare state that was called into question, but also the traditional Weberian model of bureaucracy came under attack as being slow, inefﬁcient, ineffective and unresponsive to service users. The crisis in the welfare state and the weaknesses of state bureaucracy led to the search for alternative ways of organizing and managing public services and redeﬁning the role of the state to give more prominence to markets and competition. The shift was in response to a combination of stimuli for change driven by both theoretical arguments and pragmatic rationales. This chapter ﬁrst reviews the theoretical arguments that have inﬂuenced the new trends in public service reforms, including neo-classical and new institutional economic theories. Second, it describes the more pragmatic rationales for change in the management of public services. The rest of the chapter provides an overview of the new approaches to public sector management in order to situate the rest of the book in a wider international context beyond the case study countries and sectors.

Theoretical bases of new approaches to public management Several theories provide the theoretical underpinnings and justiﬁcation for the changing role of government in general and for the new 31

32 The Changing Role of Government

approaches in public management in particular. These include neoclassical economic theories and new institutional economic theories, which have inﬂuenced market-oriented reforms by emphasizing choice and competition. This section reviews some of the relevant theories. Neo-classical rationalism – the limited case for government intervention The classical economic view is that, by giving free expression to selfinterest, choice and competition, markets offer productive1 and allocative efﬁciency,2 delivering what consumers would choose to have at the minimum price. In that case the public sector should intervene only where there are reasons why the market will fail. Natural monopolies, inability to exclude free riders, the failure to impose the full social costs of consumption on direct consumers, charging where the cost of extending the service is zero, ignorance on the part of consumers or producers about what is available or what they need – these are all reasons why the market, unchecked by the state, will fail. It is under such circumstances that there is a limited case for government intervention; otherwise markets are best suited for service delivery. Box 1 in Chapter 1 summarized the market failure arguments. Even though these ideas provide a set of considerations against which to consider the case for government intervention, there is need for caution. There are few services which fall neatly and wholly into any category of market failure, and where they do belong depends on local circumstances and judgements (Malkin and Wildavsky, 1991). The theoretical approach does, however, have the virtue of bringing these considerations into the open, forcing us to identify the reasons for intervention. For different services in differing national contexts there will be different reasons for government to be involved, and these imply diverse levels and forms of involvement. The country and service sector case studies referred to in this book explore the public sector response to market failures in speciﬁc country contexts. By identifying the speciﬁc causes of market failure, those who argue generally for restricting the role of government can limit the scope of its intervention. Even where there is a case for intervention, government does not necessarily have to assume the entire responsibility for the provision of a service. The case for governments assuming responsibility may be reduced by separating the elements of service provision. The World Bank (1994, 1997) describes this as the ‘unbundling’ of services. The idea is that, if the processes of production and delivery are itemized, only some elements may require public intervention. Thus,

Changing Approaches to Public Sector Management 33

the electric power industry can be ‘vertically unbundled’, separating generation (which can easily be competitive) from distribution (more likely to be monopolistic). Alternatively, there may be scope for ‘horizontal unbundling’ in which a service can be broken up geographically (like British water supply) or by category of service (freight as opposed to passenger trains). Breaking up services in these ways, reduces the costs to new providers of entering the market and allows performance comparison between them. Even if no new rivals actually enter the market, there is always the threat of ‘contestability’, that is that they may do so (Vining and Weimer, 1990). Neo-classical theory addresses the question why market failures may occur. Can it also deal with the question how service provision should be organized where there is market failure? There have been attempts to deduce appropriate organizational arrangements (involving the public, private and community sectors) from the economic characteristics of goods and services (Batley, 1996; Picciotto, 1995; Savas, 1987 and 2000; World Bank, 1994 and 1997). However, this is a purely technical approach that assumes that there are universal solutions regardless of the political and social context. The next set of theories directly addresses the question how organizations are structured; they underlie the new approaches to public management. Theories on organization of government intervention How should government intervention be organized? The traditional twentieth century response, which new institutional theories put to question, was that it should be by public ownership managed through hierarchical bureaucratic administration. Public ownership purports to replace the individual interests of owners (operating in malfunctioning markets) with the wider interests embodied in the state, to pursue social goals as opposed to private beneﬁt. The underlying assumption was that the objectives of government could be transmitted into action through administrators who would pursue policy, guided by their expertise in operationalizing policy, by their own professional commitment to the public interest, and by the command structure of bureaucracy. The implementation of policy was according to a bureaucratic ethic of neutrality and impersonality, without regard to the personal preferences of post-holders (Du Gay, 2000; Dwivedi and Olowu, 1988; referring to Weber, 1978). This thinking forms the basis of the traditional or ‘old public administration’ which has been the object of the new public management reforms that are discussed later in the chapter.

34 The Changing Role of Government

Public choice theory: The New Right critique of the welfare state and of the public administration based on it, was strongly inﬂuenced by the ideas of economic liberals such as Hayek (1973), and by public choice theorists such as Niskanen (1971), Buchanan (1975) and Mueller (1979). Public choice theory is one of the most inﬂuential in the new approaches to public management (Jordan, 1995). The central criticism of public choice against traditional pubic administration is that reward systems in the public sector do not promote effective performance and that politicians and bureaucrats have no incentives to control costs. This often leads to waste of resources and an in-built tendency for expenditure to grow and for delivery to take precedence over productivity. In the absence of any automatic disciplining mechanism such as competition and market forces, government agencies oversupply collective goods in order to maximize budgets and reap rents (Niskanen, 1971; Tullock and Eller, 1994). Thus, as Dixon et al. (1998) point out, bureaucracy is said to be self-serving leading to ‘opportunism’, and even deceitful and dishonest behaviour by bureaucrats, their clients and politicians (Dixon et al., 1998). This is because either environmental uncertainty makes contracts incomplete or ‘principals’ cannot effectively monitor the behaviour of their ‘agents’, who do not have identical interests and who have information that is not accessible to them. (ibid.: 165). The tendency for bureaucrats to promote the growth and expansion of governmental functions leads to overextension of the state which then creates an ever expanding bureaucracy that requires a hierarchical authority structure based on rational rules (Hayek, 1960, cited in Dixon et al., 1998). Over time, however, the capacity for top-down control is weakened as bureaucratic expansion gets to a point where it becomes impossible to fully control or even co-ordinate large organizations, leading inevitably to bureaucratic failure (Dixon et al., 1998; see also Breton and Wintrobe, 1975). The usual response to bureaucratic failure is ‘to create another bureau to oversee those who have lapsed into sin. Bureaux are piled on bureau and the bureaucracy grows on’ (Perlman, 1976:76, cited in Dixon et al.). Confronted with the bureaucratic failures of old public administration, politicians (as principals) face the task of creating organizational arrangements (incentives, sanctions and monitoring) that minimize the costs of the undesirable behaviour of agents and of the activity undertaken to control it (Weimer and Vining, 1991:132). The resulting growth in the complexity of rules limits the use of initiative, creativity and innovation, with the result that ‘good people are trapped in bad

Changing Approaches to Public Sector Management 35

systems’ (Osborne and Gaebler, 1992). Public choice thinking underlay the ﬁrst wave of neo-liberalism, economic adjustment and state withdrawal in the 1970s and 1980s. It provides some of the theoretical inspirations for reforms such as contracting out, privatization and performance management as cures for the ‘ills’ of bureaucratic arrangements for service delivery. The ‘new institutionalism’ can be seen as a reaction to the ‘reductionism’ of public choice theory and as the theoretical basis for recent approaches to public management (see Lane, 2000). It addresses the question how individual interests can be coordinated or aligned to shared goals, ‘the issue of how best to structure organizations that consist of individuals pursuing their own self-interest’ (Harding and Preker, 2002). New institutionalism’s contribution is to see that established institutional frameworks or organized ways of doing things govern individual interests and interactions. Organizations evolve within a framework of institutions. Institutions are considered to be the ‘rules of the game in a society’ (North, 1990) that shape the way that individuals interact. For this discussion, institutions include rules and norms about structures of authority, contracts, organizations and property rights. Principal-agent theory (Lane, 2000; Stiglitz, 1987; Walsh, 1995) examines organizational relationships as a tension between the ‘principal’ who demands a service and the ‘agent’ who provides it. Principal-agent theory assumes that actors anywhere are governed by economic self-interest. The question is then how principals can manage the selfinterest of those empowered to act on their behalf, their agents, so that it is aligned with the purposes that they (the principals) wish to achieve. The problem arises not just from conﬂicts of interest but also from the privileged access of the agents to information – the problem of asymmetric information. In public policy, ‘principals’ are ultimately citizens and ‘agents’ are politicians and bureaucrats, but the whole structure of a public bureau can be seen as being governed by chains of principal-agent relationships. Each has a measure of autonomy and each has their own interests to advance. The likelihood of the principal effectively controlling the agent depends on (i) how much information the principal has about the performance of the agent, and (ii) how far the principal can structure the relationship so as to control the agent or give incentives so as to make the agent’s interests correspond to the principal’s. There are general problems of control by principals in the public sector, resulting from the multiplicity of principals with diverse objectives, the

36 The Changing Role of Government

difﬁculty of constructing incentives where proﬁts do not accrue to agents, procedural constraints, and the inadequacy of democratic accountability as a means of checking the performance of agents. The relationship between principal and agent can be structured in different ways, in an attempt to overcome these problems and to motivate the agent to act in the interests of the principal. The principal-agent problem is particularly severe in the classical bureaucratic form of administration with its long bureaucratic hierarchies, monopolistic agents and difﬁculties of motivating them through long term and unspeciﬁed contracts. The traditional approach to controlling the bureaucracy is by (i) separating the sphere of policy-making by politicians from administration by public servants, and (ii) developing a public service ethic governed by professionalism and neutrality. New approaches to public management aim to further clarify the relationship between policy-makers and service providers, for example through performance contracts. They seek to enhance the accountability of managers to policy-makers, give managers greater autonomy in day-to-day operations, and strengthen their incentives to respond to clients’ demands. To achieve these objectives, the drive is from bureaucratic chains of command to contractual and market relationships. Transparent contractual relationships within the public sector or between the public and private sectors are ways of making the principal’s demands clearer and giving the agent incentives to fulﬁl rather than frustrate the principal’s policy. Organizational arrangements for service provision might be seen as more or less appropriate to the degree to which they can achieve this. Under market conditions, efﬁcient contractual arrangements will evolve through competitive pressures; but public sector organizations have no such impulse. Reformers therefore argue for the importation of business practices into government. The principles of agency theory have motivated reforms in public sector management with emphasis on performance measurement and incentives. Transactions cost economics draws a distinction between (i) contractual relationships between separate organizations in the market, and (ii) the vertically integrated internal organization of a ﬁrm or bureau (Lane, 2000; Williamson, 1975 and 1985; North, 1990). Each has advantages in terms of efﬁciency. The advantages of contractual relationships are that they specify the basis on which performance will be monitored and offer powerful incentives of proﬁt and loss. A range of different sorts of contract can be negotiated to deal with the characteristics of the service that is

Changing Approaches to Public Sector Management 37

required – detailed and legal or loose and reputational, long term or short term, very speciﬁc or broader franchise arrangements. However, there are difﬁculties in contracting that arise from the limited information (‘bounded rationality’) available to those negotiating about the terms and contingencies, the self-interested opportunism of the parties, and the uncertainty of the environment. Performance requirements are difﬁcult to specify, monitor and enforce, especially for the more qualitative services such as education, which are characteristic of government. So transaction costs are incurred in negotiating and managing contracts. Transaction costs may be so heavy as to justify internalizing the service. Institutions such as the ﬁrm or government bureau come into being for this reason; they ‘have the main purpose and effect of economizing on transactions costs’ (Williamson, 1985:1). While they can offer only relatively weak incentives for good performance to employees – salaries, sacking or promotion in place of proﬁts and losses – on the other hand, they avoid the problems and uncertainties of contracting. New management approaches have favoured contractual arrangements, but the balance of advantage between the greater incentive structure given by contracts and the saving on costs given by vertical integration will vary by country context. It is less costly to make contracts where there are the trust, legal system, information and skills to do so. The balance will also vary depending on the complexity of the good or service being exchanged; the transaction costs of ‘buying a hospital inpatient episode are likely to be higher than those associated with buying a kilo of rice’ (Mills et al., 2001:11). Property rights theory offers a way of understanding the incentives for performance that exist in private ownership and for identifying whether they can be introduced into the public sector. Owners often do not have full control of their assets; for example their rights may be circumscribed by other shareholders’ claims, by legal obligations, by the claims of creditors, or by the rights of employees. They have only ‘residual control’, in return for which they have a right to residual revenue, that is to the revenue remaining after other claims have been met. Owners’ interests, though limited by other claims, are nevertheless aligned to those of the ﬁrm. Managers and employees may also share in the residual revenue of the ﬁrm through bonus schemes, share options and pay increases. They therefore have an incentive to perform. On the other hand, in the public sector, typically, there is (i) no ﬁnancial return, and (ii) no match between ownership, responsibility for decision-making and the distribution of returns. However,

38 The Changing Role of Government

organizational reforms have been proposed to mimic the property rights of the private sector by aligning managers’ incentives with the performance of the organization (Harding and Preker, 2002; Lane, 2000). This brief review has indicated the theoretical origins of recent public sector management reforms. They have been the basis of arguments not just for contracting-out and privatization but also for altering relationships between the public and private sectors and for new approaches to the internal management of public organizations. They indicate some of the key principles of the ‘new public management’ approaches that will be described later in this chapter. The question for this study is whether such approaches are relevant to the delivery of services in developing countries. Paradoxically, while the new institutionalism provides much of the thinking underlying new approaches to public management, it warns against simple notions of organizational design or the re-structuring of organizations ‘according to formulas by means of policy ﬁat’ (Lane, 2000:223). Organizational arrangements exist in an institutional environment, and cannot therefore simply be transplanted between environments where, for example, the structure of transaction costs, property rights, legality and authority and the political condition are different. Chapter 3 will examine how the forces for and against change have played out to produce different results in our case study countries and sectors. Reformers, including international agencies, may put forward technical arguments for organizational change but it is important to recognize that their recommendations enter a political and institutional arena in which they are only some among the bargaining parties. The same organization in a different cultural or institutional setting will itself take on a different meaning. Similarly, the same type of reform may produce different results in a different context.

Pragmatic rationales for public sector reforms Although the above theories provide the intellectual arguments for reforms in the public sector, actual reforms in poor countries have been driven more by pragmatic rationales. First, and perhaps the most prominent of the stimuli for change in the public sector world-wide, are economic and ﬁscal crises in both rich and poor countries dating back to the 1970s and 1980s characterized by increased budget deﬁcits and external indebtedness. Crises led to the urgency for change. This point is illustrated by the economic crisis in Asia, which led to calls for

Changing Approaches to Public Sector Management 39

structural reforms in the relationship between government and business. New Zealand’s reforms in the 1980s were triggered by balance of payment problems and ﬁscal crisis (Flynn, 2002). The second driver for change was the increasing public criticism of the effectiveness and efﬁciency of public services in the face of rising taxation, especially in advanced industrial countries such as the United Kingdom. The third rationale for change has been the emphasis on good governance from the 1990s, which provided some of the impetus for new approaches to public management. An efﬁcient, accountable and responsive public service is seen as a key element of good governance (World Bank, 1992, 1997; Larbi, 1999). For most poor countries the conditions attached to IMF and World Bank lending provided the avenue for the introduction of some of the new approaches to public management.

From old public administration to new public management The arguments for limiting the role of the state were in essence against traditional public administration. The public choice arguments reviewed above were the most critical of the Weberian model of bureaucracy, which had been the dominant model for organizing public administration systems around the world and was the model passed on to newly independent countries in Africa and Asia. The strength and merit of this model lay in its claim to be based on merit, stability or permanency, predictability, political neutrality, professionalism and commitment to duty, accountability through hierarchy and responsiveness to rules. However, by the late 1970s criticisms of the over-extended state, coming mainly from neoliberal and new institutional writers, attracted much attention not only among academics, but also among politicians and policy makers. As reviewed earlier in this chapter, both the state and the public administration that served it were seen as monopoly providers of services and as fundamentally inefﬁcient, and regardless of their customers. In summary traditional public administration was diagnosed as suffering from ‘bureaupathologies’ – characterized by waste, inﬂexibility, the impenetrability of hierarchy, the rigidity of dense internal rules and unresponsiveness to service users (Caiden, 1991; Flynn, 2002; Hughes, 1998). Traditional public administration was widely perceived as not working and not able to meet or anticipate rapidly changing demands on the public services. The above diagnoses presupposed the

40 The Changing Role of Government

solution – new approaches to public management as a response to the dysfunctions of old public administration. In the quest for efﬁciency and effectiveness in government, the adoption of private sector management techniques and practices was advocated to deal with the problems of the old public administration by replacing hierarchy with markets in the hope this would minimize the dysfunctions of traditional public administration. In emphasizing management and markets instead of administration and hierarchy, efﬁciency is considered more important than adherence to rules, effectiveness and results take precedence over legality, ﬂexibility and adaptation are more vital than predictability and responsibility, and competition with private sector delivery is preferred to delivery by government monopoly and bureaucracy. The new public management (NPM) is a set of reforms meant to shift the public sector from bureaucratic to post-bureaucratic organizational arrangements (Kernaghan et al., 2000). The new management approaches are sometimes presented in the literature as the anti-thesis or even as a replacement of the Weberian bureaucratic model. Compared to previous attempts to tackle speciﬁc problems in public administration (e.g. Management by Objectives and Zero-Base Budgeting), NPM is a more holistic challenge to traditional public administration (McLauglin et al., 2002; Kaboolian, 1998). However, the traditional public administration framework cannot be easily replaced by the new public management. Some qualities of traditional public administration, such as stability, predictability and due process remain important. Some scholars (McLauglin et al., 2002:10–11) have questioned whether new public management has enough ‘conceptual coherence to provide a sustainable alternative to public administration either as a theoretical construct for academic research or as an approach to the management of public services’ (see also Lynn, 2001; Riccucci, 2001). In this context, it is perhaps more appropriate to view the new public management approaches as complementary to traditional public administration; an attempt to reform or modernize public administration from within by bringing in new ideas and practices to improve organization and management in government and the delivery of public services.

The era of new public management The meaning and key elements of the new public management (NPM) have evolved since the term was ﬁrst used by Christopher Hood in his

Changing Approaches to Public Sector Management 41

seminal article ‘A Public Management for All Seasons?’ published in 1991. The reforms he described dated back to the late 1970s through the 1980s (see Barzelay, 2002 for a discussion of the intellectual origins of NPM). There are variations in the usage of NPM and in what constitutes its key elements, which are summarized in Table 2.1. Hood’s original conception of NPM was as a convenient ‘short hand name’ for a set of broadly similar administrative doctrines which dominated the public administration reform agenda of most member countries of the Organization for Economic Cooperation and Development (OECD) in the 1980s. Thus the term encapsulated most of the structural, organizational and managerial changes taking place in the public services of these countries aimed at bringing their management systems and methods closer to that of business (Dunleavy and Hood, 1994:9). Current theory therefore draws mostly on the experiences of OECD countries. NPM can thus be deﬁned as a set of particular management approaches and techniques, borrowed mainly from the private forproﬁt sector and applied in the public sector. It is sometimes perceived as an ideology based on the belief in the efﬁcacy of markets and competition, and in business-like management ideas and practices (Ferlie et al., 1996:9; Hood, 1991, 1995). Separating the elements of reforms under the new public management poses some difﬁculties as some components are necessarily and inextricably linked. The inter-connectedness of elements of NPM does not however prevent a typology of its components for practical and analytical purposes. As Table 2.1 depicts, NPM’s key elements may be categorized into two main, but not mutually exclusive strands. On one the hand are ideas that emphasize managerial improvement and restructuring and on the other hand are techniques and practices that emphasize markets and competition. The former include restructuring by decentralizing the management of public services to emphasize managerial autonomy and professionalism. These types of reforms in our case study countries are discussed in Chapter 4. It also involves dis-aggregation, downsizing and emphasis on organizational and individual performance management. The second set of ideas emphasizes markets and competition as a way of giving ‘voice’ and ‘choice’ to users and promoting efﬁciency in service delivery. Market-type mechanisms, as they are called, include contracting out and other forms of private sector participation, creating synergy between public and private sectors, introducing user charges for public services, and emphasizing quality and

Closer focus on quality, feedback from clients and other interests groups

Stress on private sector styles of management practice

A split between public funding and independent service provision

Elaborate and developed quasi-markets as mechanism for allocating resources within the public sector

Creation of competitive environments within and among public sector organizations; ﬂexibility to explore cost-effective alternatives to direct provision and regulation

Shift to greater competition and mixed provision, contracting relationship in the public sector; opening up provider roles to competition

Decentralization: organizational unbundling; new forms of corporate Governance: move to board of directors model Split between strategic core and large operational periphery

Decentralized management environment to replace centralized, hierarchical structures

Hands-on professional management

Ferlie et al., 1996

Shift to disaggregation of units into quasicontractual or quasimarket forms

OECD, 1995

Key elements of NPM by different authors

Hood, 1991

Table 2.1

Creating synergy between the public and private sectors

Maintaining receptiveness to competition and private sector participation

Increased managerial autonomy; reduce central controls

Borins, 1997

Driven by mission/goals rather than rules and regulations

Competition within public services: may be intra-public or with a variety of alternative providers

Catalytic government: ‘steering not rowing’

Decentralized government authority – ﬂexible, less layered forms of organization

Osborne and Gaebler, 1992

42

Increasing focus on results – efﬁciency, effectiveness and quality More focus on efﬁciency with establishment of productivity targets Strengthening the ability of the centre to guide and respond better to change and diverse interests at least cost

Greater emphasis on output controls

Explicit standards and measures of performance

Stress on greater discipline and parsimony in resources use; reworking budgets to be transparent in accounting terms

Source: Adapted from Larbi, 1998a.

OECD, 1995

Anticipatory government: prevention of problems rather than treatment

Deregulation of the labour market

Enterprising government: earning money not just spending it

Result-oriented government: emphasizing outputs not inputs

Customer driven; empowering citizens

Osborne and Gaebler, 1992

Leveraging change through market rather than bureaucratic mechanisms

Provision of human and technological resources that managers need to meet their performance targets

Demanding, measuring and rewarding both organizational and individual performance

Providing high – quality services that citizens value: service users as customers

Borins, 1997

Downsizing

Strong concern with value-for-money and efﬁciency gains

More transparent methods to review performance

Stress on provider responsiveness to consumers; major concern with service quality

Ferlie et al., 1996

Key elements of NPM by different authors – continued

Hood, 1991

Table 2.1

43

44 The Changing Role of Government

responsiveness to users. These reforms in the case countries are analyzed in Chapters 5, 6 and 7. New public management may be seen more as a menu of reforms with different countries adopting different elements of the ‘menu’ to a greater or lesser extent than others. However, despite the variation in responses it is possible to generalize that most countries are at different stages of: • Rethinking and reshaping government and its role • Restructuring and repositioning organizations to enable greater managerial autonomy • Redesigning and improving service delivery to be more responsive and efﬁcient • Reframing systems of performance and accountability; and • Revitalizing human resource capacity and organizational performance. It is worth pointing out here that not all the administrative and management reforms taking place in the public sector fall under the rubric of new public management. For developing countries nonNPM reforms include various efforts at capacity building, political decentralization or devolution, and ethics and anti-corruption reforms (Polidano, 2001; Bangura, 2000). Our research did not examine these types of reforms and therefore they are not discussed here.

A review of new public management in practice As was noted in Chapter 1, the research on which this book is based examined the extent to which the new organizational arrangements for service delivery have been implemented and how these reforms are changing the role of government in particular countries and sectors. This section will locate the new approaches to public management reforms that we examined among much wider international examples. These include various forms of organizational arrangements (decentralized and autonomous agencies) and the use of market type mechanisms (contracting out and charging for services). This will be followed up in subsequent chapters with more detailed analysis of the case study countries. Decentralizing the management of public services One of the key planks of the new management approaches is decentralizing the management of public services by giving public sector

Changing Approaches to Public Sector Management 45

organizations and managers autonomy over operational matters, i.e. not just the freedom to manage but also enabling them to manage. This has taken different organizational forms – executive or autonomous agencies, corporatization, and deconcentration, all involving different levels of delegation of ﬁnancial and other controls to decentralized units. The creation of executive or autonomous agencies is a major trend in the management of public services, based on the idea that the policy arms of government can be separated from the executive or implementation arms. It is an example of reform based on principal-agent theory. The main thrust of reforms in this direction is hiving-off operational arms of ministries or departments to form special purposeagencies. The United Kingdom’s ‘Next Steps’ Agencies best illustrate this form of management decentralization. Once created, agencies relate to the central or parent ministries not through the traditional hierarchy but by some form of performance-based agreement and on an arms’ length basis. These agencies generally have greater managerial ﬂexibility in the allocation of ﬁnancial and human resources in return for greater accountability for results. As Jervis and Richards (1995:10–11) have argued, the executive agency idea was born out of ‘the desire to remove the framework of governance for public services from the arena of contested democratic politics. Placing public services at arms’ length from politicians was intended to give managers sufﬁcient space to get on with management, within the broad framework laid for the public service.’ The United Kingdom has created about 140 executive agencies since the late 1980s with about 75 per cent of civil servants working in them. New Zealand provides another example of agency-type reforms. Policy advisory functions have been separated from operational responsibilities assigned to agencies. As in the United Kingdom, managers on ﬁxed-term performance-related contracts head these agencies with considerable autonomy over ﬁnancial and personnel matters. Australia also has similar initiatives, though on a much smaller scale compared to the United Kingdom. The trend in agenciﬁcation is also found in developing countries. A recent study by Talbot and others (2002) found that Jamaica initially created four agencies in 2000, and has since created four more with plans for more to follow (see also McKoy, 2001). Tanzania has nine agencies out of a target of 46 planned by 2004, whilst Latvia has a staggering 150 agencies (Talbot and Cauﬁeld, 2002). A common trend in some African countries is the hiving-off of revenue collection

46 The Changing Role of Government

departments from ﬁnance ministries to form autonomous authorities as in Ghana, Kenya, South Africa, Uganda and Zambia (Hope, 2002; Larbi, 1999a). This has led to signiﬁcant improvement in revenue collection in some cases but in some others there have been signiﬁcant problems of corruption. The research on which this book is based found cases of management decentralization in all the countries and sectors studied though the degree and types differed. These are analyzed in Chapter 4. It is worth noting that there are some differences between the United Kingdom model of agency, which requires no legislation for its creation, and the model adopted by some other countries, which require some form of legislation. Also the rationale for the creation of agencies in some developing countries (e.g. Tanzania and Jamaica) seems to be driven more by the desire to generate revenue and for agencies to recover cost whilst in the United Kingdom agency creation was driven more by the quest to improve management, efﬁciency and performance in service delivery. Despite the potential beneﬁts of decentralized management through agency creation, there is need for governments to exercise caution. Experiences of some of the developing countries cited above suggest that agencies require the existence of a credible system for monitoring performance and ensuring ﬁnancial accountability before relaxing controls over inputs. Where these controls are weak or undeveloped and arbitrary behaviour cannot be checked, introducing greater managerial ﬂexibility may only increase arbitrary and corrupt behaviour (World Bank, 1997:90; see also Talbot and Cauﬁeld (op. cit.). A critical dimension of management decentralization is the devolution of budgets, ﬁnancial control, and human resource management functions to managers. The belief is that delegation of ﬁnancial responsibility and human resources gives managers some leeway to consider alternative methods of ensuring the provision of required services, and ﬂexibility in meeting targets (Walsh, 1995). Singapore put in place a system of devolution of ﬁnancial management as a prelude to creating autonomous agencies. From 1996, ministries and departments were assigned operating budgets based on target outputs where outputs are quantiﬁable and measurable (Lim, 1997). Borins (1997) points out that in the United Kingdom managers have freedom to manage consolidated running costs of their programmes, and the practice of creating semi-autonomous cost or budget centres is prevalent in higher educational institutions and other public organizations. Namibia permits year-to-year budget transfers. In the research for this

Changing Approaches to Public Sector Management 47

book we found cases of delegation of budgets to semi-autonomous bodies and agencies, including teaching hospitals in Ghana, water agencies in Bolivia and semi-commercial agencies in agriculture (e.g. Cocobod in Ghana) and industry (e.g. Zimtrade in Zimbabwe). These will be analyzed in detail in Chapter 4 where it will be argued that there is a link between organizational autonomy and performance but that managerial autonomy needs to be granted within a clear policy framework and accompanied by appropriate accountability arrangements. Improving and emphasizing performance In return for increased managerial autonomy associated with decentralized management there is an increasing emphasis on output controls in resource allocation and rewards linked to measured performance both at the individual employee and organizational levels. There is a tighter deﬁnition of what organizations and their managers have to produce, whilst easing or even removing controls and constraints upon how they carry out their tasks. There has been a shift in the form of control and accountability mechanisms from inputs and processes to outputs, requiring public organizations, managers and employees to work to performance targets and output objectives. Performance management systems are quite advanced in countries like the United Kingdom and New Zealand. A number of developing countries have also embarked on performance management initiatives, including Ghana, Sri Lanka, Tanzania, and Uganda. In Sri Lanka, Pasteur (1992) noted that the Urban Programme Unit had developed a fairly extensive range of indicators for the purposes of assessing grant entitlement and adjudicating competition among local authorities for performance awards. In Ghana (Dodoo, 1997) and in Tanzania and Uganda (Clarke and Wood, 2001) performance measures are being used to govern the relationship between ﬁnance ministries and line ministries. Uganda’s Result Oriented Management Programme involves the use of output oriented budgeting, whilst Tanzania has also been experimenting with performance-based budgeting; ministries are expected to prepare annual plans with budget ceilings and set clear targets, including performance indicators (Therkildsen, 1999, cited in Bangura, 2000). Similarly, as part of Ghana’s Civil Service Performance Improvement Programme (CSPIP) launched in 1995, ministries, departments and agencies are required to prepare performance improvement plans and deﬁne their mission.

48 The Changing Role of Government

A logical development from the emphasis on measuring and rewarding performance is that audit institutions in most advanced countries, and more recently in some developing countries, now undertake performance auditing in addition to the traditional compliance auditing. Both the National Audit Ofﬁce (for central government institutions) and the Audit Commission (for local governments) in the United Kingdom are very advanced in the use of performance auditing. A number of countries have also introduced performance agreements or contracts between public enterprises and government and between funders and providers of services. These include Bolivia (Mallon, 1994), Ghana (Larbi, 2001; Christensen, 1998), Senegal and other francophone countries in Africa, Pakistan and India (Islam, 1993; Shiley and Xu, 1997) and several OECD countries, including Australia, New Zealand and France (OECD, 1999). A World Bank (1995) study identiﬁed 385 performance contracts in 28 countries spread across Asia, Africa, Latin America, the Middle East and North Africa, and Central Europe. Experience suggests that performance contracts or agreements help to clarify goals, increase managerial autonomy, and encourage dialogue between government public managers. However, as Nellis (1990) has observed ‘their beneﬁts have been oversold’. In most developing countries the potential beneﬁts are undermined by various implementation problems, including the lack of ‘hard budgets’ for enterprises, poor management of information, lack of enforcement of rewards and sanctions and weak government commitment to contract obligations. Alongside organizational level performance, a number of countries have also introduced performance evaluations and incentive systems at individual levels. Chief executives of agencies in the United Kingdom and New Zealand are on performance related ﬁxed term contracts. In Ghana chief directors (formerly permanent secretaries) are not only employed on ﬁve-year renewable contracts but are also expected to sign performance agreements with their respective ministers. Zimbabwe introduced a similar scheme for its top civil servants in 1994 (Therkildsen, 1999). Experience, however suggests that top civil servants in these countries do not usually have control over operational resources and have limited managerial freedom due to centralized control over employment and expenditure. This makes it difﬁcult to hold them accountable for the poor performance of their ministries, and thus provides an escape route for failure to meet the terms of their contracts. The limited experience in performance appraisal in developing countries, however, suggests that there are problems in deﬁning and

Changing Approaches to Public Sector Management 49

measuring performance targets for employees in some cases. The system may also be undermined by political patronage in employment and promotions processes and lack of enthusiasm in the system in some cases. Market-type mechanisms: contracting and charging for services Closely associated with the managerial emphasis in government, a number of countries are using market-type mechanisms (OECD, 1993a; Walsh, 1995). This trend is manifested in several forms – increasing focus on the customer, user fees or charges, contracting out of service delivery to the private and voluntary sectors, public-private partnerships, and outright privatization. Our research covered contracting out and user fees, hence the discussion that follows will be limited to these. Contracting out has become one of the key forms of public services delivery. This and other forms of private sector participation are being used in those public services that cannot be privatized for strategic, political or other reasons. Privatization (sale and transfer of assets to the private sector) is sometimes treated as part of ‘new management’. However, in fact what it does is to change the portfolio of activities carried out by the government, with the impact of reducing the size of the public sector. New public management is more about how to improve the management of activities that remain under public sector ownership by applying private sector management practices. Much has already been researched and written on privatization in the 1980s and 1990s in both developed and developing countries. Thus in this study privatization is not discussed, except in as much as it changes public sector practice. More signiﬁcantly it requires government to adopt new ways of achieving policy goals by regulation, as will be discussed in Chapter 8. Contracting out is increasingly being applied as a management tool, especially where a complete market solution or outright privatization cannot be employed. As a market-type mechanism, it is used to foster competition for the market with governments contracting with private and voluntary sector organizations, which competitively bid to provide services (Savas, 2000; Walsh, 1995). Contracting out thus puts competitive market forces directly at the service of government. Though contracting out or outsourcing of public service delivery is not new, what is new is the extension of the practice to cover activities that have traditionally been carried out by in-house arrangements. Where outputs are easily speciﬁed but direct competition is impossible, competition managed through various forms of contracts such

50 The Changing Role of Government

as service contracts, management contracts, leases, and long-term concessions, can yield efﬁciency gains. In the United Kingdom, outsourcing of services has been extensively used in local government, following the Local Government Act of 1988, which introduced compulsory, competitive tendering. The range of services delivered through outsourcing has extended from blue collar services such as garbage collection and cleaning to cover professional services such as accounting and auditing, legal services, information technology and in some cases even human resource management services such as recruitment (e.g. Lambeth Council in London). Contracting out is also widely used in other OECD countries. Our research shows that, in the health sector, contracting out of some non-clinical services is common in India, Mexico, Thailand and South Africa, whilst some clinical services are contracted out in Zimbabwe and South Africa (Mills et al., 2001; Bennet and Mills, 1998). Tanzania is piloting the contracting of cleaning services in one department with the intention of extending it to other ministries and departments and to other services. In Brazil, the World Bank (1997:88) notes that contracting out of road maintenance to private contractors led to 25 per cent savings over the use of government employees. A number of African countries have also contracted out urban infrastructure development. The case of Agences d’Exécution des Travaux d’Intérêt Public (AGETIPS), nonproﬁt, non-governmental agencies, is widely cited as very successful. Operating in francophone African countries, AGETIPS enter into contractual arrangements with governments to implement infrastructure projects. With increased managerial autonomy, transparent procedures and protection from political interferences, these organizations have, inter alia, regularly obtained unit prices ﬁve to 40 per cent lower than those previously obtained and have been able to complete projects on schedule with relatively minimal cost overruns (World Bank, 1994). Most of the above examples are service contracts, which are usually short term. Various forms of management contracts are also in use in different countries. Leasing and concessions, which are medium to long-term contracts, are widely used for infrastructure projects and in the management of state assets for service delivery in utility sectors. The World Bank (1997) cites the case of Malaysia, where the leasing of the Port Kelang has increased operational efﬁciency. Similar arrangements have been used for water supply in several Latin American countries, and in most francophone African countries such as Guinea, La Côte d’Ivoire, Gabon and Mali with some improvement in technical

Changing Approaches to Public Sector Management 51

efﬁciency. Management contracts are used in electricity supply in the Gambia, Gabon, Guinea-Bissau and Rwanda (Olowu, 1999). Chapter 7 analyzes the use of different forms of contracting arrangements in the four sectors (agricultural marketing, business development, health and water). Sufﬁce to note here that the degree of application of contracting differs according to sector and service characteristics, as well as the political sensitivity of the service. Whilst contracting out is becoming popular in the public sector, there is need to exercise caution in its application in a developing country context. Introducing contracting out assumes that the capacity exists within the private sector to take on the provision of contractedout services. This is not always the case in developing countries and for some services where both markets and government capacity are weak. Weak contract enforcement, corruption and other institutional factors tend to undermine the efﬁcacy of contracts, particularly in countries where markets are weak. In general, greater and effective use of market mechanisms must be accompanied by effective regulatory capacity. As will be shown in Chapter 8, regulation of service provision is one of the new roles that governments in developing countries are struggling with. User fees or charges are another market-type mechanism that has increasingly become common in developing countries and transitional countries, particularly those that have undergone IMF/World Bank structural adjustment reforms. The intention is to recover fully or partly the cost of service delivery by charging users. Where charges already existed (e.g. water and electricity tariffs), these have signiﬁcantly increased as central and municipal government providers of services attempt to balance their ﬁnancial books and stabilize funding for service delivery. Public utilities such as water and electricity have seen increases in their charges to consumers in recent years in a number of developing countries. Before reforms in the 1980s and 1990s, the provision of social services such as public education and public health care in most developing countries were almost universally based on free access. They were ﬁnanced through direct budget allocation. Even where fees were charged, these were minimal. However, in many countries, services are now subject to user charges in order to recover part of the cost of service delivery. For example, fees have been introduced at different levels of education in Ghana, Uganda, Malawi and Kenya, apart from basic education. There is evidence that, in comparison to other developing regions, user fee reforms have been most extensive in sub-Saharan Africa (Mills et al.,

52 The Changing Role of Government

2001; Shaw and Grifﬁn, 1995, cited in Adams and Hartnett, 1996). This is because the gap between resources and health and education needs, and the inﬂuence of international donors have been greatest in Africa. There are similar developments in some Latin American countries (e.g. Venezuela) where user fees in health and education have been introduced. The need to raise additional revenue to supplement government revenue in the face of increasing demand for services is the key rationale for the introduction of user fees or charges. However, user charges do not only place emphasis on raising funds, but also on preventing over use of services by consumers by making them more cost conscious. They are also meant to make providers more efﬁcient by improving quality (Adams and Hartnett, 1996:7). The assumption here is that since users are paying for the service they will only use it when they really need it and will insist on better value for money. This is part of the move towards a market-orientation in the provision of public services. However, the introduction of user fees has brought its own problems, ranging from corruption and mismanagement of revenue collected to issues of equity and access to services by the poor. For example in South Africa, Shore (2002) reports that the introduction of user fees to recover cost in the rural water sector had a devastating effect on health in KwaZulu-Natal province. Thousands of poor households who could not afford the charges resorted to the use of rivers and stagnant ponds, which became a major contributory factor to the outbreak of cholera and consequent loss of hundreds of lives. He argues that the free supply of water and sanitation to poor people in South Africa would cost signiﬁcantly less than the medical and economic costs of cholera. In Ghana, Zambia and other African countries the introduction of user fees in health has led to a drop in hospital attendance. These issues will be discussed further in Chapter 5.

Conclusion This chapter has provided an overview of theoretical and practical arguments for the new approaches to public management. The intellectual arguments for new approaches to public management are rooted in neo-classical as well as new institutional theories, in particular public choice and principal-agent theories. New public management represents a menu of reforms; there are different aspects or dimensions to it. These centre around restructuring to allow for

Changing Approaches to Public Sector Management 53

increased managerial autonomy and the introduction of market-type mechanisms in the public sector. Whilst new public management may be seen as complementary to traditional public administration, it certainly does not replace it. The application of new approaches to public management is not limited to their heartland of developed OECD countries. Many developing countries are applying market-based reforms to their public services in order to improve management in government and the delivery of public services. However, this does not justify the claims of convergence or of the universality of new public management reforms. In developing countries hierarchical bureaucracy has not been substantially replaced and there is much to do to build the capacity of traditional bureaucracy to meet the preconditions for the implementation of the new management reforms (Schick, 1998; Manning, 2001). Even within the OECD countries, comparative studies show that there are variations in the adoption and application of new public management. Ridley (1996:19) for example concludes that the extent to which countries adopt management reforms in government may relate to ‘each country’s administrative culture and the resistance to change of its established order.’ Thus context is important. There are mixed experiences in applying NPM reforms. Whilst best practices indicate improvements, other experiences suggest constraints and capacity limitations in applying these practices to the public sector, particularly to developing countries. The rest of this book will analyze and present evidence on the application of new approaches in four core developing countries (Ghana, India, Sri Lanka and Zimbabwe) and several reference countries. What new internal management restructuring or organizational forms have been introduced across the four sectors (business services, agricultural marketing, health, and urban water supply) in the four countries? To what extent have reforms been implemented and how well have the new organizational forms performed? What are the implications of reforms for government capacity to manage services, to control or regulate providers? What are the key capacity constraints? These are some of the questions to be addressed in the rest of the book.

3 The Politics of Service Reform

Introduction This chapter asks whether citizens and politicians (the ‘principals’) in developing countries have been instrumental in demanding, designing and directing reforms in service delivery.1 It has been argued that this was the case in the advanced countries that led the process of reform towards new public management models – such as the United Kingdom, New Zealand and Australia. In those cases, an underlying factor was growing public expectations of improved and ﬂexible public services in the context of economies that were offering increasing choice to consumers (Hood, 1991; Manning, 2001). Underlying economic weaknesses were triggered into ﬁscal crisis and government budget deﬁcits by the rise in oil prices in the early 1970s. As noted in Chapters 1 and 2, the feasibility and effectiveness of maintaining the level of state activism in economic planning and social welfare was called into question (Larbi, 1999a; Caiden, 1991; Dunsire and Hood, 1989; Walsh, 1995). Governments were able to nurture changed public expectations and build a constituency of support in favour of an assault on conventional bureaucratic service provision and on the vested interests associated with it (Caiden, 1991). They drew on resurgent neo-liberal ideology and economic theory of ‘state failure’ (Pollitt, 1993; Walsh, 1995). Public service reforms were often not abrupt but protracted through a continuous cycle of incremental learning and modiﬁed reform, often guided by small reform units with private business advisers (Flynn, 1997, p. 36). In the case of developing countries, less is known about the reform process, particularly in the social sectors. Are their paths to reform a 54

The Politics of Service Reform 55

response to changing public expectations, triggered by crisis, oriented by ideological commitment, led politically but steered by small teams of policy advisers, incremental and adaptable to experience? Who are the principals – the advocates and leaders of change? Who are their agents and how far are these motivated to act in the principals’ interests? Table 3.1 indicates the types of reform that were analyzed in the four service sectors: privatizing and contracting out the management of public services, decentralizing management to semi-autonomous units within the public sector, the application of charges to users of services, and the development of enabling and regulatory roles by public agencies. For Ghana and Sri Lanka these were ‘second generation reforms’ Table 3.1

Reform types analyzed by sector

Type

Health

Urban water

Business services

Agricultural marketing

Decentralizing Autonomous Corporatization Autonomous management hospitals of utilities agencies within public sector

Corporatization of marketing boards

Charging users

Charging users

Charging for quality control services

Contracting out

Contracting Franchise and ancillary and concession clinical services

Other private sector participation

Water tariffs and charges

NGO and informal provision

Enabling private sector

Tax breaks, loans, grants and subsidies

Regulating private sector

Regulation Environmental of hospitals, and economic professions regulation; and contracting pharmaceuticals

Charging for technical services

Contracting of services

Divestiture of Liberalization state textiles of markets manufacture Marketing, Market business information promotion services and advisory services; representation Quality assurance

56 The Changing Role of Government

that followed the earlier thrust during the 1980s to reduce state intervention in the management of the economy. For the later reformers, India and Zimbabwe, ﬁrst generation reforms occurred later and in parallel with the reform of public services.

Political stakes in the reform of public management Underlying the question ‘who directs the reform process’ is the question ‘who participates in it at all’. Grindle and Thomas (1991) distinguish reforms that become matters of wide public mobilization from those that generate responses largely within the bureaucratic arena. They argue that the stakes are higher in the ﬁrst case; determined political support is needed to drive them through. In the second case, the political stakes are lower; the crucial issues are within the competence and compliance of the bureaucracy. The factors that Grindle and Thomas identify as determining whether reforms become openly political or are managed internally are summarized in Table 3.2. These include the distribution of the concrete costs and beneﬁts of reforms between government and sections of the public, and also factors to do with the ‘visibility’ of reforms, their administrative complexity, whether public support is required for their implementation, and the duration of the process of implementation. So, reforms such as the introduction of user fees are likely to become matters of open public debate – the beneﬁts are most obviously to the public purse, the costs are to consumers, and their impact is immediate and visible. Decision-makers therefore confront high political stakes in pushing such reforms. According to this framework, many of the management reforms listed in Table 3.1, for example contracting out or management decentralization, are unlikely to generate a public reaction. They are a matter of detailed working out and interaction within the government system, requiring a high degree of sustained technical competence and commitment. They may have great signiﬁcance for the population but only in the longer term and not in direct and immediate costs or beneﬁts. The arena of reform is instead likely to be within the bureaucracy where the interests and behaviour of ofﬁcials and professionals are affected. Grindle and Thomas argue that, in these cases, the political stakes are relatively low: the risk to government is only of failure to achieve bureaucratic compliance rather than of loss of public support.

The Politics of Service Reform 57 Table 3.2 reform

The public and bureaucratic arenas of response and resistance to

Characteristic of reform

Features of reforms in the public arena, requiring political support and stability

Features of reforms in the bureaucratic arena, requiring bureaucratic compliance

Dispersal of the costs

Costs have wide impact among the population

Costs focus on government institutions

Dispersal of the beneﬁts

Beneﬁts are focused on government

Beneﬁts are not immediately felt by bureaucracy and only in long term by public

Technical and administrative complexity

Reforms have low administrative content and can be done quickly

Reforms are administratively complex

Level of public participation

Reforms require wide public involvement and are ‘visible’

Reforms require limited public involvement and are ‘invisible’

Duration and visibility of reform process

Reforms can be achieved quickly and are visible

Reforms require sustained effort with few immediate visible returns

Examples

User fees Privatization of services

Contracting out Decentralized management

Source: Adapted from Grindle and Thomas (1991).

This chapter suggests a modiﬁcation of Grindle and Thomas’s argument in two main respects. First, as Chapter 1 showed, there is not such a clear distinction between the bureaucratic and public arenas. In weaker political-economies, particularly the African cases in this study, the bureaucratic arena is itself highly politicized and inter-connected with societal interests; it is where power, employment and patronage are concentrated. So, reforms that in more advanced countries have had relatively low political salience, in poorer and weaker government systems may be highly politicized. Second, where policy reform is led by external agencies (donors) rather than by government, the public arena is weakened and policy-making becomes more exclusive and closed. So, reforms may be both politically salient and covert.

58 The Changing Role of Government

Principals and agents in public service reform The following sections use a broad principal-agency framework, as described in Chapter 2, to analyze the relationship between the actors in the case of the studies of reform in the four sectors – particularly health and water, but also agricultural and business development services. Some reference is also made to education. In a democratic polity, the ultimate principals are the citizens, or consumers of speciﬁc services. In principal-agent theory, they are ‘principals’ in the sense that politicians, as agents, seek their mandate from and act as the representatives of the public. In their turn, appointed ofﬁcials are, in principle, the agents of political leaders in executing policy. Each has a measure of autonomy and each has their own interests to advance. The likelihood of the principal effectively controlling the agent depends on • how much information the principal has about the performance of the agent, and; • how far the principal can structure the relationship so as to control the agent or give incentives so as to make the agent’s interests correspond to the principal’s. The following sections ﬁrst consider the balance of inﬂuence between external bodies – the international ﬁnancial institutions and donors – and national governments. The chapter then considers interests and institutions at a national level, breaking these down into two categories: (i) the ‘principals’, that is the public and the politicians who in formal political theory would decide the priorities of government; (ii) the ‘agents’, that is the senior core government ofﬁcials, the ministry-level ofﬁcials and professionals, and public sector workers;

External bodies – agents or principals? As Chapter 1 showed, economic crisis was a key catalytic factor in bringing about reform in the case study countries. This is probably true always and everywhere, and not only in developing countries. Crisis generates both a need for change and also possible ‘windows of opportunity’ by throwing the normal rules of the game into ﬂux (World

The Politics of Service Reform 59

Bank, 1997; Grindle and Thomas, 1991). There were internal reasons why collapses of government spending power should lead to radical proposals for change, but, particularly in the case of the poorer countries studied here, these conditions also created a susceptibility to external pressures. Among these are most strongly the multilateral or international ﬁnancial institutions, the International Monetary Fund and World Bank, which act not only on their own behalf but also as bodies that inﬂuence the agenda of bilateral donor agencies and the credit policies of commercial banks. They present themselves as ‘agents’ of nationally determined programmes. It is difﬁcult to disentangle the contribution of the different internal and external actors in bringing about shifts in policy. However, in Ghana, Sri Lanka, India and Zimbabwe, multilateral lenders played a key role with bilateral donors in support. What is distinct about the involvement of external actors in developing countries is that they come not just as advisers but as the ﬁnancial sponsors of reform, which gives them particularly great inﬂuence (Corkery et al., 1998). There are cases where reforms have been advanced in the absence of real local support, but few where the international agencies have been wholly absent, even if only in the wings. However, governments are, of course, not without inﬂuence; the balance of power between external development agencies and government varies by country, service sector and the speciﬁc reform issue. The inﬂuence of the IFIs and donors, even in the absence of real local support, has been more readily asserted in the immediate reforms associated with stabilization – divestiture in industry, abolition of external trade controls, cuts in civil service expenditure. This was not so much as part of sector speciﬁc reform programmes but as a response to the more general requirements of structural adjustment, orchestrated by the IMF and World Bank. Fiscal crisis and the IMF’s stabilization package required the divestiture of state-owned enterprises, and the removal of controls on imports and on prices. Ghana began the process of privatizing public enterprises in 1983, dismantling import licensing, and, with the support of USAID, reforming the regulation of enterprises. Zimbabwe’s structural adjustment programme removed state controls of prices on most industrial (and agricultural) goods and liberalized trade including in the textiles and garments sector. Sri Lanka’s autonomous programme of liberalization and divestiture was given further impetus when it signed up to an IMF/World Bank Policy Framework in 1987. Many of these were ‘stroke-of-the-pen’ reforms whose implementation required little more than a change of policy.

60 The Changing Role of Government

More complex were the ‘supply-side’ reforms designed to promote the development of industrial and agricultural market producers (Jackson, 2002). Aspects of agricultural trade were also directly affected by IMF and World Bank conditions, particularly in the African countries where the crisis was deeper. Import controls and price subsidies were challenged and swept away where the international ﬁnancial institutions had sufﬁcient inﬂuence; monopoly marketing boards were privatized, corporatized or lost their monopoly status (Hubbard, 2003). However, where reform affected only domestic agricultural marketing, there was less international pressure for change. So, in Ghana, World Bank reform efforts focused on the cocoa sector and to some extent rice marketing; much less exposed to pressure was maize, which is internally traded. In India which is much less subject to donor pressure, the ‘reform process has largely bypassed the agricultural marketing sector’ (Ahluwalia, 1996 in Hubbard, 2003). The water sector was not immediately affected by structural adjustment programmes. It was initially bilateral donors that supported reforms leading to more efﬁcient and equitable management of water resources. For example, in 1994, four donor countries (Germany, The Netherlands, Norway and the United Kingdom) supported Zimbabwe’s development of a water resource management strategy. Together the bilateral donors and World Bank commissioned studies and supported proposals for pricing and structural reform in the water sector, although national leadership of the process was emphasized (Batley, 1998). However, by the end of the 1990s a new international convention had developed and governments everywhere were ‘under considerable pressure from the multilateral funding agencies to move towards concessions or face a block on further funding to the sector’ (Nickson and Franceys, 2003). In general and well beyond these particular case study countries, it took much longer for public management reform to affect state intervention in the social sectors – health and education. In these areas, the pressure of donors, the IMF and World Bank has been much more incremental. Certain aspects of reform in the social sectors could be introduced quickly under conditionality and without real political support, as long as there was little organized political resistance – for example, rises in tariffs and introduction of user fees. But reforms in the organization and roles of government bodies, in changing attitudes to and the working relationship with the private sector required local political and administrative commitment over the long term. This was

The Politics of Service Reform 61

particularly the case in the health and education sectors but also in water and sanitation. In all these, large administrative and professional apparatuses implied political resistance, against which the model being proposed by reformers was less well worked out and clear than in the economic sectors (see also Nelson, 2000). The social sectors – and also internal agricultural trade – have been relatively freer from the attention of the international ﬁnancial institutions, and therefore generally slower to reform. In the health sector, chronic ﬁscal crisis, and then the new poverty agenda and sector-wide approaches of donors, eventually forced re-consideration of the role of the state, particularly in Africa. However, in this sector, the technical advisers of donor and international organizations generally played the role of encouraging, ﬁnancing and endorsing rather than of imposing reform. Pressure was greatest where the crisis was deepest, with the effect that there were more elaborate plans for reform in Zimbabwe and Ghana, and even in Thailand after the 1997 crisis, than in South Asia. Donors encouraged hospital autonomy in Ghana, Zimbabwe and Thailand, user fees particularly in Zimbabwe where fee increases formed part of the ﬁrst structural adjustment programme, and contracting out of services, again particularly in Zimbabwe (Mills et al., 2001). Donors and particularly international ﬁnancial organizations have been fundamentally important actors in reform, most directly in the case of demand-side, stroke-of-the-pen, economic reforms. However, there were some cases where governments had been the initiator of liberalizing reforms. Among the four core countries, this was most clearly the case in Sri Lanka which, in 1977, launched its own liberalization programme involving the privatization of state enterprises and the promotion of foreign inward investment. India has been a modest liberalizer but it made its own decision for compliance with the terms of the World Trade Organization when, in 1995, it undertook a tariff reform that opened its previously highly protected textiles industry to imports. Among the reference countries, South Africa and Argentina launched wide-ranging programmes of reform in the early 1990s, backed by the international institutions but clearly primarily in response to internal dynamics: respectively, the fall of apartheid, and political crisis arising from hyper-inﬂation. Even where the international ﬁnancial institutions and bilateral donors asserted inﬂuence, governments were not powerless to modify or resist unwanted changes. The effect of health reforms was little more than ‘nibbling away at the fringes of the state, rather than fundamentally

62 The Changing Role of Government

changing its role in health’; countervailing forces or inertia preserved the dominance of the state (Mills et al., 2001). In spite of pressure from the IMF and World Bank for full liberalization of both the external and internal trade in cocoa, the Ghanaian Government successfully defended its choice to move slowly so as to minimize the risk to farmers (Hubbard, 2003). The following sections will examine the powerful internal forces that stood against change. Politicians and ofﬁcials had to weigh the balance of advantage and decide whether to work with donor proposals, openly to resist them or simply to let them slide. While governments were not powerless, the international ﬁnancial institutions and donors were almost always a signiﬁcant enough presence in policy determination to hold in question any straightforward view of the relationship between national principals and national agents. IFIs and donors were usually a deus ex machina that produced policy extraneously, breaking the direct relationship between citizens, politicians and service providers.

The principal-agency characteristics of services The leading question is ‘who, in practice, is the principal?’ Who determines policy preferences and mandates agents to implement them? The agency problem is that the nominal agents frequently have little incentive to serve the goals of the nominal principals. The next section will explore the role in the reform process of the nominal principals – the public (the citizen-consumer) and political leaders. The following section will look at the role of the nominal agents – public ofﬁcials in core and line ministries or public bodies. This section will ﬁrst indicate how the relations between these actors may differ by sector. The incentives to responsive service delivery differ, since the characteristics of sectors inﬂuence the capacity of principals to organize and assert control. Drawing on the research on Ghana, Zimbabwe, India and Sri Lanka, Table 3.3 summarizes the balance of power between principals and agents in three service sectors. The table is schematic, ignoring the difference between country contexts, and grouping business development and agricultural services into one category for comparative purposes. It also ignores differences between parts or stages of each service from policy-making, through ﬁnancing, to delivering and monitoring. Here we focus on the policy-making or reform process.

The Politics of Service Reform 63

The text below identiﬁes the elements of control by principals and agents as being composed of the following factors: Control by principals Capacity of client groups to organize is greater where • the service is used regularly and predictably, not only in crisis • the service is area-based Clients' capacity to exercise inﬂuence is greater where they have • information on the quality of the service • choice about whether to use the service Policy-makers' capacity to control and incentivate provider organizations is greater where

Control by agents Capacity of provider groups to organize is greater where • the service has a high information or professional content • professional organization is strong • unionization is strong • contractors are large, few and have speciﬁc skills and assets Agency structures that favour provider control are • monopolistic • led by professional staff

• service providers' effort and outcomes are measurable • information on the provider's performance is available • the provider's contract is speciﬁable and enforceable

The following sections set out the analysis in more detail, but Table 3.3 indicates the broad conclusion. Control by citizens and clients is greater where they can organize themselves and can exercise informed choice, and by policy-makers where they can gather information and measure the performance of providers. The providers or agents are more likely to be able to assert their own control where they can form professional groups or can unionize, where there are few agents with the necessary capacities and where the service is monopolistic. Taking these factors together, the table shows that

Strong: Users have choice about whether to use the service.

Business and Medium/Strong: agriculture Services are for services speciﬁc user groups. Stronger in industry than agriculture

Weak: Information asymmetry limits choice, but alternative providers exist.

Medium: Clients have information on service, but no alternative suppliers of piped water.

Weak: Clients are scattered and use service in crisis.

Curative health

Capacity of clients to exercise inﬂuence

Urban piped Strong: Service is water area-based and regular, facilitating client organization.

Capacity of client organization

Medium: Service effort and output difﬁcult to assess; information asymmetry. Difﬁcult to specify contract

Strong: Service effort and output is easy to measure and monitor. Relatively easy to specify contract.

Weak: Service effort and output difﬁcult to assess; information asymmetry. Difﬁcult to specify contract

Dominant: Monopolistic with a high degree of autonomy in management, or large contractor.

Dominant: Large direct deliverers with high autonomy.

Weak: Small Weak: Nonorganizations with monopolistic. a relatively weakly established professional base and low unionization.

Medium/strong: Engineering dominates provider or ministry; moderate unionization; big contractor interests

Strong: Strong unionization and professional interests in direct provider organizations and ministry. Some major suppliers/ contractors.

Agency structure

Capacity of control by agents

Policy-makers’ Providers’ control of performance organization

Capacity of control by principals

Principal-agent relationships by service

Service

Table 3.3

64

The Politics of Service Reform 65

principals (clients and policy-makers) are weakest and the agents (the service providers) are strongest in the case of curative health services. In urban water supply, there is a greater possibility of balance between the two sides. In business and agricultural services, the agent-providers have relatively lesser possibility of dominating the principals.

The ‘principals’ – citizens and political leaders This section ﬁrst surveys the broad characteristics of the engagement of citizens and politicians in policy development and then examines how their behaviour differs between sectors. The citizen-consumer as principal A fundamental ﬁrst point with regard to the involvement of the public is that most of the reforms investigated here were of little direct interest to users or citizens in the case study countries. Managerial reforms – the decentralization of internal management, contracting out of functions, developing support and regulatory systems – were hardly likely to inspire public passion. Given a low level of public engagement with policy in the countries under study, reform generally remained within the ‘bureaucratic arena’. Some reforms had a more immediate political impact – user fees, ‘privatization’ and the liberalization of markets. However, public demand for reform was rare; more often, there was pressure on politicians for the defence of existing rights and privileges than for change. In practice, with few exceptions (see below), the public and particularly the poor, were largely outside the policy process. Producer groups (farmers, traders and industrialists) appeared sometimes as lobbyists against change. Consumers of social services were mostly an implicit (not organized) pressure on politicians to preserve existing rights. In most cases, consumers were the ‘silent stakeholders’ in reform. The research programme’s surveys of users’ opinion of health and water services in Ghana, Zimbabwe, Sri Lanka and India found strong critical opinion about the price and quality of the services but widespread ignorance about the proposals for reform of those sectors. The reform issue that did arouse respondents’ comment in the Ghana and Zimbabwe surveys was that of user fee and tariff rises (Rakodi, 2000). In spite of the failings of public services, the surveys showed that users accustomed to public sector provision generally supported its continuation and opposed alternative arrangements. This seemed to be based on a belief that entitlement to basic services could best be

66 The Changing Role of Government

assured by governments, and on experience of privatizations which had left users worse off and with less say in decisions. Reforms have often threatened public institutions in which users have a basic conﬁdence. Moreover, users have not been convinced that the reforms addressed their concerns about the low quality, queues, rationing, staff shortages and disrespect for customers often associated with public services (Rakodi, 2000). In spite of these concerns, there were few instances in which surveyed attitudes were translated into active intervention in the policy process. Political leaders Who then are donors’ local allies for change? Previous sections showed that, in almost all cases, reform was associated with structural and sector adjustment programmes. Formally, therefore, governments had ‘bought in’ to the reforms, although they might often have little real commitment. This level of agreement, or at least acquiescence, was concentrated at the level of the president or prime minister and the key ministers, particularly ministers of ﬁnance. It was backed by core central agencies associated with the political leader, such as the public service commission and head of the civil service. The focus of reform leadership at the highest political level is noted also by Grindle (2001 and 2003) and Nelson (2000). In the core case study countries in South Asia and Africa, political leadership at other levels – sector ministers, parliaments and individual politicians – rarely played an important role in advocating and driving through reforms, although it might obstruct them. There was often formal political tolerance of the reform process, combined with informal inertia or sabotage by politicians in the legislature and even in government. Reform design did not much involve sector ministers except to ratify the proposals of the political leadership. Exceptions among reference countries were Argentina and South Africa where there were coherent party programmes and a broader based political commitment to reform including among service sector ministers. In Ghana, ministers responsible for water under the Rawlings régime and under the current government argued for private participation in urban water supply but backed down in the face of the resistance of the powerful trade union and public mobilization (Larbi, 1998a).2 Political acquiescence may be enough to allow donors with other local actors to initiate reform, but probably not to sustain it. Politicians, of government or opposition, outside the core executive had more inﬂuence on the implementation than on the design of liberalizing

The Politics of Service Reform 67

reforms. Usually their posture was negative, particularly where change was disruptive of the established distribution of beneﬁts and systems of patronage. Sector differences As indicated in Table 3.3, sector characteristics may affect the capacity of citizens and service clients to control the performance of service providers: i)

Organization and deﬁnition of client groups: While agricultural marketing and business support services are oriented to production and apply to speciﬁc producer groups, health care and drinking water supply are consumer services applying to the general public. Water consumers are deﬁned territorially and can organize on that basis, but health care users encounter the need for services individually and in a state of vulnerability. ii) Exercise of choice and inﬂuence: The sectors are differentiated between those where users have some choice about whether to use services at all (agricultural and business support services); one (health care) where they have some choice about which public or private service provider to use but inadequate information on the quality of the alternatives; and one (piped water) where they may have little or no choice. iii) Measurability of performance: Policy-makers and users have stronger possibilities of control where they have information on and can assess the performance of producers. Water supply is the most clearly quantiﬁable, tangible and measurable; health care, business and agricultural support are qualitative services with greater difﬁculty of performance measurement. These criteria indicate a scale of capacity of citizens and clients, as the ultimate principals, to control policy-makers, and for the latter to manage the performance of the provider organizations that are supposedly their agents. Industrial promotion and agricultural marketing services present the greater possibility of the recipient group organizing to demand services: ﬁrst, industrial (and perhaps agricultural) producers are a deﬁnable group and, second, they have some choice about whether to apply for the service. Piped water supply services are typically monopolistic but, if they are supplied locally, consumers may organize territorially to hold providers to account for measurable physical outputs. The weakest clients are those of curative health services

68 The Changing Role of Government

who may have some choice but little information about how to exercise it and little capacity to organize. Agricultural marketing The ideologies supporting state intervention in agricultural markets are of national food security, the protection of farmers and consumers from the market, and post-independence redistribution to previously disadvantaged producers. There is little political attraction in challenging these conventions and liberalizing markets, where to do so would unsettle the beneﬁts of subsidy and protection for electorally or economically powerful groups. The social pressures against change were particularly organized and articulate in the agricultural services sector. They were effective in slowing liberalization and the reduction of government subsidies, especially in South Asia where peasant farmers form an important political bloc and where the inﬂuence of the international ﬁnancial institutions was less strong. In India, politicians face a line-up of support for the status quo – urban consumers, farmers capable of producing surplus, traders, millers, parastatal personnel and government ofﬁcials. The political interest is in the defence of entrenched welfare rights rather than in the promotion of uncertain gains from freer trade and competitive markets. Politicians ‘aspire to be identiﬁed as the guardians of the poor’ (Kohli and Smith, 1998). In Sri Lanka, where rice production employs around half the labour force, farmer protests had a powerful effect in slowing and limiting the liberalization of external trade and in reducing farm input subsidies. The Ghanaian Government’s caution in following the World Bank and IMF proposals to open the cocoa trade can be attributed partly to the political sensitivity of exposing farmers to abrupt change. The social interests in retaining state control were not monolithic. Potentially, urban consumers had contradictory interests to those of producers, and traders to those of producers. However, as in Sri Lanka, it was easiest to seek to square the circle: to defend high prices to farmers and subsidies to consumers, the losses being covered by government, rather than to advocate change out of which nobody could be sure that they would gain (Smith and Ellis, 1997). Zimbabwe was one of the few cases where there was a lobby – the large-scale commercial farmers – in favour of the liberalization of trade and privatization of agricultural services. However, they received temporary political toleration rather than support and their gains were short-lived. Under pressure of political and economic crisis, in 2001 the government

The Politics of Service Reform 69

resorted to export prohibition and compulsory grain purchases at ﬁxed prices to feed the population. Urban water Large portions of the population are typically outside the range of public water supplies. Even in the wealthiest of the countries studied, Argentina, in the early 1990s around 30 per cent of the population of Greater Buenos Aires lacked access to the public water supply and instead made their own illegal connections or developed their own water supply. During the previous ﬁve decades the viability of the water company had been undermined by party campaigns for a combination of low tariffs and politicized investment decisions. The excluded population depended on patronage to obtain water connections; existing recipients beneﬁted from the low tariffs. The political risks of radical change, in raising and collecting unpaid tariffs and in de-politicizing investment decisions, were certain while the gains in a more efﬁcient and extended supply system were long term and uncertain. The patronage basis of decisions about urban water access was a major obstacle to management reform in all the core countries. Nickson and Franceys, 2003 describe a ‘low-level performance equilibrium’ between urban consumers and producers of ‘unsafe drinking water delivered at unpredictable times, in exchange for a minimal tariff which may not always have to be paid.’ Where consumers have organized to affect policy on urban water, it is to defend or improve existing arrangements. In Ghana a movement supported by international NGOs resisted the ‘commodiﬁcation of water’ through leases to foreign companies in Accra and Kumasi. In Cochabamba, Bolivia, riots led to the abandonment by the government of its concession of water management to a private company (Nickson and Vargas, 2002). In the major cities of Zimbabwe, residents’ associations have been active in pressing for improved services, though urban water supply is good by international standards. In Bulawayo, Widespread protests have been organized by residents’ associations, complaining not only about the service and its cost but also that meters were not being read properly but only estimated. Mass refusals to pay bills gave way to a pre-election moratorium granted by councillors on further payment of arrears while the matter was reviewed. This led quickly to a new attempt by councillors and ofﬁcials to explain tariff levels … (Batley, 1998, p. 48)

70 The Changing Role of Government

The high expectations of urban consumers in Zimbabwe seem to derive from a surprising combination of ideologies: the established expectations of the previous colonial population and the expectations raised by the redistributive policies of the post-colonial government. In addition, the fact that water is provided by local government in Zimbabwe offers a political focus for popular organization. As described in the next section, radical reform to change the organization of water supply resulted not from public demand but from highlevel political leadership that mobilized rumbling discontent into an awareness of crisis. Health care The reform of health care was not generally driven by the political leadership or by sectoral ministers, nor was it supported by legislators. The health sector is complex, with multiple sub-sectors and services, and does not provide easy focuses for political or social mobilization for or against change. In the African countries where the reform proposals were most radical, they had been externally inspired so there was little sense of political ownership. In all countries, the initiators and supporters of change among ofﬁcials in the core government agencies, in the ministries of health and in international organizations regarded the support of politicians as very uncertain and unpredictable. Civil servants therefore proceeded opportunistically in the face of inconsistent policy direction: …politicians make ad hoc decisions, not rational for the health sector, and often change their minds; we have to cope with these decisions (Ministry of Health ofﬁcial in Sri Lanka, quoted in Russell and Attanayake, 1997, p. 82). As in the water sector, health reform was more likely to bring political risk than gain. The risks were different for the various reforms. In Sri Lanka and India, decentralized management of hospitals would threaten the structure of political control and meet the opposition of trade unions and health professionals. Politicians opposed user fees in all case study countries but particularly in Sri Lanka, where there was widespread commitment to free public health care amongst middle classes as well as trade unions, the radical left, health professionals and managers. Donors’ pressure for the introduction of charges was greater in the African countries, where also public opposition was less mobilized and had fewer channels of expression (Herbst, 1993):

The Politics of Service Reform 71

one-party politics and more limited democratic accountability meant that politicians were less fearful of an electoral backlash or the ‘political suicide’ associated with fees in South Asia (Mills et al., 2001, p. 99). There were very few cases of positive public pressure in favour of reform in health care. NGOs in India and the media in India and Thailand have campaigned for improved regulation of health care, but remained a weak counter-balance to the considerable professional and business interests lined up against regulation. In general, rather than press for change, users have found alternatives in the private or traditional sectors where the public health service was inadequate. Exit rather than voice was their approach (Mills et al., 2001).

The ‘agents’ – public service administrators, professionals and workers Public ofﬁcials of core ministries Previous sections have established that, in the context of economic crisis and structural adjustment, international development agencies usually had a key role in the development of proposals for reform. The formal support of governments was, of course, essential; and, sometimes, governments played an important part in initiating their own liberalization programmes or in modifying reforms presented by international agencies. However, political support for reform was often only at the highest level, with a narrow base of support among lower level political leaders. There was little evidence of coherent party programmes of reform and little attempt to mobilize public support, except in Argentina and South Africa. Moreover, public demand for reform was rare; more often, there was pressure on politicians for the defence of existing rights and privileges than for change. So, international agencies often had little guarantee of stable support for liberalizing reforms from government or politicians, citizens or users. As long as there was basic political acquiescence, relatively little support was necessary for the immediate policy reforms associated with structural adjustment – divestiture and de-regulation in the productive sectors, civil service cuts and one-off tariff rises. However, for the longer term ‘supply-side’ reforms (in organizational structure and process, efﬁciency orientation, and changed relationships with the private sector), there was one group whose support was essential to

72 The Changing Role of Government

the implementation of reform – the senior ministerial administrators and professionals. In formal terms, these are the ‘agents’ of citizens and politicians; in practice, together with the international agencies, they were often closer to being the ‘principals’. Where reform was successful, it had high level political and donor support and a working combination of senior ofﬁcials and external advisers. Similarly, Grindle (2001 and 2003) identiﬁes the importance of small ‘design teams’ attached to the political executive, and Nelson (2000) of ‘change teams’. The initiators of change were usually outside the ministry that was subject to reform. Crisis and adjustment put the following in the driving seat: the ministry of ﬁnance, and agencies subject to the president’s or prime minister’s ofﬁce – for example, planning and public service commissions. They had negotiated the commitment to the World Bank, IMF and donors, reported to them, and had an across-the-board responsibility for liberalization and civil service cuts. Typically, where radical reform was really implemented, these core public ofﬁcials led and were supported by top civil servants of the line ministries working with international advisers.3 The degree to which core ofﬁcials were involved depended on the salience of the sector to macro-economic adjustment; there were also differences between sectors with regard to the respective roles of administrators and professional staff. Sector differences – the inﬂuence of core and line ministry ofﬁcials The ﬁrst wave of reform in the industrial sector was a direct consequence of IMF and World Bank conditionality – privatization of state-owned enterprises and the de-regulation of production and trade. Ministries of industry could do no more than comply, although they might procrastinate. It was the second wave of reform which left room for the initiative of senior ministry ofﬁcials. This was the development of a new raft of agencies and departmental units concerned now to ‘enable, promote and facilitate’ private industry. For example, having dismantled its own import licensing function, the Ghanaian Ministry of Industry and Commerce struggled, without much success, to give technical advice to ﬁrms it had previously regulated. In Zimbabwe, the Ministry of Finance together with donors opted to create a semiautonomous Zimbabwe Investment Centre out of parts of ministries. Completely autonomous support organizations (for export promotion and information support) were set up with their own or donor funding in Ghana and Zimbabwe. In India, among a vast number of traderelated and semi-autonomous industrial research and training insti-

The Politics of Service Reform 73

tutions, some gained a new lease of life as the government released its controlling grip. The inﬂuence of senior ofﬁcials in the core institutions of government over the agricultural sector was weaker. State intervention in agriculture was less immediately exposed to structural adjustment and was defended, particularly in India, by its historical importance to national food security and to the needs of poor consumers and producers. Nevertheless, the driving forces for change in all countries were again the core ministries of ﬁnance, planning commissions and presidential ofﬁces that were obliged to pursue agendas of cost-cutting and de-regulation in response to economic and ﬁscal crises. The sectoral ministries and agencies generally acted to restrain reform (Box 3.1). In the service sectors – water and health – core government ofﬁcers, with international agencies, were again the main initiators of change. Crisis and structural adjustment agreements set the broad policy agenda for shrinking budgets, staff cuts, raised tariffs and fees, contracting out and privatization. However, unlike in agriculture and industry, where the most signiﬁcant part of the reform was for state withdrawal, in water and health the state was bound to have a continuing role. Here, the development of detailed new management practices could not easily be directed from the public service commission, ministry of ﬁnance, president’s or prime minister’s ofﬁce. Unless

Box 3.1 Line ministries have often had little inﬂuence on sector liberalization In India, it was the Minister of Finance’s 1993 budget rather than national or state ministries of agriculture that lifted restrictions on the movement of agricultural products. But it was not possible to shift India’s labyrinth of national and state ministries, and public sector undertakings (for procurement, warehousing, costs and prices) from an attitude of rationing and control. In Ghana, the president’s ofﬁce entered into agreements with the World Bank to liberalize Ghana’s leading export crop, but accepted the case of the cocoa board (Cocobod) for a slow process of compliance. In Zimbabwe, treasury constraints precipitated the need to cut back on the Ministry of Lands and Agriculture’s services and price subsidies. But political recognition of the implications of this reduced and more market-oriented role was not forthcoming. When staff of the ministry proposed, with donor (World Bank, EU and DfID) support, to re-organize the ministry in recognition of its changed functions, the permanent secretary was sacked and the process stopped (Hubbard, 2003).

74 The Changing Role of Government

conditions could be created for an abrupt and radical transfer to private provision, new arrangements would have to be worked out incrementally within the public sector. In the latter case, the sectoral administrators and professionals had a necessarily strong role in reform, or in stopping it. Professional staff – engineers in water and doctors in health – were much more important in the direction of these sectors than they had been in ministries of industry and agriculture. Nelson (2000) comes to similar conclusions about education reform in Latin America. Radical reforms challenged the control of water supply by professional engineers and public administrators, in favour of the private sector and ﬁnancial managers. Where radical proposals for water sector reform were successfully implemented, they were driven politically. In the case of Argentina, the new government of President Carlos Menem in 1989 declared its commitment to liberalization and reform of the state. A rapid programme of privatization of utilities culminated in 1992 with a State Reform law that spelled out future arrangements for gas, electricity and water, including the 30-year concession of Buenos Aires’ water supply. Similarly, the president of the Philippines pushed through a concession arrangement for Manila by demanding radical change in response to a declared ‘water crisis’. Both presidents managed to raise the political salience of sector reform, giving it a similar degree of urgency as had been assumed by macro-economic stabilization reforms. By contrast, in Ghana, sector ministers under two governments have proposed putting Accra’s water supply onto a lease contract, with the support of development agencies (DfID and World Bank) and the senior staff of the water board. Unlike Argentina and the Philippines, Ghana adopted a negotiated approach that gave time for water sector unions and NGOs to mobilize and resist change. Incremental water sector reforms were often led by ‘lower level alliances’ of water agency managers usually with, but sometimes without, the pressure of donors and their technical advisers. These generally retained existing professional (engineering) control while freeing agencies from the constraints of being part of government. At a national level, in Zimbabwe it was the professional staff of the Department of Water Resources who in 1985 put the case to the cabinet for the corporatization of the state bulk water supply organization, as a way of escaping civil service controls on investment, management and recruitment. This meshed with World Bank structural adjustment and was supported by international consultants and donors (particularly GTZ), although the proposal always had weak

The Politics of Service Reform 75

political backing. A similar attempt to turn the Sri Lankan National Water Supply and Sewerage Board into a consumer-oriented, costconscious organization, with USAID support, foundered on the resistance of leading engineers and central government politicians to the loss of their control to professional managers (Franceys, 1997). In some countries, the local management of water delivery allowed alliances for change to develop between water managers and politicians. Gweru and Bulawayo municipalities in Zimbabwe have developed efﬁcient local water supply systems, and were on the brink of conﬁrming public private partnerships – concessions or buildown-operate schemes with foreign companies – before the country’s ﬁnancial crisis intervened (Plummer, 2001; Batley, 1998). In India, the World Bank has worked with state and local ofﬁcials to introduce commercial practices and contracting out in state and metropolitan water boards (Franceys and Sansom, 1999). However, cases of effective change are rare; normally professional engineers and public sector workers have resisted private sector involvement and new management approaches. Crisis and the adjustment programmes agreed by core government agencies with the international ﬁnancial institutions indirectly affected the health sector, particularly in Africa. Health sector reform was often a spin-off from broader national commitments by public service commissions and ministries of ﬁnance to contracting out, staff cuts, and the raising of user fees. For these to be operationalized, however, the support was required of the senior ofﬁcials of the ministry of health, often backed up by technical advisers of donor and international organizations. Commitment by the top professionals and ofﬁcials of ministries of health was greater in Africa where donor inﬂuence was more complete and where economic and ﬁscal crisis had led to a near collapse of health services. The charging of fees was accepted by doctors and managers in Zimbabwe and Ghana as a way of maintaining health spending. It was rejected by the same people in Sri Lanka and India as an affront to free care and, perhaps, as a threat to the informal charges which they had enjoyed. Even where there was high-level commitment to reforms of this type, in practice there were major difﬁculties with implementation. Many of the proposed health reforms entailed a weakening of powerful internal interests and erosion of professional autonomy. In South Asia where health systems were somewhat more secure than in Africa, ministry ofﬁcials commonly resisted decentralization of control to hospitals; contracting out and privatization were resisted by unions; and

76 The Changing Role of Government

the medical profession opposed the regulation of private practice in which it usually had a stake. The health sector was, in this respect, the most impervious to radical reform. It was not easily dealt with by the sort of ‘stroke-of-the-pen’ privatization that international agencies and some political leaders had demanded of the productive sectors or, in a few cases, water supply. Reformers had to persuade the medical professionals and public sector unions who saw change as a threat, who had a large place in the organization and delivery of health services, and who had an important political voice (Mills et al., 2001).

Emerging constituencies for change Reform was often constrained by lack of political commitment and by the interests embedded in existing organizational arrangements. Even where reforms were introduced, governments and public agencies could easily slip back into previous practices. However, in some cases the ﬁrst round of reform built up momentum for a further round of change, supported by the agents and beneﬁciaries of the reform process. Grindle (2001 and 2003) and Nelson (2000) describe how educational and health reformers have achieved change, in some Latin American countries, by understanding the political constraints and opportunities and calculating opportunistically to build alliances and outwit the opposition. This research also found some examples of how reform can sow the seeds of further change. Liberalization, the privatization of state enterprises and deregulation of industries and agricultural trade are unlikely to be reversed. They have created a new set of incentives for entrepreneurs and, sometimes, led to the creation of private and public support agencies whose services are in demand. In Sri Lanka, early encouragement of foreign direct investment created support for further liberalization to encourage foreign investment and employment. In Zimbabwe, textile ﬁrms and trade unions pressed for further freedom from government controls, and the private sector established its own new agricultural and industrial support services (Jackson, 2002). Government ofﬁcials in Kenya who had opposed the removal of the state monopoly in maize marketing eventually came to support it after seeing its positive effects. The millers and traders who began to deal in imported maize and rice in Kenya and Sri Lanka became a constituency for further liberalization. Urban consumers who beneﬁted were a latent source of demands for the freeing of agricultural trade if it could deliver cheaper and better food (Hubbard, 2003).

The Politics of Service Reform 77

The Argentinean Government learned from earlier utility privatizations, developing a capacity for negotiation that led to better and clearer contracts in gas, electricity and water. Aguas Argentinas, the company that held the concession for water supply to Buenos Aires, overcame suspicion of the privatization of management by taking some quick steps to satisfy the opposition. It ensured early water connections to the previously excluded poorer population by cross-subsidies from other water consumers, and made contracts with communities to supply labour in return for earlier and cheaper connections. It also improved the conditions of employment of staff transferred from the public sector water company, and worked with NGOs and small local ﬁrms to build infrastructure and extend services (Nickson and Franceys, 2003). In the health sector, the development of capacity could also accumulate: Once a degree of decentralization has taken place and hospital staff have the opportunity to learn new systems and skills, there will be stronger systems and a pool of experienced staff in place which will be able to cope with greater degrees of decentralization. (Mills et al., 2001, p. 93) A key issue for reformers is then how to phase this development of skills and build constituencies of interest in favour of reform. One of the problems in the professionally dominated health (doctors) and water (engineers) sectors of developing countries was how to make the ﬁrst incursion. Whereas in countries such as the United Kingdom reformers could call on the support of ﬁnancial and managerial staff as ‘the instrument through which many of the changes have been forced’ (Ferlie et al., 1996, p. 7), in developing countries these groups are often weak. Reform might be phased to build this management cadre as a core group of advocates of change. Secondly, organizational reform is likely to receive more support if it is dissociated from direct threats to the status of employees. Where re-organization was going on after the earlier stages of structural adjustment were out of the way there was less likely to be resistance. Thus, Ghanaian health and water workers, no longer threatened by further major employment cuts, could support reforms which would decentralize control, introduce new sources of fee revenue and raise salaries. In Zimbabwe, on the other hand, health sector reforms were associated with the general staff cuts and suppression of salaries of the earlier stages of structural adjustment, and were met with opposition or suspicion.

78 The Changing Role of Government

Conclusion The concern of this chapter has been to identify who have been the leaders, the supporters and the resisters of public service reform. It has adopted a broad principal-agent framework, comparing reality with an ‘ideal’ situation where (i) citizens are the principals over political policy-makers as their agents, and (ii) policy-makers are the principals over public service ofﬁcials as their agents. Public service reform is generally complicated by the fact that public service ofﬁcials are both the agents and the objects of change. Reform in most developing countries is further complicated by an additional set of external actors in the shape of international ﬁnancial institutions and donors. The analysis has identiﬁed the interests involved in reform, indicating how the balance between them is affected by institutional factors including: (a) the importance of the statist model to the stability of weaker states, (b) the pre-eminence in reform given to international organizations by economic and ﬁscal crises, (c) the effect of the characteristics of different service sectors on the power balance between clients, policy-makers and provider organizations. Chapter 1 identiﬁed the institutional barriers to change. Pre-reform practices across the sectors were a product of colonial practices, postcolonial nation-building and the current international conventions. Undoing these has often meant a profound challenge not only to vested interests but even to the stability of the state. Economic crisis has forced change but also created stress conditions which make resistance to change especially strong. The losses from reform for powerful groups (politicians, public sector workers, and sometimes the professionals and administrators) are more certain than the gains to an illserved public. The ‘constituency for reform’, in a new managerial class and new beneﬁciaries, takes time to emerge. Organizational reform is a process that requires long term strategy and commitment. The ﬁrst generation of reform in the 1980s and early 1990s, under pressure of crisis and structural adjustment, focused on reforms concerned with macro-economic stabilization. Several factors made for relatively quick implementation – the immediacy of the ﬁnancial crisis, the availability of ready-made models of neo-liberal economic reform and the ‘stroke-of-the pen’ nature of many of the policy changes. The more recent organizational reforms in basic service delivery, particularly in health, education, water and sanitation present a much harder reform task. They have less clear-cut goals, offer uncertain beneﬁts, involve multiple actors, challenge existing provider groups,

The Politics of Service Reform 79

and require long-term commitment. In the social sectors, citizen or client awareness and capacity to organize to make demands for improved services are weak, and policy-makers have relatively little capacity to assert control over the performance of providers. This is particularly the case in the health sector; infrastructural services (water), on the other hand, appear to offer greater opportunities for control by policy-makers over provider organizations. A second point with regard to the nature of these reforms is that their concern with the organization of service delivery was of little direct interest to users or citizens. The struggle for change generally remained within the bureaucratic rather than the public arena. Where reform took place it was a covert, administrative rather than an overt, political process. As Grindle (2001) has found in other studies of reform, the political mobilization of social demands was not the determining factor; reform is generally an élite process ‘generated by the executive rather than by legislatures, political parties, interest groups or think tanks’. Political leadership, where it existed, was concentrated at the highest levels – the president or prime minister and Minister of Finance. Otherwise, political engagement was usually weak and largely in defence of existing interests and arrangements. The political risks of promoting change were much greater and more certain than the possible gains. The interests of service recipients were more often experienced by policy-makers as a passive drag on change than as a source of active demands. Producer interests were more assertive than the voice of the consumers of health and water services. Paradoxically, it was normally the supposed ‘agents’ of the policy process who were the key leaders or ‘principals’ of change – international and key domestic ofﬁcials. This was most clearly the case in the reforms directly associated with the ‘conditionalities’ of structural adjustment – privatization and de-regulation of the productive sectors, increases in tariffs and fees, cuts in stafﬁng. These could be driven through by international agencies with the acquiescence of political leaders and top ofﬁcials of core ministries. This analysis has shown that, by comparison with macroeconomic and industrial reform, there had been much less high-level political and core ofﬁcial involvement in health and water reforms. Here, line ministry ofﬁcials often had a key role in managing change and were likely to let it lapse. Professional staff – engineers in water and doctors in health – are much more important in the direction of these sectors than they are in ministries of industry and agriculture, and were likely to have a continuing role

80 The Changing Role of Government

whatever the reform (see also Nelson, 2000 on this point). Their involvement often had the effect of blocking change. Where social reform was more successful (e.g. health care in Ghana), there was a small reform team that included donors, core ministry and line ministry ofﬁcials with high level political support. The deep involvement of international lenders or donors in the policy-making of countries in crisis can lead to the ventriloquizing of policy through national political leaders. This can give the impression that the local ‘ownership’ of reform is without substance, and can undermine the relationship of accountability between national citizens, policy-makers and provider – principals and agents. External bodies (IMF, World Bank and donors) have the greatest force with regard to the weakest governments, with the greatest dependence and the least capacity to negotiate. As a result, proposals for reform, including in the social and utility sectors, have often been most sweeping and radical in the countries with the deepest crises. The consequence has been a large gap between radical reform design and modest outcomes, particularly in Africa.

4 Decentralizing Organizational Arrangements for Service Delivery

Introduction The key assumption that underlined the study was that organizational and management reforms in the four sectors studied were pushed by the new public management-type reform agenda reviewed in Chapter 2. These reforms tend to ignore the speciﬁcities of sectors and the institutional contexts of poor countries often characterized by state dominance, weak market institutions, ﬁscal crisis, poor incentives and political sensitivities. This chapter will examine the nature and the extent of ‘the new management’ organizational reforms for service delivery across the four sectors, using evidence from the case study countries (Ghana, Zimbabwe, India and Sri Lanka) and reference countries where possible. The focus in this chapter is on internal management reforms rather than externally oriented market-type reforms, which are analyzed in Chapters 6–8. To put the reforms in context, the ﬁrst section brieﬂy reviews the pre-reform organizational arrangements for service delivery. The second section then outlines the types of organizational arrangements, and the third section examines the nature and extent of reforms in organizational arrangements. The fourth section reviews the available evidence on the performance of reformed organizational arrangements. The conclusion teases out cross-country and crosscutting issues in organizational arrangements and performance.

Pre-reform organizational arrangements The context in which reforms were introduced in the four core countries was one of centralization and direct provision of public services by 81

82 The Changing Role of Government

the state, with minimal private sector participation and only lip service paid to improving performance. Sub-Saharan Africa and South Asian countries had similar levels of high and active state involvement in the four sectors of the economy, especially in the immediate postindependence period until reforms in the 1980s. They also shared a similar tradition of being Commonwealth countries with the organization of their public services more or less following the British tradition before reforms. Ministries, departments and agencies for service delivery were quite hierarchical in structure and consequently very bureaucratic in their operations. In curative health, where formal systems were largely government funded and provided, the organizational arrangements for service delivery were found to be broadly similar (Mills et al., 2001). The depth of hierarchical structures and bureaucratic controls were more noticeable than in other sectors. This may be explained by the fact that historically there was a stronger state presence and direct role for government in curative health care provision than in other sectors in all the four core countries. As Mills et al. point out: The Ministry of Health (MoH) was at the apex of a pyramid, with lower management levels below it, and a structure of services – at least in theory – consisting of peripheral outpatient facilities, local hospitals, general hospitals and central hospitals with referral and supervisory relationships between them. (Mills et al., 2001:3) The lower management levels referred to above typically took the form of deconcentrated regional and district ofﬁces of the MoH and health facilities accountable through the hierarchy to the ministry. Large hospitals, especially teaching hospitals, existed outside the deconcentrated organizational structure, but had some delegated management authority from the MoH, as was the case in Ghana and Zimbabwe before reforms. Delegation of ﬁnancial and other management powers to the lower units of the health system were limited; decision-making was centralized in the MoH and other government agencies. In India the relationship between the federal and state governments further complicated the issue of centralization (Mills et al., 2001:35). Over-centralization was closely associated with vertical programmes and professional cadres. In India the federal government had signiﬁcant authority over most priority services such as malaria and AIDS control (ibid.: 36). In the case of Ghana, for example, vertically organized programmes necessarily encouraged the development of separate,

Decentralizing Organizational Arrangements for Service Delivery 83

vertical management structures (Cassels and Janovsky, 1992; Larbi, 1998c). It is worth pointing out that there has always been a signiﬁcant involvement of non-state organizations in curative health in poor countries. This is more noticeable in India and Zimbabwe both of which have signiﬁcant private allopathic medical markets with the public sector accounting for only 22 per cent and 35 per cent of total health expenditure respectively. This compares with Sri Lanka and Ghana both of which have a dominant public sector, accounting for about 75 per cent of health expenditure in the case of the former. In addition church mission hospitals (subsidized by government) are signiﬁcant providers of curative health services in most African countries, as are NGOs in India. Indigenous medicine providers are also signiﬁcant in all countries, but more so in India and Sri Lanka. Thus for curative health, the pre-reform organizational arrangements were predominantly through hierarchically organized structures with some degree of delegation to deconcentrated health administrations, hospitals and other health facilities. Although the private and voluntary sectors were outside this hierarchical structure, they were important providers of health services to supplement the state sector. Like health, urban water supply has predominantly been delivered by state organizations for decades in the four countries. There were varied combinations of three forms of pre-reform organizational arrangements for direct state provision at both national and sub-national levels. At the national level were two types of organizational arrangements – a semi-autonomous national water board (Sri Lanka) or a national water corporation (Ghana). Both Ghana and Sri Lanka transformed their water departments into public corporations in 1965 and 1975 respectively, with their own boards of directors and delegated autonomy over a number of operational issues. However, in practice they operated in a near civil service mode with their autonomy constrained by their parent ministries and other central control agencies over operational issues such as appointments, budgets and borrowing, plus formal and informal political interventions. In India, water supply was the responsibility of sub-national governments – state or municipal water boards or departments. Zimbabwe combined national and sub-national government responsibilities for water – a national water department supplying bulk water to municipal water departments for distribution. The critical difference in all these arrangements was the extent to which these organizations had operational autonomy from parent ministries and departments. Like health,

84 The Changing Role of Government

water is a politically sensitive service in most countries, meaning that governments are often reluctant to let go their involvement in and control over certain operational issues such as setting of fees and tariffs, and appointments. Pre-reform state involvement in the economic sectors – agriculture and industry – was also high, though less than the two social sectors. In the industrial sector, there was widespread state involvement in ownership, management and production, justiﬁed on infant industry or import substitution grounds. This was more prominent in Ghana and Sri Lanka than in India and Zimbabwe. The former two countries had weaker industrial bases and private sectors and thus actively promoted industrial development through public ownership and attraction of foreign direct investment. In contrast, India and Zimbabwe had a signiﬁcant private sector presence in industry and sought to protect existing industry and guide its development. The tracer industry used in this study is the textile sub-sector, which has traditionally been seen as a take-off industry and one of the key areas where governments intervened to set up state-owned importsubstitution industries. This was the case in Ghana, India and Sri Lanka after independence and in Zimbabwe during the UDI era (Jackson, 2002). The main organizational arrangements for state involvement in the manufacturing sector in general and the textile sub-sector in particular took two main forms – state-owned enterprises, which operated under a parent ministry of industry and joint ventures by the state and the private sector. Whilst the private sector was involved in all four countries, large and small-scale private textile producers were more signiﬁcant in India and Zimbabwe than in the other two core countries. Ghana and Sri Lanka, driven more by ideology than economic rationale, pursued an active policy of nationalization of manufacturing ﬁrms up to the late 1970s. Governments in all countries put in place policies to protect their ‘infant’ industries through export/import licensing arrangements and regulations. These controls later became the objects of reforms in the 1980s and 1990s and the manufacturing sector became the most liberalized and deregulated among the four sectors examined in this study. In agricultural trade, the organizational arrangement for government intervention commonly took the form of marketing and produce boards or corporations. This was justiﬁed on a combination of grounds, including food security, consumer and producer welfare and price stability (in the case of staple foods like grains) or on national development and tax raising grounds, in the case of export crops, (e.g. cocoa in

Decentralizing Organizational Arrangements for Service Delivery 85

Ghana). For example, the Cocoa Marketing Board (Cocobod) in Ghana had monopoly over the purchasing, processing and exporting of cocoa until the 1980s. Although it delegated this role to a farmers’ cooperative in 1961–66 and to private licensed agencies (LBA) in 1966–1977, these arrangements failed due to growing indebtedness of the agencies and the public monopoly over purchasing of cocoa was reinstated. A newly created Produce Buying Company (PBC), a subsidiary of Cocobod became the sole buyer of cocoa in Ghana. India, Sri Lanka, Kenya and Zimbabwe had strongly interventionist policies with regard to food supplies. For example in Kenya, the National Cereals and Produce Board (NCPB) had monopoly over various aspects of maize marketing, including storage, distribution and price stabilization. In India the Food Corporation of India purchased wheat and rice and sold them through Fair Price Shops, leaving the surplus to the competitive market. Sri Lanka had similar schemes of state-controlled import, procurement and rationed distribution for rice and monopoly control over wheat imports, which it used to stabilize prices through the Paddy Marketing Board, the Food Commissioner’s Department and the government-controlled Cooperative Wholesale Establishment. In contrast to other countries, the state did not exercise monopoly over the marketing of any food crop in Ghana. The main government food agency was the Ghana Food Distribution Corporation, which operated drying and storage facilities for grains (mainly maize), but this amounted to a maximum of 10 per cent or less of grains. In summary pre-reform organizational arrangements for service delivery in the four sectors were largely by public enterprises or boards with some degree of operational autonomy (industry water and agriculture), devolution to sub-national government departments, largely to municipalities (water in Zimbabwe and India), deconcentration to ﬁeld ofﬁces of central ministries and departments (health) and delegation to semi-autonomous bodies. Until reforms in the 1980s and 1990s, corporatization was rare, except in industry. Common features of the prereform organizational arrangements in all four case study countries were the dominance of the state, particularly in health and water, and limited operational autonomy in practice.

Types of organizational arrangements In order to appreciate the extent of organizational reforms in the four sectors, this section brieﬂy outlines the different categories of reformed

86 The Changing Role of Government

organizational arrangements. The main organizational forms involve some form of decentralization and may be categorized as follows: Political devolution to sub-national governments, i.e. political decentralization. Whilst this was not new, there were attempts to transfer more responsibilities to local governments or their departments in health. In this arrangement, the sub-national government has the authority and the responsibility for the provision of the service and any changes may not be effected without changes in the legal provisions that transferred such responsibilities. This may be a real or functional decentralization. The former refers to the transfer of responsibilities for speciﬁc functions such as water supply, sanitation, primary health care and basic education (e.g. water delivery by municipalities in Zimbabwe and India). The latter refers to the transfer of broad responsibilities for a whole range of public services to subnational level governments (Larbi, 1998a:99; Mills et al., 1990). Deconcentration is administrative decentralization, which delegates the responsibility for the delivery of services to ﬁeld units of central ministries, departments or agencies. Such ﬁeld units are then accountable to their headquarters through the hierarchy and have limited authority. As noted in the previous section, this form of decentralization has been common in the health sector in all countries and also applies to ﬁeld ofﬁces of public enterprises and boards in both water and agriculture. Autonomous/executive agencies: As described in Chapter 2, this is a typical new public management reform, which is based on the assumption that there can be a separation between policy/strategic core and operational arms of government. Responsibilities for speciﬁc operational functions are delegated to units that are either outside the central line ministries or granted some semi-autonomous status from the parent ministry. Those operational units that are given agency status usually have signiﬁcant levels of operational autonomy over issues such as budgets and human resources management, but in return are held responsible for results. This creates a principal-agent relationship whereby the central/parent ministry is the principal and the managers of the decentralized units are the agents. Where it works well, it is expected to provide incentives for improved performance in service delivery by giving managers not just the freedom to manage, but also allowing them to manage by reducing central controls and interferences over issues such as budgets and human resources management, whilst at the same time holding the agencies and their managers accountable for results. This implies establishing standards, measures

Decentralizing Organizational Arrangements for Service Delivery 87

and methods to review performance. The purpose is to free units involved in service delivery from the constraints of direct controls through hierarchy and politics. Examples of this organizational arrangement were found in the health sector in Ghana and also in industry. Public enterprises, public corporations and boards: These were common pre-reform organizational arrangements in all sectors, except health, and have been the object of reforms either by corporatization or privatization. Most of these organizational arrangements were set up by statutes, separate from the civil service, but usually treated as part of the public services. The statutes deﬁned their responsibilities for particular services and they usually combined social and political policy objectives with economic objectives. In many cases the status quo has remained either due to the failure of privatization or the reluctance of governments to disengage themselves for strategic reasons (see World Bank, 1995). Corporatization has become one of the new forms of organizational arrangements designed to free state or public enterprises, boards and similar organizations from political control and interference, as well as grant them some commercial freedom. Corporatization brings external orientation to an enterprise by granting a new legal personality with responsibility and authority according to private company law. This involves the strategic reduction of the direct role of government, whilst at the same time increasing allocative and productive efﬁciencies by replacing support through the budget system with elements of the market, e.g. by commercialization and borrowing from banks. The shares of a corporatized ﬁrm are usually held by the state as in French Telecom, and Post and Telecom in Holland (Thynne, 1998). There were examples of corporatization in the water sector (Ghana) and in agricultural trade (Ghana and Zimbabwe). In some cases corporatization becomes a step towards privatization, as was the case in dairy and cotton trade in Zimbabwe. Performance agreements: Corporatization, executive agencies and other forms of decentralized management are expected to give more operational freedom to managers and organizations, and free them from political and bureaucratic controls. Performance contracting or agreements, deﬁned here as a written or negotiated agreement between government and public enterprises or private providers of public services, have become one of the key instruments to restructure and manage the relationship between governments (as principal) and managers (as agents) (Larbi, 1998a, 2001; World Bank, 1995). The contract

88 The Changing Role of Government

typically deﬁnes objectives and sets targets for the agent and determines resources and autonomy that the principal has to grant the agent. Performance agreements and contracts have been widely used in India, Pakistan and several francophone Africa countries. Whilst they are not organizational arrangements for delivering public services per se, they help to facilitate arrangements such as corporatization and executive agencies, by clearly setting out the responsibilities and expectations of both principals and agents. The next section will discuss the extent of reforms in organizational arrangements in the four sectors.

Reform of organizational arrangements The research found evidence of varying degrees of restructuring of service delivery arrangements towards decentralization in all four sectors. However, these were largely in organizational form, with little evidence of real changes in processes that deliver actual operational autonomy and improved performance in service delivery. In the health sector, whilst government remains the dominant provider, there have been efforts towards management decentralization in the form of creating arms length agency structures (Ghana), granting more operational autonomy to big hospitals (Ghana and Sri Lanka) and delegating more budgetary and ﬁnancial control to deconcentrated units of the MoH. In the urban water sector there were examples of attempts to reform existing service delivery arrangements through corporatization, commercialization, the use of cooperatives and, in some cases, partial privatization. Similar changes towards commercialization and corporatization of public organizations are discernible in agricultural marketing. In the business sector the trend is towards the creation of new agencies that play enabling and promoting roles, the conversion of old agencies and the adjustment of the role of ministries to enabling and advisory roles. Using the above categories of organizational restructuring, the sections that follow will discuss the extent and nature of reforms in the four sectors. Health Decentralization is a common theme in health within a broader context of political devolution. Three main types of decentralized management were found in the health sector – deconcentration, devolution and executive/semi-autonomous agencies. Both Ghana and

Decentralizing Organizational Arrangements for Service Delivery 89

Zimbabwe had restructured their central ministries of health to allow for more decentralization. Ghana replaced vertically organized technical and programme structures characterized by multiple layers with functional management divisions that cut across programmes (e.g. Policy, Monitoring and Evaluation Division, and Human Resources Management Division). Whilst deconcentration and devolution are not new organizational arrangements in health, it is worth pointing out that all four countries have made attempts to decentralize authority down to either deconcentrated ﬁeld units of central ministries or sub-national level governments. For example, reforms in Ghana had strengthened district health management teams, which dated back to the 1970s, with more resources and management functions delegated to them than before. To illustrate, the control of non-salary recurrent expenditure by the MoH headquarters reduced by more than half, from 66 per cent in 1992 to 28 per cent in 1996, whilst the share of the district health administration and the teaching hospitals increased from 10 per cent and eight per cent to 23 per cent and 17 per cent respectively, during the same period (Smithson, Asamoa-Baah and Mills, 1997:19). However, like other countries in sub-Saharan Africa, the policy intention of devolving health management and delivery to local governments by integrating district health systems and staff into the local government structure has not been realized due to the reluctance of the central Ministry of Health to cede control over staff and other resources to the districts. Zimbabwe’s reforms similarly involved devolving the management of service provision to district health management boards (for rural districts) and to hospital boards in the case of provincial and central hospitals. Although Sri Lanka created a department of health at the provincial level headed by a provincial minister, health management remained centralized in practice because of the reluctance of the central bureaucracy to let go controls. Similar resistance by the central bureaucracy in India had delayed the delegation of health management responsibilities to local government (Mills et al., 2001). Thus although decentralization is a common reform theme one also sees a common problem of resistance from central bureaucracies which undermines implementation. Apart from water services in India and Zimbabwe, the transfer of functions and resources to local governments has not occurred as expected; what has taken place is more deconcentration to ﬁeld ofﬁces of central ministries, departments and agencies.

90 The Changing Role of Government

Another form of organizational arrangement identiﬁed by the study is the creation of autonomous and semi-autonomous organizations within the public health sector, following reform examples in the United Kingdom and other OECD countries. Two forms were identiﬁed in the health sector – autonomous hospitals and executive agency-type organizations. The autonomous hospital concept is gradually being institutionalized in a number of developing countries. This applies mainly to large hospitals and is modelled along the lines of the United Kingdom’s National Health Service’s trust hospitals. The incentives for creating autonomous hospitals vary. They include the desire to make large hospitals more efﬁcient, responsive and less dependent on government resources by granting the management more operational freedom from the constraints of civil service rules and bureaucratic controls. In return the management is expected to be held accountable for their resource allocation decisions. Among the four core countries studied only Ghana had a national policy for hospital autonomy in place at the time of the research, whilst Zimbabwe had a policy in the making. Two Ghanaian teaching hospitals had been granted autonomy under legislation.1 This effectively separated them from the core civil service and from the Ghana Health Service. The reconstituted boards of the two hospitals have more operational freedom than before, including determining policies of the teaching hospitals, ﬁnancial management and control, and decisions on the level of user charges. In line with the purchaserprovider split and agency model, government funding for the hospitals is to be increasingly based on the number of patients requiring specialist treatments and other referral services (Larbi, 1998b,c). It should, however, be noted that even though Ghana’s policy on hospital autonomy was in place by 1988, actual implementation did not start in earnest until 1995/96 and has been very slow. As noted above, Zimbabwe’s policy of hospital autonomy was still in the making at the time of this research. The four main tertiary hospitals had considerable autonomy before 1980, but came under the direct control of the central Ministry of Health after independence. In response to donor advice and inﬂuence, there were proposals to decentralize management authority back to the hospitals. These included making a separation between hospitals as providers (agents) and central and provincial ofﬁces of the Ministry of Health as purchasers (principals), with a contractual relationship between the two sides.

Decentralizing Organizational Arrangements for Service Delivery 91

Unlike the two African countries, India and Sri Lanka had no policy on hospital autonomy and, consequently, tertiary hospital management was highly centralized (Russell and Attanayake, 1997). However, both countries had a number of autonomous health organizations. In India these included the Tamil Nadu Medical Supply Corporation, the Tamil Nadu State AIDS Control Society and Andrah Pradesh Council for Hospital Management, all set up under separate legislation to give them corporate status (Mills et al., 2001). Sri Lanka had one teaching hospital, which could set and retain its own user fees, which was quite radical compared to autonomous hospitals in Ghana (see Chapter 5). Another organizational form is the delegation of responsibility for services to voluntary (not-for-proﬁt) organizations and cooperatives. This form of organizational arrangement for service delivery is not new in most developing countries. Religious groups, NGOs and other nonstate providers have been involved in health care provision in developing countries since pre-independence. Whilst state restrictions and nationalizations may have reduced their role in the decades immediately after independence, especially in Africa, the economic and ﬁnancial crisis of the 1980s and the emphasis on good governance in the 1990s provided the context for and increasing role of the voluntary sector in health care provision. In many African countries church mission hospitals have become a permanent feature of the health system. What used to be informal arrangements with governments have become more formalized into either contracts or memoranda of understanding under which mission hospitals and other not-for-proﬁt providers deliver speciﬁed services on behalf of government and in return government pays for part of their cost (e.g. salaries of staff). Ghana has gone a step further in decentralizing health management than the other three countries by creating an NPM-type executive agency – the Ghana Health Service (GHS) – out of the Ministry of Health. This is inﬂuenced by the United Kingdom National Health Service model and has its own Chief Executive. The GHS is effectively the operational or implementation arm of the health system and the MoH remains the policy arm. However, teaching hospitals are not part of the GHS since they have a semi-autonomous status granted by legislation in relation to the MoH and the GHS. In summary, although there have been reforms towards further decentralization of the health sector in all countries, the traditional forms of deconcentration and to some extent devolution, are the most dominant types of structuring. NPM-type executive agencies and corporatization are very limited in practice. Within these decentralized

92 The Changing Role of Government

organizational structures, the extent of actual management autonomy tends to be limited in practice and has implications for the capacity to perform. These issues will be discussed later in the chapter. Urban water supply Organizational arrangements in the urban water sector have seen some transformation in recent years in many countries. However, state ownership and provision remain the dominant organizational form in the four case study countries, whilst private sector involvement is increasingly the trend in some of our reference countries (e.g. La Côte d’Ivoire, Guinea, Argentina, Bolivia and Chile). The main organizational arrangements are public or state enterprises – both corporatized (Ghana) and non-corporatized (Sri Lanka) – devolution to sub-national governments in the form of either multi-purpose municipal corporations (India) or a municipal department (Zimbabwe), with varying degrees of managerial autonomy. These arrangements largely keep ownership, ﬁnance and delivery within the public sector and enable varying degrees of operational autonomy. However, as will be discussed in Chapters 6 and 7, recent reforms in response to external pressures show a trend towards the introduction of private sector participation in the urban water sector in some of our reference countries. Reforms in the water sector have generally been slow in all four case study countries. It was noted earlier in the chapter that one of the prereform organizational arrangements for delivering water was through a public enterprise or board as was the case in Ghana and Sri Lanka. Among the four case study countries Ghana presented the only example of corporatization as seen in Box 4.1. The only evidence of performance contracts was in Ghana where the GWC management has been signing a performance contract with the government (Ministry of Works and Housing) since 1989 under pressure of the World Bank. This involves annual performance targets negotiated with the State Enterprises Commission (SEC) – the monitoring agency – with provision for rewards and sanctions. Key performance indicators include ﬁnancial and economic indicators (e.g. water production and net surplus/deﬁcit), efﬁciency and productivity indicators (e.g. number of staff per 1000 connections, staff/cost ratio and unaccounted for water) and management improvement/projects indicators (e.g. regular updating of corporate plan, submission of ﬁnal accounts and redeployment/retrenchment of a certain number of staff by speciﬁed dates). As will be discussed in the next section, GWC has made moderate improvements since the introduction of performance

Decentralizing Organizational Arrangements for Service Delivery 93

Box 4.1

Corporatization of the water sector in Ghana

The Ghana Water and Sewerage Corporation had undergone some reforms in the 1970s and 1980s, which delegated some management and operational responsibilities to regional managers/directors. Despite a policy reversal in 1990,2 regional directors were given some leverage over ﬁnancial and personnel matters – apart from training, capital, chemical procurement and a few other expenses which are handled at head office, all other expenses are devolved to the regional offices; some virement is allowed within non-salary recurrent budget (Larbi, 1998a). Following a change in the governance framework of economic state enterprises, GWSC was corporatized and became a public limited company3 under a new name – the Ghana Water Company Limited (GWC). In principle, this gave the board and senior management greater operational autonomy from government, including the ability to hire and fire its own staff, and removal of restrictions over borrowing and investment decisions. One area where its autonomy was constrained was the setting of tariffs, which was subject to political approval and delay. However, this control was removed by the creation of a semi-autonomous Public Utilities Regulatory Commission (PURC), which negotiates tariffs with utility companies. Corporatization was expected to pave the way for private sector participation as a next step in reforms. However, since the mid-1990s this has been stalled due to government indecision in response to the political sensitivity of water in the face of a very vocal coalition of groups opposed to private sector involvement in the water sector.

contracts. However, overall the performance contract system has been dogged with problems such as the government’s failure to meet its obligations, the lack of hard budgets and weak enforcement of sanctions (Larbi, 2001). Like Ghana, a government department of water supply and drainage provided Sri Lanka’s water until 1975 when it was converted to a semiautonomous central government corporation called the National Water and Drainage Board (NWSB). Responsibilities for water supply are shared between municipalities and the NWSB. Despite being a corporation, the NWSD continued to operate in a civil service mode under the Ministry of Housing, Construction and Public Utilities, with its operational autonomy constrained, as was the case in Ghana before corporatization. Like GWSC, NWSD delegated some management functions to its regional and district managers as part of reforms sponsored by USAID in the 1985–90 period. However, unlike Ghana, Sri Lanka has not made progress on corporatization due to a combination of factors, including the lack of a dynamic management team which is dominated by engineering expertise and culture, the reluctance of the

94 The Changing Role of Government

parent ministry to relinquish control over key operational matters such as tariff setting, capital investments, and human resource management and lower donor pressure for reforms. In recent years, however, there has been an increasing trend towards commercialization in the NWSD (Nickson and Franceys, 2003). The organizational arrangements in Ghana and Sri Lanka largely enable provision of urban water by central government-owned organizations. However, in India and Zimbabwe the responsibility for water provision is devolved to sub-national governments. For example in the Indian states examined in this research the responsibility for water was largely by state and municipal corporations or councils. There were indirect controls by the state and national governments that set the policy and regulatory frameworks. In the case of Pune, for example, the water supply department had only limited control over its own income, tariff setting and stafﬁng levels, and has to accommodate external political priorities, which undermined its ability to make and implement business decisions. In the case of Zimbabwe local and municipal councils delivered urban water supply. This was usually the responsibility of a department within the municipal government and thus subject to direct hierarchical control in principle. As part of a municipal government department, one would have expected the water unit to have little autonomy over its operations. In practice, a case study of Bulawayo showed that the Engineering Services Department, which is responsible for water, had a considerable degree of managerial and ﬁnancial autonomy. It is able to negotiate the setting of policy objectives, determine its own organizational objectives and budget priorities and, within the limits of its budget, it could vire expenditure between budgetary headings without political and bureaucratic interference (Batley, 1998). Again, as in the health sector the public sector arrangements for delivering water have not been substantially reformed, apart from Ghana’s corporatization of the water provider and the introduction of performance contracting. Chapters 6 and 7 will highlight some private sector oriented reforms such as service and management contracts and concessions both in and beyond the case study countries. Agricultural marketing As has been noted earlier in this chapter, the organizational arrangements for agricultural trade in the four case study countries were largely centralized and state dominated prior to reforms. This was typical of most developing countries where concerns for food security,

Decentralizing Organizational Arrangements for Service Delivery 95

price stabilization, promotion of welfare and substitution for private intermediaries were the main motivations for state intervention. Parastatals in the form of public enterprises or marketing boards dominated key crop purchasing, distribution and exports. Most of these organizations were subjected to reforms driven by structural adjustment in the 1980s and 1990s leading to the creation of some new organizational arrangements. Reform trends have been mainly towards corporatization and commercialization of state organizations with only little movement towards private sector participation. This section will describe the reform of organizational arrangements for agricultural trade. Reforms aimed at bringing in elements of the market are discussed in Chapter 7. The traditional forms of organizational arrangement for agricultural marketing in most developing countries were state/public enterprises and marketing boards with monopoly powers or bureaucratic government departments under the ministry of agriculture. Crop marketing boards operated in all four case study countries. As has been noted elsewhere in this book, liberalization was a key feature of structural adjustment programmes. This had a strong inﬂuence on the restructuring of agricultural trade in countries that adopted SAPs, more so in the two African countries than in the south Asian countries. In Ghana and Zimbabwe, the two main organizational trends are corporatization and commercialization, underpinned by unbundling of huge monolithic bureaucracies and parastatals and/or their exposure to competition or even privatization in some cases. Corporatization and commercialization of parastatals make for clearer roles. In Ghana, the Cocoa Board (Cocobod) underwent massive restructuring, corporatization, commercialization and downsizing in the 1980s and 1990s. The Produce Buying Company (PBC) had earlier been set up in 1977 as the subsidiary company (corporatized) of Cocobod and as the sole buyer of cocoa, following the failure of private licensed buying agencies. By 2000 Cocobod had undertaken further corporatization of several subsidiary companies and agencies.4 The restructuring included ﬂoating of majority shares in PBC to producers and private investors in 1999. In addition, liberalization opened up the domestic purchasing and, later, the export of cocoa to licensed private ﬁrms, thus exposing state-owned corporatized companies to competition with the private sector. In contrast, the Ghana Food Distribution Corporation, the main government food grain agency, remained unreformed and by mid-1990s was virtually bankrupt.

96 The Changing Role of Government

The adjustment-driven reforms in the 1980s led to some liberalization of agricultural trade in Zimbabwe. As was noted earlier, in the prereform era the country had separate boards for grains, dairy goods and cotton, as well as the Cold Storage Commission, all hierarchically organized under the Agricultural Marketing Authority, which was also under the Ministry of Lands and Agriculture. Liberalization and deregulation in the late 1980s and during the 1990s transformed the organizational arrangements for agricultural marketing. For example the Grain Marketing Board (GMB) was commercialized and government had to pay for non-commercial activities it required the GMB to undertake (Hubbard, 2003:112). This contrasted with Ghana’s Food Distribution Corporation which mixed commercial objectives with political and food security concerns without compensation from government. By the late 1990s the deregulation of the domestic maize market ended GMB’s monopoly over buying and selling of maize, followed later by the removal of its monopoly over maize imports, opening the way for private sector competition. Kenya’s Produce and Grain Marketing Board went through similar reforms. It must be noted, however, that recent food problems in Zimbabwe have led to some policy reversals; for example the monopoly of the GMB on all maize trading was re-imposed in 2001. This illustrates the vulnerability of reforms to difﬁcult institutional contexts, including political exigencies. Zimbabwe has also undertaken the corporatization and commercialization of some boards and then gone on to privatize them (e.g. cotton and dairy marketing boards). Others like the grain and meat marketing boards and the Cold Storage Commission were corporatized without privatization. Overall, Zimbabwe seems to have undertaken more reforms in agricultural trade than in Ghana, though the political and economic crises of the late 1990s and early 2000s have set back some reforms. Industry/business services Perhaps more than the other sectors, structural adjustment and liberalization led to radical reforms in organizational arrangements in industry. Adjustment programmes prompted a re-think of direct state involvement in manufacturing industry, and more speciﬁcally the textile sub-sector. The competition and marketization elements were more prominent in industry than in any of the other sectors. Stateowned textile and other manufacturing ﬁrms were exposed to competition and privatization from the 1980s, and in some cases to

Decentralizing Organizational Arrangements for Service Delivery 97

liquidation. In addition several semi-autonomous and independent regulatory, promotion and enabling agencies were set up (Jackson, 2002) in recognition of the fact that the state has a role to play in providing the institutional environment in which markets can function effectively (e.g. by providing information and co-ordinating investments). These are further discussed in Chapter 8. Sri Lanka, Ghana and Zimbabwe, all embarked on adjustment reforms, though at different times. India’s reforms of industry were more conservative and restrained. The emerging NPM-type organizational arrangements for state involvement included: • Corporatization of existing state-owned industries, i.e. turning them into public companies, with ownership and management remaining with government; • Corporatization of state-owned enterprises with private management; • Joint ventures (e.g. Zimtrade in Zimbabwe); • Privatization/divestiture of state-owned enterprises. Initially this was very ambitious in Ghana, Sri Lanka and Zimbabwe, but the result has been disappointing; • Semi-autonomous agencies – mostly regulatory and promoting/ enabling agencies (e.g. Zimbabwe Investment Centre, Export Development Board in Sri Lanka and the Council for Scientiﬁc and Industrial Research in India). Reforms included policies to promote private sector investment in the respective economies and deregulate prices. New agencies (e.g. onestop shops) have been set up to promote investment or existing ones had their mandate revitalized to streamline the procedures for private sector participation in industry. These included the transformation of the Greater Colombo Economic Commission into a national investment promotion organization in Sri Lanka, and the establishment of the Ghana Investment Promotion Centre and the Zimbabwe Investment Centre. Sri Lanka and Ghana have set up Export Processing Zones (EPZs) to further liberalize and encourage investment in industry. In the case of Sri Lanka, the EPZs have been very active in the production and export of garments. These EPZs enjoy a signiﬁcant level of operational autonomy from direct government control. As far as the textile industry is concerned the two South Asian countries have been very much more successful than Ghana and Zimbabwe, where the industry has been in decline since liberalization and exposure to competition from cheap imports.

98 The Changing Role of Government

Apart from state-sponsored enabling and promoting agencies, there were also agencies that were private and independent of government. Examples of these were the Private Enterprise Foundation in Ghana and the Zimbabwe Enterprise Development Centre.

Have reforms improved performance? One of the key rationales for organizational and managerial reforms in government was to improve performance in service delivery. This section will examine the evidence of the effects of reforms on performance. It will ﬁrst examine the extent to which planned reforms were carried out and then whether reforms have improved the efﬁciency and outputs of the organizations. The analysis focuses more on technical than on allocative efﬁciency because the evidence from the research was weaker on the latter. Were planned reforms carried out? It is apparent from the above that there have been efforts towards the introduction of NPM-type organizational arrangements in all sectors and case study countries, but the take up rate has been both slow and varied. Industry and agriculture have gone much further in reforms than water and health; they were exposed more to structural adjustment conditions and liberalization than the other two sectors. Thus some of the more ‘advanced’ reforms – commercialization, corporatization and even privatization – are found in industry and in agriculture. In all sectors, governments are tending to restructure their organizations rather than to make a real change in what they do; governments are still very much in control and in charge of service delivery. This is more noticeable in health and water and to some extent grain marketing, where governments have been reluctant to let go controls even where organizations have been decentralized and granted some autonomy (e.g. the water sector in Ghana and grain marketing boards in Kenya and Zimbabwe). These are sectors where there are still a lot of political sensitivities and welfare considerations, and thus reluctance to change, as Chapter 3 showed. Delays in reforms were also common across sectors and countries. This was obvious in the case of the corporatization of the water board in Sri Lanka, the introduction of private sector participation in the water sector in Ghana, the creation of Ghana Health Service as an executive agency, and the implementation of the autonomous hospital policy in Zimbabwe.

Decentralizing Organizational Arrangements for Service Delivery 99

In summary there has been more change in structures than in the processes that would enable existing and newly created organizational structures to operate. The rhetoric of reform does not match the reality, thus creating a signiﬁcant gap between planned and actual reforms. Sector organizational performance Health – mixed performance Among the four case study countries, Sri Lanka had better health status indicators, particularly in the reduction of Infant Mortality Rate (IMR) by almost half (15.35 per 1000 births) whilst progress in Ghana (72.02 per 1000 births) and Zimbabwe (56.2 per 1000 births) has been much slower during the period of reforms. Zimbabwe had made good progress in improving the health status of its population in the post-independence period by redressing inequalities and expanding coverage of basic services (Mills et al., 2001). This achievement has stagnated and then declined in the past decade due to a combination of AIDs, the brain drain and the declining macro economic and political situation in Zimbabwe. Sri Lanka’s better performance in overall health outcomes is explained largely by factors such as focus on primary health care and female literacy, rather than by organizational reforms per se. In the case of Ghana, the decentralization of decision-making and the greater share of non-salary budget devolved to decentralized units had brought resource allocation closer to the frontline. This has empowered frontline managers, though there was no evidence that this translated into signiﬁcant improvement in health outcomes. In general, poor performance in the sector is explained by weak management systems (planning, budgeting and human resource management, ﬁnance and information systems). There was evidence of good planning systems in Sri Lanka and at the central government level in India, but weak or absent systems at the state level. Also corruption and political interference undermined performance in all countries. Agricultural Marketing – mixed performance In agricultural trade, performance was mixed and inconsistent. For example in Zimbabwe, the Ministry of Lands and Agriculture was initially successful in its post-independent policy of developing and supporting communal farmers. However, the improved performance in output was encouraged by subsidies, which could not be sustained in the face of crisis, especially since 2001.

100 The Changing Role of Government

One common feature in all countries is continued subsidies to loss-making marketing boards and other parastatals. For example, in Zimbabwe, Hubbard (2003) points out that the losses of the grain marketing board increased to ﬁve per cent of GDP in 1993 due largely to overpricing of grains purchased from farmers. The performance of the sector in market development has been held back by policy uncertainty over the opening up of the external grain market. Despite the above, there was some evidence that the restructuring of organizational arrangements has led to signiﬁcant improvement in efﬁciency and effectiveness in some cases. The two leading examples in the agricultural trade sector are Cocobod in Ghana and Dairyboard of Zimbabwe Limited (DZL). Both organizations were commercialized and corporatized and, in the case of the latter, privatized. With full commercialization in 1994 and freedom from statutory responsibilities and political control, DZL became increasingly proﬁtable with a proﬁt of over $Z32 million in 1994 and over $Z70 million in 1997 when it was successfully privatized. In the case of Cocobod, Hubbard (2003:136) describes it as a ‘rare example of a substantially successful public marketing agency based on small holder producers.’ Its subsidiary company, CMC, has performed very well in the export market with a very good reputation for the quality of its cocoa and has a triple ‘A’ credit rating among international ﬁnancial institutions with ability to sell on the London futures market. Like DZL, its success is explained by commercial independence, lack of political interference and competent staff. The performance of other organizations that were commercialized and corporatized has been less impressive. These include the Produce Buying Company (PBC) and the Food Distribution Corporation in Ghana, the Cotton Company, the Cold Storage Commission and the Grain Marketing Board in Zimbabwe, and the Paddy Marketing Board in Sri Lanka. Common explanations for the less impressive performance of these organizations included corruption and mismanagement, lack of ﬁnancial independence and continued political interference. Underlying most of these weaknesses was the lack of adequate and clear frameworks within which corporatized organizations can operate as commercial entities (ibid.). Urban water supply – generally poor performance The performance of the urban water sector was generally poor. As Table 4.1 illustrates, among our case studies only Zimbabwe (Bulawayo) had performed well on efﬁciency, effectiveness and equity

Decentralizing Organizational Arrangements for Service Delivery 101 Table 4.1

Summary of performance in urban water supply

Operational efﬁciency Staff productivity index Unaccounted for water

Ghana

India

Sri Lanka

Zimbabwe

13 per 1000

24 per 1000

28 per 1000

4 per 1000

53%

40–60%

45%

25%

Financial efﬁciency Operating ratio 0.88 Collection efﬁciency

Effectiveness Water availability

Equity % dependent on vendors Estimated urban service coverage (government ﬁgures in italics)5

2.53

0.63

0.7

24% connections not billed

76%

Days receivable ratio – 161 days

N/A

N/A

1–2 hours per day

Only 30% have 24 hr service

24 hours, but rationed

34% of low-income households

N/A

N/A

Minimal

60% (87%)

N/a (92%)

63% (91%)

100% (100%)

Source: Nickson and Franceys, 2003.

indicators, with a staff productivity index of four per 1000, unaccounted water of 25 per cent, almost 24 hours water supply and 100 per cent coverage. In the case of Ghana, reforms produced moderate improvements in some areas like staff productivity, which improved from about 30 in the late 1970s to 13 staff per 1000 water connections by 1995 due mainly to major cuts in the number of employees. As Table 4.1 shows, operational efﬁciency in the form of unaccounted for water, was relatively poor in Ghana, India and Sri Lanka with performance well below the industry average of about 30 per cent. Financial efﬁciency was relatively better in Sri Lanka and Zimbabwe than the other two countries. Performance has been inﬂuenced by a number of factors including the lack of any substantial reforms in organizational arrangements, continued political

102 The Changing Role of Government

interference and lack of ﬁnancial independence. Also in Ghana, and more recently in Zimbabwe binding macro economic conditions like depreciation of the local currency and weak government commitment to reforms have had adverse effects on performance.

Institutional and capacity issues Decentralizing the management of public services involves disaggregating, fragmenting and reorganizing structures, functions and roles within government ministries, departments and agencies. It has signiﬁcant implications for and impact on government capacity. This section brieﬂy examines some of the institutional and capacity issues arising from the case studies, which have wider implications for the design and management of reforms. Operational autonomy is linked to improved performance The sector studies point to a strong link between organizational or operational autonomy and performance. Where managers have the freedom to manage, not just in theory, but also in practice, there seems to be improvement in efﬁciency and effectiveness (e.g. Cocobod and Cocoa Marketing Company in Ghana; Zimtrade and Dairy Board Zimbabwe Limited in Zimbabwe). Nickson and Franceys (2003) cite the case of SAGUAPAC, a water co-operative in Bolivia, as an example of a water provider whose good performance is due largely to its operational autonomy and accountability to users. Those agencies that demonstrated the ability to generate funding independent of government seem to have enhanced their operational autonomy by reducing their dependence on government and minimizing the use of ﬁnance as a lever for intervention. Management autonomy does not only support efﬁciency, but also enables the development of a strategic view. Low levels of organizational autonomy in practice (e.g. the water boards in Maharasthra state in India and grain marketing boards in Kenya, Sri Lanka and Zimbabwe) seem to correlate with poor performance. Government and its bureaucratic machine have a role in enabling decentralized management in public services to work by ensuring that decentralized agencies have adequate organizational resources, are free from unwarranted interventions and operate within an ‘enabling’ institutional environment. Managers should not only have the freedom to manage, but also be allowed to manage.

Decentralizing Organizational Arrangements for Service Delivery 103

Human resource capacity is crucial Human resource capacity is a major issue in the reform of organizational arrangements. In the health sector, skills and professionalization of personnel seem to be better at the central level than at lower levels of government, though some countries (e.g. Ghana) have made signiﬁcant improvement in health management skills at lower levels. The general problem for all sectors seems to be the ability to attract and retain skilled and professional staff, though this varies from country to country. There is a strong link between the competence and quality of staff and the performance of agencies. High performing organizations cited in the sector studies tend to have better salary conditions than comparable government organizations. However, where a strong professional ethos compensates for relatively poor salaries, performance also tends to be high. This was the case in urban water supply in Bulawayo where, although salaries were higher than central government, they were lower than in the private sector. Institutional constraints The application of management decentralization in varying contexts and in different forms suggests that there are some institutional constraints with implications for the capacity of government and its central agencies to manage the process. These may be outlined as follows: • Inadequacy of available technical systems e.g. accounting information systems in hospitals and other decentralized agencies, making accountability and transparency difﬁcult. • Weak personnel management systems, usually politicized and based on patronage. Competent people are trapped in bad systems. • Unreformed institutions such as public service commission rules and ﬁnancial administration regulations that prevent managers of reformed organizations from having control over operational resources, especially ﬁnance and personnel. It is not enough for governments to create decentralized structures if other institutional processes that will enable those structures to work effectively are not reformed as well. A culture of centralization is a disabling factor, reducing government’s ability and willingness to grant autonomy to its operational arms, especially in health and water in the four core countries. Experience elsewhere suggests that unless devolved management and control involve a substantial change in power

104 The Changing Role of Government

structure, devolution of control by itself will only have limited impact (Walsh, 1995). There is a continuous risk that autonomy will be subverted or eroded by ministers and top bureaucrats at the centre. This is evident in the reluctance of central agencies to approve tariff increases in water and user fee increases in the health services of Ghana, India and Sri Lanka. • The tension between autonomy and control, which is evident in the case studies, is related to another institutional problem – the lack of clarity of authority relationships between principals and agents. The result is often conﬂicting and politicized relationships between principals and agents, or between different levels of principals seeking to control a single agent. This was evident in the case of the water company in Ghana, which had multiple principals to report to, including the State Enterprises Commission, the sector ministry, the Ministry of Finance, the board and sometimes donors, each with different motives. Although decentralized management has been touted as good practice with a potential to improve organizational performance, there is need for caution. The experience from other countries suggests that the decentralization of management within public services requires the existence of a credible system for monitoring before relaxing controls over ﬁnance and inputs. Where these controls are weak or undeveloped and arbitrary behaviour cannot be checked, introducing greater managerial ﬂexibility may only increase arbitrary and corrupt behaviour (Nunberg, 1995; World Bank, 1997:20). One key lesson is that the centre needs to be strengthened to perform its integrating, co-ordinating, monitoring and evaluating roles perhaps before or at the same time as decentralized agencies are created (Larbi, 1998a:204). The problem of capacity is not only limited to central agencies but is even more acute at the level of decentralized agencies. Planning, budgeting, and management systems within decentralized units are often weak, whilst ﬁnancial and human resources at these levels are often lacking. If governments are to make progress in further decentralization of management and service delivery then attention needs to be given to issues of accountability and monitoring.

Conclusion Organizational and management reforms have taken several forms aimed at achieving greater operational autonomy. Decentralization of

Decentralizing Organizational Arrangements for Service Delivery 105

management responsibility is a common theme in all four sectors and countries, although the degree of adoption and implementation vary. The best performance often goes with organizations that are ﬁnancially more self-sufﬁcient and/or accountable to their users. Hard evidence on performance was difﬁcult to ﬁnd, but where this was available it pointed to a strong link between managerial autonomy and performance. Greater managerial autonomy, especially over issues like the ability to determine and manage fees and tariff increases, budgets, investment planning and human resources without undue interference from politicians and central ministries, are associated with better performance in terms of productive efﬁciency and effectiveness in achieving outcomes. This is irrespective of sector and of ownership. Autonomy is not so much a legal status as a real practice. This was evidenced in the case of water supply bodies, which were both ﬁnancially self-sufﬁcient and locally accountable to either elected local representatives (Zimbabwe) or cooperatives (Bolivia). However, managerial autonomy needs to be within a clear policy framework and matched by ex-post managerial accountability to users or elected representatives. It was also evident that for economic sectors such as industry and agriculture, exposure to market competition helped to promote efﬁciency and good performance where managers were relatively free to manage. Whilst reforms looked comprehensive on paper, actual implementation was limited in practice. Pre-reform organizational arrangements in health and water have not signiﬁcantly changed due largely to welfare and equity concerns and the traditional attachment to public provision. Thus in health, water and, to some extent, grain marketing, governments still dominate in provision with some shift to arms-length and decentralized agencies. Capacity is a major issue in reforms, especially human resource capacity and adequacy of available technical systems such as accounting information systems in hospitals and other decentralized agencies, making accountability and transparency difﬁcult. The institutional context of reforms is crucial in affecting reform outcomes and needs more attention in planning reforms.

5 The Experience of Charging for Public Services

Introduction This chapter examines the experience of ﬁnancing public service delivery by charging users directly through user fees and tariffs and/or withdrawing government subsidies to the provision of services. As was noted in Chapter 2, charging for public services is part of the new approach to public management. Like other reforms examined in this book, charging raises a number of issues relating to the appropriateness of reforms to poor countries, the ‘new’ roles of government in implementing a policy of user charges, and the capacity implications for government and public agencies in managing the new roles associated with charging. The chapter begins with a review of the context and rationale for the introduction of charging and for withdrawing subsidies. This is followed by a discussion of the application of user charges in health, urban water and agricultural marketing in Ghana, Zimbabwe, India and Sri Lanka. Where possible evidence from other countries is used for comparison and to highlight particular issues. The performance and impact of user charges are analyzed next. The main concerns here are whether the policy of charging has achieved the purposes for which it was introduced in the various sectors in terms of revenue generation, equity, efﬁciency and effectiveness. It highlights some of the capacity issues that help to explain performance. The concluding section of the chapter highlights key lessons and broader issues arising from the discussion.

Explaining user charges Charging for services takes different forms. In the health sector it has mainly taken the form of user fees either at the national level, 106

The Experience of Charging for Public Services 107

involving all parts of the health system, or a community-based charging system, which usually operate in primary health care (Mills et al., 2001; Arhin-Tenkorang, 2001). Whilst user fees are out-of-pocket payments at the time of use of a service, a number of prepayment schemes such as health insurance are increasingly becoming important in developing countries. For water services, charges are often in the form of consumption tariffs and connection fees charged to households. The application of charging in the economic sectors has taken the two forms. First is the direct payment for services rendered to users (e.g. issuing of permits or agricultural extension services). The second is the withdrawal of subsidies so that users pay the market price or the full cost of services (e.g. inspection and grading of crops before exports, and fertiliser and other farm inputs supplies). The policy of charging users for services is not new in the sectors studied. What is new is the widening of scope and application to a broader number of services and countries from the 1980s through the 1990s. Efﬁciency, equity and quality arguments were used to justify the introduction of user charges. Universal basic services, whilst desirable, became increasingly unsustainable for poor countries, and in practice became inequitable as both the rich and the poor were subsidized. In the case of water, for example, the World Bank, (2003a, 2003b) argues that there are more equitable and efﬁcient ways to provide subsidies to the poor than to impose low tariffs on all users. More targeted approaches would include subsidizing the cost of household connections in low-income areas as a way of improving access by the poor. Advocates of charging also argue that free or subsidized public provision of services, where ﬁnance is also largely public, encourages excessive use or inappropriately high demand. There is need to manage the demand side through charging and to improve equity by reallocating and targeting revenue to provide services for poorer communities (Donaldson and Gerard, 1993; Adams and Hartnett, 1996). For example in health, cascading or scaling of user charges according to the level of health facility, should provide the correct price signal to encourage better use of the referral system for more serious cases. In the case of economic services to agriculture and business, it was argued that subsidies sent the wrong signal to both producers (to over produce) and to users (to over demand) (Hubbard et al., 2003). Thus user charges were seen as instruments to address inefﬁciencies and inequities in service provision, whilst recovering at least part, if not all of the cost of delivering services from users.

108 The Changing Role of Government

User charges represent attempts to diversify the ﬁnancing of public services and reshape public spending. These have become widespread in the health sector in Sub-Saharan Africa but less so in South Asian countries (Mills et al., 2001; Adams and Hartnett, 1996). In public utilities such as water, where tariffs already existed, they have seen signiﬁcant increases in the past few years in some countries in Africa and Latin America following private sector involvement or as part of reforms, as in Bolivia and Argentina (see Chapter 7 of this book). As has been noted elsewhere in this book, a number of factors explain the introduction of reforms in the public services. First, charging was a policy response to difﬁcult macroeconomic conditions and crisis in public ﬁnances which imposed resource constraints on the ability of governments either to ﬁnance the maintenance and expansion of infrastructure for the delivery of public services or to continue to subsidize inputs to agriculture and other services. The second factor explaining the introduction of charges was the adoption of IMF and World Bank stabilization and structural adjustment programmes prompted by crisis. There were speciﬁc conditions attached to adjustment lending and donor support to sector reforms, some of which explicitly demanded the introduction of new user charges or increases in existing ones in health (World Bank, 1993), water and other public utilities (World Bank, 1994). Governments had to stabilize public ﬁnances by cutting back on public expenditure, raising taxes, eliminating subsidies, which could not be sustained under ﬁnancial crisis, and introducing price controls. Thus expenditure cuts in public services became a common feature in many countries that adopted structural adjustment programmes (Breman and Shelton, 2001; Jayarah and Branson, 1995). In addition charging was seen as one way of addressing issues of inefﬁciency and ineffectiveness in the delivery of services and to recover costs. It is worth noting here that in recent years there has been a policy rethink in the World Bank and some donor agencies, which call for caution in the introduction of user charges in health and basic education, because of their effects on poor people’s access (World Bank, 2003b). For economic sectors such as agricultural marketing and industry, the move towards increasing commercialization, corporatization (and privatization in some cases) meant that charging users for the cost of services became a key instrument. In the water sector, the traditional view of water as a public good has been challenged by recent reforms. Moreover, with its competing uses, water has economic value and therefore needs to be recognized as an economic good. Charging users

The Experience of Charging for Public Services 109

to cover both operational and capital costs is seen as a necessary step not only to distribute costs fairly among consumers but also to attract private sector participation (Nickson and Franceys, 2003). The varying contexts of case study countries All the four case study countries – India, Sri Lanka, Ghana and Zimbabwe – are post-colonial countries that had commitments to government provision of public services in general and free social services in particular. Even when services were paid for these were often subsidized by the state. As in most countries, introducing charging raised two difﬁculties for governments – the political undesirability of raising revenue from an already poor population and the difﬁculty of collecting charges or fees in services such as health (Donaldson and Gerard, 1993). The motive for the introduction of charges in the case study countries has been varied. As noted elsewhere in this book, new management approaches tend to see user charges as a way of increasing allocative efﬁciency in the use of resources and increasing accountability to clients. However, the practice in the case study countries suggests that the objectives for charging were more prosaic – to raise revenue and to service debts (Batley and Larbi, 1999). In terms of context, the two sub-Saharan African countries, particularly Zimbabwe, were more susceptible to external donor pressure to introduce charges than India and Sri Lanka. This may be explained by the depth of ﬁnancial and economic crises in African countries in the 1980s and early 1990s, which gave the international ﬁnancial institutions more leverage to press for the adoption of structural adjustment programmes, the introduction of, or increases in user charges and the abolition of subsidies. In the case of health, the severity of the crisis was evidenced, for example in Zimbabwe, by the fall in the Ministry of Health’s real expenditure by 14 per cent between 1990/91 and 1991/92, and by a further 29 per cent in 1992/93 (Russell et al., 1997). In the case of Ghana, real government health spending at the start of the structural adjustment programme in 1983 was only 20 per cent of its 1975 levels (Smithson, Asamoa-Baah and Mills, 1997). In addition to deteriorating infrastructure and lack of essential equipment and supplies, salaries and other conditions of service for public sector personnel had deteriorated such that a signiﬁcant number of professionals were leaving the public sector for either the private sector or abroad. Also the political environment in Ghana and Zimbabwe enabled the introduction of charges without much opposition. In the absence of

110 The Changing Role of Government

electoral competition in the two countries, the political cost of user charges was relatively low in Ghana and Zimbabwe. Populations in both Ghana and Zimbabwe were also used to some form of charging, even if informally, so fees did not come as a big shock. In contrast to the African countries, user charges have not been part of the reform policy agenda in India and Sri Lanka, despite World Bank pressure and a long experience of economic liberalization and adjustment in the case of the latter. Part of the explanation was that India and Sri Lanka (and other south Asian countries such as Bangladesh) have a long political tradition and strong ideological commitment to free basic social services. In addition opposition to user fee policy was much stronger in India and Sri Lanka than the two African countries; opposition came from a cross section of society, including trade unions, professionals and politicians. The political environment was more competitive with a high risk of political backlash if charges were introduced and or increased in services such as health and water. Thus charging for services, especially in health, has become more widespread in sub-Saharan Africa than in India and Sri Lanka. The next section will examine the extent to which user fees have been introduced and worked in practice.

Charging – what and how? The previous section provided an overview of the context and rationale for charging public services. This section discusses the extent to which the policy of charging was implemented in the case study countries and sectors. It has to be noted, however, that industry receives light attention in the rest of this chapter due to the little available evidence on charging. It is worth remembering that both health and water are treated as ‘merit goods’; they are neither pure public nor pure private goods. Governments have traditionally stepped in to provide these services at subsidized rates because of the market failure arguments noted in Chapters 1 and 2. Charging is therefore sometimes highly controversial, though adjustment and liberalization challenged this perception. Health As noted above, user charges have not been part of the reform policy agenda in India and Sri Lanka. The focus in this section will therefore be on the two African countries where user charges have been widespread.

The Experience of Charging for Public Services 111

Ghana was a relatively earlier candidate for structural adjustment than Zimbabwe and introduced user fees by legislation in 1985, compared to Zimbabwe in 1991. The World Bank encouraged the government to recover up to 15 per cent of recurrent health expenditure from fees and also to set up a drug revolving fund (‘cash and carry’) in 1988 with the objective of recovering the full cost of drugs from fees. Full cost recovery for hospital feeding was also introduced in 1994. Whilst user charges were introduced as instruments to raise revenue and to manage demand, they also gave managers more control over resources generated from fees; revenues were retained at the facility level. Implementation of the user fee policy in Ghana was not without problems. Whilst the enabling legislation allowed for the revision of drug fees in line with inﬂation, it failed to make any provision for the revision of non-drug fees. As a result, inﬂation eroded the value of the non-drug fees such that by 1995 the real value of the ofﬁcial fee level had dropped by more than 90 per cent (Smithson, Asamoa-Baah and Mills, 1997:24; Nyonator and Kutzin, 1999). This encouraged facility managers to collect unofﬁcial fees, sometimes with the knowledge and tacit support of the Ministry of Health. Anecdotal evidence also suggested that unofﬁcial or illegal charges by health professionals were common. Zimbabwe had a long tradition of user charges in government health facilities before independence in 1980. However, a desire to redress historical imbalances in the immediate post-independence period meant that cost recovery received low priority and exemptions were introduced in 1981. Unlike Ghana, exemptions in Zimbabwe were initially based on an income threshold of about US$12 (Z$150) per month. Crisis in the late 1980s led to a new emphasis on cost recovery as an integral part of structural adjustment. The objectives included to raise revenue from two per cent of recurrent cost to ﬁve per cent by 1993, and to eight per cent by 1995; and to improve efﬁciency by strict enforcement of fees in order to discourage frivolous use and encourage appropriate use of the secondary and tertiary hospitals. Revisions in 1994 raised the income threshold to about US$30 (Z$400) in order to improve equity. However, in practice managers did not have adequate information to allow means testing exemptions to work effectively (Russell et al., 1997) and the poor were the losers. User fee rates in Zimbabwe were relatively lower than in Ghana despite revision in 1994. All the same, fees negatively impacted on the poor and had to be withdrawn from government health centres in 1995 due to low utilization and the high cost of administration. Unlike

112 The Changing Role of Government

Ghana, revenue from fees was remitted to the Treasury in the early years of the policy, which was a disincentive to managers to be serious about charging and collecting fees at the facility level. It was not until 1996 that retention of fees was allowed at the cost centre level for nonsalary recurrent items, thus giving some degree of ﬁnancial autonomy to managers of the Health Service Fund. Even though revenue retention was decentralized, this was only to the district, provincial and central hospitals rather than to the level of every health facility. Thus Ghana’s system was more decentralized than that of Zimbabwe. Water Charging was not new in the water sector in the four case study countries, even though governments have always subsidized provision across the board. The level of tariffs varied across countries and attempts to increase tariffs have been resisted, especially in the two Asian countries where political sensitivities to increases in tariffs were more noticeable. Public expectations of free water supply or low tariffs, and poor collection rates by providers have been the product of the history of those countries after independence when most governments were committed to free public services. Before reforms, the water sector in Ghana was heavily subsidized. A combination of low tariffs and poor collection rates meant that the then Ghana Water and Sewerage Corporation (GWSC) was unable to cover its operational cost. Increases in tariffs were an essential part of the cost recovery policy adopted in Ghana under the auspices of structural adjustment in the 1980s. As Table 5.1 shows, 1986 saw a 355 per Table 5.1

Cost recovery as a percentage of recurrent health expenditure

Year

Ghana (%)

Zimbabwe* (%)

1985/86 1986/87 1987/88 1988/89 1989/90 1990/91 1991/92 1992/93 1993/94 Average

5.2 7.6 12.4 10.3 7.7 8.2 7.8 7.9 9.6 8.5

N/A 2.6 2.5 2.3 1.3 2.0 3.5 N/A N/A 2.4

*Zimbabwe’s recovery ﬁgures are based on non-salary recurrent health expenditure, whilst Ghana’s are based on recurrent health expenditure.

The Experience of Charging for Public Services 113

cent increase in tariffs for the ﬁrst time in many years as part of World Bank-supported water sector reforms. Following the adoption of corporate planning and the signing of performance contracts by the GWSC, the World Bank got the government to agree to allow increases in water tariffs to recover operational cost and to take account of inﬂation and debt servicing. In practice this was difﬁcult as increases required political approval, which was often either delayed or refused for fear of a backlash from the public. The situation has changed since the late 1990s with the creation of an independent Public Utilities Regulatory Commission. This has taken tariff setting out of the hands of politicians, led to signiﬁcant increases in tariffs and enabled the water company to recover more than its operational and maintenance costs. Tariffs are now adjusted annually, almost in line with inﬂation. This is part of the attempt to get the sector ready for private sector participation. In Zimbabwe, the need to redress past imbalances meant that cost recovery from users was de-emphasized after independence in 1980. There were two levels of tariffs. The Department of Water Resources charged ‘bulk water tariff’ to retailers of water (mainly local authorities) and another consumer tariff, which was charged by the retailers, subject to ministerial permission. The bulk water tariff was usually below cost in order to allow for the cross-subsidization of non-viable areas. Cross subsidization was done in the name of equity, but this tended to ignore the real costs and, in practice was inefﬁcient and inequitable as it beneﬁted most the large consumers in subsidized areas, such as commercial farmers. There were indications at the time of this research that both the ministry and the local authorities intended to increasingly recognize true costs in the tariff setting. Unlike Ghana, water tariffs charged by municipalities (cities and big towns) were set and collected by local authorities with the approval of the Ministry of Local Government. Thus there were two levels of tariff controls: one by ministers restricting the rates charged by the Department of Water Resources for bulk water to municipalities and the other by the Ministry of Local Government to control the rates charged by municipalities to consumers for their household water supply. The only exceptions were places where water was supplied directly by the Department of Water Resources (DWR), in which case it was the DWR that set the tariffs (Mudege, 1997). The tariff structure typically included the cost of capital works, operation and maintenance. By law, urban councils were required to set up separate water accounts and to charge tariffs that enabled them to balance that

114 The Changing Role of Government

account. At the time of this research there were plans to de-regulate tariffs and the Minister of Local Government could permit increases of up to 30 per cent per annum if councils provided good justiﬁcation (Batley, 1998). In Sri Lanka, the National Water Supply and Drainage Board (NWSDB) had the power to charge for water and to determine the appropriate level of tariffs, but subject to the political approval of the Minister of Finance and other relevant ministers. However, collection of charges started only in 1984 due to public resistance. It took 10 years for NWSD to reach a respectable level of 95 per cent bill collection rate (Franceys, 1997). Tariff levels were generally low and did not cover all operating costs. In the early 1990s, the World Bank and other donors argued for reduction in government subsidies to NWSD and for the board to recover a greater proportion of its costs, including debt servicing. This was to be achieved by increasing tariffs and by gradually phasing out capital subsidy. However, the government feared this might increase the cost of household water supply beyond affordable limits and was therefore reluctant to implement the increase. Perhaps more than any of the other three countries, water supply in India has been treated as a social service, close to a public good. According to Franceys (1999) consumers have been led to believe that water supply is a right for which no payment is required. This view has been reinforced over the years by the attitude of sector staff that ‘government is responsible for ﬁnancing all aspects of their work and that the most that can be expected is for consumers to pay for part of the operational and maintenance costs’ (ibid.:10). The setting of tariffs is the responsibility of state governments. Whilst both the national and state governments recognized the need for cost recovery to address the ﬁnancial needs of the sector, they have been slow to change the tariff policy to respond to those needs. Elected representatives, who are reluctant to increase the water rates in order to avoid public disfavour, run the local bodies responsible for delivering water. Just to illustrate, at the time of this research, tariffs in Orissa started from about two US cents (Rupees 1.5) per m3 for domestic supplies and even lower in Goa at less than two US cents per m3. In the case of the latter, tariffs were increased in 1996, for the ﬁrst time since 1988 (Franceys, 1999:28). In recent years, there has been an increasing pressure and trend to increase charges in some states and municipalities in India. For example Hyderabad increased water prices by about 30 per cent. For water, as in health, a survey of users suggests that they are willing to pay more only if the service will improve to justify the

The Experience of Charging for Public Services 115

increase (Rakodi, 1998; World Bank, 2003b). However, as the World Bank (ibid.) rightly observes, the risk comes when there is a mismatch between low quality of service and substantial increases in prices. This can result in public protests as was played out in the case of Cochabamba in Bolivia, leading to the termination of the contract of the private sector provider (Nickson and Vargas, 2002; Nickson and Franceys, 2003; World Bank, 2003b). Charging in agriculture As an economic sector one would have expected charging to be widespread in agriculture. However, this was not the case before reforms. The state subsidized services to the sector in all the case study countries. Adjustment and liberalization brought some changes in policy in favour of reducing or even eliminating subsidies and introducing charging. In Ghana, some subsidiaries of the Cocoa Marketing Board (Cocobod) charged for some services such as those provided by the Quality Control Division (QCD). With the liberalization of the cocoa trade and the subsequent introduction of private licensed buyers in domestic buying and export sales, it was important that Ghana’s long-standing reputation as the world’s top quality cocoa producer was not compromised. The QCD extended its services (e.g. inspection, grading, sealing and fumigation) to the private licensed agencies for fees charged at the various stages from the warehouse to the ports (Hubbard, 2003; Shepherd and Onumah, 1997). The QCD provided similar services for coffee – grading and sealing bags inland, inspection at the ports, fumigation and the eventual issuing of a certiﬁcate. As at 1995/96 charges ranged from about six US cents (100 cedis) per inspection to about 18 US cents (300 cedis) per bag for grading. The Ministry of Agriculture inspectors also charged six US cents per MT to test sample insect damage (Shepherd and Onumah, 1997). The QCD has increasingly become self-ﬁnancing with a trend towards full cost recovery. However, despite commercialization, the lack of autonomy meant that other parts of Cocobod, such as the Produce Buying Company, were not allowed to operate on a full cost recovery basis. Kenya (one of our reference countries) introduced, with donor support, a crop purchasing revolving fund to improve the cash ﬂow of the National Cereal and Produce Board (NCPB) and to help pay farmers promptly. However, like the drug revolving fund and user fees in health, the scheme was introduced without any clear protocol and ring-fencing arrangement to ensure the fund was used only for food purchases. This resulted in leakages and misapplication of the funds.

116 The Changing Role of Government

The Central Warehousing Corporation and the various State Warehousing Corporations in India also operated charges (rents) for the use of their warehouses by the Food Corporation of India. Similarly, in Sri Lanka the Food Commissioner’s Department (FCD) charged commercial rents for the use of its storage space by private traders, licensed bondsmen, Multipurpose Co-operative Societies and sometimes by the Paddy Marketing Board. The Marketing and Food Policy Division of the Ministry of Agriculture charged for commissioned work. Both organizations were able to generate additional resources to cover part of their operational cost (Smith and Ellis, 1997). In Zimbabwe, the Department of Veterinary Services charged fees for some of its services such as inspection of meat and issuing of permits for milk and meat product exports and imports. Similarly, the Department of Research and Specialist Services (DRSS) charged users for its meat grading and other services. However, fees accounted for only 30 per cent of the actual cost of meat grading. Overall, charging generated only ﬁve per cent of the budgetary allocation of the DRSS, and this went to the Treasury. It was estimated that full cost recovery for its meat grading and other services would have generated up to 29 per cent of the department’s budget and made a signiﬁcant difference in resolving its ﬁnancial problems. Areas such as meat grading, soil analysis and fertiliser recommendations, pest and disease identiﬁcation and advisory services, seed analysis and germination testing, were all identiﬁed as potential candidates for full cost recovery (Hubbard, 1999). In summary, the incentives for charging do vary according to sector. Charges and fees were most easily introduced in the case of sectors or services with lower equity concerns and where the service being offered (e.g. domestic tap water, meat grading) was measurable and a private rather than public good. Apart from India and Sri Lanka, charging in water seems to be increasingly reﬂecting the true cost of service delivery though political sensibilities understandably do persist. In water there is a high equity concern and this can be accommodated technically, quite simply by exemption from connection charges and by lifeline and stepped tariffs for consumption, but only if water is on meter.

Charging: performance and impact The policy of charging was introduced in the different sectors in principle to meet various objectives, including equity, efﬁciency and

The Experience of Charging for Public Services 117

effectiveness of services. However, in practice revenue generation to help address ﬁnancial crisis was the major motivating factor. This section examines the performance and impact of charging against these objectives. Charging and equity One of the main concerns in the introduction of charging in public services is its potential impact on access to services by the poor who may not be able to afford the cost. Different services have dealt with the issue of equity by providing some sort of safety net for the poor. Equity is more of an issue in social services such as health and basic water supply than in economic sectors, although in agriculture there are food security concerns. Health: In the health sector charging was introduced with the expectation that the poor would be exempted. Although exemption systems were set up in Ghana and Zimbabwe, the research ﬁndings reinforce the view that exemption systems often fail the poor and that user fees become inequitable without such protection. In Ghana and Zimbabwe there was lack of preparation in setting up exemption, monitoring and evaluation systems with negative impact on equity of access. For example in a survey in the Volta Region of Ghana it was found that the exemption system was virtually non-functional with less than one in 1000 contacts granted exemptions (Nyonator and Kutzin, 1999). Even for the exemptions granted, 71 per cent of these went to health service staff. Similarly, in Zimbabwe, a signiﬁcant proportion of the poor and vulnerable were not effectively protected due to stringent requirements for proof of income (Chisadza, Mapong and Nazerali, 1995; Russell et al., 1997). The user fees system included a system of waivers for the poor based on an income threshold of about US$33 (Z$400), but in practice fewer than 20 per cent of the eligible poor actually received waivers. In both Ghana and Zimbabwe, it was apparent that health managers were more interested in raising revenue from user fees than in granting exemptions, which translate into loss of revenue not fully compensated by government. Similar studies in Uganda (Konde-Lule et al., 1998) also found that exemptions and waivers from user fees at the local level were not effective, and that managers had little motivation to extend exemptions and waivers. Reinforcing the inequity of user fees is the evidence of a signiﬁcant reduction in utilization of health services by the poor in reaction to the introduction of user fees. In Ghana, a study in the Volta Region

118 The Changing Role of Government

found a 50 per cent drop in outpatient attendance (Waddington and Enyimayew, 1990), with the decline in utilization being more sustained in the rural areas, where the majority of the poor live, than in the urban areas. In Zimbabwe, equity of access and utilization improved in the period up to the mid-1980s, but the concentration of resources in large hospitals created input disparities to the disadvantage of peripheral facilities, which faced acute input shortages. Overall, user fees had negative impact on utilization as fewer people visited clinics and hospitals because they could not afford the user fees. This is consistent with similar ﬁndings in other countries (see Gilson et al., 2000; Nanda, 2002). Poor patients in Zimbabwe responded to charging by delaying treatment, refusing certain investigations, prescriptions and procedures in order to reduce their hospital bill and thus reduce their utilization of health care. However, fees had relatively less impact on utilization in Zimbabwe than in Ghana due to lower fees and weak enforcement. Overall, the impact has been that resources have somewhat increased leading to improvements in quality, principally in the form of increased drug availability, but that the poor have been further excluded. In contrast to the evidence from Ghana and Zimbabwe, other studies in Africa suggest improvement in the utilization of health services, despite user fees. In the case of Niger, Mauritania and other countries, this is partly due to a combination of community ﬁnancing and insurance, and improvement in quality (Gilson et al., 2000). Kenya also presents another example where a carefully planned introduction of user fees has worked reasonably well. An initial attempt to introduce user fees in 1989 failed due to a number of factors, including poor planning and implementation strategies, poor cost recovery and high cost of administration, and 50 per cent decrease in outpatient numbers, with the poor most affected. The policy was re-introduced in 1991 with donor support, using a cascading fee structure so that fees were highest at the referral level and declined at lower levels of the health system. The emphasis was on exploiting the biggest revenue sources and speeding up the payment of national insurance reimbursements to users. With an incremental approach to implementation, user fees were gradually increased and gained increasing user acceptability as managers focused on using the revenue to improve the most visible aspects of quality (Collins et al., 1996, cited in Mills et al., 2001). Water: In the water sector the World Bank has argued that generalized subsidies to all users in the form of low tariffs have often not beneﬁted the poorest, who pay more for water because they frequently

The Experience of Charging for Public Services 119

do not have access to public supplies (World Bank, 2003a). Instead, public subsidies for low income groups should be more targeted by covering part of the connection cost as a way of improving access. Connection costs are often high and unaffordable to the poor; but the payment of consumption charges after connection may not be such an issue for low-income groups since they normally pay more than the ofﬁcial rates for supplies purchased from private vendors and landlords of multi-tenancy households. However, pro-poor consumption charges require a scaled tariff structure to ensure that the poor get the minimum supply of water required at affordable rates. As Plummer (2000a) has argued, the structure of consumption charges can be a big problem for the poor because they are not adjusted to the irregular and uncertain sources of the poor’s income. One common way of addressing equity in the water sector is through the ‘lifeline provision’ of the basic necessary water at low rates. This varies from 6 m3 per month in South Africa to 20 m3 per month in some other countries. However, in multi-tenancy or large family households the poor tend to pay more per unit of water than in higher income groups because their gross consumption level takes them collectively above the lifeline rate. There was evidence of inequity in tariffs in the water sector. For example in India, a survey of 35 urban centres in Pune and Maharashtra states suggested that ordinary consumers paid between US$0.47 – US$1.90/m3 for unmetered water, compared to an average metered water tariff of just US$0.06 – US$0.07/m3 for middle and high-income households (Rakodi, 1998). In Sri Lanka, despite lifeline/basic needs tariff rates, middle and high-income households in Colombo, the capital city, enjoyed more subsidized tariffs because of the reluctance of the government to raise tariffs signiﬁcantly above the rates applying to the lifeline provision. In the case of Ghana, an average family spent about US$0.50 a day on water, more than half the minimum wage of US$0.90 which many poor people do not earn (Rakodi, 1996). Lifeline provision of the quantity of water required for basic needs assumes that low-income groups are connected to the network, but this is often not the case. The cost of buying from private vendors and connected multi-tenancy homes tend to be 5–20 times higher than the ofﬁcial tariffs. Public subsidies to low-income groups should be targeted in order to cover part of the connection fee. However, as the case of Bulwayo (Box 5.1) suggests, tariff structures can be adjusted to increase equity if consumption can be measured. This is technically much simpler than in the health sector.

120 The Changing Role of Government

Box 5.1 Addressing equity and efﬁciency in tariff reforms: the case of Bulawayo Bulawayo is the second largest municipal council in Zimbabwe. Its Engineering Services Department is responsible for the delivery of water. Reforms in the water sector have attempted to address the issue of equity and allocative efficiency at the same time in response to changing circumstances. The general thrust of reforms since the 1992 drought has been to increase both the efficiency and the equity or fairness of provision. There were some marked inefficiencies and inequities in the previous tariff structure which (i) set a higher fixed charge for low-income people in high density areas (HDAs), but which then encouraged high and wasteful consumption levels by (ii) not charging the residents of HDAs for consumption, and (iii) charging others at a rate which did not vary by level of consumption. Under conditions of drought, rationing had a reforming effect. On the side of equity, a universal ration of consumption per household was established to cover basic needs; above that level it was mainly the better-off who were affected and ‘the brunt of cuts fell on outdoor use, especially garden watering’ (Winpenny, 1993:33). On the other hand, because their households tend to be larger, lower income groups were more constrained by the universal ration per household. With respect to allocative efficiency, the fixed limit on household use had the positive effect of allocating water to basic survival needs from inessential uses. With regard to non-domestic use, according to Binnie and Partners (1993:46), most industrial ﬁrms restrained their use of water for civic reasons rather than because they were seriously affected by the cost of water. In that sense, industrial tariffs and penalties remained well below their true economic value and were therefore allocatively inefﬁcient. The post-drought tariff structures strengthened some of the tendencies in favour of the equitable and efﬁcient allocation of water. ‘Rationing’ was now by price not imposition. Equity was offered by the imposition of lower fixed charges in the central city (mainly lower income areas), a low charge per unit of water was introduced for essential consumption levels, followed by a progressive sliding scale, with higher fixed and consumption charges for the low density suburbs and industry. In terms of allocative efficiency, the sliding scale (or rising block tariff) ensures the allocation of water to higher value uses and discourages wasteful consumption. But, logically, above the basic essential level of consumption, economically efficient allocation would require the same scale of tariffs for all uses so that rational choices are made between them (Binnie and Partners 1993:49). Instead, Bulawayo City Council has chosen to favour high density inner city areas by applying a higher scale to low density areas and a fixed rate, high charge to nondomestic uses. Source: Batley, 1998.

The Experience of Charging for Public Services 121

Agriculture: In agricultural marketing, safety nets for the poor are required when food supplies are entrusted to the market, in order to ensure all can buy staple food. There have been attempts to shift to market-based food security, by state withdrawal from the purchase and distribution of food and the introduction of income-based relief in the form of cash or vouchers. Hubbard (2003:115) cites the case of Sri Lanka in the 1980s, which moved half of the population off rice ration schemes and onto a food stamp programme. Other countries such as Botswana and Namibia operate income-based safety programmes during drought, whilst the Indian state of Maharashtra gave cash payments in their famine relief programmes in the 1980s (ibid.). Has charging improved efﬁciency and effectiveness? The introduction of user fees in health and increases in water tariffs have had limited impact in easing the ﬁnancial burden on governments in the case study countries. This is partly because of popular resistance to increasing user fees and tariffs due mainly to political sensitivities to such increases. As Table 5.1 shows, on the average Ghana has recovered about 8.5 per cent of recurrent health expenditure compared to only 2.4 per cent in Zimbabwe, though there were signiﬁcant variations from this average from year to year. It is worth noting, however, that in the case of Ghana the bulk of user fee revenue (80%) was accounted for by drug charges. There was evidence that some of our reference countries performed better. For example, Thailand was able to generate 80–90 per cent of non-staff recurrent expenditure at hospitals. This is partly explained by the fact that historically Thailand has never provided health services for all free at the point of delivery and there is a cultural acceptance of paying for services. Besides, the government has speciﬁc selective schemes (as opposed to comprehensive subsidies) to deal with difﬁculties in payment and concerns for the poor (Mills et al., 2001; Bennett et al., 1998a). In the water sector, Zimbabwe (Bulawayo) and some of our reference countries (Argentina and Bolivia) performed better at raising a substantial proportion of their ﬁnance through tariffs. In the case of the latter two countries, this was partly because of the involvement of service users in the management of the service and of government commitment to and acceptance of change. The private sector is very much involved in the delivery of water in both countries. In recent years the Ghana Water Company has been able to cover its operational and maintenance cost by increasing tariffs with the approval of the Public Utilities Regulatory Commission. Cost recovery from tariffs

122 The Changing Role of Government

in India and Sri Lanka is relatively lower because of the reluctance of governments to allow realistic increases. It was difﬁcult to ascertain the impact of charging on the quality and efﬁciency of services in all the sectors because of either the lack of evidence or weakness in the data available. In the case of health, there was evidence that the cascading or scaling of the user fee system in Zimbabwe encouraged a more efﬁcient use of secondary and tertiary hospitals, i.e. the referral system. The withdrawal of fees at government health centres in 1995 reinforced these incentives. In contrast Ghana had a ﬂat rate user fee system across health facilities, which did not encourage appropriate use of hospitals and the referral system. However, it was evident that Ghana’s drug revolving fund had enabled hospital managers to improve the quality of services by redressing the perennial problems of scarcity and erratic supply of drugs. It is worth pointing out that charging created perverse incentives and inefﬁcient provider behaviour in both Ghana and Zimbabwe, including the reluctance to grant exemptions (because this meant losing income) and over prescription of drugs since that generated more revenue. In the other three sectors, the introduction of or increases in charges have generated demands from users for better quality and reliable services, particularly where users can organize themselves. Example of this was the organization of water consumers in Bulawayo to demand improved services. Users are increasingly voicing demands for ‘value for money’ services. Increasing user voice through participation and systems of complaints has a potential to make providers more responsive and accountable to users, particularly the poor. Service providers cannot afford to ignore this trend. However, continued subsidies and government reluctance to eliminate them in sectors such as agriculture undermine efﬁciency. One key and common problem is government reneging on its ﬁnancial commitments to service providing organizations, for example by late or even non-payment of bills. In a study of Ghana, for example, public institutions were found to be several months and even years in arrears in the payment of water bills (Larbi, 1998a). This is a signiﬁcant disabling factor in their ability to generate revenue through charging. This is also evident in the water sector and in grain marketing in Kenya and Zimbabwe, where state agencies tend to owe substantial arrears in bills and promised investments are either delayed or even shelved. In the case of agricultural trade, government ﬁnancial crisis is cited as one of the key obstacles to managing commercially contracted food security stocks in Kenya and Zimbabwe (Hubbard, 2003). To the above

The Experience of Charging for Public Services 123

constraints can be added political interventions and delays in approving increases in user fees in health and water tariffs. The effectiveness and efﬁciency of charging are partly based on the capacity of public organizations to manage them. The case studies demonstrate that poor administration and management of user fees and charges were not only reﬂected in lack of working exemption systems, but also in defective billing and collecting systems which leave a signiﬁcant proportion of bills unpaid as in India, or create loopholes for illegal charges and corruption in the case of user fees in health. Besides government accounting systems are often weak and unable to generate information for frontline managers.

Conclusion In principle a policy of user charges can be used as an instrument to promote equity provided that effective and functioning exemption systems for the poor and vulnerable groups are built into the policy or subsidies are speciﬁcally targeted to reach the poor. In practice, the evidence from the case studies shows that exemption systems fail the poor and are the weakest link in the policy of charging. This applies more to health than to the water sector where lifeline provision and a scaled tariff structure protect the poor to some extent. Thus the policy of charging may be well intentioned, but it may have unintended consequences in implementation. This requires that subsidies for the poor be carefully constructed and targeted, a message also emphasized by the World Bank (2003b). The key dilemma for most governments is how to balance equity issues with the economic rationality of cost recovery. Underlining this dilemma is the inability to devise effective safety nets for the poor, particularly in health. The impact and performance of user fees suggest that on their own they are unlikely to achieve equity, efﬁciency, or sustainability objectives in health. Evidence from our reference countries (e.g. Kenya and Thailand) and from other countries in Africa (e.g. Niger and Mauritania) suggests that user fees in health tend to work better if backed by prepayment insurance schemes, though these schemes are not inherently pro-poor (ibid.). Across the board subsidies in water are inequitable – big consumers beneﬁt most. This is consistent with the World Bank’s (2003b) observation, and in line with other studies of the incidence of public spending in health and education, which show that the beneﬁts accrue largely to the high and middle-income groups. The share going to the poorest 20 per cent (where spending can make a difference in improving

124 The Changing Role of Government

quality and access) is always less than 20 per cent. Our water sector studies showed that where equity concerns can be dealt with by scaled consumption charges or ‘lifeline provision’, then charging could have an equitable effect by removing regressive subsidies. In the health sector this is more difﬁcult and the effect of charging has been inequitable. The introduction and implementation of charging exposes weaknesses in government capacity. User charges were introduced in response to ﬁnancial crisis and, consequently, rushed without adequate preparation for implementation. The policy design, particularly in health, was ﬂawed by a narrow focus on raising revenue with inadequate attention paid to efﬁciency and equity issues (Larbi, 2003). Also the relevant public organizations did not have adequate organizational and management capacities to effectively manage the implementation of user charges and exemption systems. As Mills et al., (2001) point out, in the case of health, there was lack of a dedicated unit at the centre to oversee implementation and to produce and monitor national guidelines on exemption and how to use revenue. Charging also imposes demands on government capacity to design and manage exemption systems that work for the poor and vulnerable. It is also apparent that charging is a political issue and therefore its design and implementation need to take account of the political economy of each country. Some services (e.g. health and water) are politically more sensitive than others. For example cost recovery by user charges or withdrawal of subsidies in grain marketing are held back by continued government control and unwillingness due partly to the fear of backlash from the public and partly to a genuine concern to make services affordable for the poor. Taking account of the institutional context or the ‘rules of the game’ in play in each country or service sector, as well as developing the capacity of public organizations to design and manage user charges are important for effective policy implementation.

6 Working with Private Partners

The case for privatizing management The previous two chapters described reforms within public management. In some respects these have ‘imported’ market approaches and values into the public sector. In this chapter we examine the experience of ‘exporting’ functions to the private sector, and the impact this has on the functioning of public administration. Whether through the import of practices or the export of functions, these can be seen as different ways of addressing the principal-agent problem of public administration. That there is such a problem will be accepted for the purposes of this chapter, although it can well be counter-argued that this critique of public administration is less a description of reality and more a mobilizing device to generate impetus for reform (Salamon, 2002; Joshi and Moore, 2002). Lane (2000) represents the principal-agent problem and alternative routes to its reform on two dimensions as in Table 6.1. In the classical form of public administration, politicians are at the hub of two Table 6.1 From bureaucratic administration to charging, short-term contracts and the market Contracting on the supply side Long term Contracting on the demand side

Short term

Taxes

I

Bureau

II

Charges

III

Concessions, vouchers

IV

Source: Based on Lane (2000, p. 193).

125

Internal markets, agencies Firm

126 The Changing Role of Government

relationships. First, they receive the demands of citizens (their electorate) and, second, they fund bureaucracies through taxation to supply the response. Principal-agent problems arise in both relationships: citizens have weak control over politicians, and politicians have weak control over lengthy bureaucratic hierarchies. The public bureau (the supposed ‘agent’) may exercise power over the public and the politicians whom it should serve, based on its compulsory funding (taxation), its monopoly status, and the long-term contractual security of permanent employment. By contrast, it is argued that a ﬁrm is truly an agent because, in a competitive market, it is governed by shareholders’ search for proﬁt and by consumers’ choice and willingness to pay. Some of the management reforms described in Chapter 4 and 5 involved a shift from type I to type II in Table 6.1. In those cases, the public service remains the direct provider, paid for out of taxation, but, in the shift to type II, managers are made accountable for their performance within shorter-term contractual agreements instead of having long-term employment contracts within the public service. The approaches described in this chapter add the second dimension of reform identiﬁed by Lane. They seek to change also the taxversus-charges axis, towards types III and IV. They put the provider onto a more wholly marketized footing, with short-term contracts and payment for services rendered. However, this may stop short of full privatization: government may retain ownership of assets, the service may be charged to clients or government may itself ﬁnance private providers out of taxation. The important point is that accountability of providers to government or to citizens is made more direct; the World Bank (2003b) describes this as strengthening shorter routes of accountability. The cases of private participation referred to in this and the following chapters range from the complete privatization of the ownership of assets, through the contracting or licensing by government of private operators, to the private ﬁnancing of public services by investors. The term ‘private participation’ is used because in almost all cases there is a mixture of public and private roles. Usually, the private partner assumes the function of directly providing or producing the service while government keeps overall indirect responsibility for ensuring that the service is delivered adequately. Indirect roles include tasks such as analyzing policy options, setting standards and monitoring their enforcement, raising and allocating ﬁnance, managing budgets, contracting, regulating and creating incentives for private producers (see Chapter 1). Under these forms of provision, government retains

Working with Private Partners 127

responsibility for the protection of the public interest while service managers are freed from the constraints of public administration and exposed to the competitive pressures of the market. The goal orientation of public leaders, in principle, is to the satisfaction of the ‘public interest’, that is to the wider social interest rather than to private ends. Interpretations of what is in the public interest arise from a complex amalgam of institutionalized practices and political competing demands. Public management therefore involves a balancing of multiple objectives, including changing (and perhaps conﬂicting) policy goals, efﬁciency and equity in delivery, due process and legality in procedures of operation. Procedures are supposed to be open, transparent and subject to scrutiny. They have therefore to be based on institutionalized rules that govern ways of operating both internally in the management of staff and other agencies, and externally in relation to the public. Public organizations are also rulebased in relation to each other; they are supposed to coordinate under the authority of government, respecting each other’s spheres of responsibility (Lane, 2000; Du Gay, 2000). While these features of public administration are necessary to the core policy-making functions of government, they are widely thought not to be appropriate to the management of service delivery. They give no encouragement to operational ﬂexibility, few or no incentives for the achievement of results and efﬁcient use of resources, and rarely any clear basis of accountability for performance. They are also held to divert resources towards people with inﬂuence and away from the poor (Devarajan and Reinikka, 2002; World Bank, 1997 and 2003b). One response has been to clarify the roles of public service managers by giving them more deﬁned responsibilities, within semi-autonomous units but still within the public sector. Such reforms were analyzed in the previous chapters. Marketizing the delivery of public services is seen by its advocates as further freeing managers from the encumbrance of public service conditions (World Bank, 1997, p. 87). Within the framework of commercial contracts, goals and objectives can be speciﬁed, and managers freed to act ﬂexibly and entrepreneurially to achieve given ends, focusing on the satisfaction of consumers rather than obedience to rules of procedure. Principal-agent relations are clariﬁed through more direct and speciﬁc contractual relations between government (as principal) and independent operators (as agents). This does not remove the case for public administration but shifts it back from service delivery to the management of other deliverers: the

128 The Changing Role of Government

public interest, policy objectives, transparency and coordination are now to be achieved by contract, monitoring and regulation. After a decade where there has been a focus on ways of reducing the role of government in the economy, there is now recognition that a smaller role for government in the direct provision of services may mean a bigger role for government in policy development, coordination and regulation. (Bennett and Mills, 1998) Ideally, public-private relationships combine the virtues of both public administration and of private management. But, this can only be true if the public and private partners have the capacity to perform their new roles. The danger is that the efﬁciency gains from contracting the private or voluntary sectors may be outweighed by the additional transaction costs of doing business between multiple actors. Setting policy frameworks, coordinating, contracting and monitoring come at a price. If the price is higher than the gain, then there is a better case for the full integration of the service within either the public or private sectors. This and the following chapters examine how far such models of public service reform are being applied in developing countries, whether their application differs between sectors, what roles are implied for government, and how far they are constrained by the capacity of governments and their private partners.

Organizational arrangements for service provision Privatization in the sense of complete divestiture has applied mainly not to public services but to the ‘productive’ and competitive activities that were previously organized as state-owned enterprises. State ownership of commerce, industry, agricultural production and marketing was common but has been internationally largely undone, often under pressure of structural adjustment (see Chapter 3). However, even in these productive sectors, the dismantling of state ownership has often been accompanied by the emergence of new roles for the state in advising, promoting, enabling or regulating the non-government sector. The grounds for government’s continued involvement is that there is a residual public interest whether in developing market providers or in restraining the impact of the market. Some examples will be examined in Chapter 8. From the beginning of the 1990s, a further wave of reform began to affect developing countries – for the transfer of the delivery of

Working with Private Partners 129

social and infrastructural services to the private and community or NGO sectors (Odle, 1993; Adam et al., 1993; Cook and Kirkpatrick, 1995; Batley, 1996). The ﬁrst candidates included services such as street cleaning, waste collection, road maintenance, parks, hospital cleaning, catering, security guards, and public utilities such as water and power supply, airports and ports. Later candidates for liberalizing reform were the social services, particularly health and education. These were often difﬁcult cases to undertake since they frequently confronted political and ideological objections as well as being technically difﬁcult to put out to contract or to regulate. They often affected basic human needs or strategic national resources, involved large groups of public sector workers, were key spheres of political inﬂuence, and were undoubted areas of ‘market failure’ that required some level of state intervention (see Chapter 1, Box 1.1). Apart from a few countries such as the United Kingdom and Chile, the complete privatization of social and infrastructural services is rare. Much more common is some sort of contractual relationship where government retains ownership and control, while the private or NGO sectors take on operational roles subject to public control. Under these arrangements, private ﬁrms or NGOs are invited to tender for a contract or licence let by a government organization. The contract or licence may be small-scale and short-term for a speciﬁc activity ﬁnanced by government, or large-scale and long-term where the contractor ﬁnances and manages an entire government function (such as the delivery of water supplies). Recognition that there is not a straight and simple choice between public and private provision opens up the possibility of combining the two sectors in various organizational arrangements. These can be seen as alternative ways of allocating the risks and responsibilities of ownership, ﬁnancing, operation and maintenance between government and private or community operators (World Bank, 1994 and 1997; Savas, 1987 and 2000). Table 6.2 shows how ownership and the ﬁnancing and operating functions may be split between public and private or non-governmental sectors, and gives examples drawn from the research.1 The cells in light gray indicate the functions performed by the public sector; the more heavily shaded cells identify those functions where there are shared public and private responsibilities; and the white cells those where the private sector, NGOs or users take on roles. Under different organizational arrangements for service delivery, government and the private

Asset ownership Public

Public

Public

Public

Private to public

Public

Works/supplies contract (one-off)

Service contract (< 3 years)

Management contract (3–5 years)

Lease contract (affermage) (10–12 years)

Build-operatetransfer (BOT) (20–30 years)

Concession (20–30 years)

Private debt repaid by revenue

Private debt repaid by revenue

Public

Public

Public

Public

Capital ﬁnancing (investment)

Private from revenue

Private from revenue

Private from revenue

Public

Public

Public

Current ﬁnancing (operations)

Private

Private

Private

Public or private

Public

Public

Tariff collection

Distribution of functions in different organizational arrangements

Contract type (typical duration)

Table 6.2

Private

Private

Private

Private

Private for speciﬁc services

Public with private inputs

System operation

Grain mill

Water supply

Water company management

Hospital cleaning, water billing

Drug supply

Example from case studies

As lease, with Water supply, responsibility Food storage for investment

Major investment projects

Extended operation and maintenance contract

Management tasks over a short period

Speciﬁc technical task

One-off -e.g. building or supplies

Types of activity

130

Asset ownership

Public and private

Public and users

NGO/charity

Private

Joint venture (indeﬁnite)

Co-production

Licensed NGO or charity

Licensed private ﬁrm

Public and/ or users

Public and/ or private

Current ﬁnancing (operations)

Private

System operation

Types of activity

Private, maybe Private with public subsidy

Private

NGO/charity

Business and agricultural services

Example from case studies

Private Medical operation of practitioners services, under licence

Under contract Mission or licence, hospitals especially in social service

Public and/or Public and/or Local users users infrastructure and development

Public and/or Public and/or Infrastructure private private and development

Tariff collection

NGO/charity, NGO/charity, NGO/charity maybe with maybe with public subsidy public subsidy

Public and/or users

Public and/or private

Capital ﬁnancing (investment)

Distribution of functions in different organizational arrangements – continued

Contract type (typical duration)

Table 6.2

131

132 The Changing Role of Government

sector2 retain different degrees of responsibility for the ownership of assets, for ﬁnancing investment and operations and for managing operations. The main possible organizational arrangements for service delivery are described more fully in Box 6.1. On a scale of declining government control, they range from the pure public management of services, through various levels of contracting of functions to the private sector, through the licensing and regulation of private activity, to informal arrangements of joint venture and collaboration. With regard

Box 6.1

Public-private organizational arrangements

Pure public service delivery – Government owns, ﬁnances and directly delivers services. Particular oneoff works or supplies may be purchased from contractors. Public ownership and ﬁnance with limited private operation – Government retains ownership of and responsibility for service provision but ﬁnances private providers to give particular support services through management or service contracts. Public ownership, some private ﬁnance and private operation Under these arrangements, government passes responsibility to the private sector for the production and delivery of the service and has ultimate ownership of the assets, with the following differences – Lease: the contractor ﬁnances the operation of the service from revenue, whilst government covers investment costs – Concession: the contractor ﬁnances the operation of the service and investment costs from revenue – Build-operate-transfer: the contractor ﬁnances the operation of the service and investment costs from revenue, and retains ownership until the end of the contract. Private ownership, ﬁnancing and operation under government regulation or support – Provision and ﬁnancing of a monopoly service by private owner/operators under licence, with regulation by government, – Competition between government-licensed private producers or deliverers of the service, – Government ﬁnancial support of private consumption and provision, through, for example, subsidies, vouchers or loans. Co-production – Joint venture: Formal partnership between independent public and private providers through joint ownership, joint ventures and investment, – Informal understanding between organized groups of citizens/clients and state agencies to make resource contributions to the joint production of a service.

Working with Private Partners 133

to the state’s role across this range of possible arrangements, there are four main ways in which it may be involved: 1. The state directly provides services, owning, operating and ﬁnancing operations without private involvement. 2. The state purchases the services of contractors. 3. The state licenses and regulates private, NGO or community providers. 4. The state collaborates with private, NGO or community providers through formal or informal joint arrangements. This and the next two chapters are concerned primarily with 2 and 3, where government seeks to exercise its responsibility indirectly but retaining authority. This chapter sets out the case for the alternative arrangements, considering how the nature of services and the capacity of actors may inﬂuence choice between them. The following two chapters look at the experience of contracting and regulating nongovernment service providers in the four research sectors.

The application of public-private arrangements for service delivery Before looking at real cases of contracting in the next chapter, this section indicates how and why, in principle, alternative organizational arrangements may be appropriate to different public services. It adopts an essentially technical approach, suggesting a correspondence between the characteristics of services and how they might be organized. It focuses on four broad categories of organizational arrangement – contracting out, lease and concession, licence, and joint venture or partnership. Contracting out Within the framework of public ownership, there is the contracting of services paid for by the public sector. This may be through one-off contracts, or ﬁxed period service or management contracts. Being relatively short-term and speciﬁc, they safeguard the interest of the public authority, which retains its control over the delivery of the service. These arrangements are therefore likely to be appropriate where there is a strong public interest and government does not wish to cede its responsibility because it has little conﬁdence in its ability to handle long-term contracts. This may apply, for example, in situations of

134 The Changing Role of Government

monopoly, or where there are strong equity considerations, or where there are important effects not only on direct consumers but also on wider society (see Box 1.1, Chapter 1). On the other hand, piecemeal contracts may have the disadvantage of creating problems of fragmentation and difﬁculties of coordination. These arrangements are therefore most easily manageable where tasks are naturally divisible and discrete: for example, construction and maintenance works, neighbourhood services, refuse collection, ancillary services in the health sector, and aspects of internal administration such as billing, salary payments and vehicle maintenance. Lease and concession of monopolies Lease and concession contracts are longer-term arrangements that allocate managerial and ﬁnancial responsibility to the private sector. In leases, the contractor covers running costs from revenue; in concessions, it must also ﬁnance investment in ﬁxed plant. They are ways of managing natural monopolies whilst avoiding the concentration of power in either the public or private sectors. They achieve this end by separating the direct and indirect aspects of provision, subjecting the private producer to public contractual control. They are most clearly applicable to non-public goods with a monopoly tendency and where there is sufﬁcient public interest (on grounds of welfare or external effects) to justify continued government ownership. Examples are the public utilities (water, electricity, gas) and collective services such as refuse incineration, bus terminals and municipal market management. Build-operate-transfer schemes can be seen as a variation of lease and concession contracts, where the primary purpose is to achieve the construction of a particular capital investment project, rather than to manage the system. The investor designs, ﬁnances and builds an asset – for example, a highway or a water production and transmission system. The investor then operates the asset for typically 25–30 years, in return for payments made by government, or sometimes by the users; after that period the asset is transferred to the government body. Licensed competition between producers Licensed or regulated competition between producers is technically appropriate to ‘private goods’, that is those which it is possible to charge for and which individuals compete to consume. The argument for public intervention is not so strong as to demand direct public sector entry into the production or delivery of the service. In the absence of monopoly, the case for government intervention might rest

Working with Private Partners 135

either on the need to promote equitable access or to mitigate the effects of unrestrained competition on society at large. An example of the ﬁrst sort would be the management of bus transport to ensure that an entire city is covered at similar tariffs; of the second, that market stall-holders fulﬁll basic health requirements or that builders comply with building regulations. There are many cases where government licenses private providers but also keeps a stake as a direct provider itself. This applies particularly in the case of ‘merit goods’, where there is a policy commitment to equitable access and government seeks to guarantee this by itself offering the service. Basic education and health are common cases of this sort. In other services, raising less profound equity issues, governments sometimes use licensing as a step towards reducing their direct role in service delivery. Joint venture or partnership Several sorts of joint arrangement can be identiﬁed where state agencies and private bodies act in a mutual endeavour, with or without joint ownership. Public-private partnership in joint ownership or investment schemes is most likely to occur where there is a strong possibility that opportunities for private investors will be generated by government involvement. The public sector’s contribution might be either to undertake necessary investments which private ﬁrms are unable to perform (due to their large scale, high risk, or difﬁculty of charging to consumers) or to facilitate private action (by covering risk or using the coercive powers of the state). Typical circumstances would be the acquisition of land and the installation of infrastructure for social housing and commercial development. Joint venture companies are at the formal end of the spectrum. Government enters into contractual relationships with the private sector both in setting up the company and in awarding the joint company the contract to undertake the work. Unlike pure contracting out or concession arrangements, there is the complication that government participates on both sides of the contract, leading to possible conﬂicts of interest: government is there to ensure the public interest but also has an interest in the contracting companies’ gains. Apart from the collaboration of public and private sector investors, another possibility is that the beneﬁciaries of public services may participate in their production and delivery. Again the possibilities of user participation range from the more formal and contractual to the more informal and casual. Cooperatives formalize the ownership by the

136 The Changing Role of Government

beneﬁciaries of the organization that supplies the service, as in the case of the water systems in Santa Cruz, Bolivia (Nickson and Vargas, 2002). More common are looser arrangements where users collectively contribute to the management of a service in some form of institutionalized agreement that is not legally binding or formalized. The term ‘co-production’ has been coined to deﬁne these looser arrangements where the lines between state agencies and non-state actors are blurred (Ostrom, 1997; Joshi and Moore, 2003).

Matching organizational arrangements to capacity There are risks and costs to be set against the possible advantages of public-private partnerships. Successful contracting depends upon there being genuine competition between competent ﬁrms, social control to guard against nepotism, and governmental capacity to specify and enforce contracts. Where the conditions exist, contractual arrangements may make for efﬁciency; but the conditions are exacting. Weak public administrations may lack the capacity to undertake the new (indirect) roles which follow from contractual arrangements: setting broad frameworks of policy, managing contracts, regulating providers, ﬁnancing and supporting providers and consumers. If they fail to do so, particularly in the case of monopoly services, the situation may be worse than under direct public provision. The transaction costs of making and monitoring contracts may not be justiﬁed by the possible efﬁciency gains in service provision. Factors to consider in assessing governmental capacity to manage contractual relations with private partners can be summarized as • The characteristics of the service, • The complexity of alternative contractual relationships, • The capacity of government and the contractors to manage the relationship, • The institutional environment. These factors inter-relate as Chapter 7 will show. Different types of service are more or less suited to alternative contractual arrangements that require different degrees of capacity on the part of government and contractors. Moreover, capacities have to be deployed within an institutional environment that may either support or limit the conditions for effective public-private working. Institutional factors include the stability of the economy and the political system, the functioning

Working with Private Partners 137

of the judicial and ﬁnancial systems, and the relationships between government and the business sector. As the World Bank (1994 and 1997) implicitly suggests, there is an uncomfortable conundrum: precisely in those countries where the public provision of services has failed, the conditions for effective private participation are likely also to be absent. The Bank’s answer is to select for private involvement ﬁrst those services that present the least problem for contracting whilst also building government’s capacity to manage more complex cases (World Bank, 1997, p. 97). The effect of service characteristics on the ease of contracting Chapter 3 described some of the characteristics of service sectors and how these may affect the balance of power between principals and agents, or, in this case, governments and contractors. Among these, Williamson (1987) has described three factors that determine the ease of contracting: • The degree to which the service can be speciﬁed in advance: Is it possible to adequately specify the requirements to which the contractor is subject? This is more difﬁcult in the case of services where the objectives are multiple, long-term, non-measurable or qualitative – for example social or health services; • The degree of asset speciﬁcity: Are the assets (equipment, knowledge) required by the contractor so speciﬁc to the contract that choice is limited and the winning contractor can hold government to ransom? Large scale infrastructure investments (for example in the water sector) or highly professional or technological services in the health sector present this sort of problem; • The practicality of gathering and measuring information about performance: Is the contractor’s performance measurable? Does government have the capacity to gather the information that it needs in order to monitor contract compliance? Information asymmetry in favour of the private provider may lead to ‘regulatory capture’, where the provider effectively controls the public regulator. Information gathering and assessment may be a general problem for a weakly organized government but is particularly likely to arise in the case of social (including health) and professional advisory services where goals are difﬁcult to quantify and measure. The greater the difﬁculty of specifying and measuring performance, the more the principal will have difﬁculty in controlling the agent

138 The Changing Role of Government

according to simple contractual rules. The argument then grows for a relationship between principal and agent that is more based on understanding than on speciﬁed contractual commitments. On the other hand, if the contract becomes too loose, it will reduce the pressures on the contracting agent to act efﬁciently. The effect of contract types Different types of contract present different sorts of cost or capacity constraint. In principle, the risks to the contracting body and the contractor rise with the growing duration and complexity of contracts (Williamson, 1975). Short-term, simple contracts where the object of exchange is discrete, present relatively low risk to the principal (the contracting agency) and to the agent (the contractor) who can more easily agree about the terms and outputs they expect, without concern that conditions and requirements may change over time. Short-term, service or management contracts for speciﬁc inputs are more likely to be speciﬁable and measurable. On the other hand, they present an opposite problem: a total service would have to be composed of many sub-contracts to specialist companies. This presents problems of high transaction costs in designing contracts, and coordinating and monitoring contractors. Longer term and more complex arrangements, such as concessions, have the advantage of wrapping all aspects of the service into one contract over a long period, potentially cutting transaction costs. On the other hand, the design of such contracts requires a great deal of information and experience to anticipate the relationships between all the elements of the service and to try to anticipate all the possible risks and uncertainties that may occur during the term of the contract. Since no written agreement can foresee everything that may happen over a 20–30 year period, long-term contracts are likely to be ‘relational’ – based on sufﬁcient mutual trust and understanding to allow for adjustment to circumstances (Walsh, 1995, p. 136). Sako (1992) describes these as ‘obligational contractual relationships’ by comparison with the ‘adversarial contractual relationships’ that try to specify all possible eventualities in legal terms on the assumption that the partners seek to outwit each other (Flynn, 1997). Trust may be injected into a contractual relationship by resorting to a third party, an independent regulator, in whom the partners both have conﬁdence. In such a mediated relationship, the regulator should fairly recognize both the goals of the contracting body and the requirements of the contractor to remain viable.

Working with Private Partners 139

Particularly in the context of developing countries, all these issues are in question: the existence of trust between partners, the ability to handle complex contract design and management, and the conditions for disinterested regulation. The capacity of partners Apart from the characteristics of the service and the complexity of different contract types, there is also the capacity of the partners to manage contractual relationships between public and private partners. On the side of the private contractors the following questions arise: • Are there alternative possible providers with adequate capacity to undertake service provision? Are there enough possible contractors to provoke competition? Can contracts be speciﬁed (broken down) in such a way as to maximize opportunities for small and large contractors to compete? On the side of government: • Does it have the skills and organization to make and monitor contracts? • Are the necessary instruments for contract enforcement available, for example by legal compliance or by the threat of affecting contractors’ reputation by divulging information on their performance? Can government exercise them? Where contracting is undertaken in the absence of the necessary conditions, the danger and likelihood is the emergence of inefﬁcient and even corrupt relationships. Two studies of contracting out in Kumasi and Accra in Ghana illustrate the problem, in what might be thought to be the simplest of services – public toilet management. Even in these very small-value contracts and even where the conditions exist for competition, political practices have encouraged a lack of transparency in contract allocation and refusal by contractors to pay revenues to the municipality (King, Inkoom and Abrampah, 2000; Crook, 2002). Chapter 7 will explore this question further. Experience of private sector involvement in the case countries So far, this chapter has assumed a rational planning model in which decisions about alternative organizational arrangements are made on the basis of service characteristics and the capacities of governments

140 The Changing Role of Government

and contractors. This was a way of describing the alternatives and of identifying the factors that allow them to function. However, as Chapter 3 showed, history, institutions and vested interests are more important than rational planning in explaining reforms, particularly in the conditions of developing countries. As a prelude to the following chapters, which look at speciﬁc experiences of contracting and regulating, this section outlines the broad patterns of liberalizing reform and private sector involvement that have taken place in the research countries. The countries include Ghana, Zimbabwe, India, Sri Lanka and several ‘reference countries’ in Africa, Asia and Latin America, and the focus is the four service sectors that were studied. Business support The fullest government withdrawal from direct provision is in the industrial sector, achieved through a combination of divestiture, privatization and closure of unproﬁtable state enterprises. The textiles and garments industries were selected for study because, in most countries, they were among the ﬁrst to develop under state protection and the ﬁrst to be privatized. Governments have now almost wholly shifted from the role of direct production of yarns, fabrics and garments and very largely also from regulatory and licensing roles. From state production and control of the private sector, governments have shifted towards attempting to deliver support services to the private sector – engendering a favourable policy framework, market analysis and promotion. Having dismantled its own import licensing function, the Ghanaian Ministry of Industry and Commerce has tried to adjust to giving technical advice to ﬁrms it had previously regulated. In Zimbabwe, the Ministry of Finance together with donors opted to create a semiautonomous Zimbabwe Investment Centre out of parts of ministries. Completely autonomous support organizations (for export promotion and information support) were set up with their own or donor funding in Ghana and Zimbabwe. In India, among a vast number of traderelated and semi-autonomous industrial research and training institutions, some gained a new lease of life as the government released its controlling grip. From the end of the 1970s, Sri Lanka created a series of agencies to promote inward investment, developing a signiﬁcant and internationally linked textile sector almost from nothing. The research examined the operation of these agencies intended to promote and facilitate private ﬁrms in the four countries – these are described in Chapter 8.

Working with Private Partners 141

Agricultural marketing The policy goal in most countries is to liberalize agricultural trade and promote a market economy. Progress has been relatively fast in external trade under pressure of international competition – for example, in meat and cotton marketing in Zimbabwe and cocoa purchase and export in Ghana. However, domestic trade in staple grains has often remained protected and controlled by governments, particularly in India and Sri Lanka. The growth of private trading and private services to farmers has often occurred, not as a result of policy, but as the private sector ﬁlls gaps left by the failure of government schemes. Government bodies have tended to cede ground to the private sector through their decline rather than by the planned restructuring of public intervention. Apart from the unplanned growth of private services, two types of relationship with the private sector were found. First, particularly in the export trade, governments have sought to strengthen their support roles to traders and farmers, for example by regulating the quality of cocoa exports in Ghana. Second, they have contracted the private sector to provide speciﬁc inputs to government-managed services, for example grain storage in Sri Lanka and India, maize exporting and transportation in Kenya. Government bodies have frequently tried to hold on to their support roles in agricultural marketing, contracting inputs from the private sector, rather than putting entire services out to the market. Urban water supply Internationally, urban drinking water is a case of radical reform. Private ﬁrms are increasingly being brought into the ﬁnancing, construction and operation of water systems through short and long-term contracts. However, reform in the four core countries was more limited, with the partial exception of Ghana where there were proposals for establishing lease arrangements for urban water supply. In India, Sri Lanka and Zimbabwe, reform mainly involved the changes within the public sector that were examined in Chapters 4 and 5: increasing the autonomy of state agencies, decentralizing water supply to local government, strengthening management, operating on a commercial basis and establishing new approaches to regulation. The pressure on stateowned utilities to operate more efﬁciently had led some to contract out speciﬁc functions to the private sector – for example, meter-reading, billing and managing pumping stations. The water sector offers a range of types of contract – service and management contracts, lease, concession, and build-operate-transfer – that

142 The Changing Role of Government

will be analyzed in Chapter 7. But, the research had to go beyond the four core countries to examine the more radical cases of privatized management, in Argentina, Chile, Malaysia, the Philippines, South Africa, Trinidad and Tobago. The major research question is whether such reforms can operate in poorer countries. Health care The broad structure of curative health services as between the public and private sectors remains largely unchanged by reform in the four core countries. The private sector is dominant in primary and informal care and is, at best, weakly regulated, while the public sector is dominant in hospital and formal provision. The main reforms again largely focus on re-structuring within the public sector – decentralization from and reorganization of ministries of health – rather than in changing its role. With regard to private providers, policy statements declare the intention to improve regulatory frameworks, to support private sector development and to contract out services but are weakly implemented in most cases. On paper, there has been more radical reform in the African than in the Asian cases with proposals for the contracting out of support and clinical services. Subsidies to non-proﬁt (church) hospitals were also common in Africa. However, in practice, the Asian countries, including Thailand, presented more examples of successful contracting out. These appeared to be less rooted in (donor-led) ideology and more in pragmatic decisions than in the African countries. Contracting is much more often of support services than of the clinical aspects of health as Chapter 7 will show.

Conclusion Public administered services face two principal-agent problems: citizens have weak control over policy-makers, and policy-makers have weak control over lengthy bureaucratic hierarchies. Theory suggests that introducing the private sector into the delivery of public services can help to overcome these problems. This may be by full privatization, or government may retain ownership of assets, and itself contract and ﬁnance private providers out of taxation. By putting the provider (the agent) onto a marketized footing, under contract and with payment for services rendered, the citizen and policy-maker (the principals) are given more control over the performance of agents. Within the framework of a contract, goals and objectives can be speciﬁed, and managers are freed to act ﬂexibly and entrepreneurially to achieve given ends.

Working with Private Partners 143

Full privatization, in the sense of complete divestiture, has applied mainly in the productive sectors – industry, agriculture and commerce. The complete privatization of social and infrastructural services is rare. Much more common is some sort of contractual relationship where government retains ownership and control, while private ﬁrms or nongovernmental organizations deliver all or part of the service. Under different contractual arrangements, government and the private sector retain different degrees of responsibility for the ownership of assets, for ﬁnancing investment and operations and for managing operations. There are technical grounds why different arrangements may be suitable to the provision of different services: contracting out where tasks are naturally divisible and discrete; lease and concession where services have a monopolistic tendency; licensing where government needs to promote equitable access or to mitigate the effects of unrestrained competition; and joint ventures where there are complementary interests between the public and private sectors. However, the judgement about whether and how to contract depends also on the capacity of government and private partners to handle alternative contractual relationships. Services where it is difﬁcult to specify the desired outputs (for example, because they are qualitative) and to gather information to measure their effects are intrinsically the more difﬁcult to contract. Short-term contracts may be easier to agree and to monitor and less risky than long-term concessions but, on the other hand, they may impose higher management costs on government. Governments may or may not have the skills and organizational capacity to design and implement contracts, and the private sector to compete for them. The capacity of the partners will in its turn be inﬂuenced by the context in which they operate – for example, the degree of political and economic stability and respect for the rule of law. The real decisions of government to engage in one or the other form of partnership with private ﬁrms depend on the history and institutions of particular countries and of particular service sectors within countries. This chapter has described the broad pattern of liberalization in the research countries. The industrial sector and external agricultural trade are generally the most fully liberalized, while formal urban water supply and health care remain more resistant to change. However, in all cases there is a tendency to move towards a stronger involvement of private service providers. Chapter 7 goes on to examine the experience of contracting out services to the private sector, while Chapter 8 examines how governments regulate and enable private providers.

7 The Experience of Contracting

Introduction This chapter explores evidence from the country cases studies about the experience of contracting across the service sectors.1 Table 7.1 classiﬁes the forms of contractual arrangement examined in the text. Down the left-hand column, contractual types proceed from shorterterm and simpler forms to longer and more complex forms, as described in Chapter 6. It can be seen that the business sector is almost absent from the table because here privatization has been most complete. The private sector operates independently rather than under contract, and governments have removed themselves to a support role – as will be described in Chapter 8. The chapter begins by describing how far the different contractual types are used in each of the services – business support, agricultural marketing, urban water supply and health care. The next section provides the raw material on which the rest of the chapter is based. It describes the forms of contract that have been adopted, the policy trends and the experiences of governments in implementing different approaches. The following section compares experience across the different service sectors. What has been the experience of implementation? Do governments gain by shedding risks and costs onto contractors? Does contracting lead to better service performance, as the theory would suggest? Are governments prepared to contract out some sorts of activity more than others and, if so, why? Are governments more prepared to adopt some forms of contract than others and, if so, why? Underlying these last two questions is whether governments confront feasibility problems in contracting out and, in particular, whether they face limits of their own capacity. The chapter then assesses the capacity 144

The Experience of Contracting 145 Table 7.1

Contract types and examples mentioned in the text

Contract type

Business support

Agricultural marketing

Water supply

Health care

Works/supplies ‘spot’ contracts (one-off)

Maize export and transportation – Kenya

Transport and tankers – India

Construction, equipment maintenance, drug supply – general

Service and management contracts (< 5 years)

Grain storage – India and Sri Lanka

Sewerage treatment and pumping plants – India. Water utility – Chile

Cleaning, catering, security – India, Sri Lanka, Thailand. Diagnostic services – India, Thailand. Clinical services – South Africa.

Lease (10–12 years)

Water utilities – Ghana (proposal)

Build-operatetransfer (BOT) (20–30 years)

Wheat mill – Sri Lanka

Concession (20–30 years) Joint venture (indeﬁnite)

Water supply – Malaysia Water utility – Argentina

Zimtrade – Zimbabwe

Water utility – Colombia

Church and mining company facilities – Ghana, Zimbabwe

and institutional constraints presented by contracting within and across the different service sectors and countries.

The use of contractual arrangements One-off works or supplies contracts The contracting by governments of speciﬁc, short-term inputs is long-established practice. Building or engineering works, supply of

146 The Changing Role of Government

materials, specialized professional services are the sorts of inputs that government departments have routinely contracted on a one-off basis, as and when the need occurs. Apart from the development of longerterm contractual arrangements, there is also a tendency across all sectors to extend the range of speciﬁc inputs that can be put out to tender and that are not at the heart of the governmental operation. In general, it is the generic, support services – cleaning, catering, transport, security and maintenance – that remain more likely to be put out to contract than professional and managerial functions, perhaps because of the greater capacity for senior staff to resist. However, there are exceptions. Agricultural marketing From 1993, the Kenyan Government began to liberalize external trade in maize. It ﬁrst allowed private ﬁrms to import maize and, from 1995, required the National Cereals and Produce Board to put maize exporting out to tender. The NCPB had previously managed the export of maize that was surplus to domestic requirements. The ﬁve international commodity dealers that won contracts in June 1995 bought negotiated amounts of grain and took responsibility for shipping and export. Separate contracts were awarded for the transport of the maize to depots and for the inspection of the quality of the grain. NCPB manages the tendering process as each lot of grain is released, ensures the grain is delivered, and undertakes fumigation. Although there were some problems in contract management, NCPB and government ministries responsible for monitoring demonstrated that they had the capacity to handle such contracts. The arrangement has brought beneﬁts in terms of reduced costs of export; more rapid payment to the food chain, including farmers, truckers, ports authorities and NCPB; fewer delays; and the passing of the risk of price ﬂuctuations to the contractors. From the point of view of the contractors, the main problems were uncertainties in the broader policy environment, including the unpredictability of taxation on imports, and poor transport infrastructure (Lewa, 1998). Urban water supply The contracting out of engineering works and the laying of pipe-lines has been long-established practice in the management of public water companies everywhere. For example, in Bulawayo, Zimbabwe, all but minor works are carried out by contractors. In several of the core countries, there were examples of the extension of these practices into the routine operations of public companies, as a ﬁrst move towards the commercialization of their operations. For example, Chennai

The Experience of Contracting 147

Metropolitan Water Board was one of the few Indian water boards to cover its operating costs, aided by contracting out ancillary services, for example by hiring staff cars and water tankers. Also in India, the Hyderabad Metropolitan Water Supply and Sewerage Board contracted out bill delivery and cash collection to the private sector. Some of these arrangements shade off into medium-term service and management contracts and will be dealt with under that heading. Health care Contractual relationships are not new in the health sector: goods and services have always been purchased from private providers. Based on the case of Ghana, Larbi (1998c) has classiﬁed the types of activity that, traditionally, have been contracted out as follows: • • • •

Feasibility studies for new projects; Construction and major maintenance of new buildings; Supply, installation and maintenance of capital equipment; Supplies, especially of drugs.

Typically ‘contracts’ have consisted of one-off purchases in the market by central tender boards or ministries, by districts or hospitals, rather than formal contractual arrangements. However, there is evidence from the case study countries that, underlying these one-off purchases is a sustained relationship based on repeat purchases. Informally, relationships with suppliers are often quite persistent, given the small pool of ﬁrms that can be drawn upon in most developing countries. There have been attempts to extend the range of activities that are contracted out, particularly in the African countries that have been most exposed to donor pressure. However, contracting continues to apply largely to support services as opposed to clinical services. Even in the area of support services, the extension is mainly in the area of non-core ancillary activities rather than in core functions that may threaten the employment of the more qualiﬁed and organized workforce. Even for ancillary staff, all countries showed great reluctance to make staff redundant, preferring to apply contracting to new services that had not previously been provided. Service and management contracts Agricultural marketing In the agricultural sector, India and Sri Lanka both scaled down their direct management of strategic grain reserves used as a buffer against

148 The Changing Role of Government

food insecurity. They moved towards the contracting out of grain storage, but they did it in different ways. Sri Lanka was much the more radical, completely removing the Food Commissioner’s Department (FCD) from its previous monopoly role in the importation of rice, sugar and wheat. Within this liberalized market, it was nevertheless felt necessary to retain a basic reserve of rice in government-owned bonded warehouses. Since 1990, 11 private companies (only one nationally owned) have been contracted to import rice and hold it in FCD stores without paying duty until the rice was sold. Their obligations are to pay the FCD for the use of the stores, manage the stock, and maintain a minimum amount as a buffer against need. They may only sell surplus at prices above a minimum set by FCD or re-export it with permission. The arrangement works well from the government’s point of view. The greater risk appears to be to the private companies: external factors can easily threaten the sustainability of the arrangement. First, as Sri Lanka attains self-sufﬁciency in rice production import requirements will fall. Second, the proﬁtability of rice imports can also easily be threatened by the Government of Sri Lanka’s periodic decisions to subsidize imported wheat (Smith and Ellis, 1997). India’s is a much more limited withdrawal from government responsibility for maintaining large stocks of grain. The Food Corporation of India continues to directly purchase rice and wheat and to store them. However, in the face of criticism of the high cost and wastage of its own stores, FCI has begun to enter into long-lease contracts with public and private bodies for the storage and management of stocks in the contractors’ stores. Contractors are awarded 10-year leases whether or not their facilities are used. However, the contracting process is not competitive, and there are built-in biases in favour of public contractors – the central and state warehousing corporations. There are low incentives for private investors to invest in new storage facilities when they would have to compete with subsidized public sector competitors, especially when the latter are represented on the FCI’s board (Kohli and Smith, 1998). Urban water supply There were few cases among the four focus countries (Ghana, Zimbabwe, India and Sri Lanka) of longer term and more complex contractual arrangements getting off the ground. One example in India is the Hyderabad Metropolitan Water Supply and Sewerage Board, which has introduced private sector management of its sewerage treatment plant. Chennai Metropolitan Water Board had gone further, since 1993

The Experience of Contracting 149

contracting out the operation of 14 sewage pumping plants, in order to bypass the poor performance and absenteeism of its public sector staff. Small ﬁrms were able to take on these opportunities with minimal capital outlay, since they received cash payments up front. Savings of 20 per cent were reported (Franceys, 1995). Going beyond the focus countries, the case of Santiago, Chile illustrates successful contract management but, at the same time, the extreme complexity of managing virtually an entire water company on the basis of piecemeal contracts. Santiago’s public water company juggled multiple service and management contracts to achieve a very high level of efﬁciency (Box 7.1). Nevertheless, it was fully privatized in 1999, the entire infrastructure and assets being sold off to a private company regulated under a government licence. Health care In the Asian countries studied – India, Sri Lanka and Thailand – there have been long-standing practices of contracting non-clinical support services, in ﬁelds such as catering, laundry, cleaning, security, equipment and grounds maintenance. Generally, the decision to contract had not been driven so much by policy reform agendas as by pragmatic decisions made incrementally (Bennett and Mills, 1998). However, in some cases, there were broad governmental attempts to promote contracting out that affected the health sector. The Thai Government in 1994 adopted a Cabinet resolution to limit the civil service by systematizing existing practices of contracting out. In the same year, Zimbabwe’s Public Service Commission instructed the Ministry of Health to contract out all the support services mentioned above, but here the effect of this abrupt step was to over-load the capacity of staff leading to a very slow pace of implementation. Non-clinical support services are commonly contracted out for one year. In principle, contracts are awarded competitively, but in practice, in all the countries the number of ﬁrms that actually bid is small, even where, as in India and Thailand, there is a large number of potential bidders. Possible explanations include collusion between ﬁrms to restrict competition, the non-transparent award of contracts, and low prices ﬁxed by government against which contractors can bid (in Thailand and Mumbai, India). The research concluded that, where there were gains from contracting out, these were likely to come less from competition between suppliers, and more from the effective speciﬁcation and monitoring of contracts (Mills et al., 2002).

150 The Changing Role of Government

Box 7.1

A service contract – an example from Santiago, Chile

There is a long tradition of contracting-out water services in Chile. Even prior to its privatization in 1999, EMOS, the public water company of Santiago, had one of the highest levels of labour productivity among Latin American water companies. A large part of its operations were contracted out. EMOS retained only the functions that it considered strategic or which it carried out more efﬁciently than the private sector. Virtually all other operations were contracted out. These included: •

•

•

•

• •

• •

•

•

long-term planning. Although tariff planning was carried out internally, the studies to (determine the cost of capital were contracted out to private specialists. feasibility studies for new projects. EMOS contracted outside consultants for most of its new project work and companies were invited to tender for the work. construction contracts. EMOS contracted out all of its new infrastructure works as well as improvements and repairs to existing works, normally through international competitive tendering. maintenance contracts. Permanent contracts were used for maintenance and renovation of the network and pumping-stations. Temporary contracts were used for rehabilitation of infrastructure, maintenance of electrical and mechanical equipment, civil works, and site maintenance. Three contractors were awarded the 2–3 year contracts for maintenance of the distribution network, each with responsibility for a geographical section of the EMOS jurisdiction. Each contractor supplied the equipment necessary to carry out a wide range of up to 700 tasks, for each of which there was a ﬁxed price. EMOS inspectors supervised the performance of these contractors. quality control contracts. Laboratory analysis of water samples was carried out by contractors and carried out in the laboratories of EMOS. information technology contracts. The computerized billing system was fully integrated with a system of automatic meter reading. The operation and maintenance of this system was contracted out on an annual basis. ﬁnancial contracts. Most billing, transfer of cash, and payment to staff and contractors was contracted out. administration contracts. Three-quarters of the vehicle ﬂeet was contracted out through an open register of approved drivers. Cleaning, photocopying, and ofﬁce maintenance services for all EMOS installations were contracted out as well as its catering, training and ﬁre safety services. commercial contracts. The following commercial services were contracted out: meter reading, distribution of bills and receipts, repair of water meters, installation and replacement of water meters, disconnection and reinstallation of water supply, calibration of water meters, inspection and collection from business premises, and debt-collection. EMOS required commercial contractors to guarantee minimum rates of pay and conditions for their staff. It also enforced quality standards on its contractors, requiring that all meter repairers should hold a relevant technical qualiﬁcation. industrial relations contracts. EMOS contracted consultants to introduce the following innovations: performance evaluation, incentive systems, study of future management needs, and quality circles.

Source: Based on Nickson and Franceys, 2003.

The Experience of Contracting 151

However, contract speciﬁcation was usually deﬁcient, providing a poor basis for assessing performance, and monitoring was rarely effectively undertaken (Box 7.2). Among support services, formal contracts are more effectively made and monitored in relation to services that are more easily speciﬁable: for example, security and the maintenance of grounds and equipment. ‘The key difﬁculty was where quality standards were needed as in the case of catering’ (Mills et al., 2001). Although service contracts in the sphere of clinical services are much fewer, they also focus on services that can quite easily be speciﬁed and monitored. In Tamil Nadu, India, and in Thailand, the use of high technology diagnostic services is contracted by hospitals from private ﬁrms. However, [F]or many clinical services it is not possible to specify in advance the service required, in which case the ease of monitoring performance is critical…In reality, however, for most clinical services in developing countries it would seem that adequate information systems to monitor contract performance do not exist. (Bennett and Mills, 1998, p. 323) Problems arising from the contracting of qualitative clinical services are illustrated in Box 7.2. This case of contracting for the services of private GPs in South Africa shows how contracts that appear to be formal may in practice become relational in nature if terms are not or cannot be speciﬁed and monitored. That is, they depend on a relationship of trust and on the desire to retain reputation between contractor and contracting agency, rather than on the setting and policing of contract speciﬁcations. But trust is a scarce commodity. Build-operate-transfer (BOT) Agricultural marketing Wheat is not produced in Sri Lanka but is imported, processed and distributed under a state monopoly. The state monopoly remains but its operation has been almost wholly contracted out, partly under pressure from USAID, which until the mid-1990s was the principal contributor of food aid. The Cooperative Wholesale Establishment (CWE) imports the wheat, contracts out the milling, contracts the government’s Food Commissioner’s Department to store the wheat-ﬂour, and contracts transporters to distribute the ﬂour to around 8000 ‘multi-purpose cooperative societies’ around the country.

152 The Changing Role of Government

Box 7.2

Contracting for the services of private GPs in South Africa

In South Africa, Provincial Departments of Health contract with private GPs to provide clinical and medico-legal services to patients within a deﬁned geographic area. Most of this work is the provision of curative primary care services, usually in a separate part of the GP’s private surgery. A study examined contractual relationships in eleven private GP practices in two provinces, Western Cape and Eastern Cape. The contracts are based on a fee-for-service payment mechanism governed by a standard charge for a list of services. This is recognized by both Provinces and GPs to result in high volumes of patients seen quickly in order to maximize hourly income. There is no speciﬁcation of quality, or of treatment protocols. The contracts are monitored by monthly reports submitted by each GP detailing number of patients seen, diagnosis and drugs dispensed. The principal sanction speciﬁed in the contract is termination by the purchaser in case of ‘breach of a material condition’, ‘serious negligence or default in performing a duty’ or ‘misconduct calculated to discredit the medical profession’ but these are not further described or ever resorted to. On the providers’ side there is also a lack of effective sanction despite occurrences of late or insufﬁcient payment. Few GPs can produce a copy of the contract. Monitoring by the Provinces is patchy. Where it occurs, it focuses on volumes of services and does not address issues of quality. The researchers concluded that the contracts were ‘relational’ in nature – relying less on formal written contract provisions and more on a continuing working relationship between GPs and provinces. However relational contracts are generally understood to be underpinned by mutual trust. In this case, trust was argued to be lacking to a signiﬁcant degree both because the history of the relationship extended to the apartheid era and because some GPs had been found guilty of fraud in both provinces. There were therefore signiﬁcant contractual difﬁculties, and dissatisfaction with the contract on both sides, but no clear alternatives to ensure primary care provision to this population. Source: Based on Palmer and Mills, 2002; in Mills et al., 2002.

Well before it came under donor pressure, the Government of Sri Lanka had embarked on its own programme of public-private partnership. In 1980 it granted a twenty ﬁve year monopoly wheat milling contract to a private company, PRIMA Ceylon Ltd. The company headquarters of PRIMA are in Singapore. Besides the milling agreement, PRIMA was also contracted under a BOT arrangement to build and operate the mill, and then to transfer ownership to government in 2005. The contractual relationship with PRIMA is carefully constructed to ensure that it does not beneﬁt from monopoly rents. Legal ownership of the ﬂour remains with the CWE and

The Experience of Contracting 153

quality standards are imposed. PRIMA has incentives to perform well: on its performance depends the possible renewal of this lucrative contract. Moreover, it is motivated to operate efﬁciently through its right to keep any excess of ﬂour above an agreed rate of extraction of 740 kg of ﬂour per 1000 kg of wheat. However, the risk to PRIMA is low: there are agreed minimum quantities of wheat to be milled annually, procurement is ﬁnanced by CWE, and sales are guaranteed (Smith and Ellis, 1997). Urban water supply Build-operate-transfer contracts are used to ﬁnance capital investment in infrastructure. The private investor takes on the limited operational responsibility of managing the particular asset until it is transferred to the public sector, rather than of providing the service as a whole. The private sector incurs the debt and is repaid over 25–30 years by the public utility through the latter’s guaranteed purchase of water. The private operator does not have direct access to the revenue generated by the project, as it would in a lease or concession contract, but depends on the public body. Such arrangements require that the private company and its creditors should have conﬁdence in the capacity and commitment of the public utility to fulﬁl its side of the bargain – to raise revenue and repay the capital cost. The level of investment required for a major water supply scheme makes the risk particularly acute. The risk is augmented by any weakness in the legal infrastructure of the country, the stability of the currency in which revenues will be earned, and the capacity of the public utility to earn the revenue by collecting customers’ payments. A BOT to develop new water supplies for Hyderabad Water Supply Board failed on these grounds. The Board was unable to convince the international bidders that the revenue and payments for their water would be forthcoming. Similarly, in Zimbabwe several proposals for the development of bulk water supplies through BOTs have collapsed due to investors’ lack of conﬁdence. Build-operate-transfer schemes have been more successful in middle than low-income countries. Nickson and Franceys (2003) cite the case of Malaysia where the Water Department of the Ministry of Infrastructure Development undertook a BOT for development of water supplies for Kota Kinabalu. A private operator took over existing water supply and transmission facilities and constructed a new dam and water treatment plant. Even here, the state government that negotiated the contract and took a 20 per cent stake in the company was unable

154 The Changing Role of Government

to obtain adequate revenue collection from users and had to cover its annual deﬁcit by a soft loan from the federal government. As Nickson and Franceys point out, BOTs do not overcome the fundamental difﬁculty of governments which is not to handle capital investment but to handle their clients – that is, to deliver a service and get payment for it. Indeed, the cases above indicate that BOTs often founder on these governmental incapacities. Lease and concession Urban water supply By comparison with BOTs, lease contracts involve less risk to the private sector. Leases cover operation and maintenance but not new capital investment, and the private operator manages the collection of the revenue from which it can take its share. Concessions, on the other hand, combine both elements – capital investment as well as the operation and maintenance of the entire water delivery system. This makes for a more straightforward and comprehensive solution for governments, but higher risk for the private investor and operator. In three of the research countries, lease and concession contracts were proposed or being negotiated, but all had been beset by delays. Lease contracts are proposed for urban water supply in Ghana, but have been bogged down in uncertainties about the transparency of the contracting process and in public opposition to what is perceived as the ‘privatization’ of water. In Sri Lanka, concessions have been proposed for two tourist areas and a freeport industrial area, but there are doubts whether concessions can be viable in such limited areas. In Tamil Nadu State, India, a concession is planned for the water supply and sewerage of Tiripur Area Development Project, through a complex partnership between a BOT operator, a ﬁnancing institution, and local, state and central governments. A successful case is that of Buenos Aires, where a concession agreed in 1993 was the ﬁrst of a new global wave of concessions. An international water company, Ondeo, led a consortium which successfully managed the extension of water and sewerage systems, reaching poorer sections of the population at reduced tariff rates, and achieving a good return on its investment (Box 7.3). The Buenos Aires case suggests that a large-scale concession, managed by a major international operator in a relatively rich city where water is plentiful, is indeed capable of delivering [beneﬁts]. (Nickson and Franceys, 2003)

The Experience of Contracting 155

Box 7.3

A concession contract – an example from Buenos Aires, Argentina

A consortium called Aguas Argentinas bid to supply water at a tariff 27 per cent below the water rate charged by the previous public company to consumers. The consortium which won the concession was led by a French water company, Ondeo, which became the service operator as well as principal shareholder. A regulatory body, ETOSS, was set up to review the tariff structure, oversee contract compliance, control water quality, and protect customers. The contract required the concessionaire to lift coverage of the population from 70 per cent for water and 58 per cent for sewerage to 100 per cent and 95 per cent after 30 years and then to return all the assets to the government. Within seven years coverage had reached 85 per cent and 63 per cent, revenue had increased by 69 per cent and expenses only by three per cent in real terms; ﬁnancial subsidies from central government had been eliminated. ‘The ﬁnancial information illustrates that it is possible to operate a viable water utility whilst investing heavily in quality and service improvements’ (Nickson and Franceys, 2003). Moreover, Aguas Argentinas managed to reach the hitherto excluded poor population of the city, by cross-subsidizing the payment of connection fees by poorer households, allowing the phased payment of fees, and working with NGOs to reduce costs by organizing voluntary labour. (Based on Nickson and Franceys, 2003).

However, these favourable circumstances are not typical of developing countries, and were not to last in Argentina. The macro-economic crisis of 2001, followed by a 300 per cent devaluation of the local currency, left the international operator appealing for reparations through an arbitration tribunal. The same company has suffered losses from its operation in Manila. In the even more risky environment of Mozambique, another large operator, Saur International, has withdrawn from a consortium, selling its share for a symbolic one Euro to Aguas de Portugal. Saur argues that as a result of ‘a marked increase in risk in developing countries there is a need for ﬁnancial solutions including substantial grants and soft funding…’ (Talbot, 2002). Joint venture and public-private partnerships Business promotion In the sphere of business promotion, several agencies were described in Chapter 4 as cases of decentralized management, with relatively high levels of autonomy from the civil service apparatus. Many of these also

156 The Changing Role of Government

include at least an element of public-private partnership, in respect of one or more of the following: • ownership: government, private or shared ownership of the joint venture company; • ﬁnance: by government or private contributions or from the earnings of the company; • decision-making: boards with different balances of public and private representation, appointed by government, by government and business or self-selected. Jackson (2002) studied agencies that provide support services to business, with a focus on the textiles and garments sectors. Services included research and development, training, market information and export promotion. Of these business development agencies in India, Ghana, Sri Lanka and Zimbabwe, most were governmentowned with government appointed boards, but several had private representation on their boards and sources of funding that depended on services rendered. In broad terms, the ﬁnding of the research was that agencies were more likely to be effective in supplying business support services if they had both more autonomy from government and also more private participation in their ownership, ﬁnancial sources and decision-making structures. These characteristics made the agencies more ﬂexible and responsive to their business clienteles. The clearest case of a successful joint venture (covering ownership, ﬁnance and decision-making) was Zimtrade, established by the Zimbabwe Government in the structural adjustment period to develop export opportunities. One half of its Board was appointed by the private sector and one half by government. Its staff were a combination of people drawn from the public and private sectors and joined by short-term consultants. Their salaries were at private sector rates. Zimtrade received funding from the government and from the European Union which sponsored its development, but these sources were being replaced by its own sources derived from a levy on the exports of voluntarily registered companies. As a consequence, within the framework of plans agreed with government, Zimtrade had both a high level of autonomy in its decision-making and an incentive to serve the companies that paid for it.2 However, the research found a limit to the virtues of autonomy and private involvement. If agencies go too far in this direction, they risk

The Experience of Contracting 157

acquiring their own proﬁt motive and are in danger of losing sight of their public objective – to sponsor the business sector as a whole. Another organization in Zimbabwe had gone further in this direction. The Industrial Development Corporation had been set up to incubate infant industries and to recuperate ailing ﬁrms, but had come to act as a holding company hanging onto successful businesses and neglecting its broader mission (Jackson, 2002). Urban water supply Nickson and Franceys (2003) cite the case of a joint venture to provide water and sanitation services to the Municipality of Cartagena in Colombia. The joint venture company, Aguas de Cartagena, is owned jointly by the municipality and an international water company, Aguas de Barcelona, together with a small participation of local capital. Aguas de Barcelona operates and maintains the service, and manages a major investment programme ﬁnanced by the World Bank and the Inter-American Development Bank. The joint venture gives Aguas de Barcelona the operational functions that it would have under a lease or concession, but without risking its capital in an unstable environment. While the involvement of the municipality gives it some protection from that environment, on the other hand, there are potential conﬂicts of interest between the two parties. The advantage of the joint venture for the municipality is that it retains its participation and an increased income from its shareholding. The risk is in the conﬂict of interest and the ‘regulatory capture’ that is implied by the municipality’s position as both a shareholder and the protector of consumers (Nickson, 2001a). Health care The relationship in many African countries between government and church-based (and in Zimbabwe, mining company) hospitals is one of informal, trust-based arrangements. Traditionally, governments have given block grants (Zimbabwe) or subsidized the staff of church health facilities (Ghana) without any serious speciﬁcation of the service to be provided in return. The ministries of health have counted on the goodwill and competence of the recipients who, in turn, have expected to be trusted. These are ‘relational’ contracts, governed not by speciﬁc terms but by an understanding that their informality was justiﬁed by faith-based commitment and motivation. However, there is a delicate balance between trust and abuse of

158 The Changing Role of Government

freedom. For example, the arrangement between the Government of Zimbabwe and the mine hospital entailed the payment by government of the fees of indigent patients. The hospital had no incentive to check the capacity of its patients to pay; so the numbers of free patients and the costs to government mounted. In both countries there were proposals to move towards more explicit contracts. The Government of Ghana had agreed with the church authorities that there would be a block grant with a two-year rolling contract, specifying the services to be provided and the monitoring arrangements.

Assessment of the experience of contracting The effect of contracting on performance Within each sector, the research showed that, where contracts of any sort are fully implemented, they usually have the effect of improving performance by comparison with direct public provision. Performance here covers efﬁciency (the cost of providing the service) and effectiveness (the coverage and quality of the service). Table 7.2 compares contracted out services in the health sector with those provided by government and, for Sri Lanka, competitively contracted services versus those that were put out to selected providers without competition. Contracted clinical services are generally cheaper and better quality than the directly provided services, and never worse. Support service contracts – for cleaning, catering and maintenance – present a more qualiﬁed conclusion. Costs are lower but quality is also sometimes lower than the public service. The research identiﬁed a ‘cost quality trade-off’ where contracts were being let by governments at such a low price that the quality of the service was being driven down and few bidders were willing to compete for the contracts (Mills et al., 2001, p. 131). In the case of urban water supplies, research (Nickson and Franceys, 2003) reported much higher levels of productivity, improved cost recovery, increased coverage of the population and improved service quality resulting from • service contracts for the management of sewage plants in Chennai, India; • a series of service and management contracts for water supply in Santiago, Chile;

India

High technology equipment contracts: • Cost: borne by users; much cheaper than private services • Quality: longer opening hours; functioning equipment; staff dissatisﬁed with pay • Managers’ comments: private sector better able to secure higher staff productivity

Catering contracts in Mumbai hospitals: • Cost: lower • Quality: lower • Managers’ comments: contracting had advantages e.g. reduced administrative burden

Equipment maintenance: • Cost: no data • Quality: managers complained of slow response time • Managers’ comments: contract expensive but no convenient alternative

Agreement with Mining company hospital: • Cost: contractor price for inpatients lower than government; for outpatients higher • Quality: at least as good • Managers’ comments: little attention paid to contract

Agreement with church organizations: • Cost: Lower • Quality: largely similar • Managers’ comments: church hospitals had advantages of greater staff commitment and more autonomy

Zimbabwe **

Clinical contracts Agreements with church organizations: • Cost: similar • Quality: largely similar • Managers’ comments: church hospitals had advantages of greater staff commitment and more autonomy in resource use decisions

Support service contracts

Evidence on health policy outcomes: comparison of contracted-out service with directly-provided government

Ghana *

Table 7.2 service

159

High technology equipment contracts: • Cost: varying pattern: come cheaper, others more expensive • Quality: better maintenance and productivity • Managers’ comments: advantages of higher productivity and improved access

Thailand

Source: Mills et al., 2001, pp. 132–133.

Note: ‘Cost’ represents the cost to the government of contracting-out (not the cost to the contractor of providing the service); ‘managers’ are those at hospital level.

* None studied ** No evidence *** Comparison was between competitively awarded contracts and directly negotiated contract.

Cleaning service contracts: • Cost: probably slight lower • Quality: probably better • Managers’ comments: more satisﬁed with contracted service

*

Clinical contracts

Sri Lanka Competitive contracts for provision of hospital food:*** • Cost: negotiated contract more costly • Quality: lower food quality with competitive contracts • Managers’ comments: cost minimization is main aim; achieved by competition

Support service contracts

Table 7.2 Evidence on health policy outcomes: comparison of contracted-out service with directly-provided government service – continued

160

The Experience of Contracting 161

• a management contract for the water and waste system in Trinidad and Tobago; • a BOT contract to increase water production in Kota Kinabalu, Malaysia; • concession contracts for water systems in Argentina; • a joint venture for the water system of Cartagena, Colombia; • a cooperative water supply system in Santa Cruz, Bolivia. In agricultural marketing, Hubbard (2003) found • the contracting of maize exporting and transportation in Kenya generally led to lower costs and reduced delays in both operations and payment for services; • lower cost of privately contracted grain storage in Sri Lanka and India; • savings in the cost of ﬂour-milling in Sri Lanka, together with the eventual transfer of the mill to government ownership. In the business sector, there was only one clear case of a contract – a joint venture for export promotion in Zimbabwe. However, the research found that business support services generally operated better (more accessibly and usefully to ﬁrms) where there was some distancing of the agency from government itself. This might be through private sector ownership or shareholding in the operation, private sector representation on governing boards, or access to ﬁnancial and human resources that were independent of government (Jackson, 2002). If contracted out services generally perform better, why is contracting not more common? Chapter 3 identiﬁed the institutional and political obstacles to reform. The change from a statist model threatened existing power structures and offered only uncertain gains. Political leaders stood to lose control of established instruments of patronage. There was little client or citizen demand for change. Public service employees felt threatened, and the future possible supporters of liberalizing reforms – managers and new beneﬁciaries – were not yet in place. Nevertheless, this chapter has shown that sometimes change does occur. The following sections analyze the evidence provided by the case studies in Ghana, Zimbabwe, India, Sri Lanka and some other countries to identify why some types of activity have been contracted out and why some forms of contract have been preferred over the others.

162 The Changing Role of Government

Activities that have been contracted out Chapter 6 referred to the key characteristics identiﬁed by Williamson (1987) as those that determine the ease of contracting a service: • The degree to which the service can be speciﬁed in advance: How easy is it to specify the requirements to which the contractor is subject? • The degree of asset speciﬁcity: Are the assets (equipment, knowledge) required by the contractor so speciﬁc to the contract that choice is limited and the winning contractor can hold government to ransom? • The practicality of gathering and measuring information about performance: Is the contractor’s performance measurable? Does government have the capacity to gather information to make the assessment? The evidence from the case studies is, indeed, that it is mainly the more speciﬁable, measurable and competitive activities that are contracted out. Across all sectors the most widely contracted activities were the generic support services – cleaning, laundry, maintenance, transport, supplies, and building works. In these areas, governments can relatively easily design contracts and control the contractor. Certain other activities that were also sometimes contracted out had a similar speciﬁable and measurable character. In the water sector, almost all key outputs are reducible to contractual speciﬁcation and to quite easy assessment of performance – water volume, coverage and quality. In agricultural marketing the speciﬁc and discrete functions of grain purchase, shipping, storage and milling were contracted out in several countries. In health, the hire of diagnostic equipment was another case. However, while all of these have a high degree of speciﬁability and measurability, some of them clearly require the contractor to have rather speciﬁc assets in terms of equipment, professional and managerial skills, and access to large-scale capital. This gave the big, and often international, companies that took on major contracts in the grain trade, storage and milling, in developing and running water systems, or in supplying high technology diagnostic equipment strong leverage in negotiation with government. The more qualitative services and less discrete activities were rarely contracted and, where they were, presented problems to the government that contracted them. It was difﬁcult to establish quality

The Experience of Contracting 163

standards for hospital catering in Mumbai, India, and in Sri Lanka. Agreements with church and mining hospitals in Ghana and Zimbabwe were not speciﬁed at all and instead left to broad understandings. Attempts in the agricultural sector of Zimbabwe to contract out professional services – tsetse control and veterinary services – failed. Business services, which would have been difﬁcult to specify, were not contracted out by government to the private sector but operated separately by government, the private and non-proﬁt sectors, or through various forms of loose partnership. Even in the areas where contracting was relatively straightforward, there was often much less contracting out than would have seemed conceivable. It was only in Sri Lanka, and in some of the reference countries – Thailand, Argentina – that something like a ‘contract culture’ had emerged. In Ghana, Zimbabwe and India the pace was slow, often due to resistance from public sector workers or to mutual suspicion between the public and private sectors. The feasibility of contractual types Governments have far more experience of short-term contracts: one-off or ‘spot’ contracts for the undertaking of works and purchase of supplies, and slightly longer-term service and management contracts. In all of these, the service remains in the ownership and control of the public sector, although there is some disposal of speciﬁc (usually ancillary or support) activities and non-core staff. Indeed, one attraction of these sorts of contracts for public sector managers is that they increase managers’ control by enabling them to specify price, delivery dates and quality of the contractor’s work, which they can rarely achieve with their own direct employees. However, among the disadvantages of short-term and speciﬁc contracts for public sector managers are: the transaction costs of repeated contracting and of coordinating contractors; the difﬁculty of achieving a coordinated service; the reduced interest that large ﬁrms may have in periodic and uncertain work, especially where they have to build up speciﬁc assets and skills to undertake them; and the inability to fully engage the private sector’s management skills and ﬁnancial resources. The difﬁculty of operating a whole service on the basis of multiple contracts is illustrated by Santiago’s pre-privatization water supply (Box 7.1). More comprehensive, long-term contracts present the opposite advantages and disadvantages. Public sector managers are relieved of operational functions, becoming freer to set policy requirements and

164 The Changing Role of Government

standards and then to purchase the services of a provider. Splitting the roles of public purchaser and private provider through contracts eliminates the conﬂicts of interest that exist where the public service is both ‘poacher and gamekeeper’ – both ﬁnancing the service and managing its provision. Moreover, through such comprehensive contracts the public service gains access to a full complement of private sector management skills, motivation and control of human resources, and access to capital. In concession contracts, the service is totally integrated, with responsibilities for investment and service management linked in the same contractor. This obviates a common problem where infrastructure is built by a contractor to standards that cannot be maintained by the operator or paid for by customers. Another advantage of comprehensive, long-term contracts is that more of the risks are passed to the private contractor. The risks of unexpected construction and operating costs, delays, uncertain levels of demand and unwillingness to pay by customers are passed from the public to the private sector (Nickson and Franceys, 2003). Why, then, were long-term contracts so rarely adopted in the developing countries that were examined? The following main explanations emerge from the case studies: • They threaten the employment of core public sector staff. • They provoke ideological objections to ‘privatization’. Although public sector ownership remains, it is the private managers who are in direct interface with customers, at least in the case of concessions, leases and joint ventures, and possibly with management contracts. • International companies are often the key contenders for these contracts, adding ‘de-nationalization’ to ‘privatization’ as an ideological objection. Small and local companies are much less likely to have the necessary experience and capacity. • There are questions about the capacity of governments to design and manage long-term and more complex contractual arrangements. • The environmental and institutional conditions that allow such contracts to operate are not prevalent. Companies that are to take on these contracts and the associated risks will need to be sure that the conditions exist for them to beneﬁt from their investment. The last two bullet points are examined further in the next section. Table 7.3 makes some broad comparisons, drawing on the case studies. It establishes eight broad categories against which the alternat-

The Experience of Contracting 165

ive contractual arrangements can be compared. The italicized statements indicate conditions that are likely to make a contractual arrangement more feasible, effective and beneﬁcial, especially in developing countries with lower levels of government capacity: i)

The contract is explicit. It deﬁnes and separates the roles and responsibilities of the partners. ii) The contract is for speciﬁc, measurable services. It is not complex and difﬁcult to specify. iii) The contract is largely independent of external conditions (political, economic, ﬁnancial and legal stability). Its application does not depend greatly on these institutional conditions being in place. iv) The contractual relationship has the effect of increasing access to capital. It brings in private sector access to capital rather than merely the management of existing capital sources. v) The contractual relationship brings in private sector management skills. It harnesses private sector management and reduces the operational demands on public sector managers. vi) It integrates capital investment with the management of operations. These are not separately managed, leading to investment without regard to the sustainability of its management. The table also includes other considerations that affect the legitimacy or acceptability of contracts. Contracting out will certainly challenge public service employees, to a greater or lesser degree, but will it appeal to other constituencies? vii) The contractual arrangement gives new opportunities to local businesses. Or, is it likely primarily to attract international companies? viii) The contract does not appear to ‘privatize’ the sector. Is the public sector perceived as remaining in control? ‘Privatization’ is a cause of popular resistance to contracting out. In Table 7.3, the ratings indicate the comparative advantages presented by different contractual types; the shaded areas indicate where the advantages are strongest. The table shows that, while longer term and more comprehensive contractual arrangements present clear investment and operational advantages, on the other hand they present problems of feasibility. Short-term, simple contracts are more feasible. First, they set very speciﬁc tasks. Second, they are less dependent on the existence of necessary political,

Y

Y Y Y N

Build-operatetransfer (BOT)

Lease

Concession

Joint venture

N

Y

Y

Y

YY

YY

N

NN

N

N

Y

YY

N

YY

Y

YY

N

N

Y

YY

YY

Y

Y

Y

Key: Y = positive response; YY = very positive; N = negative response; NN = very negative. The shaded areas indicate where the differences are clearest.

Y

Service and management contracts

N

YY

Y

N

N

N

N

NN

NN

N

N

Y

N

N

N

N

Y

Y

Y

N

N

Y

YY

YY

Retains Promotes Avoids public local ‘privatization’ service business jobs

Is explicit Sets Is not Increases Increases Integrates about roles speciﬁc dependent access to access to capital and and terms tasks on external capital private operations conditions management

Works/supplies

Local legitimacy

The contract…..

The feasibility of contract types

Contract type

Table 7.3

166

The Experience of Contracting 167

economic and legal conditions. Third, they are less likely to confront problems of ‘legitimacy’ because they do not seem to privatize or internationalize provision, and are less likely to threaten public service employment.

Capacity to contract Apart from the question what sorts of contract have been adopted and what advantages and problems of feasibility they have presented in practice, there is also the question whether governments have the ‘capacity to contract’. Incapacity may be an explanation for not contracting or for contract failure. Short-term ‘spot’ contracts are likely to be easier to design and manage; on the other hand they may present transaction costs in the shape of frequent contracting and of the need to coordinate multiple contractors. Longer term and more complex arrangements, such as concessions, have the advantage of wrapping all aspects of the service into one contract over a long period, potentially cutting transaction costs. On the other hand, the design of such contracts requires a great deal of information and experience to anticipate the relationships between all the elements of the service and to try to anticipate all the possible risks and uncertainties that may occur during the term of the contract. Because such complete knowledge does not exist, they are likely to depend on goodwill and trust between the partners to work out appropriate solutions over time. The analysis of capacity to contract is based on the three dimensions set out in Chapter 1: 1. The internal organizational factors needed to undertake contracting and manage contractual relationships, including human resources and incentive systems; management systems, roles and relationships; capital and ﬁnancial resources. 2. Coordination and clarity of relations between the ‘task network’ of organizations that together are responsible for contract design and implementation. 3. The external environment and how it affects (enables or constrains) the mobilization and deployment of personal and organizational capacity, including the public sector institutional context of rules, regulations, laws, policies and power relationships; and the broader social, economic and political context within which public institutions and partners have to act.

168 The Changing Role of Government

The separation between internal, organizational factors and external or institutional factors is, of course, not hard and fast. The ‘external’ can only be effective through its impact on the way that organizations behave. The following sections survey the capacity issues that were identiﬁed in the three service sectors where contracting was practised, recognizing that the task of contracting may vary in its complexity between sectors. It then identiﬁes the common and distinct aspects of sector experience. Table 7.4 outlines the ﬁndings, grouping the key capacity constraints for each sector into three broad categories – the internal, inter-organizational and external factors. Capacity issues in the health sector Internal aspects of capacity In Ghana, the lack of skills in and experience of contracting was an important constraint within the ministry of health both with regard to the development of service contracts and the attempt to put the relationship with church hospitals onto a clearer contractual basis. In Zimbabwe, there was also a need to develop contract design skills, but the ministry of health had long experience of support service contracts and was able to build on this by using external consultants and overseas training. The two African countries had poor information systems on the costs of government-run services, meaning that they had no ﬁrm basis on which to decide whether to contract, to evaluate bids or to monitor performance. Financial information systems were being developed but the staff of ministries of health were poorly motivated to maintain the information systems or to act on them. One result was that late payment to contractors was common, acting as a deterrent to potential bidders. The organizational gap between central ministries that made contracts and local agencies (such as hospitals) that received and monitored the service gave little incentive for central ofﬁcials to design the contract well or for local ofﬁcials to supervise and monitor its execution. Zimbabwe had taken steps to correct this by involving hospital staff in tender boards. The Asian countries – India, Sri Lanka and particularly Thailand – generally had the experience and skills to design and negotiate contracts for support service contracts. Typically, however, the process of contracting was concentrated in the central civil service and bureaucratized. Governments had established standard contract forms, which ensured that contracting took place in a systematic way but left little room for ﬂexible adjustment to local circumstances. Standardized

Health care Limited experience in developing policy Weak information systems Weak management and ﬁnancial systems Weak basic administration Staff resistance and lack of incentive

Poor coordination of task network Gap between contract design and implementation Weak ‘contract culture’ Lack of political support Civil service rules Centralized, rigid government systems Corruption in contract award Limited private sector development in some cases

Internal factors

Interorganizational factors

External factors

Weak adaptive capacity Socio-political opposition to ‘privatization’ Economic and ﬁnancial instability Weak legal framework Private sector risk Weak indigenous ﬁrms

Absence of pressure for change in organizational arrangements Weak ‘contract culture’

Limited experience of complex contracting Weak information on service operation Weak revenue-raising for payment to contractors Weak regulatory skills and autonomy Staff resistance and lack of incentive

Urban water

Key constraints on governments’ capacity to contract

Factors

Table 7.4

Weak adaptive capacity Lack of political coalition for reform Unpredictable policy environment Lack of trust between government and private sector

Difﬁculty maintaining task network of government bodies Weak ‘contract culture’

Weak contract design and ﬁnancial control in Ghana

Agricultural marketing

169

170 The Changing Role of Government

terms, ﬁxed prices, slow payment of contractors led to restricted private sector interest in competing for contracts. Particularly in Sri Lanka, this led to low incentives for hospital staff to monitor contracts over which they had no clear authority. Studies across a wider range of countries – India, Zimbabwe, South Africa, Mexico, Papua New Guinea and Thailand – have conﬁrmed the frequent lack of the management and information systems necessary to contracting. Where capacity was most limited, ‘the most basic administrative tasks such as ﬁling documents relating to the contract’ were not undertaken (Bennett and Mills, 1998, p. 315). More common, even in middle income countries such as Mexico, South Africa and Thailand, were failures in the ﬁnancial and information systems necessary to the monitoring and payment of contractors. These were most acute in relation to contracts for clinical services but were also problems in contracting simpler support services. The key weaknesses were in respect of: • Cost and quality information to enable decisions about what services to contract and to negotiate contracts; • Budgetary frameworks and ﬁnancial control mechanisms that can be used to monitor expenditure against plans; • Performance indicators to permit monitoring; • Clarity about the responsibilities (negotiation, contract speciﬁcation, monitoring, payment) of organizations within the ‘task network’. One result was the lack of inﬂuence of the recipients (hospitals) of services in framing contracts or monitoring contractors (Bennett and Mills, 1998). External aspects of capacity In spite of these internal weaknesses, the research on health sector contracting concluded that the more fundamental capacity problems were external to ministries of health. If internal skills and systems were lacking, they could be built up over time with experience and support. But these internal capacities would never be effective if the environment in which they had to operate remained unchanged. Lack of political backing for contracting: Contracting out was opposed by public service unions in all countries and politicians were rarely prepared to confront this in situations where unionized labour often had a strong political voice. Zimbabwe was the exception, where the policy of contracting out was (at that time) adopted by the Cabinet and support was given to displaced workers.

The Experience of Contracting 171

Weak private sectors: This was particularly a problem in Ghana where, even for support services, there were few companies willing and able to compete for contracts. For clinical services, requiring greater investment and expertise, it was a more general problem. In India and Sri Lanka, government price controls and bureaucratic procedures also often restricted the willingness of ﬁrms to compete. Mistrust between the public and private sectors about each other’s motives and reliability was a factor in Ghana and India. Bureaucratic systems: Centralized and rigid systems of control by government of line ministries and hospitals limited the freedom of local managers to decide what to contract and on what terms, especially in the Asian countries. The public sector institutional context was hierarchical, contradicting the ‘ﬂat’ and decentralized mode of management implicit in marketized approaches. Corruption: In the context of weak systems of political and social accountability, contracting may enhance opportunities for corruption through non-transparent tendering procedures. Much of this amounts to saying that, although contracting had been formally adopted in most countries, a ‘contract culture’ (Walsh, 1995) was still missing both within the organizations and in their external environment. Capacity issues in the water sector Internal aspects of capacity Among the four focus countries (Ghana, Zimbabwe, India and Sri Lanka), there were very few cases of longer term and more complex contractual arrangements getting off the ground. This is in spite of factors that should favour contracting in the water sector: there is strong international advocacy of and support for contracting in the sector, the measurability of water supply facilitates contract design, and forms of organization for water delivery are typically ‘mechanistic’, universal and standardized. Indeed, in all of the core countries, there had been moves towards the development of new forms of publicprivate partnership. However, usually they had stalled after the policy had been agreed and the legislation passed and before implementation. This was the case with the BOT and concession proposals in India, the concessions in Sri Lanka and the lease arrangements in Ghana. Failure to implement declared policy was due partly to the capacity of vested producer interests – engineers and unionized workers – to prevent the loss of public sector employment. In rare cases, a performance-oriented water board (Chennai, India) had overcome

172 The Changing Role of Government

internal objections and embarked successfully on the use of service contracts. The capacities demanded by higher level contractual arrangements do put strains on government. The piecemeal contracting out of service and management activities to multiple contractors, as in the Chilean case (Box 7.1), demands not only a sustained capacity to maintain a continuous cycle of contracts but also to monitor and coordinate them; the information demands are high. Build-own-operate and concession contracts, in India and Sri Lanka, have stalled on the incapacity of governments to demonstrate that the legal, ﬁnancial and market conditions exist to justify private investment. Making 20–30 year contracts requires government to think in unaccustomed ways so as to specify the outputs expected, the measures of performance that will be used, the way in which risks will be distributed between government and contractor, and arrangements for regulation and conﬂict resolution. Bidders require evidence, based on information that rarely exists, of the condition and commercial performance of the existing water system. For example, negotiations for the Buenos Aires concession were completed in the ‘record’ short time of two years, with the government being required to assemble operational information and ensure the legal conditions that would give investors conﬁdence. Even after the contract had been agreed and was in operation, the Argentinian Government and municipalities continued to work with the private operator to ﬁnd ways of adjusting their agreement so as to make the terms more accessible to the poor. An effective ‘relational contract’ was established in a situation where both government and contractor had the capacity to analyze, adapt and negotiate. In spite of these complexities, as in the case of the health sector, the research concluded that the capacity to contract was not mainly limited by internal but by external factors. Policy and legislation could be developed, perhaps with external advice; and contract design and management skills could be learned or augmented by hiring in consultants. The internal capacity that was more difﬁcult to develop was the long-term regulatory role, where not only new professional skills have to be developed but also new legislative frameworks and, above all, an often unaccustomed culture of detachment from politics. External aspects of capacity In some respects, contracting in the water sector is facilitated by the relatively straightforward and measurable requirements of a water

The Experience of Contracting 173

service – to deliver a sufﬁcient quantity of clean water to the whole of the population at a price that it can afford. On the other hand, it presents major institutional barriers to change; there are great political risks for governments and heavy ﬁnancial risks for ﬁrms. As observed in the health sector, long-term contracts also imply a ‘massive cultural shift’, both internally and externally. Clients have to be willing to pay, staff to meet performance standards, ﬁrms to compete, and governments and ﬁrms to maintain relationships built on trust combined with distance. The research emphasized as the fundamental problem not so much the lack of capacity to operate contractual arrangements but the incapacity of the core countries to bring about change in organizational arrangements. Civil society opposition: Particularly in Ghana and Sri Lanka, contracts with the private sector have been challenged by mass opposition movements to the ‘privatization’ of water. This has led to delays in the establishment of leases in Ghana, the abandonment of a concession in Colombo, Sri Lanka, and the termination of private management contracts in countries including Bolivia and Trinidad and Tobago. Social surveys undertaken for this research demonstrated a widespread public suspicion that ‘privatization’ might lead to the loss of rights to water (Rakodi, 1998 and 2000). Lack of political backing for contracting: In the face of public sector employees’ resistance and public opposition, political support for contracting, in practice, was weak or absent in India, Sri Lanka and Ghana. Politicians may have their own interest in retaining a service in the public sector which is subject to their patronage. This may lead them to favour limited change towards short-term service contractual arrangements where political control can be retained and even extended to the contracting process. A further point is that political support is often fragile, depending on a particular political leader without clearly institutionalized party backing. Private sector capacity: Short-term contracts for particular works and service inputs can be taken on by local ﬁrms, and it is clear that such capacity exists in most of the countries studied. The scale of investment and the risk is much greater in the case of BOT contracts, leases and concessions, and is likely to rule out all but large international companies, even in most middle-income countries such as Argentina. Firms assume the risks of construction cost over-runs, uncertainties of demand and willingness to pay, uncertainties about the quality of the assets they are assuming, and concerns that governments may not follow their side of contracts or that regulators may not operate

174 The Changing Role of Government

disinterestedly. In the case of concessions, foreign companies also face the risk of devaluation of the local currency in which consumers’ payments are made. Government capacity to limit private sector risk: The point of contracting is to pass risk to the private sector, but within limits of acceptability. The research indicated that poorer, and even middle-income countries, have difﬁculty in guaranteeing the conditions of macro-economic and political stability that could assure long-term private investors of their security, or even of the legal framework within which their contracts would be embedded. This was the case for a number of planned BOT contracts in Zimbabwe. Even the Indian and Malaysian cases had some difﬁculty in guaranteeing capacity to collect revenue to repay BOT investors. The highly reputed and effective Philippine and Argentinian concessions were by 2002 causing foreign exchange losses to the contractor, Ondeo. As Plummer (2002b) has observed, models of public-private partnerships for water supply have been transferred from high-income countries with a strong institutional capacity to low-income countries with …chronically weak public institutions, deteriorating infrastructure limited to core urban areas, an immature and unproven formal private sector, low rates of cost recovery and a weak regulatory context. Capacity issues in agricultural marketing Internal aspects of capacity The experience of contracting out agricultural purchasing and marketing was mainly limited to ‘spot’ and service contracts, except in the case of the 30-year BOT for the wheat mill in Sri Lanka. The story was similar. Except in the case of Ghana, which was attempting to make new forms of contract with communities for the construction of feeder roads, there was little problem of technical capacity to make contracts and this capacity could itself be contracted where necessary. Even so there were sometimes difﬁculties in sustaining contractual relationships. In Kenya, it was found to be demanding to maintain coordination between the ‘task network’ of separate government bodies responsible for storing maize, contracting exporters, and monitoring and evaluating contracts. External aspects of capacity The main capacity problems were found in the broader environment within which contracting took place: Lack of political backing for reform: In all countries except Sri Lanka the political drive to contracting out was weak and reluctant. In the

The Experience of Contracting 175

African countries, it occurred mainly under donor pressure. In India, it was under pressure of inadequate public sector facilities. The research identiﬁed ‘unwillingness to reform organizational arrangements’ as the main limitation on contracting and other new forms of management. It was the capacity to adapt rather than the capacity to implement that was the key. Unpredictable policy environment: Contracting arrangements were frequently contradicted by other policies that were not conducive to the maintenance of a contractual relationship. In Zimbabwe, policy in favour of contracting out had reversed. In Sri Lanka, a policy of subsidizing wheat and the imposition of rice import duties undermined the proﬁtability of contracted rice imports. In Kenya, a tax was imposed to discourage maize exports while maize export contracts were being implemented, and transport systems were not developed as expected by the traders. In India, public grain storage facilities were often favoured even where their prices were higher than those of private contractors. Private sector capacity: There was limited capacity among national ﬁrms to compete for and undertake large-scale contracts. In itself this is not a capacity limitation so long as there are others to compete for work; the export contracts in Kenya and the milling and import contracts in Sri Lanka almost all went to international companies. National and local companies in India were able to undertake the construction of large-scale storage facilities. Public-private relationships: An underlying problem in many countries is mistrust between the public and private sectors. Donor pressure had levered in the private sector in Kenya and Ghana, against a long history of governmental suspicion of the motives of the private sector. In India, public ofﬁcials remained dubious of the quality of ﬁrms’ grain stores, and ﬁrms were uncertain whether they could invest with any conﬁdence that their facilities would be used. Long-lease contracts and ﬁnancial assistance have helped to provide a basis of investor conﬁdence.

Capacity issues across the sectors The sector studies made it clear that contracting out is not a new phenomenon for governments. There is, at least, experience of making simple, short-term contracts for works and supplies as inputs into public sector programmes, and some indication that governments are seeking to put more such work out to contract. There is much less

176 The Changing Role of Government

experience of making long-term and more complex contractual arrangements, particularly where these involve (i)

responsibility for managing whole service outputs, as opposed to inputs into a public service; (ii) services that are more difﬁcult to specify and monitor, that is services that are measured qualitatively (e.g. clinical or professional services) rather than quantitatively (e.g. water supply); (iii) large-scale capital commitments by the investor, as in major water supply contracts. Even where there is experience, contracts are often designed and managed poorly due to basic administrative failures, including weak ﬁnancial, accounting and information systems, and unclear roles and responsibilities. In fact, contracting often exposes administrative incapacities that already existed, because contracts require precision in deﬁning procedures, standards and roles. Without information on the cost and coverage of services, it is not possible to design contract requirements, assess bids or measure performance. But, information systems were sometimes so poor that contract documents could not be found. Poor budgetary and ﬁnancial control mechanisms led to uncertainty about how much was to be spent, what had been spent, and delays in payment to contractors. Often central government bodies were responsible for designing and issuing contracts on behalf of agencies that had no part in making them; local ofﬁcials then had little incentive to monitor implementation. Contracts might be designed that assumed unlikely levels of revenue as a basis of payment to contractors. However, across the sectors the conclusion was that the main problems limiting contracting were not at the organizational level but in the wider environment. Where there were deﬁciencies – in human resources and skills, ﬁnancial management and information systems, and the distribution of organizational responsibilities – these could be developed. The more profound cultural and institutional constraints on capacity were Social and political resistance to change: This could be described as a problem of ‘adaptive capacity’. Even where the policy commitment existed, political backing for change was often lacking in the face of the vested interests described in Chapter 3. The water, health and agricultural sectors all present strong ideologies as well as vested interests against private sector involvement, especially where this moves towards core functions.

The Experience of Contracting 177

Incapacity of contractors: Local ﬁrms were often incapable of taking on large and complex contracts, but their incapacity was often compounded by a legacy of mistrust between ﬁrms and government and by rigid contracting and pricing practices. Across the sectors large-scale and long term contracts frequently went to international companies. This might introduce more efﬁciency but also sometimes renewed ideological resistance. Resistance to the ‘contract culture’: Contracts are based on an ethic of performance, choice and competition that challenges established relationships between politicians, bureaucrats, citizens and ﬁrms. Even where new policies and organizational arrangements were formally adopted, there was often a gulf of understanding in practice. Policy commitments to contracting carry little meaning where the basic conditions of the ‘contract culture’ do not exist. Centralized systems: The health sector, particularly in the Asian countries, illustrated the problem that ministries of health limited the freedom of agency managers to determine the design and terms of contracts. Contracts were bureaucratized instead of being an instrument of management control. Contradictory policies: The agricultural sector showed how contractual agreements could be undermined by decisions taken elsewhere in the management of the macro-economy. The terms on which contractors had agreed could be transformed by the introduction of new taxes and duties. Macro-economic and political instability: This throws into question any long-term contract that involves heavy capital investment and repayment out of revenues. The studies illustrated several cases where expectations of entering into BOT, lease and concession contracts had fallen through because governments could not guarantee the necessary conditions: political and economic stability, a legal system that would ensure contractual rights, and clear government capacity to raise the revenue so as to pay the contractor.

Conclusion Contractual arrangements for service delivery are already wellestablished in all the countries under study and across the sectors. Contracting is not a completely new role for governments. What is new is that they are moving further down this road, often under pressure of donors and their own incapacity to maintain in-house services but also by governments’ own choice. The moves are towards more

178 The Changing Role of Government

contracting out (if not full privatization), deeper and longer-term contracts, and a replacement of loose by more formal agreements. However, short-term and simpler contracts remain much more common. Most contracts are for purchases of speciﬁc supplies or public works, though a longer-term relationship is often maintained with selected contractors through repeat orders. Also common are service and management contracts, where particular inputs or particular aspects of the service (for example, maintenance or catering) are supplied to the public sector over a period of roughly one to ﬁve years. What remain uncommon are long-term arrangements, such as lease, franchise and concession, where all or part of the service are ﬁnanced or managed or both ﬁnanced and managed. Where these do apply, it is mainly in the water sector or where there are other large-scale capital works. But, there is more discussion about moving public services into private management than action, at least in the poorer countries reviewed here. The evidence of this research is that there are advantages to be gained from contracting out all or part of services. There are gains to governments or taxpayers in off-loading risk and costs to the private sector, and gains in managerial control of work-forces that were previously employed directly and are now on contract through companies. Almost all the cases presented in earlier sections of this chapter show gains in managerial efﬁciency by comparison with direct public provision. They also showed that there were widespread reported gains in service performance to customers. However, there are some caveats. First, particularly in the health sector, there was sometimes a cost/quality trade-off, meaning that driving down costs through contracts was in some cases at the expense of the quality of the service (for example, catering). This could be guarded against by more careful contracting. Second, contracts work where they are effectively designed and managed; this chapter has mentioned several cases where contracts have not been implemented or have failed. Third, where contracts succeed in raising performance, this may not be due to the greater efﬁciency of the private sector but to the contractual process itself. Contracts have the beneﬁt of requiring a clearer speciﬁcation of expectations and of providing a basis for monitoring of performance. So, if contracts offer these beneﬁts, why is contracting-out not more often adopted by governments? And why are they so reserved about moving towards longer-term and deeper contractual arrangements? In general terms, we have seen that policy-makers face resistance from

The Experience of Contracting 179

their own employees and from the public where there is any suggestion of shedding staff or of ‘privatization’. More complex and comprehensive contracts present greater potential advantages of mobilizing private management and capital, and of freeing government of coordination problems, but precisely for that reason they present the greater problems of ideological opposition. They also present far greater challenges to the capacity of governments and their partners in both contract design and implementation. This study has shown that capacity problems exist often at the most basic level – weaknesses in skills and experience, information, ﬁnancial management, organizational coordination and administration. More profound constraints exist in the wider institutional environment – centralized government systems, unclear policy frameworks, weak private sectors, unstable economies and polities, and resistance to the competitive ethic of the ‘contract culture’. Governments have usually chosen more easily speciﬁable and measurable activities for contracting – generic support services (for example, transport, maintenance and security) and services involving large capital works (water supplies, ﬂour mills). More qualitative services (clinical services) and less discrete activities (managing professional services) were rarely contracted out, and where they were often presented problems. Governments have also more often chosen to embark on short-term contracts which present fewer problems of capacity and resistance. These seem to be realistic choices.

8 Regulating and Enabling the Private Sector

Introduction Chapters 4 and 5 showed how internal management reforms within the public sector have been designed to release public managers from the hierarchical control of government, allowing them to make their own operational decisions about the raising and use of their resources. The ‘price’ that managers pay is that they are made more clearly accountable for their performance, according to policy frameworks and performance contracts. Their greater freedom is matched by a clearer statement about what they must achieve. This chapter will show how the same principle applies in the new relationship between government and the private sector or semi-independent public providers. In return for its increased involvement in service delivery, often funded by government, the private sector is subject to governmental controls about its outputs and standards. The instrument of public policy shifts from directly providing services to the contracting, regulation and support of alternative providers. Chapter 7 showed how contracts can operate as one instrument of governmental control, setting the outputs that are demanded of contractors. In this case the courts would be the default resort where contracts are not respected. Regulation is another instrument of control that would normally apply in the absence of contracts, where the provider is operating under licence. However, a ‘belt and braces’ approach is quite commonly adopted where there is both a contract between government and the provider, and a regulatory body that mediates between them. Support or enablement policies shift the focus from the control to the encouragement of market actors, but, in practice, new approaches to regulation also often operate on the basis of incentives rather than controls. 180

Regulating and Enabling the Private Sector 181

In this chapter, regulation will be deﬁned as systems of laws, directives, rules and licences that seek to impose outcomes which would not be reached by the operation of free market forces and private legal rights (Ogus, 1994). In this sense, regulation can be seen as one instrument of correction for market failures. More broadly, regulation covers all the state rules that govern the behaviour of public and private organizations and citizens. Even in the narrower sense, this is a broad territory. Paradoxically, while recent public management reform often speaks of ‘de-regulation’, in practice the new approaches have carried their own requirements for ‘re-regulation’. Indeed, some writers have described the emergence of the regulative or regulatory state (Lane, 2000, p. 135; Loughlin and Scott, 1997). This chapter will set out a framework for analyzing regulatory and enabling roles and the capacity to perform them, and then apply the analysis to some types of practice.

Forms of regulation The forms of regulation with which this chapter is principally concerned are those that purport to deal with some aspect of market failure. They seek to make markets work more efﬁciently or to make monopoly providers operate as if there were a competitive environment. In practice, in many countries there is also an earlier history of economic regulation that was developed on ‘anti-market’ assumptions, rather than to compensate for speciﬁc failures. In this view, markets themselves were seen as essentially exploitative and damaging to the public interest. Even in many states that were not formally socialist, state provision was (and often is) preferred to private provision regardless of whether the latter could operate in a fully competitive environment. That posture is universally under question and in retreat. Chapter 1 described that background and Chapter 3 described the politics of its undoing. Table 8.1 outlines alternative forms of regulation and particularly those that might be described as, in principle, ‘market-friendly’. Economic (or ‘entry’) regulation applies primarily to industries with monopolistic tendencies, where there is a case for regulating suppliers so as to make them behave as if they were in a competitive environment. This might be by making them compete for a contract (concession or franchise) to operate a public monopoly, where the contract is offered to the ﬁrm that offers the best price and quality conditions. The contractual terms are then the basis of the regulation by the

Motive Limitation of the sphere of the market Control of exercise of monopoly power

Avoidance of market dominance and control of the abuse of dominance Improving operation of the market by setting standards of products Protecting against negative external effects; promoting positive external effects of services Strengthening the efﬁciency and equity of markets

Anti-market regulation

Economic or entry regulation

Anti-trust policy

Social or product regulation

Regulation of external effects

Enabling policies

Informing and supporting market actors • Information to consumers • Information and support to producers

Penalties on producers imposing costs on non-beneﬁciaries. Regulation of examination systems to ensure common standards

Health, safety, and service quality standards imposed on producers

Imposition of limits on share of the market, and integration between ﬁrms

Entry conditions by: • Competition for the market and control through contract • Licence: control and incentive through regulation of performance, price and quality

Prohibition or direction of private producers

Examples of measures

Forms of regulation and enablement of the private sector

Form of regulation or enablement

Table 8.1

Consumer organizations Ministries Agencies for training, research, export promotion and investment

National examination system

Pollution control board

Health and safety agencies Education inspectorate Professional and consumer bodies.

Competition or price commission

Independent regulator

Government department as contractor

State planning commission Ministries

Typical agency

182

Regulating and Enabling the Private Sector 183

government body letting the contract. Alternatively, it might be by licensing a private monopoly provider to operate on condition that it complies with the price and quality terms of the licence. A government department or independent regulator then audits and enforces compliance. Anti-trust policy acts to avoid the emergence of monopoly conditions by preventing any single company or small group of companies becoming dominant and abusing their position, for example, by colluding to ﬁx prices or share out the market. Typically a public agency is set up to monitor market conditions, with statutory powers to intervene to prevent monopolistic behaviour. Product or social regulation opens the ﬁeld more widely but is still rooted in the principle that government intervenes only where there is market failure. In this case, the failure may be in the operation of competitive markets as well as monopoly. Regulation may act to correct ‘asymmetric information’ between the purchasers and suppliers of goods and services. The sphere of medicine is a classical case. Purchasers are protected from their ignorance and inability to negotiate when government lays down requirements about professional conduct, qualiﬁcations, or the quality and safety of products. Procedures may be laid down about the information supplied to consumers. Regulation of external effects: Another case for social regulation is to protect people affected indirectly by the provision (or non-provision) of services or products to others. The ‘negative externalities’ argument would apply, for example, to the polluting effects of certain industries. Regulation seeks to prevent these effects by law or to make the producer and consumer pay for them. On the other hand, government might seek to maximize the ‘positive externalities’ for society as a whole of having an educated and healthy population by ensuring national systems to guarantee access to and the quality of education and environmental health services. It might do this by regulating private providers or, where the risk of market failure is high, by directly providing a public service. Enabling policies cannot be properly classiﬁed as forms of regulation, but they have some of the same purpose in that they seek to make markets operate more efﬁciently. They might operate on both sides of the demand-supply equation: • Informing consumers about the choice, quality and price of providers;

184 The Changing Role of Government

• Facilitating the access of providers to information about market opportunities, to certiﬁcation of their standards, to credit, and to improved working practices, technology and training. In its purpose (balancing information asymmetries and increasing market entry), enablement has some similarity with social regulation and anti-trust policy, but it seeks to achieve these ends by increasing the capacity of actors to operate in the market rather than by controlling or protecting them. Government intervention is justiﬁed on the basis that the private sector will not provide these services where the market is too small or where the service has public goods characteristics (for example the diffusion of technological developments).

De-regulation and support to market development There are two linked arguments for de-regulation. The ﬁrst is the tendency for regulation, however justiﬁed in the ﬁrst instance, to lead to over-regulation at the cost of the lower proﬁtability of ﬁrms and inefﬁciency of the economy. Regulatory growth may be a product of elaboration by regulatory agencies keen to extend their role or to defend themselves from criticism for ineffectuality. Organized interests may demand additional protection and guarantees for example to control prices, restrict competition or to increase compensation. The regulation of product standards or of external effects, being more open in their motive and more in the realm of popular politics, are particularly prone to over-extension (Wildavsky, 1988; Francis, 1993). The second argument is that the very regulation that is intended to correct market failures may itself create market distortions. Even perfectly calculated rules and regulations impose new costs on producers. But regulators are far from having perfect information – they act on limited knowledge or ‘bounded rationality’. They may be ‘captured’ by vested interests and, particularly, by the very business interests that they are supposed to regulate and who have so much more information about the real costs of their operation. Clearly, the establishment of the rules of regulation is not simply a technical but also a political process involving self-seeking pressure groups and politicians (Lane, 2000). Reforms which eliminate over-elaborate and ill-directed controls can be a major source of gains in economic productivity, by reducing the burdens on ﬁrms and allowing them to be more ﬂexible (OECD, 1997; Gausch and Hahn, 1999). The OECD recommends

Regulating and Enabling the Private Sector 185

that regulations should be continuously held under review for their real effects and that, where possible, alternative approaches should be adopted. The ideal alternative is to substitute regulation with competition wherever possible. One way of achieving this is to ‘unbundle’ services and limit public intervention to the parts of the service that are indeed monopolistic. For example, the generation of electricity is potentially competitive and can be free of control, but its distribution is usually a natural monopoly. Apart from this restructuring approach, there are possibilities that public policy can seek to make or develop markets where they are weak. De-regulation aims to achieve this by removing state intervention that restrict markets.

Re-regulation under the new management Ultimately, however, liberalization does not remove the case for regulation. Indeed, the case for some sort of regulation is strengthened by ‘new public management’ approaches. If government’s role is ‘to steer, not to row the boat’ (Savas, 1987; Osborne and Gaebler, 1992) then it requires means for monitoring and guiding the rowing agencies. If there is to be a ‘separation of purchasing of public services from their provision or production’ (Foster and Plowden, 1996, p. 46), then government in the purchaser role has to have means of ensuring that the provider provides what is required. Particularly in the case of natural monopolies, privatization or the creation of armslength public agencies give the single provider immense market power that can only be countered by public regulation (Vickers and Yarrow, 1989). Hood et al. (1998) argue that regulation is central to the new management and has expanded rapidly in developed countries. At the same time, there has been some change in the nature and purpose of regulation. This is not to say that previous approaches to regulation are swept away but that, particularly in the delivery of public services, new approaches have been developed. They do not so much add to the forms of regulation listed in Table 8.1 but rather add new instruments or ways in which regulation is undertaken, moving towards a ‘lighter touch’ and more incentive-based approach. The purpose has been to release managers from compliance with detailed procedural rules and instead to measure them by results (Hood, 1995). There is a shift from a concern with regulating probity in the process of delivery to a primary concern with regulating efﬁciency and value for

186 The Changing Role of Government

money in the outcome. Instead of imposing rules of procedure on service providers, the new approach to regulation seeks to align the incentives of providers with the goals of public policy (Lane, 2000, p. 131). Typically, this is by setting performance and efﬁciency standards whose achievement by providers is rewarded in higher ﬁnancial returns, reduced taxes or charges, freedom from sanctions or greater autonomy in the management of their affairs. Mechanisms to assess this include performance audit, value for money audit and incentive price-setting. In turn, these depend on there being an independent regulator who is trusted by all sides to make a competent and disinterested judgement. The appointment of free-standing regulators is a major feature of the new approaches to regulation. Regulatory roles are a considerable challenge to the capacity of any government. Indeed, the general case for de-regulation is based partly on disbelief in the capacity of governments to regulate without introducing new distortions. There are doubts about their capacity to obtain the necessary information, to exercise critical judgement about market conditions, to exercise their role independent of self-interested pressures, and even to analyze what form of regulation is appropriate (Gausch and Hahn, 1999). Paradoxically, the new management approaches seem to require regulators to have even more expertise, sensitivity to market conditions, and understanding of the behaviour of market actors. The ‘lighter touch’ approaches to regulation may be more resource intensive. The chapter goes on to examine the performance and capacity of the study countries in undertaking alternative forms of regulation. From the list of alternative forms of regulation mentioned above, three forms of regulation will be analyzed: • Economic or ‘entry’ regulation of monopoly services in the case of urban water supply; • Social regulation of the private operators in the health sector; • Enablement of market actors through the provision of information and support services to private businesses.

Economic regulation of urban water supply1 Regulatory approaches The regulation of utilities in public or private ownership includes, on the one hand, environmental and water resource management and, on the other, the economic regulation of the monopolistic features of the

Regulating and Enabling the Private Sector 187

utility. This section deals only with the latter, where the purposes are to • protect consumers from monopoly abuse with regard to price, service quality and coverage; • reduce the risk for producers so as to encourage appropriate levels of investment. Regulating the relationship between consumers’ and producers’ interests involves the following activities: i)

licensing operators to ensure that they have the necessary knowledge, skills and capital to undertake the function; ii) price-setting to ensure that the producer can provide and invest in the service at a level that is affordable by consumers, and has incentives to improve efﬁciency; iii) performance-monitoring to ensure the fulﬁlment of contract or licence commitments to levels and quality of service provision, and to levels of price and proﬁtability that permit investment; iv) adjudication of disputes in the interpretation of agreements between government and the operators, and re-negotiation of agreements; v) sanctions for failure to comply with undertakings agreed with the regulator. Some of these functions are likely to be undertaken by government or by existing courts and competition authorities, rather than directly by economic regulators. Licensing or contracting operators and setting the policy framework of performance standards involve political decisions on which regulators may advise government. The regulator may participate in adjudication and decisions about sanctions but government departments and courts are the ultimate source of authority. The core roles of an economic regulator cluster around price-setting and monitoring. If government is the policy-maker and the negotiator of the contract or licence, who should regulate? One answer is that government as the contracting agency can regulate by monitoring the performance of the operator against the terms of the licence or contract, because the contract forms the agreed basis on which performance can be assessed and enforced. However, the case for an independent regulatory body rests on three main arguments. First, no long-term contract, such as a

188 The Changing Role of Government

concession or franchise, can be complete enough to provide the basis of all future action. Unforeseen events and new requirements require ‘just what regulation involves – a continuing task of contract monitoring, enforcement and renegotiation’ (Armstrong, Cowan and Vickers, 1994). Second, the regulator needs to be demonstrably independent of the parties to the contract – the government and the utility operator – so as to have the conﬁdence of both (and of the third party, the consumers) in its judgements. Third, the regulator needs sufﬁcient organizational autonomy to be able to build the skills and retain the knowledge required to make sophisticated and delicate judgements. In practice the division of tasks between the regulator, government as policy-maker, and the utility as provider is not clear-cut under any national system, but ‘a lot more complex and messy’ (Helm, 1999). A major question for developing countries is whether they can manage the boundaries between the three spheres and, particularly, retain the impartiality of the regulator and, therefore, the conﬁdence of investors. A second major question is whether they can manage the technical and informational requirements of alternative approaches to regulation. The two main alternative methods are outlined in Box 8.1. Each of these two approaches to price regulation requires expertise and information to assess the costs, capital requirements and cash ﬂows of an industry. For example, Price Waterhouse (1994) identify the following steps by which a regulator would set the price cap under the ‘incentive price setting’ system: • deﬁning the part of the business that is subject to regulation and identifying its business costs, investment plans and revenue projections; • estimating demand for the service at different prices; • deﬁning investment needs of the utility given its own previous investment, demand from clients and the requirements of government; • examining business efﬁciency in the achievement of previous targets and benchmarks for future targets; • establishing the formula, including the price index, efﬁciency targets and allowable costs; • identifying an acceptable rate of return, given the costs of equity capital and borrowing. Parts of this operation and potentially also of the monitoring of performance can be contracted out to consultants, but in the end the exercise is only valid if the information is available, the regulator can apply and enforce the model, and government respects the regulator’s decisions.

Regulating and Enabling the Private Sector 189

Box 8.1

Models of price regulation

Rate of Return Control is the longer established mechanism. Developed in the USA, it permits price increases provided the implied rate of return (or proﬁt) does not exceed a predetermined maximum, usually about 4–5 per cent. It involves detailed scrutiny of the company’s operating and investment costs and consideration by the regulator every time a tariff revision is requested. Excess proﬁts may be ‘clawed back’ by tariff reductions. As this model allows companies to recover allowed cost increases, it gives investors a degree of certainty about the returns they can expect. From the point of view of government and consumers, however, it has a serious handicap in that there are no in-built incentives to reduce costs and raise efﬁciency. Indeed, it may encourage them to raise costs in order to justify price increases. Incentive Price Setting systems, developed principally in the UK, aim to provide incentives by allowing ﬁrms to retain any proﬁts if they out-perform expectations by achieving greater efﬁciency. Prices are controlled by reference to an efﬁcient ﬁrm’s costs adjusted by reference to increases in a general price index, such as the Retail Price Index (RPI). In the British case, a price cap is established that is based on the RPI minus an efﬁciency target (X), plus adjustments for the cost of service quality improvements demanded by government. Every four or ﬁve years, X is adjusted to reﬂect increasingly rigorous efﬁciency targets. Any increase in efﬁciency beyond X is appropriated by the ﬁrm as proﬁt, thus providing the regulator with information about the appropriate level for future targets. Source: Price Waterhouse, 1994 and Nickson and Franceys, 2003.

Regulatory performance There are moves towards economic regulation of urban water supply in most of the countries studied, but experience is still very limited. After a brief survey of this experience from the less to the more reformed examples, the case of Buenos Aires will be referred to as one of the longest established systems in a middle income country. Government regulation of publicly owned suppliers The typical starting point, with almost no effective economic regulation, is the situation as described by Price Waterhouse for Kenya in 1994. Tariffs were set by the Ministry of Water Development but without any formal procedure, except political judgement of what was considered acceptable to the public. Tariffs did not reﬂect costs and were not adjusted systematically for inﬂation. Investment for expansion of the service was therefore funded not from revenue but from the periodic allocation of treasury or donor funds. The quality of water

190 The Changing Role of Government

supplies was set according to World Health Organization standards and formally applied by the Ministry of Water Development and the Ministry of Health ‘but enforcement is not always effective’ (Price Waterhouse, 1994). By comparison, the Zimbabwe case indicates at least the will to move towards a form of price regulation that recognizes the need for viable tariffs as a condition for investment. Water is distributed by the urban local authorities but tariffs are regulated by the central Ministry of Local Government, Rural and Urban Development. Until the deepening of the economic chaos from 2000, the ministry was relaxing previously strict price controls and releasing local authorities to charge economic tariffs. However, consumer tariffs remained subject to ministerial control and, as inﬂation accelerated, anti-economic restriction of tariffs was restored. Likewise, in spite of proposals to create an autonomous national water authority (ZINWA) for water management outside direct ministerial control, the setting of bulk water tariffs charged to local authorities was kept under the control of ministers (Batley, 1998). The Government of India has plans to subject state and local government water boards to regulatory bodies that would be ‘independent, autonomous and accountable and include consumer participation’ (Kapoor, 2001). An equivalent system of electricity regulators is already in existence, operating an American-style, rate of return regulation where decisions have to be justiﬁed in a court of law. Although the courts have defended the autonomy of the regulators, in practice the regulatory bodies operate with little independence from the state governments that control the electricity enterprises. The indication from these three cases is that where government is itself the regulator, regulation will easily be politicized and distorted. Independent regulation before privatization This section refers to three countries – Sri Lanka, Ghana and Zambia – that are setting up independent regulatory bodies in situations where the service continues to be provided by a public sector operator, but where there is an intention to involve private investors and operators through concession or lease contracts. The Government of Sri Lanka is establishing a series of regulatory commissions for each of the utility sectors including water and sanitation. The government’s plan for urban water is to achieve universal service coverage and, at the same time, ﬁnancial viability through full cost recovery. The regulatory commission for water is being set up with the advice of the Asian

Regulating and Enabling the Private Sector 191

Development Bank but is not yet operational (De Silva, 2001). Funded by levies on water charges, the water regulator is to come under a small body of three independent commissioners with responsibility through a director general for economic regulation, water quality and consumer relations (Fernando, 2001). The regulator’s role is to advise government on policy, license service providers, approve contracts for private operators, review and approve tariffs, set and monitor service quality and performance indicators, promote consumer rights and undertake dispute resolution. In Zambia and Ghana, regulators already operate. The Ghanaian Government established an independent Public Utilities Regulatory Commission in 1997 to regulate tariffs and customer service standards in the electricity and water sectors. Even before the establishment of lease contracts for water supply, PURC has begun to exercise its powers by authorizing tariff increases by the public sector Ghana Water Company to ensure that revenue is sufﬁcient to cover costs. In return for the increases, PURC has required improvements in the water company’s service quality and billing. A users’ committee reinforces PURC’s access to service quality information. The research conclusion was that these regulators were already having positive effects in pressing for improved services and for tariffs that reﬂect real costs (Nickson and Franceys, 2003). They were able to retain some autonomy from service providers, at least as long as international funding agencies were in the wings. Independent regulation of private operators Independent regulators have been established for most cases where private sector operators have become involved. The exceptions are the earlier concessions in, for example, Macau and the French West African countries, which adopted the French model where regulation is managed through the contract. In some cases, for example Jakarta (Indonesia) and Cordoba (Argentina), the regulator was created several years after the concession had been granted, in recognition of the weakness of using the contract alone as the basis of regulation. However, most independent regulators were established well in advance of privatization, allowing the existence of the regulator to be embedded into the contract with the private company – for example, in Mexico City and Manila, and La Paz, Bolivia. The same applied in Chile, the only country apart from England and Wales to adopt the full privatization of the ownership (by divestiture) of urban water supply systems.

192 The Changing Role of Government

The Buenos Aires concession (described in Box 7.3, Chapter 7) was the ﬁrst of the new wave of concessions. In this case the regulatory arrangement was established in 1993 at the same time as the concession contract was being negotiated. The relationships between government, the private company and the regulator have therefore evolved through experience. Box 8.2 describes the experience of regulation in Buenos Aires, referring to the ‘boundary’ problems that have occurred between government, the regulator and the private company. In particular, the regulator has frequently been bypassed by political interventions. Nevertheless, the research concluded that, in spite of the weaknesses of the regulatory system, it had produced some major beneﬁts. First, it made decisions about tariffs and service extensions more transparent than they ever had been before. Second, it gave

Box 8.2

Boundary problems in water regulation in Buenos Aires

The tasks of ETOSS, the water regulator for the Buenos Aires concession, are to review the tariff structure, oversee contract compliance, control and maintain service compliance, protect customers, control the quality of drinking water, and approve and supervise the execution of expansion plans. It is managed by six board directors representing each of the federal, provincial and municipal governments. They have six-year contracts in order to minimize their subjection to external pressures, but in practice membership has changed as party control of the different levels of government has changed. Two ‘boundary’ problems have emerged in practice. Firstly, there have been disputes with the company (Aguas Argentinas) about the degree to which the regulator should intervene in management decisions. The contract speciﬁes output targets, but in addition it indicates the regulator’s role in monitoring the fulﬁlment of investment plans and the maintenance and renovation of equipment. ETOSS’s staff, drawn from the old public company and with engineering backgrounds, were inclined to intervene more in these procedural issues than the company thought appropriate. Secondly, the national government overrode decisions of the ETOSS directors in a series of negotiations with Aguas Argentinas about service extensions and revised tariffs. In 1994, the government pressed for extension of water services and allowed large increases in tariffs and connection charges. In 1997, it bypassed ETOSS to agree to a universal service charge to subsidize connection charges for new customers. In 1998, it intervened in negotiations between ETOSS and the company about tariff revisions, proposing a compromise. Given that the board members were political appointees, pressure could be applied to them to back down. Source: Based on Nickson and Franceys (2003).

Regulating and Enabling the Private Sector 193

government a means to set targets, monitor progress and negotiate. Third, it opened these decisions to public information and debate (Nickson and Franceys, 2003). Regulatory capacity Fully effective regulation requires expertise, organization and institutions that are more typical of the circumstances of developed than developing countries: If regulation is difﬁcult for rich countries, it will be desperately hard for poor ones. Part of the problem is cultural: not many developing countries have good ant-trust regulation, or are used to the idea of regulating to promote competition rather than to restrain it. If you need to create a body with autonomy and expertise, you are quickly talking about 50 professionals…. That is a big problem for countries with limited human and institutional resources (The Economist, 1997) The cases indicate that the internal organizational requirements of a regulatory body are demanding. The constraints included: • Problems in establishing an organization with a clear mandate, speciﬁed duties and powers, with autonomy to take decisions and respected in their implementation. Typically, these conditions exist in law but not fully in practice as illustrated by the Buenos Aires case; • Lack of staff with expertise in accounting, economics and ﬁnance. Typically regulatory bodies have been staffed by administrators from government or by staff with an engineering background from the old public company. The new skills are in short supply and salaries that will attract and retain are unlikely to be paid in the public sector. These were constraints even in Chile, a middle income country with long experience in privatization (Bitran and Serra, 1996). Some of these skills can be contracted for speciﬁc tasks, as in the British case, but there also has to be a core competence; • Problems of access to information to make judgements about price, proﬁtability and performance. Regulation of prices requires a high quality of information about the economy and about the ﬁrm being regulated. This is difﬁcult to obtain where accounting standards and economic indices are weak and, particularly, in conditions of economic instability (Price Waterhouse, 1994).

194 The Changing Role of Government

The institutional conditions external to the regulatory organization but necessary to it are even more exacting, particularly in the context of developing countries. Typical problems were: • Instability of the political economy: Developing countries are, by deﬁnition, high-risk settings where the regulator and the regulated ﬁrm need protection against economic and political instability. Within limits, the regulator may make technical judgements to accommodate economic instability by adjusting for devaluation or allowing higher levels of proﬁtability where the risk is high (Price Waterhouse, 1994). Political instability and intervention are more intractable, as the Buenos Aires case indicated; • Difﬁculty in establishing a leadership and organization that is independent of political control, disinterested between the interests of the government and operator, and also not self-serving. In the Buenos Aires case, ‘political autonomy’ was sought by establishing a political balance on the board. A danger, identiﬁed in India, was of ex-civil servants taking on the function and acting as an extension of the government service whilst also not being subject to political accountability; • Problems in maintaining separation from political inﬂuence: Maintaining the autonomy of the regulator from political inﬂuence is difﬁcult in any situation, including in the British case (Helm, 1999; Flynn, 1997, p. 168). Keeping that distance in less transparent political systems is unlikely; • Weak demands of civil society: The demands of the media, consumer associations, NGOs, academics and independent researchers can act to shield the regulator from interference by government (Nickson and Franceys, 2003). These are more likely to exist in middle and high-income countries, though the regulators in Ghana and Argentina showed how user committees could play this role; • Doubts about the independence of the judiciary: For the enforcement of contracts and the ﬁndings of regulators, the ultimate resort is to an impartial and accessible legal system. Investors will lack conﬁdence in a regulator unsupported by an independent judiciary and regime of contract law (Price Waterhouse, 1994); • The imbalance of knowledge, experience and power with the ﬁrm being regulated: This is particularly a problem where a regulator with little experience or basis of comparison confronts an international private operator with global experience, greater information and capacity to negotiate. Information asymmetry leads to uncertainty ‘whether the

Regulating and Enabling the Private Sector 195

price was reasonable or the private operator was beneﬁting unduly’ (Nickson and Franceys, 2003). It is cited as a problem even in Chile which has relatively well-established capacity (Bitran and Serra, 1996). However, the cases indicate that gains from regulation are possible, even where the organizational and institutional conditions are not perfect. Any separation of the role of regulator from those of policymaker and provider helps to hold operators to account for the achievement of performance, to de-politicize decisions about tariff levels, and to reduce the risks for private investors. The condition for greater effectiveness is that the regulator has at least some autonomy both from political control and from the water supply company, even where the latter is publicly owned as in Ghana. The clearest separation was where the regulator was outside the apparatus of government and the company was privately managed; such conditions have so far been met primarily in middle and upper income countries rather than the poorest which were the main focus of this research.

The social regulation of health services2 Regulatory approaches There are two main theoretical reasons why public authority would intervene in the relationship between the consumers and providers of personal health care.3 The ﬁrst is to do with the informational asymmetry or imbalance that is typical of professional services. Consumers may lack the information on which to make judgements and assert control over their ‘agents’ (the providers) so as to obtain appropriate, good quality and efﬁcient care. The second reason is that private providers will tend to focus their attention on the consumers who will pay and on the services that can most easily be measured and charged, rather than on the achievement of allocative efﬁciency and equity in access to services. Policy responses to these speciﬁc problems may be to enable more providers to enter the market and give them incentives to fulﬁl public policy goals, to provide information to the consumers about their choices, or to regulate the health providers. This section focuses on the regulation of formal non-state organizations (hospitals and nursing homes), of pharmaceutical production and drugs control, and of the professional staff within public or private health systems. Private health care delivery has grown since the 1980s in all the countries under study – India, Sri Lanka, Thailand, Ghana and

196 The Changing Role of Government

Zimbabwe. The research found that this was generally not due to attempts by governments to promote the participation of the private sector, but principally to broad macro-economic factors. In the Asian countries and particularly in Thailand, growing private incomes had led to increased capacity to pay for private services and disenchantment with the quality and availability of public hospitals. In India, this was compounded by the fact that frequently informal payments had to be made to get access to services even within the public sector (Bennett and Muraleedharan, 1998; Bennett et al., 1998). In the African countries the situation was typically the converse; macro-economic stagnation or decline had resulted in the under-funding of public health services and the loss of public conﬁdence in them. This had led patients increasingly to resort to private providers in Zimbabwe, and to traditional or informal providers in Ghana. A comprehensive survey of public attitudes makes the same point for Nigeria (Ayoola et al., 2000). In both continents, the non-state health sector is a resort for both the relatively rich and the poor. For the rich, the ‘organized’ private sector is an alternative to poor public health provision; for the poor, the ‘unorganized’ informal sector may be an alternative to lack of access to public provision (Mills et al., 2002; Bloom and Standing, 2001). By the early 1990s, private exceeded public health expenditure by two or three times in India, Thailand and Zimbabwe, but was much lower in Sri Lanka and Ghana where government doctors had been banned from private practice, at least until recently (Mills et al., 2001, p. 30). However, governments have only slowly recognized and reacted to the growth of private sector provision: In the Asian countries, it appeared that the Ministries of Health had for many years been positively uninterested in regulatory issues and policy on the private sector. Consequently, private sector expansion had occurred in a policy vacuum with no concurrent reforms to ensure adequate quality. (Mills et al., 2001, p. 149) Governmental recognition of the need to revise regulatory arrangements came late in the day, usually after the private sector had consolidated its expansion and was able to resist change. The case for reform was (therefore) often framed in the liberal language of ‘partnership and coordination’ rather than regulation. Donors also came rather late to a recognition of the need for more control of private practice, and ‘concrete proposals on how best to strengthen regulation are largely absent’ (Mills et al., 2001, p. 144).

Regulating and Enabling the Private Sector 197

This may be partly because the regulation of private actors in the health sector presents a more complex and contextually speciﬁc set of problems than economic regulation of the water sector. Fundamentally, it is a sector predominated by people rather than by infrastructure, leading to much greater problems of monitoring, control and motivation. By comparison with the monopoly of piped urban water, in the health sector there are multiple small-scale providers offering a wide range of health products and services. Moreover, there is no neat division between the public and private sectors, but an intermeshing of public and private, formal and informal providers across what Bloom and Standing (2001) describe as the organized and unorganized health care economies. Public doctors may also practise privately and refer patients from one to the other realm; publicly purchased drugs may seep out through private markets; medical councils may both regulate the profession and also informally represent its interests. The way that these institutional arrangements have evolved varies from country to country. For example, in Ghana public doctors were still prohibited from working in the private sector which had become a refuge for poorly qualiﬁed doctors. In the Asian countries, most doctors in private practice also held public posts, and in Sri Lanka even worked in private ‘channelling clinics’ where they referred to themselves in public or private hospitals. The problems and possibilities of regulation clearly vary depending on the nature of these public-private relationships, affecting the degree of trust, information and common interest between supposedly distinct government and private bodies. The regulation of the health care sector is complex for other reasons than the multiplicity of services and providers, and the way that their interests may overlap with those of government and regulatory bodies. First, it is concerned with difﬁcult-to-measure issues of the quality and quantity of services, where information is difﬁcult to collect and regulation is a matter of judgement, for example: • Creating and implementing laws requiring registration of providers including clinics, hospitals, pharmacies, diagnostic centres; • Setting and applying standards required of providers; • Prosecuting providers in case of negligence or malpractice; • Supporting and monitoring professional regulatory bodies, such as medical councils; • Regulating the standards of medical schools.

198 The Changing Role of Government

Second, the organizational arrangements for regulation are themselves, typically, complex, reﬂecting the multiplicity and diversity of the sector, with its professional specialisms. Table 8.2 exempliﬁes the case of the state of Tamil Nadu, India, showing how regulatory functions are shared across government, statutory bodies, the judiciary and civil organizations. Regulatory performance The key distinction to be made here is between the existence of quite elaborate regulatory frameworks and their weak implementation. All the study countries had regulatory frameworks governing the standards of medical facilities, the professions permitted to practise different aspects of medicine, and the importation, manufacture, wholesale and retail of drugs. The cases of Sri Lanka and Zimbabwe are presented as examples in Table 8.3. As also for Ghana and India, they indicate a pattern of quite comprehensive legislation, but much of it dating back to pre-Independence times with little amendment until very recently. India was exceptional in the weakness of its regulation of the private health sector. Until 1997, only three states (Maharastra, Karnataka and the Union Territory of Delhi) had legislation requiring the registration of private hospitals and allowing for their inspection. A meeting of the Central Council of Health and Welfare in 1997 urged all states to pass such legislation but this was only slowly being implemented. On the other hand, India was also exceptional in that it had Table 8.2

Actors involved in regulation in Tamil Nadu, India Types of actor

Government

Statutory regulatory body

Legal

Civil society

State Department of Health

Indian Medicine Council Pharmacy Council

Judiciary Consumer Protection Courts

Consumer organizations

Implementing agencies – e.g. Drugs Control Authority

Tamil Nadu Medical Council Tamil Nadu Nursing Council Tamil Nadu Homeopathic Council

Source: Based on Bennett and Muraleedharan, 1998.

Nursing Homes Regulation Act 1949, Ayurveda Act 1961, Homeopathy Act 1970 allow for registration and quality monitoring of facilities New legislation: Private Medical Institutions Act for all private medical institutions

Medical Dental and Allied Professions Act 1971 (plus later amendments) provides for regulatory authority, compulsory registration of facilities with beds and structural standards New legislation: Medical Services Act expands MOH’s powers to regulate prices of state-aided hospitals, and referrals between public, private and non-proﬁt organizations

Sri Lanka

Zimbabwe

Source: Mills et al., 2001, pp. 154 and 155.

Governing hospitals and nursing homes

Medical Dental and Allied Professions Act (plus later amendments) provides for establishment of Health Professions Council, registration of professionals, monitoring and control of practice standards

Medical Ordinances Act, Ayurveda Act and Homeopathy Act allow for establishment of medical councils, control of professional education and registration for each profession

Governing individuals

Key regulatory statutes in Sri Lanka and Zimbabwe

Country

Table 8.3

Drugs and Allied Substances Act 1969 establishes Drug Control Council for drug registration, licensing, control and inspection of pharmacies

Cosmetic Devices and Drugs Act 1980/4 establishes licensing and quality controls Opium and Dangerous Drugs Act prohibits dangerous drugs and sets standards Consumer Protection Act safeguards public from unscrupulous sales

Governing pharmaceuticals

199

200 The Changing Role of Government

brought fee-paid medical services under the Consumer Protection Act, making them subject to consumer courts; this extension (beginning in 1992) had been done by decision of the judiciary and not the government. Although there were weaknesses in these regulatory frameworks, the main problem was in their implementation. Failures occurred both in government ministries of health and in professional councils. The transfer of responsibility for the execution of regulatory statutes to professional councils (of doctors, nurses etc.) was normal practice in all the countries. This amounted to self-regulation under the law. Almost universally, Medical and professional councils had tended to adopt very passive approaches to regulation, failing to inform the general public of their role and waiting for consumers to bring complaints to them. (Mills et al., 2001, p. 160) They played a minimal role in controlling poor practice. Only one case had been heard by the Ghana Medical and Dental Council in the previous 10 years; in Tamil Nadu, India, only one doctor had been struck off the list in ten years. Inactivity, professional protectionism, personal favours and corrupt approval of licences were widespread accusations of state and national medical councils in India. The one exception to the poor performance of professional councils as regulators was the Nurses and Midwives Council of Ghana, which had responsibility for registration standards, and curriculum development and examinations in training institutions. This council and the wider membership had a strong reputation for their motivation, collective pride and proactivity (Smithson et al., 1997). Underlying the poor performance of almost all the professional councils were absent or ineffectual supervision and monitoring by ministries of health. The professional councils failed to self-regulate their members because they themselves were not regulated by governments. Box 8.3 illustrates the weakness of government implementation in Zimbabwe, which was not the worst case. Three main regulatory failures on the part of governments were identiﬁed. Most basically, except in the case of Thailand, the databases of information on which regulatory action depended were not maintained; in none of the four core countries were there data on the numbers, location and characteristics of private health care providers. Secondly, while the legislation speciﬁed that ministries of health would set standards against which

Regulating and Enabling the Private Sector 201

Box 8.3

Ineffectual regulation of private health providers in Zimbabwe

A study reviewed the operation of ﬁve pieces of legislation thought to be most relevant to private/for proﬁt providers in Zimbabwe: • • • • •

The Medical, Dental and Allied Professions Act which covers the control of entry and quality of services by individuals and institutions; The Public Health Act which covers activities and services that relate to communicable diseases; The Traditional Medical Practitioners Act which provides for registration and discipline of traditional healers; The Drugs and Allied Substances Control Act which controls the pharmaceutical market; The Dangerous Drugs Act which provides further controls on the market for dangerous drugs.

It was found that knowledge of regulations was reasonable. However, there was consensus among health sector stakeholders that regulations were ineffective and unenforced. Practices running counter to regulations such as employment of unlicensed workers, unnecessary procedures and self-referrals to own laboratories and surgeries were identiﬁed. The problems were widely attributed to the failure of inspection systems. Resources for inspection systems were scarce, relative to the scale of the inspection task. For example, the Health Professionals Council which is responsible for enforcing the Medical, Dental and Allied Professions Act is staffed by one registrar and six or eight supporting staff in Harare. The integrity of inspection systems was also questioned, the supervision of the inspectors being deemed inadequate. There is a dominance of the medical profession among the regulators, potentially an instance of ‘regulatory capture’ by which the regulated exert undue inﬂuence on the regulatory process. Other problems were identiﬁed in regulatory design such as outdated legislation …; inadequate price and quality regulatory mechanisms; and unclear procedures for processing consumer complaints. Source: Hongoro and Kumaranayake, 2000; in Mills et al., 2002.

the performance of private providers would be assessed, frequently the standards had not been formulated. Thirdly, even where standards had been set, inspections were rarely carried out routinely to ensure that the standards were applied. The research concluded that the underperformance of public sector regulation in its turn allowed private operators to provide care of an unacceptably low standard. Regulatory capacity The problems of capacity to undertake regulation of health care were often similar but sometimes distinct from those of economic regulation

202 The Changing Role of Government

in the water sector. Some of the differences derive from the greater complexity of the health sector and of its regulatory arrangements. Numerous service delivery organizations lead to greater requirements of ﬁeld level regulation and to greater problems of stafﬁng, coordination and inter-penetration of public and private interests. Underlying the problems of internal organizational capacity is a gap between relatively elaborate policy and legislation and the inadequacy of organizational arrangements for implementation. Russell, Bennett and Mills (1999) attribute this partly to the exclusion of lower level ofﬁcials responsible for implementation from the élite political and ofﬁcial circles that make the policy. Moreover, as noted earlier, regulation has only recently been perceived as a priority function for government. The result is that inadequate staff and funds are allocated to the tasks of registration, inspection, monitoring and coordination of providers.4 For example, in Tamil Nadu in 1996, 87 drug inspectors were responsible for 2800 manufacturers, 11,500 licensed sales outlets and a probably larger number of unlicensed outlets. In Sri Lanka, there was ‘no unit or staff within the ministry of health responsible for registration, monitoring or co-ordination of private providers…’ (Russell and Attanayake, 1997). As was found in the case of health sector contracting, another basic impediment to effective regulation was the weakness of systems for information collection and record-keeping in government. In the case of health care, this was also a problem among private providers. Among the professional medical councils that often had the main direct responsibility for exercising regulatory powers, there were also often problems of under-resourcing and under-stafﬁng. For example, in Zimbabwe and some states in India, councils complained of underfunding and dependence on government for inadequate grants. But even where, as in Sri Lanka, the medical council had its own source of funds from fees paid by members, regulatory performance remained weak. The research found that the root cause was in the ‘inertia, lack of motivation and self-interest’ of the councils (Mills et al., 2001, p. 165). Their culture was generally one of professional protectionism. Only in a few cases – the nursing council in Ghana and the medical council of Sri Lanka – had they reinforced an ethic of high standard medical practice and public service. Coordination between actors presented particular problems of capacity in the health sector. Communication was often weak between ministries of health and medical councils, either because powers were not exercised or, as in Tamil Nadu, where the medical council had no

Regulating and Enabling the Private Sector 203

responsibility to report its operations to the state department of health. There were also often problems of coordination between levels of government where there had been decentralization of some responsibilities without any clear organizational arrangements for reporting and control. In Sri Lanka, after decentralization, responsibility for implementing the Nursing Homes Act was transferred to provinces without any new guidance. In Zimbabwe, responsibility for registering and inspecting private health institutions was transferred to the central Health Professions Council while organizational capacity remained at local level, until an accommodation was found between them. External capacity constraints to effective regulation arose from the power of private medical interests and their overlap with the public sector. Private sector interests among health care providers and professionals tended to oppose tighter regulation and could exercise their inﬂuence by their presence within government as politicians and ofﬁcials. They could exercise their inﬂuence not only within the government machine but also as business or professional pressures upon it. For example, the Sri Lankan Government Medical Ofﬁcers Association called government consultants out on strike against a proposal that doctors choose between working in the public or private sectors. Only in Ghana was the private medical sector not linked to public practice and not a powerful lobby. Only in India and, to a more limited extent, Zimbabwe was there anything like a counter-lobby for stronger regulation, based on a strong consumer movement and consumer protection law.

Enablement of private providers – business support services Approaches to enablement The forms of regulation already described are ‘market-friendly’ to the extent that they intervene to correct market failures or inefﬁciencies rather than to replace market motivations with state planning. They seek to compensate for consumers’ lack of choice under monopoly conditions (in the case of water) or lack of information to make choices (in the case of health), to protect investors from the risk of political interference, and to create a framework of standards within which the private sector can compete.5 They do so by setting the structure of rules under which consumers and producers interact. By contrast, enabling policies imply positive state intervention to advance the capacity of consumers or providers to participate in the market, to make and to exercise choices.

204 The Changing Role of Government

On the consumer side, this might include policies to • Increase information on the choice, quality and price of services and providers; • Support consumer organizations to demand responsive services; • Increase the capacity of consumers to pay for services, for example through subsidies to consumers or incentives to producers. On the provider side, policies might seek to • Improve working practices, technology and training; • Increase information about market opportunities and prices; • Offer certiﬁcation or accreditation of the quality of products or providers; • Give access to resources and credit, subsidies or tax exemptions. The research focused on the relationship between government and the private sector in the provision of support by public agencies to private organizations. These business support services might be underprovided, if they were left to the market. The market will fail to provide such ‘enabling’ services if there is little hope of gaining any return. Information services are classically a case where it is difﬁcult (and undesirable in the general interest) to restrict the service to those who pay; ‘free-riders’ can easily get access to information that others have paid for. Tax exemptions (and perhaps subsidies) are clearly also resources that only government can provide. The other services mentioned above – training, consultancy advice, accreditation – are more likely to be chargeable by market providers. However, even in these cases, the market may be too small or the returns too low and uncertain to attract providers of support services. The economic case for government to take on the role of providing these services is that there are wide social beneﬁts (externalities – see Box 1.1) from the development of a more competitive and betterinformed private sector. Access to information, technology and resources increases the opportunity for producers to assess risks and opportunities and to enter the market. In their turn, consumers beneﬁt from increased choice and competition, and there are potentially wide beneﬁts in the increase in economic activity. The performance of business support services The adoption by public agencies of enabling roles has probably gone furthest in the industrial sector where government has most retreated

Regulating and Enabling the Private Sector 205

from ownership and regulation of production. In place of their previous direct role in production, public agencies have often adopted new roles of supporting private producers. The research examined the provision of business support by government to ﬁrms in the textile and clothing industries of Ghana, Zimbabwe, Sri Lanka and India. However, most of the agencies examined had a broad brief to cover industry as a whole. A large number of public agencies exist to offer business support services in the four countries. The research focused on a sample of agencies, selected for their importance to the textiles sector. They can be broadly differentiated by the way that they had come into existence and by the type of service that they offered. Some had adapted from a previous role or organizational structure; others were created to take on business support roles: 1. Modiﬁcation of role without changing organizational structure: the Ministry of Industry and Commerce, Ghana. 2. Modiﬁcation of role and structure: the Ghana Investment Promotion Centre, the Zimbabwe Investment Centre, the Indian Council of Scientiﬁc and Industrial Research, the Zimbabwe Industrial Development Corporation. 3. Creation of new organizations to support private ﬁrms: the Sri Lanka Export Development Board, Board of Investment, Textile Training and Services Centre and the Clothing Industry Training Institute; the Indian Apparel Export Promotion Council; and Zimtrade of Zimbabwe. In the ﬁrst category, the Ghanaian Ministry of Commerce and Industry has been required by the government to shift its focus from import licensing to introducing ‘market-oriented policies which will spur businesses to adopt efﬁcient management and marketing practices’ (Republic of Ghana, 1993). But it has to do so with all its old structure and personnel. This makes it different from ministries of industry in many other countries where, as part of the liberalization process, ministries have often created separate agencies to give business support services to ﬁrms. In the second category are old-established agencies that had become part of the apparatus of state regulation and ownership but had now turned to a more arms-length relationship with government and to a modiﬁed role of supporting private industry. Their move towards a support role for the private sector had required that they distance themselves from direct control by government. In the third category

206 The Changing Role of Government

are agencies founded mainly in the 1980s and 1990s with the speciﬁc purpose of promoting industrial capacity, trade and investment. Table 8.4 shows that, except for the Ministry of Industry and Commerce in Ghana, all these agencies have some level of autonomy from the regular civil service through their company, society or agency status and the fact that they report directly to ministers or to the president. They each stand between core government and the private sector but at different ends of the spectrum on different dimensions. Several have their own sources of ﬁnance, giving them independence from budgetary allocations. Most have some sort of structural relationship with the private sector, not just in the sense that they offer services to ﬁrms, but through ﬁrms’ representation on their boards, through their membership, or through the payment of fees for services. The most extreme case is the Industrial Development Corporation of Zimbabwe which is completely ﬁnancially autonomous through its earnings from holdings in private companies. The roles that the agencies perform can be broadly classiﬁed as follows: 1. Facilitation of ﬁrms’ internal working practices, skills, research and technology: • the Sri Lankan Textile Training and Services Centre (TTSC) and the Clothing Industry Training Institute (CTTI); • the Zimbabwe Industrial Development Corporation (IDC); • the Indian Council of Scientiﬁc and Industrial Research (CSIR); • the Ghana Ministry of Industry and Commerce (MIC). 2. Promotion of ﬁrms’ awareness of opportunities for investment and export, supporting their marketing, and organizing trade missions: • the Sri Lanka Board of Investment (BOI) and Export Development Board (EDB); • the Zimbabwe Investment Centre (ZIC) and Zimtrade; • the Indian Apparel Export Promotion Council (AEPC); • the Ghana Investment Promotion Centre (GIPC) and the Ministry of Industry and Commerce (MIC). Facilitation agencies In the ﬁrst group, the Ghana Ministry of Industry and Commerce has found it most difﬁcult to adjust to new roles. Not only did it retain its pre-existing staff and structure, but it also had, at the same time, both to dismantle its old import licensing functions and deﬁne a new set of

Govt agency Govt agency Govt agency Govt agency

MIC GIPC

AEPC CSIR

BOI

EDB TTSC CITI

Ghana

India

Sri Lanka

Company Society

Ministry Govt agency

IDC Company ZIC Govt agency Zimtrade Joint venture

Zimbabwe

Status

Agency

Reports to president under Ministry Ministry of Enterprise Development, Industrial Policy and Investment Promotion

Ministry of Textiles Ministry of Science and Technology

Minister President appoints board

Owned by government Ministry of Finance Reports to president

Relationship with government

Medium – rents, fees and service charges None Medium – fees Medium – fees

High – membership fees Medium – fees

None None

Very high – earnings None High – donor funding and export levies

Level of ﬁnancial autonomy from government

The relationship of business support services with government and the private sector

Country

Table 8.4

Representation on board Service fees and representation on boards

Representation on board

Consultation. Membership Service fees. Membership

None Representation on board

Holdings in private ﬁrms None Levies. Half board appointed by private sector

Involvement of private sector – apart from as service recipients

207

208 The Changing Role of Government

market-oriented activities. It was successful in reducing the number of trade restrictions affecting industry, but assumed wide responsibilities that elsewhere were assigned to separate speciﬁc agencies: policymaking, strategic planning, collecting information on industrial activity, ﬁnancial assistance to ailing industries, developing an export zone, and promoting inward investment. It has had to do this under the severe ﬁnancial constraint of budgetary allocations that were enough to do little more than pay the ministry’s wage bill, and with just 200 staff, less than some of the specialized agencies of other countries. Among the specialized ‘facilitation’ agencies, the two Sri Lankan training and advisory agencies were both successful in developing a range of fee-paid services. Their fee-base, accounting for about half their recurrent income, gave them an increasing level of autonomy that allowed them to apply some of these resources to paying for higher quality staff than would have been possible within civil service rates. Their success in selling services also proved their utility to the textiles and clothing sectors, although there was criticism by the private sector that their services did not sufﬁciently reﬂect industrial needs. The Indian CSIR has also adjusted from an earlier disregard for the commercial value of scientiﬁc and technological development to a much more direct concern with the commercial relevance of its work to industry in general (Box 8.4). Its capacity to raise independent resources through payment for its services is seen by CSIR as one measure of its relevance to the needs of Indian industry. The Zimbabwean Industrial Development Corporation (IDC) shared with the Indian CSIR some of its original functions – to promote industrialization through research and development. Born in 1963 under the post-colonial white régime, it became an instrument for overcoming international sanctions against foreign investment and imports. New industries were to be ‘incubated’, then sold off to the private sector and the proceeds used to fund a further round of investment. In the postindependence period from 1980, similar policies were maintained in the name of state-led national development. Later, IDC was also used to take over failing private companies so as to revive them for resale. The record of IDC is of great success in developing, reviving and running companies. It became a fully self-ﬁnanced organization, generating income from the proﬁts of the ﬁrms that it ran, paying its staff private sector salaries and passing some surplus to government. However, in the process it can be argued that its motivation became distorted. Effectively, it became a holding company with a principal

Regulating and Enabling the Private Sector 209

Box 8.4

The Council for Scientiﬁc and Industrial Research, India

CSIR promotes and coordinates industrial research, establishes and develops institutions for particular industries, and disseminates research. It has a network of 40 laboratories and 80 ﬁeld centres across India, and employs around 21,000 staff, of whom 15,000 are qualiﬁed technicians and scientists, including more than 3000 PhDs. Its ‘2001 Vision and Strategy’ aims to link its research to the market, to increase the generation of funding from private clients and to re-structure the organization. CSIR has introduced a performance-related budget system and a management information system to measure ﬁnancial and qualitative targets. Control over externally generated ﬁnance has been devolved to its operating units, encouraging them to start their own development programmes. CSIR has established a set of key indicators of its performance, related to its contribution to science and technological output, human resource development, contribution to the national economy and its own revenue mobilization. During the early 1990s, it maintained an average annual increase of between 14 and 20 per cent in the number of patents ﬁled. In 2001/2, 410 patents were ﬁled in India and 580 abroad, including 100 in the USA. In the same year, the estimated value of the contribution made by CSIR’s research and development to the national economy was $1.15 bn. About 30 per cent of CSIR’s gross income was derived from contracted research and services. The remainder came from budgeted or special allocations from government. Source: Slater et al., 1998 and www.csir.res.in/Proﬁle.htm

interest in retaining successful companies rather than in developing national industry: In a sense the IDC has enjoyed too much autonomy since it has been able to retain successful companies in its own interest rather than returning them to the private sector. (Jackson, 2002, p. 125) Promotion agencies The second group of support agencies was concerned with providing information and advice to exporters and investors, supporting overseas marketing and organizing trade missions. There was a conundrum in assessing their performance: some of the most active promotion agencies operated in the less successful economies, and the more successful economies – India and Sri Lanka – sometimes had the weaker agencies. The link is that in the stronger economies such public agencies may be superﬂuous because ﬁrms have their own connections or can rely on private service providers.

210 The Changing Role of Government

The Indian Apparel Export Promotion Council is a membership organization providing information services to exporters. It has a high degree of ﬁnancial autonomy based on members’ fees. However, it reports to the Ministry of Textiles, is directed by a senior civil servant and has the character of a large, hierarchic government department. With 600 staff across nine ofﬁces, it is led by an upper cadre of 108 executives, whose promotion has depended to a high degree on seniority. It has the reputation of being slow to react to the changing opportunities presented by liberalization (Slater et al., 1998). In Sri Lanka, the Export Development Board (EDB) also reports to the minister but has a relatively high degree of operational autonomy in deciding its own work programme and in managing its staff and ﬁnancial resources. However, it depends entirely on government grants, which have been limited by the government’s programme of ﬁscal constraint. Policy direction has been weak and EDB’s workload has declined. Stafﬁng levels, on the other hand, have remained constant leading to poor promotion prospects and low morale. In spite of EDB’s lack of impetus in promoting exports by domestic companies, export industries in Sri Lanka have boomed. This has been driven principally by the investments of overseas ﬁrms rather than by the development of indigenous exports. It is the Sri Lankan Board of Investment’s role to promote inward investment including in public-private infrastructure development projects. The BOI offers grants, tax holidays, exemption from duties and exchange controls, and advice and assistance to foreign investors. It is freer from the ﬁnancial constraints affecting the EDB since it has an independent source of revenue in fees for services, and rents on land. The Board’s direct responsibility to the president of Sri Lanka has given it inﬂuence in sustaining the country’s liberal trade and investment régimes, as well as status, ﬂexibility and autonomy in its own operations. Its status and autonomy are enhanced by the fact that it acts as a single window for (prestigious) foreign investors and has a board drawn from the public and private sectors. However, the Board of Investment’s ‘single window’ is also a monopoly entry point which makes its ‘enabling role’ compulsory: overseas investors require its support to gain access to investment approval. The Ghana Investment Promotion Centre (GIPC) has a broader remit than the BOI – to promote overall investment from whatever source, domestic or foreign. However, it works in a much more difﬁcult context than BOI, both economically and politically. Manufacturing industry, including the textiles sector was in deep decline, dominated

Regulating and Enabling the Private Sector 211

still in the mid-1990s by state-owned enterprises, even after structural adjustment. There was a low level of interest among foreign investors, except in the minerals industries. There was a history of suspicion between private ﬁrms and government, based on long experience of intervention and state control. Unlike the case of the BOI in Sri Lanka, the fact that the GIPC reported to the national president worked to enhance the distrust with which it was held by the private sector. The then president intervened in the management of GIPC, appointing public and private board members, inﬂuencing its strategy and even leading overseas trade missions. Wholly ﬁnanced by the government, GIPC’s special status gave it relatively secure budgetary allocations and qualiﬁed staff but also a distance from the private sector that it was supposed to serve. The Zimbabwe Investment Centre (ZIC) and Zimtrade have, until recently, operated in a more positive economic environment than Ghana’s, in an economy with a relatively strong private business sector. Since 2000, the environment has changed fundamentally, with the accelerating decline of the economy and the external imposition of economic sanctions. In the mid-1990s, the Zimbabwe Investment Centre had made a successful transition from being an instrument of investment control and approval to the promotion of investment, particularly from overseas. It provides a ‘single window’ for investment formalities, information services to investors and a system to link investors to local companies. The main limits on ZIC’s effectiveness, besides the economic environment, were its dependence on uncertain budgetary allocations and insecure policy guidance from government. The attitude of government to foreign investment, generally, and speciﬁcally to foreign participation in the privatization of public enterprises created an unclear framework for investment promotion. Zimtrade, on the other hand, had a relatively unproblematic policy brief – the promotion of exports – and it had access to its own sources of income avoiding both ZIC’s total dependence on government and the total ﬁnancial autonomy of the Zimbabwe Industrial Development Corporation. It was probably the most effective of all the enabling agencies that were studied (Box 8.5). The capacity to enable private ﬁrms On a scale of effectiveness, Zimtrade and the Indian Council for Scientiﬁc and Industrial Research were probably the highest scorers, followed by the Sri Lankan training institutions and Board of Investment, and the Zimbabwe Investment Centre. Less effective were

212 The Changing Role of Government

Box 8.5

Zimtrade, the export promotion agency of Zimbabwe

Zimtrade was set up as a joint venture between the public and private sectors immediately after structural adjustment and liberalization with the brief of developing external trade. While it is wholly owned by the Government of Zimbabwe, half its board is appointed from the private sector and most (around 90 per cent) of its income is from charges paid by all registered companies as a levy on their exports. Its ﬁrst director came from the Confederation of Zimbabwean Industry. Its staff joined from the Ministry of Industry and Commerce as well as from the private sector, and are paid at private sector rates. Zimtrade beneﬁted from a large support programme by the European Union which paid for foreign consultants and advice in setting up services in market information, trade research, marketing and training for exporters. Zimtrade’s foundation in the early 1990s coincided with the opening of the economy and particularly the opening of trade in Southern Africa. This put it on a virtuous circle of growing exports leading to higher income which fed back into better services for exporters. The export levy apparently provided an excellent performance incentive, aligning Zimtrade’s incentives with those of its customers. But there were some perverse effects. First, all exporting ﬁrms paid the levy, but not all got access to Zimtrade’s services which were in great demand. Second, it was easier for Zimtrade to provide services to the large established exporters than to the small ﬁrms which might become exporters. Third, when the down-turn in Zimbabwe’s economic fortunes came at the end of the 1990s, Zimtrade lost much of its income though it had not been responsible for the government policies that led to the down-turn. Until then, Zimtrade had all the ingredients of success: support at the highest level in government, strong private sector involvement, a clear goal, independent funding, clear incentives to serve its purpose through the funding arrangements, well-paid, experienced and motivated staff. Source: Jackson, 2003.

the Sri Lankan Export Development Board, the Indian Apparel Export Promotion Council, and the Ghana Investment Promotion Centre. Probably least effective was the Ghanaian Ministry of Industry and Commerce which had tried to sustain roles that elsewhere had been taken on by specialist agencies. The Zimbabwean Industrial Development Corporation was an outlier that was certainly effective, but which had largely detached itself from its original enabling role so as to manage its own concerns. The factors that determined the capacity of these agencies were interrelated. Key among them were that the more successful agencies had some distance from direct governmental control, private sector

Regulating and Enabling the Private Sector 213

participation, sufﬁcient independent sources of funds to support their autonomy, and an entrepreneurial organizational structure. Success meant maintaining a balance between these external and internal elements and avoiding a situation where any was developed to excess. In terms of their internal organizational capacity, the more successful agencies had been set up for the purpose of promotion or facilitation and had clear organizational objectives. Some, such as CSIR, had adapted from earlier functions and had very clearly stated their new purpose to serve the interests of industrial clients. The Ghanaian Ministry of Industry and Commerce (MIC) was the classic case of an organization beset by multiple old (regulatory) and new (enabling) goals. Secondly, the successful organizations were free to recruit staff for the suitability of their experience and skills to the organization’s purpose and to manage their human resources without external intervention. They were also able to pay staff at rates that exceeded those of regular civil servants. This was true of CSIR and the Zimbabwean institutions. The MIC and Ghana Investment Promotion Centre (GIPC) had inherited their staff and these were paid on public sector wage scales without any special incentives or recognition of their proximity and comparability with the private sector. The GIPC did at least have greater security that it would receive its budgetary allocations than was true for ministries. The staff of the Sri Lankan agencies all had some freedom to recruit and some salary distinction from the civil service, but in practice the Export Development Board had tended to retain its established staff and not to develop their expertise Going with clarity or simplicity of purpose was the relatively small scale of most of the successful enabling agencies. The Zimbabwean organizations had between 35 and 50 staff; the Sri Lankan training agencies less than 80. This compared with the 200 of the Ghanaian ministry, 330 of the Sri Lankan Export Development Board, and the 600 of the Indian Apparel Export Promotion Council. The exceptional case was the 21,000 of the Indian Centre for Scientiﬁc and Industrial Research. However, these were spread across 120 sites where each centre had a high degree of devolved power to initiate their own programmes and raise revenue. A related but separate factor was that, hierarchical organizational structures and the absence of devolved decision-making were features of the less successful organizations. The aspirations and talents of professional staff were often frustrated by organizational rigidities and by senior staff’s dominance.

214 The Changing Role of Government

External factors are interwoven with the internal. The successful agencies were regarded by the private sector as ‘honest brokers’ of information and support. This required them to have legitimacy with private ﬁrms in terms of the technical value of the services they were offering and in terms of their empathy with private interests. At the same time, they had to maintain distance from but also the support of government. Formal private sector participation in the decision-making of enabling agencies was widespread and took the form of representation on boards, membership and the payment of fees for services (see Table 8.4). Where this really made the agencies responsible and responsive to the private sector, participation was a feature of greater capacity. However, formalistic representation on the board can give an illusion of inﬂuence and proximity to the client. The independence and inﬂuence of the private representatives was real in the case of Zimtrade, but in the case of the Ghana Investment Promotion Board representatives were not independently selected. Probably the best indicator of real proximity to the client is not representation but the payment of fees for services. The more responsive and effective agencies – Zimtrade, CSIR, the Sri Lankan training institutions – were those that partly depended on their clients for payment. Payment for services helped to align the interests of employees and the enabling agency with those of their clients. Access to private sources of income also gave agencies some autonomy from government inﬂuence and some freedom from the vicissitudes of budgetary controls. Underlying these factors are some fundamental conditions. The possibility of public agencies enabling the private sector is inhibited where there is distrust between ﬁrms and government, where government support is subject to political manipulation, and where the private sector is too weak to take up opportunities or to pay for services. The four countries had emerged from long periods of political control of the private sector. The most constrained country was Ghana, where private sector weakness and mistrust, macro-economic instability and presidential intervention in agency operations were the norm, at least until 2002. In Zimbabwe, what had been a promising situation, where government and large private businesses existed in a state of mutual respect if not affection, collapsed from 2000 throwing some welldeveloped agencies into disarray. India and Sri Lanka offered more stable conditions of growing conﬁdence and a balance of inﬂuence between government and private ﬁrms. However, even there, enabling

Regulating and Enabling the Private Sector 215

agencies often found that there was an absence of a clear framework of industrial policy within which they could operate. Paradoxically, government enablement of private ﬁrms is most necessary where it is most difﬁcult to achieve – where the technical, human resource, political, economic and legal conditions are most lacking.

A comparison of the experience of regulation and enabling This chapter has examined roles of government that are both deeply traditional and new. Regulation is a fundamental role of government but has historically tended to over-extend, with the effect of suffocating markets without achieving the purposes of public policy. The forms of regulation that have been examined here are those that have become more important after ‘market-unfriendly’ regulation had been swept away. They are based on the view that, without appropriate state intervention, sometimes markets would fail to deliver services efﬁciently. The cases of regulation that were examined involved setting and applying conditions for the allocation of monopoly rights in the case of urban water, and ensuring the quality and quantity of health care services where clients would otherwise be under-served or uninformed. The enabling policies that were examined went a step further; these anticipate and seek to avoid market failure by acting to reinforce the capacity of actors – consumers and producers – to enter and operate in the market. This chapter reported on the case of business support services where, among all the sectors, there is the greatest experience of enabling action. Both regulatory and enabling roles are a considerable challenge to the capacity of any government. They require capacity to understand and exercise critical judgement about market conditions, obtain necessary information and act independently of political and self-interested pressures. It is not surprising that, in the study countries, there were many deﬁciencies in performance. More surprising is that there were also some relatively successful cases. The study of urban water showed that there could be gains from the introduction of an independent regulator, even where the organizational and institutional conditions are not perfect. The clearest case is where the regulator is outside the apparatus of government and the company is privately managed, as in the case of Buenos Aires. Such conditions have so far been met primarily in middle and upper income

216 The Changing Role of Government

countries rather than the poorest that were the main focus of this research. In most of the countries studied regulatory roles were embryonic or poorly performed. But the case of Ghana showed that, even where the water supply company was still publicly owned, an independent regulator could have the effect of improving water standards and adjusting tariffs to reﬂect real costs. Health care regulation was poorly undertaken in all countries studied, except Thailand. Until recently, regulation has been neglected by government and by external funding agencies. Standards of professional care and of drug control have frequently not been set; where standards have been set, they have rarely been applied through inspections; and governments and professional councils rarely had an adequate basis of information on private ﬁrms and practitioners. Underlying the poor performance of almost all the professional councils were absent or ineffectual supervision and monitoring by ministries of health. Enabling services to business included agencies intended to ‘facilitate’ ﬁrms by providing consultancy, advice, training, research and development, and others intended to ‘promote’ investment, marketing and exports. A number of agencies of both types performed well, in Zimbabwe, India and Sri Lanka. They maintained a portfolio of services that was in demand by private ﬁrms, had staff and ﬁnancial resources adequate to deliver them, and had relationships with government and the private sector that gave them credibility without compromising their autonomy. The best performers had a direct measure of their relevance to industry in the fact that their services were directly paid for. Underlying the problems of performance were some common constraints on capacity running across all types of activity and agency (Table 8.5). Regulatory and enabling agencies required some common internal, organizational conditions. As agencies that stand between government and the private sector, they all needed, but often did not have, sufﬁcient autonomy to decide their own internal affairs and to operate without pressure from either side. Financial resources and freedom to recruit appropriate staff were often lacking. The skills to undertake the ‘new’ roles of regulation (accounting, economic analysis, information collection and performance assessment) and enabling (professional competence, business development, market assessment, research and training capacity) were often scarce in the economy as a whole. Agencies whose pay rates were low compared with the private sector had difﬁculty in acquiring and retaining skilled staff.

Regulation of urban water Regulator’s mandate and autonomy not respected in practice Lack of staff expertise in accounting and economics Inadequate information on price, proﬁtability and performance Blurred boundaries between government, regulator and ﬁrm Neutrality of regulator in question

Economic instability Political instability Political inﬂuence on regulator Weak demands of civil society Partiality of judiciary Greater knowledge and power of the private ﬁrm

Internal organizational factors

Interorganizational factors

External institutional factors

Overlapping public and private medical interests Political capture by professions and business Weak demands of civil society

Lack of policy framework by government for agencies Weak coordination by government of delivery and regulatory agencies Neutrality of regulator in question

Weak organization of implementation Inadequate staff numbers and skills Under-resourcing and stafﬁng of professional councils Inadequate information collection and recording Vested professional interests

Regulation of health care

Key constraints on governments’ capacity to regulate and enable private ﬁrms

Factors

Table 8.5

Economic instability Direct political control of agencies Distrust between government and ﬁrms Weak private sector capacity

Lack of clear industrial policy Weak private sector participation Services not paid for

Unclear or multiple purposes of agency Hierarchical organization Lack of autonomy in stafﬁng, ﬁnance and decision-making Low pay, motivation and expertise

Enabling the business sector

217

218 The Changing Role of Government

There were differences as well as similarities in organizational requirements. The enabling agencies needed entrepreneurial and small teams operating in a non-hierarchical organizational structure. In these cases, sources of ﬁnance that rewarded effectiveness in delivering services to industry were an important incentive. On the other hand, the regulatory agencies had to operate in a more rule-bound environment where the key failings were of a more bureaucratic nature: lack of routinely collected information, absence of records, and non-completion of inspections. These cases required, but often lacked, guaranteed government funding. The enabling agencies beneﬁted from the engagement of the private sector in decision-making and in paying for services; by contrast, the regulators needed, but often failed, to keep their distance from private companies. The key external constraint to the effective operation of regulatory and enabling agencies was the invasion of their autonomy by political and vested interests. ‘Invasion’ is probably the wrong word because it implies resistance on the side of the agency. Few agencies could assume even that their own staff would expect to operate within a ‘culture of neutrality’ where regulation and support services applied without favouritism. In the case of health regulation it is clear that the same professional interests usually operated in government, the private sector and the regulator, preventing any disruption of those interests. Political leaders overrode the independence of water and health regulators and sometimes directed enabling agencies. The capacity of civil organizations, consumer organizations or the private sector to act as a countervailing force to this political pressure was usually weak. State failure was most widespread in the case of health. This may be partly for technical reasons: the difﬁculty of exercising qualitative assessment and control of multiple services provided by many smallscale providers, often within weakly coordinated policy networks. But, the more fundamental problem is that in health care there is the greatest intermeshing of public and private interests.

9 Conclusion

Introduction The previous chapters of this book have analyzed the application of a series of new public management reforms outside their countries of origin, in Africa, Asia and Latin America. The core issue is whether reforms based on a diagnosis of the weaknesses of the ‘overinterventionist’ state in advanced countries such as the United Kingdom, New Zealand and Australia, are appropriate responses in states where the levels of public management capacity, market development, resources, political inclusiveness, legal effectiveness, political and economic stability are quite different. Our core argument, as outlined in Chapter 1 has been that governments may be illequipped to adopt unfamiliar approaches to public service provision, where the institutional conditions on which the new management practices are premised may not be present. This ﬁnal chapter ﬁrst provides a summary and pulling together of ﬁndings from each of the earlier chapters, giving an overview of the nature of and extent of the application of the new management approaches in sectors and countries we studied. The chapter then goes on to review the performance of speciﬁc reforms. This is followed by a section which questions the relevance and appropriateness of reforms to different institutional and country contexts and argues that context does matter both for the type of reform and sequencing of reforms. The ﬁnal part of the chapter identiﬁes the need for capacity development with an approach that strengthens both technical and institutional competence for reforms.

219

220 The Changing Role of Government

Summary Going round in circles? Chapter 1 traced the evolution of the ‘statist’ and ‘developmental’ models over the last half century. These two models gave pre-eminence to the role of the state in development in general and service delivery in particular. By the late 1970s through the 1980s these models were in crisis and came under increasing pressure to change, ﬁrst in some advanced economies, and later in poor countries in sub-Saharan Africa and south Asia, with similar models of reforms. Crisis led to ‘home-grown’, market-driven reforms in the management of public services in advanced countries. However, for lowincome countries in Africa and South Asia, reforms were often delayed and took place in the context of IMF-World Bank stabilization and structural adjustment programmes designed in response to economic and ﬁnancial crises. Reforms involved stabilization, liberalization – including deregulation of trade and prices – and the privatization of state-owned enterprises. The four case study countries embarked on reforms at different times. Sri Lanka started with a homegrown economic reform in 1977, and then adopted IMF-World Bank reforms in 1987. Ghana was one of the pioneers in adopting structural adjustment in Africa through its Economic Recovery Programme in 1983. Both India and Zimbabwe were later adjusters beginning their programmes in the early 1990s and postponing public service reform. Crises and adjustment provided the context in which the market approach to reforms in public services were introduced in the four countries and in the four sectors covered in this study. However, the depth of crisis was more severe in sub-Saharan Africa than in the south Asian countries. Structural adjustment and stabilization had a constraining effect on public expenditure, especially in the social sectors such as health and education. Cutback management was also applied to the public administration system especially in sub-Saharan Africa, where overstafﬁng, erosion of wages and demoralization were common. Thus reforms in public services were introduced in circumstances in which the civil service and other state implementing agencies were themselves subject to reforms and weak in capacity. By the late 1980s and early 1990s the efﬁcacy of structural adjustment programmes and the market approach to public service reforms came into question. The earlier emphasis on minimizing the role of the state gave way to a realization of the need for a capable state that is able to create the enabling conditions for markets to work (World

Conclusion 221

Bank, 1997). Markets needed the state in order to function, just as the state needed the market for the delivery of services. With the new focus on poverty reduction and attainment of the Millennium Development Goals, the state returns to the centre stage as the principal enabler and partner of non-state actors – the private sector and civil society groups. Thus the neo-liberal minimalist role of the state has given way to a more pragmatic approach in which the state and non-state actors have legitimate and different roles to play in delivering services. One of the arguments of this book has been that the new approaches to public management embodied in some of the reforms proposed for (or imposed on) low-income countries were inherently more complex and taxing of government capacity than previous arrangements. This book has focused on the following new approaches and roles – regulating markets, enabling other providers, managing service delivery through decentralized structures, contracting out and charging users. Given the state of maturity of public administration systems, the difﬁcult circumstances and the institutional contexts in which reforms were introduced, we have questioned the appropriateness of the comprehensive implementation of the new public management reforms in the countries and sectors examined in this book. Our ﬁndings suggest a more cautious and incremental approach to reform with a sequencing that allows for capacity development. Extending the new public management reform agenda to poor countries Chapter 2 provided an overview of theoretical and pragmatic drivers for the new approaches to public sector management. Theoretical arguments provided by public choice and new institutional theories questioned the dominant role of the state and the efﬁcacy and effectiveness of the traditional system of public administration. These have provided the theoretical underpinnings for market-based reforms such as privatization, contracting out, charging and performance management. These approaches were transferred to middle and low-income countries when crisis provided the opportunity for reforms by throwing into question the established ‘rules of the game’. Crisis was a common driver for reform in both developed and developing countries, although the depth of crisis was more serious in the latter. Apart from this, reforms in advanced countries were driven largely by ideological and political commitment, combined with rising public expectations and demands, whilst reforms in poor countries

222 The Changing Role of Government

were largely driven by external pressures made possible by the depth of crisis, indebtedness and lending conditions of international ﬁnancial institutions. The new public management reform agenda which evolved from the new institutionalist and managerialist theories was summarized in Table 2.1 and may be crystallized into three core central tendencies as follows: • Organizational restructuring in order to shift emphasis from centralized administration to decentralized management through the creation of autonomous and semi-autonomous agencies and the devolution of resources and operational decisions to frontline managers; • Increasing use of market-type mechanisms such as charging for services, privatization and contracting out, whilst regulating providers and making services responsive to users/customers; • Increasing emphasis on performance – an attempt to shift the methods of doing business in the public sector away from complying with procedural rules towards ‘getting results’. As noted in Chapter 2, the new public management agenda presents a menu of reforms. This book has focused on some of its key elements: decentralized management, charging for services, contracting out and public-private partnerships, enabling and regulating other providers. Regulation becomes a necessary element of reform if the state is disaggregating functions and structures, contracting out and devolving service delivery to semi-autonomous agencies and the private sector. The contribution of this book has been in bringing together evidence on the application of these new approaches to four developing countries and several reference countries in Africa, south Asia and Latin America, and from across four service sectors. Taking account of political dynamics in reforms The weak position of citizens and policy-makers as principals: In principle, the providers and the reformers of services act as the ‘agents’ of citizens and politicians. However, we found a low level of public engagement in the policy process, tenuous direct accountability of providers to clients, and few cases where political leaders came up with clear programmes of reform and a strategy for change. The research programme’s surveys of users’ opinion of health and water services in Ghana, Zimbabwe, Sri Lanka and India found strong critical opinion

Conclusion 223

about the price and quality of these services, but widespread ignorance about the proposals for the reform of those sectors. We found evidence that control by citizens and clients is greater where the nature of the service permits them to organize themselves and exercise informed choice, and by policy-makers where they can gather information and measure the performance of providers. In practice clients and policymakers are weakest and the service providers are strongest in the case of curative health services. In urban water supply, there is a greater possibility of balance between the two sides. In business and agricultural services, the agent-providers have relatively lesser possibility of controlling the principals. Given a low level of public engagement with policy in the countries we studied, reform generally remained the domain of senior ofﬁcials in partnership with international agencies – particularly in the more aiddependent African countries. Political acquiescence at the highest level was necessary, but political leadership was almost always personal rather than the product of a party programme. Since the reforms were about the organization of service delivery, the struggle for and against change generally did not engage the public but remained within the bureaucratic arena. However, some reforms had a more immediate impact on sections of the public – user charges, ‘privatization’ and the liberalization of markets. Even in those cases, public demand for reform was rare; more often, there was pressure on politicians for the defence of existing rights and privileges than for change. This was evident in the resistance to bringing private sector participation into the water sector in Ghana and to introducing or increasing health user fees in Sri Lanka and India. Developing and maintaining a constituency and momentum for change: Although the initial impetus to reform is typically narrowly based, it may accumulate support as it develops a new range of beneﬁciaries who wish to protect their interests. In Sri Lanka and Ghana, liberalization, privatization of state enterprises and deregulation of industries have created a new set of incentives for managers and entrepreneurs, and created support for further liberalization to encourage foreign investment and employment. In Kenya, ofﬁcials who had initially opposed the removal of the state monopoly in maize marketing, eventually came to support it after seeing its positive effects, whilst the millers and traders who began to deal in imported maize and rice in Kenya and Sri Lanka became a constituency for further liberalization (Hubbard, 2003). Decentralized health management also created a demand by managers for more autonomy from

224 The Changing Role of Government

central controls. Thus some reforms sow the seeds for further change through an emerging constituency of support. As others (Grindle, 2003; Nelson, 2000) also conclude in their studies of reforms in Latin America, maintaining the momentum for change requires reformers to understand the political constraints and opportunities and to calculate opportunistically to build a constituency of alliances to overcome opposition. Unlike the earlier ‘stroke of the pen’ ﬁrst generation reforms, the more recent reforms in public services present a much harder reform task as they embody more institutional elements which require changing or redeﬁning the ‘rules of the game’. They have less clear-cut goals, offer uncertain beneﬁts, involve multiple actors, challenge existing provider groups, and require long-term commitment on the part of government and donors.

Performance and capacity constraints in speciﬁc reforms Tables 9.1 and 9.2 act as a guide to the following sections. For each of the main reform types that have been examined in this book, they summarize the experience of performance and the leading constraints on capacity for implementation. The tables pick out the factors that are common between the service sectors, while the text also identiﬁes some of the main differences. Decentralizing management – more rhetoric than reality Although the research found evidence of varying degrees of decentralization of service delivery arrangements, these were largely structural changes with little evidence of the real change in processes, practices and behaviours that would allow operational autonomy and improved performance in service delivery. Despite the limited cases of real reform in practice, our ﬁndings suggest that, where it does occur, managerial autonomy (in control over ﬁnancial and human resources and policy implementation) is associated with better performance. There were improvements in efﬁciency and effectiveness where managers had practical freedom from bureaucratic rules and controls, and from blatant political interventions. This was evident in the case of a water cooperative in Bolivia, Zimtrade and Dairy Board Zimbabwe Limited, and the Cocoa Marketing Company in Ghana. Management autonomy does not only support efﬁciency, but also enables the development of a strategic view. Managerial autonomy is more difﬁcult to sustain in politically sensitive services such as health and water that attract frequent political

Application of reform Widespread formal adoption in all sectors in all countries, but wide failures in implementation

Widespread introduction of user fees in health services in Africa Recognition of need for water tariffs to reﬂect true costs Delivery of business services normally on a fee basis Widespread application of short term contracts in water, agriculture and health care Limited application of long-term contracts and concessions

Decentralized managerial autonomy

User fees

Contracting out

Gains in technical efﬁciency

Most applicable where the service is optional and consumption measurable Gains in allocative efﬁciency

Gains in technical efﬁciency and service quality Autonomy is more easily guarded where there is clear user accountability and own sources of ﬁnance

Positive effects or gains

Summary of performance of organizational reforms

Organizational reform type

Table 9.1

Failures in designing, managing and monitoring long-term and more complex contracts

Difﬁculty of application in non-measurable, politically sensitive services Severe problems of equity in charging for health care. Exemption systems normally fail

Autonomy can easily be lost especially in politically sensitive services

Negative effects or risks

225

Application of reform

Widespread formal existence of regulatory frameworks especially in health and water General shift to enabling roles in agricultural and business development

Regulation

Enabling of independent providers

Some effective market support in business sector, especially where the agency has autonomy from government

Gains in clearer separation of regulatory and supply roles in water

Positive effects or gains

Summary of performance of organizational reforms – continued

Organizational reform type

Table 9.1

Weak application and impact of enabling functions in water, agriculture and health Government support services often not trusted by private sector

Poor implementation of regulatory roles, especially in the health sector Interference by politicians and professional interests

Negative effects or risks

226

Decentralized managerial autonomy Low pay and inability to attract and retain skilled staff Lack of managerial ﬁnancial autonomy Inadequate information systems for accountability

Lack of clarity of authority relationships The problem of multiple principals Weak capacity of centre to set and monitor targets Weak central support services

Internal organizational factors

Intraorganizational factors

Weak basic administrative and ﬁnancial systems and skills Limited experience of contract design Poor information systems to compare bids and performance Staff resistance and lack of incentive

Contracting

Lack of central monitoring Poor deﬁnition and and weak accountability coordination of roles systems Gap between contract Gap between policy design design and and implementation implementation agencies agencies Weak policy guidelines on Unpredictable policy exemptions and use of revenue from fees

Poor salary and incentive systems and low motivation Weak systems for billing and collection Weak capacity to manage exemption systems in health

User fees

Key constraints on governments’ capacity to perform new management roles

Factors

Table 9.2

Blurred boundaries between government, regulators and the regulated Gap between policy and implementation agencies

Lack of staff skills in accounting and performance assessment Weak stafﬁng for selfregulation in health. Inadequate information on price and performance Vested professional interests

Regulation

227

Intervention by political Political resistance to and bureaucratic increases in charges leaders Unstable macro economic Policy instability environment Centralized control over Civil society opposition resources Unreformed legal and governance frameworks Central resistance to change

External institutional factors

User fees

Decentralized managerial autonomy Socio-political opposition to ‘privatization’ Economic and ﬁnancial instability affects large water contracts Centralized civil service rules and systems Weak legal framework Limited private sector development and trust Lack of culture of performance

Contracting

Key constraints on governments’ capacity to perform new management roles – continued

Factors

Table 9.2

Political and professional inﬂuence on regulator Economic and political instability Partiality of judiciary Business sector distrust of government Weak demands of civil society Neutrality of regulator not understood

Regulation

228

Conclusion 229

interventions to control prices and access. In the case of Kenya, Sri Lanka and India – but not Ghana – political sensitivity extended also to the marketing of staple grains, leading to agitation by farmers and consumers and pressure on politicians to rescind reforms. Managerial autonomy is more likely to be safeguarded where agencies have a degree of self-sufﬁciency in their sources of ﬁnance and where there is a routinized form of accountability both to government and to their users or local voters. Paradoxically, the autonomy of managers is best defended where it is transparently subject to formal rules of accountability, rather than to haphazard interventions by political leaders and public protest. User fees and charges: equity as a key concern The rationale for the introduction of charging has varied across sectors and countries. Whilst the new public management rationale for charging was based on increasing allocative efﬁciency in the use of resources and increasing accountability to clients, in practice charging has often been introduced simply in order to raise revenue and repay credit. It was most easily introduced in the case of sectors with low equity concerns (such as business services), where the service was optional to the client, and where the service being offered (for example, credit, domestic tap water) was measurable and a private rather than a public good. User charges can be used as an instrument to promote equity, but this depends on the ability of governments to design effective and functioning exemption systems for the poor and vulnerable groups, or to ensure that subsidies are speciﬁcally targeted on the poor. In the health sector this has been very difﬁcult to achieve in most poor countries. The evidence from the case studies shows that exemption systems fail the poor and are the weakest link in the policy of charging, except in cases where they are backed by a reliable pre-payment system. In the water sector, ‘lifeline provision’ and a scaled tariff structure can protect the poor and are relatively straightforward to set up, but indiscriminate subsidies for everyone are inequitable because big consumers beneﬁt most (World Bank, 2003b). In agricultural marketing, safety nets for the poor are required when food supplies are entrusted to the market, in order to ensure that all can buy staple food. There is a shift towards market-based food security by state withdrawal from the purchase and distribution of food and the introduction of income-based relief in the form of cash or vouchers (food stamps in Sri Lanka, cash payment for drought relief for some states in India, and income-based safety programmes during drought in Namibia and Botswana).

230 The Changing Role of Government

Charging for goods and services has an equitable effect for services only where merit good or equity concerns can be covered by effective exemption or scaled charging systems, and where charges are administered systematically. Scaled tariffs can relatively easily work where consumption can be measured, as with metered water. Charging for health and other social services requires exemption systems but these present great difﬁculties of deﬁning and identifying those who are eligible and of monitoring the abuse of charges by agencies (Kessler, 2003). Contracting out direct service provision Contractual arrangements for service delivery are not new in any of the countries or sectors under study, so making and monitoring contracts do not involve completely new roles for governments. What is new is that the practice is moving beyond the traditional areas (construction and maintenance works) to the delivery and management of actual frontline services. There is also a move towards contracting out (if not full privatization as in industry and some agricultural services) on a deeper and longer-term basis and the replacement of loose by more formal agreements as in the case of health. New forms of contracting (concessions and lease contracts) are in operation in several of our reference countries, but generally only at the level of policy intentions in the four case study countries. Although contracting can deliver clear efﬁciency gains, our research shows that it will only do so under certain conditions, including clear speciﬁcation of contracts, availability of information to allow monitoring, and the capacity to maintain an arms-length relationship from both providers and their clients. These conditions are unlikely to pertain in economies with weak private sector capacity and competition, where administrative capacity to design and manage a network of contracts is weak, and in services that are more qualitative and non-tangible (such as health care). A pragmatic way forward is to allow the experience of contracting to develop incrementally, focusing ﬁrst on simple and less critical cases (for example, cleaning and security services) and where services can be speciﬁed and measured (such as piped water). Simpler and short-term contracts also present less risk to both government and contractors. For example, contracts that require major capital investment by contractors are vulnerable to political and economic instability and uncertainty, as was demonstrated even in an otherwise successful case – the Buenos Aires water concession.

Conclusion 231

Regulatory and enabling roles Decentralizing and devolving responsibilities for the delivery and management of services to autonomous and semi-autonomous public bodies and to the ‘private for proﬁt’ and voluntary sectors do not absolve government from the responsibility of ensuring that services are provided. As Chapter 8 showed, governments have important regulatory and enabling roles to play in all sectors. The managers of decentralized service organizations are expected to be more clearly accountable for their performance, according to policy frameworks and performance contracts. Their greater freedom to manage has to be matched by a clearer statement about what they must achieve. The same principle applies in the new relationship between government and the private sector. In return for its increased involvement in service delivery, often funded by government, the private sector is expected to be subjected to governmental controls about its outputs and standards. In the particular case of natural monopolies (such as piped water), privatization or the creation of arms-length public agencies give the single provider immense market power that can only be countered by public regulation. In principle, government regulatory and enabling roles should increasingly be strengthened as more responsibility for direct service provision is decentralized or privatized. Where there remains a public interest (as in public health, water, food security and the environment), government should take on the new roles of ensuring competitive market conditions, supporting market development or community action, price-setting, evaluating performance, monitoring contracts, and intervening selectively to correct failures. In practice, however, in all sectors (except business support) regulatory and enabling roles are either poorly performed by governments or have not received much attention. Pro-market regulatory and enabling roles make heavy demands on government capacity: to understand and make judgements about market conditions and incentives, obtain the necessary supporting information and act independently of political and self-interested pressures. These conditions were least fulﬁlled in the health sector where regulation has been widely neglected or left to professions to regulate themselves with loose and ineffectual government guidance. The water sector presents a smaller number of agents to monitor and more quantiﬁable tools of assessment. Here there were examples (for example, Ghana) of improvements in performance brought about by regulation even where the organizational separation of regulator from provider

232 The Changing Role of Government

was not fully implemented. However, in all sectors there were wide failures in regulatory practice at two levels: ﬁrst, basic bureaucratic failings to collect information and keep records; second, the invasion of regulators’ autonomy by political and vested interests. The capacity of civil society organizations to act as a countervailing force capable of exposing such interests was weak. Agencies that existed to ‘enable’ the development of private businesses were freer of some of the constraints surrounding regulators. The regulators were intended to be detached from the regulated provider and from politicians but could easily become subordinated to both. On the other hand, the enabling agencies were intended to be responsive to their clients. The agencies that offered paid-for services had both incentives to attend to their clients’ needs and a basis of ﬁnancial autonomy from political interference. There were several examples of effective enabling agencies in the African and south Asian countries.

Implications for capacity development The failures of the ﬁrst wave of structural adjustment reforms contributed to a renewed concern with the capacity of states by the late 1980s. Whilst adjustment sought to minimize the role of the state, the new agenda recognized that, while there may be too much state intrusion in the economy, there was also often too little government capacity to make policy, perform basic administrative functions, work with private partners, and ensure the provision of infrastructure and public services. As Grindle (1997, p. 4) has noted ‘Only after a decade of experimentation with reducing government did economic reformers become explicit about the importance of strengthening government’. The second generation of reforms, which embodied most of the new management approaches, gave more attention to the transformation of the ‘hard to change institutions which structure political and economic life’ (World Bank, 2000). These institutions include clear property rights, the rule of law, ﬁnancial systems, an active civil society, accountable government, and efﬁcient and effective public administration (Burki and Perry, 1998). Such core institutions are either weak or just evolving in low-income countries. The new approaches are being grafted onto systems with deep institutional weaknesses that require long-term capacity development rather than ad hoc and short-term approaches. The following sections summarize our conclusions on the issue of capacity.

Conclusion 233

The need to strengthen basic public administration For low income countries improving the capacity of the public administration system to perform basic functions of revenue collection, compliance to law and order and delivery of basic services is paramount. Hierarchical bureaucracy has not been substantially reformed in these countries and there is much to do to develop the capacity of traditional bureaucracy to provide the preconditions for the implementation of the new management reforms (Schick, 1998; Manning, 2000). The complexity of new roles requires long-term capacity development The new roles of government are certainly more complex and require long-term capacity development than the traditional roles. It is paradoxical that governments that are widely regarded to have failed to deliver services adequately are now expected to take on the more complex functions of understanding markets, contracting and regulating private agents. An incremental and carefully sequenced approach to reforms may have a better chance of success than more comprehensive reform approaches that tend to overstretch the capacity of government and its agencies to implement reforms. In the case of charging, which may be seen as a simple role, it was apparent that weaknesses in capacity were due to the rush to implement without adequate preparation. In health, the relevant public organizations did not have adequate organizational and management capacities to effectively manage the implementation of user charges and exemption systems. For example, government accounting systems are often weak and unable to generate information for frontline managers. This reinforces the point made earlier about the need to strengthen basic public administration. Setting up exemption systems for the poor has proven more complex in the health sector than was expected. In Chapter 7 it was noted that the capacities demanded by higher level contractual arrangements (such as concessions and leasing) put strains on government. The piecemeal contracting out of service and management activities to multiple contractors, as in the Chilean case (see Box 7.1) demands not only a sustained capacity to maintain a continuous cycle of contracts but also to monitor and coordinate them; the information demands are high. Build-own-operate and concession contracts, in India and Sri Lanka, have stalled on the incapacity of governments to demonstrate that the legal, ﬁnancial and market conditions exist to justify private investment.

234 The Changing Role of Government

The capacity of governments to perform market-friendly regulatory and enabling roles is weak; these roles need to be strengthened and to be supplemented by strengthening direct user voice, participation and accountability, which the World Bank (2003b) refers to as ‘short route’ accountability. The cases reviewed in this book indicate that the internal organizational requirements are demanding especially for ‘the new regulatory roles’. The constraints identiﬁed included: • Problems in establishing an organization with a clear mandate, speciﬁed duties and powers, with autonomy to take decisions without political intervention; • Lack of staff with the required expertise and qualiﬁcations (e.g. in accounting, economics and ﬁnance) often compounded by the inability of the public sector to pay salaries that will attract and retain scarce skills in a competitive labour market. Even though some of these skills can be contracted for speciﬁc tasks, there is the need to have core competence in the public sector; • Problems of access to high quality information to make judgements about price, proﬁtability and performance. Paying attention to external as well as internal elements of capacity Long-term capacity development for reforms requires an approach that focuses on both the internal (organizational) and external (institutional) elements of capacity and behavioural change. As was noted in the case of the health sector, the research concluded that the capacity to contract was not mainly limited by internal but by external factors. The internal capacity that was more difﬁcult to develop was the longterm regulatory role, where not only new professional skills have to be developed but also new legislative frameworks and, above all, an often unaccustomed culture of political detachment. In the case of water it was noted that the institutional conditions external to a regulatory organization but necessary to it are even more exacting, particularly in the context of developing countries. These included instability of the political economy, problems in maintaining separation from political inﬂuence, weak demands of civil society, and doubts about the independence of the judiciary. As noted earlier, politics and power play a signiﬁcant role in shaping the course of events in public sector reforms. The institutional and political context in which reforms are introduced greatly inﬂuence the

Conclusion 235

success of their implementation. The new approaches work most effectively where there are mechanisms for citizen demand and information (for both policy-makers and clients) on the performance of ofﬁcials and private providers. But the institutional environment in most lowincome countries tends to be opaque and civil groups to be restrained and lacking information. In most cases, democratization is still fragile and needs to be nurtured in a difﬁcult resource-constrained environment. Political sensitivities act as a restraint on long-term policy reforms, which may be necessary but politically unpopular, as was evidenced in the attempts to introduce private sector participation in urban water supply in Ghana and to liberalize staple grain marketing in Sri Lanka and Zimbabwe. In a number of countries, especially in sub-Saharan Africa, the context is complicated by a heavy donor presence with multiple reform programmes in different sectors. Some of these reforms need to be rationalized and sequenced to take account of government capacity to implement them.

Conclusion The above discussion raises questions about the appropriateness of the radical and comprehensive application of new public management reform models to low-income countries such as those covered in this book. Among the countries studied, whilst reform designs were ambitious and wide-ranging on paper, actual implementation has been patchy. There has usually been little attention given to the sequencing and phasing of reforms to meet existing capacity for implementation or to match reform ambitions to realities. Given weak capacities, the approach to reforming public services would be better based on incremental steps within a long-term strategy, starting with the reform of basic incentives that strengthen accountability and improve performance. The World Bank (2003b) now suggests that, as administrative capacity develops and as ‘incentives become better aligned and internalized, more advanced reforms can be introduced to support deeper institutional change and scaling up.’ The new of roles of government are still evolving and government capacity to perform them in the countries studied is weak in most cases. Capacity to perform new roles varies by country context, sector and activity. Legislating for organizational restructuring and granting formal management autonomy are the easy steps in reform. It is in the implementation, the application of user charges, the contract management and the regulation that things go wrong.

236 The Changing Role of Government

Some of the problems inherent in the transformation from traditional to new public management and new roles of government in low-income countries may be transitional (for example, the lack of certain skills) whilst others may be more fundamental and institutional. One task of reformers is to be able to distinguish between transitional and more fundamental capacity problems. The former may require short-term capacity development interventions (for example, appropriate training and technical assistance for the establishment of adequate management information systems and improving ﬁnancial accountability), whilst the latter will need long-term capacity development, including institutional interventions that challenge the ‘rules of the game’ and are able to change organizational and individual behaviours. We found many examples where reforms had been undermined by an organizational culture that emphasized adherence to rules and regulations rather than to performance and the achievement of outcomes (Mills et al., 2001). These institutional obstacles to change need as much attention as organizational structure in the design of a reform. The picture that emerges from all the cases studied is that the new roles are proving more challenging than reformers might have thought. There is a need to re-think and re-prioritize reforms. The priority is to get basic government administrative structures and systems working before superimposing new roles on them. Capacity development needs to be targeted on speciﬁc, identiﬁed constraints. The challenge is to continuously monitor the process of implementation, identifying the processes that would support change best and improving the effectiveness of existing reform initiatives. In the light of the different contexts and conditions in the various sectors and countries, emphasis needs to be placed on adaptation. Context really does matter and cannot be ignored by advocates of reforms. Some of the new management reforms such as complex contracting may be inappropriate for some sectors such as health in some countries where the necessary preconditions do not exist. They may be appropriate for other sectors where private sector interest and capacity exist, and where monitoring and regulating mechanisms also exist or can be developed relatively easily. Appropriateness of reforms is therefore relative to context, sector and timing. Finally, as has been demonstrated in several cases in this book and others (see Grindle, 2003; World Bank, 2003b) reform is a political process and needs skilful political management to steer it through. Building a constituency of support and ownership among political and

Conclusion 237

bureaucratic élites and public sector workers is necessary. However, support for reforms needs to go beyond these constituencies if it is to challenge established practices. Reform sponsors and managers have to make the effort to raise awareness among service users and citizens of the inadequacy of their existing services and of the need for reforms.

Notes Chapter 2 1. Productive (or technical) efﬁciency: resources are used economically to produce a given output, or the given level of resources is used optimally to maximize outputs. 2. Allocative efﬁciency: services allocated correspond to consumers’ preferences and/or are charged at prices which reﬂect producers’ real costs.

Chapter 3 1. This is a fuller version of an article which appeared in the journal Development and Change, Vol. 35, No. 1, 2004. 2. Mills (2002) identiﬁes political leadership in reform in the case of the health sector in Zambia, South Africa and Colombia. See also Corkery et al., 1998, Nelson 2000, Grindle 2000 and 2003 for examples of sectoral political leadership. 3. Corkery et al. (1998) and Nelson (2000) note the importance for sustainability of reforms of involving line ministries, but that they are often excluded in programmes driven by stabilization and structural adjustment.

Chapter 4 1. The relevant legislation is the Hospital Administration Law, 1988 (PNDC Law 209) and the Ghana Health Service and Teaching Hospitals Act, 1996 (Act 525). The latter granted more independence to two teaching hospitals and separated them from the Ghana Health Service and the Civil Service. The hospitals were Korle Bu and Komfo Anokye. 2. In 1990 GWSC’s head ofﬁce reinstated controls over several management and operational functions, especially accounting and personnel functions in order to streamline its ﬁnances and staff numbers. This gave the head ofﬁce control over surpluses generated in variable regions in order to crosssubsidize non-viable regions and to settle redundancy payments and end-of service beneﬁts to a large number of retrenched workers. 3. Under the Statutory Corporations (Conversions to Company) Act, 1993 most commercial state enterprises became public limited liability companies. GWSC’s conversion took place in the late 1990s. 4. The subsidiary companies include the Cocoa Marketing Company (CMC), Produce Buying Company (PBC), Cocoa Processing Company (CPC) and the Cocoa Research Institute of Ghana (CRIG). 5. Governments ﬁgures usually overestimate the real coverage. 238

Notes 239

Chapter 6 1. See Batley 1996, Nickson and Franceys 2003, Plummer 2002, World Bank, 1994 and 1997 for fuller explanation. 2. ‘Private sector’ is here taken to include NGO and user or community provision.

Chapter 7 1. Unless otherwise stated, sources for this chapter are Mills, Bennett and Russell (2001), Jackson (2002), Hubbard (2003), and Nickson and Franceys (2003), together with the International Development Department research working papers on which these books are based. 2. The past tense is used in this paragraph to recognize the changed business-government relationship in Zimbabwe since about 2000. The case is explained more fully in Chapter 8.

Chapter 8 1. This section is based largely on Nickson and Franceys (2003) and earlier working papers. 2. This section is based largely on Mills et al. (2001) and earlier working papers. 3. The reason do not include the externalities argument for intervention in public health services, nor the problems of risk and uncertainty that lead governments to intervene in the health insurance market. 4. Similar explanations were found for the under-stafﬁng of water pollution and dam safety control in Zimbabwe: ‘The rules set high standards; the problem is in their application. At an organizational level this is partly a question of insufﬁcient stafﬁng and equipment…’ (Batley 1998, p. 79) 5. This section is based largely on Jackson (2002) and earlier working papers.

Bibliography Abernethy, D. (1988), ‘Bureaucratic Growth and Economic Stagnation in Sub-Saharan Africa’ in S. K. Commins (ed.) (1988), Africa’s Development Challenges and the World Bank, Boulder, Lynne Rienner, pp. 179–214. Adams, A. V. and T. Hartnett (1996), Cost Sharing in the Social Sectors of Sub-Saharan Africa: impact on the poor, Washington DC, World Bank. Ahluwalia, M. (1996), ‘New economic policy and agriculture: some reﬂections’, Indian Journal of Agricultural Economics, Vol. 15, No. 3, pp. 412–426. Amis, P. (1989), ‘African development and urban change: what policy-makers need to know’, Development Policy Review, 7, pp. 375–391. Amis, P. (1992), ‘Urban Management in Uganda’, Institutional Framework of Urban Management, Development Administration Group, University of Birmingham. Anderson, G. (1991), ‘The Fiscal Signiﬁcance of User Charges and Earmarked Taxes’, in D. Wagner’, Charging for Government: user charges and earmarked taxes in principle and practice, London, Routledge. Arhin-Tenkorang, D. (2001), ‘Mobilizing Resources for Health: the case for user fees revisited’, CID Working Paper No. 81, Centre for International Development at Harvard University. Armstrong, M., S. Cowan and J. Vickers (1994), Regulatory Reform: economic analysis and British experience, Cambridge, MA, MIT Press. Ayeni, V. (2000), Empowering the Customer, London: Commonwealth Secretariat. Ayoola, G. B. et al. (2000), ‘Nigeria: voices of the poor’, Country Synthesis Report for World Development Report 2001, Washington DC, The World Bank. Bangura, Y. (2000), Public Sector Restructuring, Occasional Paper 3, Geneva: UNRISD. Bardhan, P. (1984), The Political Economy of Development in India, Oxford, Oxford University Press. Barzelay, M. (2002), Origins of New Public Management: an international view from public administration/political science. New Public Management: current trends and future prospects, in McLaughlin, K., S. P. Osborne and E. Ferlie (eds), London and New York, Routledge, pp. 15–33. Batley, R. A. (1994), ‘The Consolidation of Adjustment: implications for public administration’, Public Administration and Development, Vol. 14, No. 5, pp. 421–437. Batley, R. A. (1996), ‘Public-Private Relationships and Performance in Service Provision’, Urban Studies Vol. 33, Nos. 4–5, pp. 723–751. Batley, R. A. (1997), ‘A Research Framework for Analyzing Capacity to Undertake the New Roles of Government’, Role of Government in Adjusting Economies Paper 23, Birmingham: School of Public Policy, University of Birmingham. Batley, R. A. (1998), ‘Urban Water in Zimbabwe: performance and capacity analysis’, Role of Government in Adjusting Economies Paper 33, Birmingham: School of Public Policy, University of Birmingham. 240

Bibliography 241 Batley, R. and G. Larbi (1999), ‘Summary of Research Findings’, Role of Government in Adjusting Economies Paper 41, Birmingham: School of Public Policy, University of Birmingham. Bendor, J. (1990), ‘Formal Models of Bureaucracy: a review’, in Lynn, N. B. and A. Wildavsky (eds), Public Administration: the state of the discipline, Chatham NJ, Chatham House. Bennett, S. and A. Mills (1998), ‘Government capacity to contract: health sector experience and lessons’, Public Administration and Development, Vol. 18, No. 4, pp. 307–326. Bennett, S., A. Mills, S. Russell, A. Supachutikul and V. Tangcharoensathien (1998a), ‘The Health Sector in Thailand’, Role of Government in Adjusting Economies Paper 31, Birmingham: School of Public Policy, University of Birmingham. Bennett, S. and V. R. Muraleedharan (1998), ‘Reforming the Role of Government in the Tamil Nadu Health Sector’, Role of Government in Adjusting Economies Paper 28, Birmingham: School of Public Policy, University of Birmingham. Bitran, E. and P. Serra (1996), ‘Regulation of Privatized Utilities: the Chilean experience’, Congreso Interamericano del CLAD sobre la Reforma del Estado y de la Administración Pública, Centro Lationamericano de Administración para el Desarollo, Caracas, Venezuela. Bloom, G. and H. Standing (2001), ‘Pluralism and Marketisation in the Health Sector: meeting health needs in contexts of social change in low and middle income countries’, IDS Working Paper 136, University of Sussex, Institute of Development Studies. Borins, S. (1997), ‘What the New Public Management is Achieving: a survey of Commonwealth experience’, Advances in International Comparative Management: international perspectives on new public management, Greenwich, CT: JAI Press. Boston, J. (1987), ‘Transforming New Zealand’s Public Sector: labour’s quest for improved efﬁciency and accountability’, Public Administration 65, Winter, pp. 423–442. Breton, A. and R. Wintrobe (1975), The Logic of Bureaucratic Control, Cambridge, MA, Cambridge University Press. Breman, A. and C. Shelton (2001), ‘Structural Adjustment and Health: a literature review of the debates, its role-players and presented empirical evidence’, WHO Commission on Macroeconomics and Health, available at http://www.cmhealth.org/docs/wg6_paper6.pdf May 2003. Brinkerhoff, D. (1994), ‘Institutional development in World Bank projects: analytical approaches and intervention design’, Public Administration and Development 14, pp. 135–151. Buchanan, J. M. (1975), The Limits of Liberty: between anarchy and leviathan, Chicago, University of Chicago Press. Bulmer-Thomas, V. (1992), ‘Life after Debt: the new economic trajectory in Latin America’, Institute of Latin American Studies, University of London. Burki, Shahid Javed and Guillermo Perry (1998), ‘Beyond the Washington Consensus: institutions matter’, World Bank Latin American and Caribbean Studies: viewpoints, Washington DC, World Bank. Caiden, G. E. (1991), Administrative Reform Comes of Age, New York, Walter de Gruyter.

242 Bibliography Chawla, M. and R. P. Ellis (2000), ‘The impact of ﬁnancing and quality changes on health care demand in Niger’, Health Policy and Planning, Vol. 15, No. 1, pp. 76–84. Chisadza, E., C. C. Maponga, et al. (1995), ‘User Fees and Drug Pricing Policies – a Study at Harare Central Hospital, Zimbabwe’, Health Policy and Planning, Vol. 10, No. 3, pp. 319–326. Chiu, N. Kam (1997), ‘Service targets and methods of redress: the impacts of accountability in Malaysia’ Public Administration and Development, 17, 1: 175–180. Clark, J. and D. Wood (2001), New Public Management and Development: the case of public service reform in Tanzania and Uganda, in McCourt, W. and M. Minogue (eds), The Internationalization of Public Management: reinventing the Third World state, Edward Elgar, Cheltenham, pp. 70–89. Collier, D. (ed.) (1979), The New Authoritarianism in Latin America, Princeton, Princeton University Press. Common, R. (1998), ‘The new public management and policy transfer: the role of international organizations’, in Minogue, M., C. Polidano and D. Hulme (eds), pp. 59–75. Commonwealth Secretariat (1993), ‘Administrative and Managerial Reform in Government: a Commonwealth portfolio of current good practice’, Proceedings of a Pan Commonwealth Working Group Meeting held in Kuala Lumpur, Malaysia, 19–22 April, London: Commonwealth Secretariat. Cook, P. and C. Kirkpatrick (eds) (1995), Privatisation Policy and Performance, Prentice Hall. Cook P. and C. Kirkpatrick (eds) (1988), Privatisation in Less Developed Countries, Brighton, Wheatsheaf. Corkery, J. and T. O. Daddah, C. O. ’Nuallain, T. Land (eds) (1998), Management of Public Service Reform, Amsterdam, IOS Press. Cornia, G., R. Jolly and F. Stewart (eds) (1987), Adjustment with a Human Face, Vol. 1, Oxford, Clarendon Press. Corry, D., J. Le Grand and R. Radcliffe (1997), ‘Public/Private Partnerships: a marriage of convenience or a permanent commitment?’, Institute for Public Policy Research, London. Crook, R. C. (2002), ‘Urban service partnerships, “street level bureaucrats” and environmental sanitation in Kumasi and Accra, Ghana: coping with organizational change in the public bureaucracy’, World Development Report Workshop, Eynsham Hall, Oxfordshire, November 4–5. Davey K, R. Batley, N. Devas, M. Norris and D. Pasteur (1996), Urban Management: the challenge of growth, Aldershot, Avebury Press. De Silva, J. (2001), ‘Sri Lanka Country Report Summary’ in Regulatory System and Networking, Asian Development Bank, Manila. Devarajan S. and R. Reinikka (2002), Draft of World Development Report on ‘Making Services Work for Poor People’, World Bank. Dixon, J., A. Kouzmin and N. Korac-Kakabadse (1998), ‘Mangerialism – Something Old, Something Borrowed, Little New: economic prescriptions versus effective organizational change in public agencies’, International Journal of Public Sector Management, Vol. 11 (2/3): 164–87. Dolowitz, D. and D. Marsh (1998), ‘Policy transfer: a framework for comparative analysis’, in Minogue, M., C. Polidano and D. Hulme (eds), pp. 38–58.

Bibliography 243 Donaldson, C. and K. Gerard (1993), Economics of Health Care Financing: the visible hand, Basingstoke, Macmillan. Du Gay, P. (2000), In Praise of Bureaucracy, London, Thousand Oaks, New Delhi, Sage Publications. Dunleavy, P. and C. Hood (1994), ‘From Old Public Administration to New Management’ Public Money and Management, Vol. 14, No. 3, pp. 9–16. Dunsire A. and C. Hood (1989), Cutback Management in Public Bureaucracies: popular theories and observed outcomes in Whitehall, Cambridge, Cambridge University Press. Dwivedi, O. P. and D. Olowu (1988), ‘Bureaucratic Morality: an introduction’, International Political Science Review, Vol. 9, No. 3, pp. 163–5. Ellis, F., P. Senanayake and M. Smith (1997), ‘Food price policy in Sri Lanka’, Food Policy, Vol. 22, No. 1, pp. 81–96. Elster, J. (1993), ‘Constitution-making in Eastern Europe: rebuilding the boat in the open sea’, in Administrative Transformation in Central and Eastern Europe, Oxford, Blackwell. Esman, M. J. (1988), ‘The Maturing of Development Administration’, Public Administration and Development, Vol. 8, No. 2, pp. 125–34. Esman, M. J. (1999), ‘Public Administration and Conﬂict Management in Plural Societies: the case for representative bureaucracy’, Public Administration and Development, Vol. 19, No. 4, pp. 353–366. Evans, P. (1995), Embedded Autonomy: states and industrial transformation, Princeton, Princeton University Press. Ferlie E., L. Ashburner, L. Fitzgerald and A. Pettigrew (1996), The New Public Management in Action, Oxford, Oxford University Press. Fernando, K. N. N. S. (2001), ‘Regulation of the Water Services and Sanitation Sector; the proposed regulatory commission for water services’, National Water Supply and Sewerage Board, Law and Economy Conference, Colombo. Flynn, N. (1997), Public Sector Management, Hemel Hempstead, Harvester Wheatsheaf. Flynn, N. (1999), Miracle to Meltdown in Asia: business, government and society, Oxford, Oxford University Press. Flynn, N. (2002), ‘Explaining New Public Management – the importance of context’, In McLauglin, K., S. O. Osborne and E. Ferlie, New Public Management: current trends and future prospects, London and New York: Routledge. Foster, C. and W. Plowden (1996), The State under Stress, Buckingham, Open University Press. Franceys, R. (1995), ‘Case Studies in Water Utilities Management: Madras, MDSUPHO programme’, Unpublished report, WEDC, Loughborough University. Franceys, R. (1998), ‘Sri Lanka: urban water supply’, Role of Government in Adjusting Economies Paper 17, Birmingham: School of Public Policy, University of Birmingham. Franceys, R. and K. Sansom (1999), ‘India: urban water supply’, Role of Government in Adjusting Economies Paper 35, Birmingham: School of Public Policy, University of Birmingham. Francis, J. (1993), The Politics of Regulation: a comparative perspective, Oxford, Blackwell.

244 Bibliography Friedman, M. (1962), Capitalism and Freedom, Chicago, University of Chicago Press. Gausch, J. L. and R. W. Hahn (1999), The Costs and Beneﬁts of Regulation: implications for developing countries, Policy Research Working papers, Washington DC, World Bank. Gilson, L. (1997), ‘Lessons of user fee experience in Africa’, Health Policy and Planning, Vol. 12, No. 4, pp. 273–285. Gilson, L., D. Kalyalya, et al. (2000), ‘The equity impacts of community ﬁnancing activities in three African countries’, International Journal of Health Planning and Management, Vol. 15, No. 4, pp. 291–317. Goodsell, C. T. (1995), The Case for Bureaucracy, New Jersey, Chatham House. Grindle, M. S. (1980), The Politics of Policy Implementation, Princeton, N.J., Princeton University Press. Grindle, M. S. (ed.) (1997), Getting Good Government, Cambridge, Harvard University Press. Grindle, M. S. (2001), ‘Designing Reforms: problems, solutions, and politics’, Faculty Research Working Paper No. RWP01–020, John F. Kennedy School of Government, Harvard University. Grindle, M. S. (2003), Despite the Odds: the contentious politics of education reform, Baltimore, MD and London: Johns Hopkins University Press. Grindle, M. S. and J. W. Thomas (1991), Public Policy and Policy Change: the political economy of reform in developing countries, Baltimore and London, Johns Hopkins University Press. Harding, A. and A. Preker (eds) (2002), ‘Conceptual Framework’, in Innovations in Health Service Delivery: the corporatization of public hospitals, Washington DC, World Bank. Harriss, J., J. Hunter and C. Lewis (eds) (1997), The New Institutionalism and Third World Development, London, Routledge and Kegan Paul. Hayek, F. A. von. (1973), Law, Legislation and Liberty. Vol. 1. Chicago, IL.: Chicago University Press. Helm, D. (1999), ‘Regulation and Responsibility’, The Utilities Journal, Vol. 2, November, OXERA, Oxford. Herbst, J. (1993), The Politics of Reform in Ghana, 1982–1991, Berkeley, C.A., University of California Press. Hilderbrand, M. E. and M. S. Grindle (1995), ‘Building Sustainable Capacity in the Public Sector: what can be done?’, Public Administration and Development, Vol. 15, No. 5, pp. 441–463. Hirschmann, D. (1993), ‘Institutional Development in the Era of Economic Policy Reform: concerns, contradictions and illustrations from Malawi’, Public Administration and Development, Vol. 13, No. 2, pp. 113–128. Hongoro, C. and L. Kumaranayake (2000), ‘Do they work? Regulating forproﬁt providers in Zimbabwe’, Health Policy and Planning, Vol. 15, No. 4, pp. 368–378. Hood, C. (1991), ‘A Public Management for All Seasons?’, Public Administration, 69, pp. 3–19. Hood, C. (1995), ‘Contemporary Public Management: a new global paradigm’, Public Policy and Administration, Vol. 10, No. 2. Hood, C., O. Jones and T. Travers (1998), ‘Regulation inside Government: where new public management meets the audit explosion’, Public Policy and Management, Vol. 18, No. 2.

Bibliography 245 Hope, R. K. (2002), ‘The New Public Management: a perspective from Africa’, in McLauglin, K., S. P. Osborne and E. Ferlie (ed.), New Public Management: current trends and future prospects, London and New York, Routledge. Hubbard, M. (2003), Developing Agricultural Trade: new roles for government in poor countries, Houndmills and New York, Palgrave. Hubbard, M. (ed.) (1999), ‘The Impact of Liberalization on Public Services to Agricultural Marketing: case studies of change options in Zimbabwe’, Role of Government in Adjusting Economies Paper 36, Birmingham: School of Public Policy, University of Birmingham. Hughes, O. E. (1998), Public Management and Administration, London, Macmillan Press. Hyden, G. (1983), No Shortcuts to Progress: African development management in perspective, London, Heinemann. International Labour Organization (1995), Impact of Structural Adjustment in the Public Services: efﬁciency, quality improvement and working conditions, Geneva, ILO. ISC (International Soldidarity Committee) (2001), ‘Water is Not a Commodity! Water is Life and Life is for All’, ISC for the Ghana National Coalition against the Privatisation of Water, 31 May, Integrated Social Development Centre, Accra. Jackson, P. B. (2002), Business Development in Asia and Africa, Houndmills and New York, Palgrave. Jalan, J. and K. Subbarao (1995), ‘Adjustment and Social Sectors in India’, in Cassen, R. and V. Joshi (eds), The Future of Economic Reform, Delhi, Oxford University Press. Jayarah, C. and W. Branson (1995), ‘Structural Adjustment and Sectoral Adjustment: World Bank experience 1980–1992’ A World Bank Operations and Evaluation Study. Washington DC, World Bank. Jervis, P. and S. Richards (1995), ‘Strategic Management in a Re-invented Government: Rowing 1, Steering 0’, Strategic Management Society 15th Annual Conference: Strategic Discovery: opening new worlds, Mexico City, 15–18 October. Jordan, B. (1995), ‘Are New Right Policies Sustainable? Back to basics and public choice’. Journal of Social Policy, Vol. 24(3): 363–84. Joshi, A. and M. Moore (2002), ‘Organisations that Reach the Poor: why coproduction matters’, World Development Report Workshop, Eynsham Hall, Oxfordshire, November 4–5. Kaboolian, L. (1998), ‘The New Public Management: challenging the boundaries of the management versus administration debate’, Public Administration Review, Vol. 58, No. 3. Kapoor, V. (2001), India Country Report, Presentation to the Regional Forum on Regulatory Systems and Networking of Water Utilities and Regulatory Bodies, 26–28 March, Asian Development Bank, Manila. Kessler, T. (2003), ‘Review of the World Development Report on Making Services Work for Poor People’, Citizens’ Network on Essential Services (CNES), http://www.servicesforall.org/html/tools/2004WDR_review_print.shtml Kernaghan, K., B. Marson and S. Borins (2000), The New Pubic Organization, Toronto: Institute of Public Administration of Canada. Kohli, R. and M. Smith (1998), ‘The Role of the State and Agricultural Markets in India’, Role of Government in Adjusting Economies Paper 29, Birmingham: School of Public Policy, University of Birmingham.

246 Bibliography Konde-Lule, J. K., D. Okello, M. Mobonye and F. Maharo (1998), ‘User Fees in Government Health Units in Uganda: implementation, impact and scope’, PHR Resource Centre, http://www.phrproject.com/publicat/inbrief/ib27ﬁn.htm 04.04.2003. Lane, J.-E. (2000), The Public Sector: concepts, models and approaches, 3rd edition, London, Newbury Park and New Delhi, Sage Publications. Lane, J.-E., and S. Ersson (2000), The New Institutional Politics: performance and outcomes, London and New York, Routledge. Larbi, G. A. (1995), ‘Implications and Impact of Structural Adjustment on the Civil Service: the case of Ghana’, Role of Government in Adjusting Economies Paper 2, Birmingham: School of Public Policy, University of Birmingham. Larbi, G. A. (1998a), ‘Implementing New Public Management Reforms in Public Services in Ghana: institutional constraints and capacity issues’, Unpublished PhD Thesis, School of Public Policy, University of Birmingham. Larbi, G. A. (1998b), ‘Management decentralization in practice: a comparison of public health and water services in Ghana’, in Minogue, M., C. Polidano and D. Hulme (eds), Beyond the New Public Management: changing ideas and practices in governance, Cheltenham, Edward Elgar, pp. 188–206. Larbi, G. A. (1998c), ‘Institutional Constraints and Capacity Issues in Decentralising Management in Public Services: the case of health in Ghana’. Journal of International Development 10 (3): 377–386. Larbi, G. A. (1998d), ‘Contracting-out in public health and water services in Ghana’, International Journal of Public Sector Management, Vol. 11, Nos. 2/3, pp. 154–163. Larbi, G. A. (1999a), The New Public Management Approach and Crisis States, Discussion Paper No. 112, Geneva, United Nations Research Institute for Social Development. Larbi, G. A. (1999b), Trade Unions, Public Sector Employees and New Public Management Reforms: impact, responses and reactions to change, Geneva, United Nations Research Institute for Social Development. Larbi, G. A. (2001), Performance Contracting in Practice: experience and lessons from the water sector in Ghana, Public Management Review, Vol. 3, No. 3, pp. 305–224. Larbi, G. A. (2003), ‘Financing Health Service Provision: user fees impact, performance and the role of government’, A paper presented to the Conference on Shared Governance: combating poverty and exclusion, International Institute of Administrative Sciences, Second Regional Conference, Yaoundé, Cameroun, 14–18 July 2003. Lewa, P. (1998), ‘Managing the Public Role in a Liberalized Maize Marketing Environment in Kenya’, Role of Government in Adjusting Economies Paper 30, Birmingham: School of Public Policy, University of Birmingham. Loughlin, M. and C. Scott (1997), ‘The Regulatory State’ in Dunleavy, P. et al., Developments in British Politics, London, Macmillan, pp. 205–18. Lynn, L. E. (2001), ‘The Myth of the Bureaucratic Paradigm: what traditional public administration really stood for’. Public Administration Review, Vol. 61, No. 2, pp. 144–160. Malkin, J. and A. Wildavsky (1991), ‘Why the Traditional Distinction between Public and Private Goods Should Be Abandoned’, Journal of Theoretical Politics, Vol. 3 No. 4, pp. 355–378.

Bibliography 247 Manning, Nick (2001), ‘The legacy of the New Public Management in developing countries’, International Review of Administrative Sciences, Vol. 67, pp. 297–312. McCourt, W. and M. Minogue (2001), The Internationalization of Public Management: reinventing the Third World state, Cheltenham, UK and Northampton, MA, Edward Elgar. McKoy, D. (2001), ‘Issues in Public Sector Accountability: lessons for the executive agencies in Jamaica’, Caribbean Journal of Public Sector Management, Vol. 3, No. 1, pp. 12–29. McLaughlin, K., S. P. Osborne, et al. (eds) (2002), New Public Management: current trends and future prospects, London and New York, Routledge. Migdal, J. S. (1988), Strong Societies and Weak States: state-society relations and state capabilities in the Third World, Princeton, Princeton University Press. Mills, A. (ed.) (2002), Reforming Health Sectors, London, Kegan Paul. Mills, A., P. Vaughan and D. L. Smith (eds) (1990), ‘Health Systems Decentralization: concepts, issues, and country experience’, World Health Organization, Geneva. Mills, A., S. Bennett and S. Russell (2001), The Challenge of Health Sector Reform: what must governments do? Houndmills and New York, Palgrave. Mills, A., R. Brugha, K. Hanson and B. McPake (2002), ‘What can be done about the private health sector in low-income countries?’ Bulletin of the World Health Organization, Vol. 80, No. 4, pp. 325–330. Ministry of Health, Ghana (1996), ‘Institutional Reform in the Health Sector’, Accra, Ministry of Health. Minogue, M. (1998), ‘Changing the state: concepts and practices in the reform of the public sector’, in Minogue, M., C. Polidano and D. Hulme (eds), Beyond the New Public Management: changing ideas and practices in governance, Cheltenham, Edward Elgar, pp. 17–37. Moser, C., A. Herbert and R. Makonnen (1993), Urban Poverty in the Context of Structural Adjustment: recent evidence and policy responses, Discussion Paper, Urban Development Division, Washington DC, World Bank. Mosley, P., J. Harrigan and J. Toye (1995), Aid and Power: the World Bank and policy-based lending, second edition, London, Routledge. Mudege, N. (1997), ‘Urban Water Supply in Zimbabwe’, Role Of Government In Adjusting economies Paper 18, Birmingham: School of Public Policy, University of Birmingham School of Public Policy, University of Birmingham. Mueller, D. C. (1979), Public Choice, Cambridge University Press. Mutizwa-Mangiza, D. (1997), ‘The Opinions of Health and Water Service Users in Zimbabwe, Role of Government in Adjusting Economies Paper 24, Birmingham: School of Public Policy, University of Birmingham. Nanda, P. (2002), ‘Gender Dimensions of User Fees: implications for women’s utilization of health care’, Reproductive Health Matters, Vol. 10, No. 20, pp. 127–134. Nellis, J. (1988), ‘Contract Plans and Public Enterprise Performance’, Working Paper 118, Washington DC, World Bank. Nelson, J. M. (2000), The Politics of Social Sector Reforms, Washington DC, Overseas Development Council, www.iadb.org/sds/doc/nelson1.pdf Nickson, R. A. (2001a), ‘Establishing and Implementing a Joint Venture: water and sanitation services in Cartagena, Colombia’, Building Municipal Capacity

248 Bibliography for Private Sector Participation Series, Working Paper No. 442 03, London, GHK International. Nickson, R. A. (ed.) (2001b), ‘Tapping the Market: can private enterprise supply water to the poor?’, Insights No. 37, ID21, IDS, Sussex. Nickson, R. A and C. Vargas (2002), ‘The limitations of water regulation: the failure of the Cochabamba concession in Bolivia’, Bulletin of Latin American Research Vol. 21, No. 1, January–March, pp. 99–120. Nickson, R. A. and R. Franceys (2003), Tapping the Market: the challenge of institutional reform in the urban water supply sector, Houndmills and New York, Palgrave. Niskanen, W. A. (1971), Bureaucracy and Representative Government, Chicago, IL, Aldine-Atherton. Nolan, B. and V. Turbat (1995), ‘Cost Recovery in Public Health Services in Sub-Saharan Africa’, Economic Development Institute Technical Material. Washington DC, World Bank. North, D. C. (1990), Institutions, Institutional Change and Economic Performance, Cambridge, Cambridge University Press. Nunberg, B. (1995), ‘Managing the Civil Service: reform lessons from advanced industrialized countries’, World Bank Discussion Papers No. 25. Nunberg, B. and I. Nellis (1995), ‘Civil Service Reform and the World Bank’, World Bank Discussion Paper 161, Washington DC, World Bank. Nyonator, F. and J. Kutzin (1999), ‘Health for some? The Effects of User Fees in the Volta Region of Ghana’, Health Policy and Planning, Vol. 14, No. 4, pp. 329–341. Odle, Maurice (1993), ‘Towards a stages theory approach to privatization’, Public Administration and Development, Vol. 13, pp. 17–35. Organization for Economic Cooperation and Development (OECD) (1993), Managing with Market-type Mechanisms, Paris, OECD. Organization for Economic Cooperation and Development (OECD) (1997), The OECD Report on Regulatory Reform, Paris, OECD. Organization for Economic Cooperation and Development (OECD) (1999), Performance Contracting: lessons from performance contracting case studies/ framework for public sector performance contracting, Paris, OECD. Ogus, A. (1994), Regulation: legal form and economic theory, Oxford University Press. Olowu, D. (1999), ‘Public Service Delivery’, in Adamolekun, L. (ed.) Public Administration in Africa: main issues and selected country studies, pp. 123–138, Boulder, CO and Oxford, Westview. Osborne, D. and T. Gaebler (1992), Reinventing Government, Reading, Mass., Addison Wesley. Ostrom, E. (1997), ‘Crossing the Great Divide: co-production, synergy and development’, in Evans, P. (ed.), State-Society Synergy: government and social capital in development, Berkeley, International Area Studies, University of California. Panos Institute (2002), Reducing Poverty: is the World Bank’s strategy working? London, Panos Institute. Payer, C. (1982), The World Bank: a critical analysis, New York, Monthly Review Press. Peters, B. Guy (1996), The Future of Governing: four emerging models, University Press of Kansas.

Bibliography 249 Picciotto, R. (1995), ‘Putting Institutional Economics to Work: from participation to governance’, World Bank Discussion Papers 304, The World Bank, Washington. Plummer, J. (2002a), Focusing Partnerships: a sourcebook for municipal capacitybuilding in public-private partnerships, London, Earthscan. Plummer, J. (2002b), ‘Developing Inclusive Public-Private Partnerships: the role of small-scale independent providers in the delivery of water and sanitation services’, Paper for a DFID/World Bank Seminar, Eynsham Hall, Oxfordhire. Polidano, C. (2001), Administrative Reform in the Core Civil Services: application and applicability of the new public management, in McCourt, W. and M. Minogue (eds), The Internationalization of Public Management: reinventing the Third World state, Edward Elgar, Cheltenham, pp. 44–69. Pollitt, C. (1993), Managerialism and the Public Service: the Anglo-American experience, second edition, Oxford, Blackwell. Price Waterhouse (1994), ‘Study on the Regulation of Private Monopoly in Developing Countries’, A report by Price Waterhouse Corporate Finance, London. Rakodi, C. (1996), ‘The Opinions of Health and Water Sector Users in Ghana’, Role of Government in Adjusting Economies Paper 10, Birmingham: School of Public Policy, University of Birmingham. Rakodi, C. (1998), ‘The Opinions of Health and Water Sector Users in India’, Role of Government in Adjusting Economies Paper 32, Birmingham: School of Public Policy, University of Birmingham. Rakodi, C. (2000), ‘Getting the pipe laid is one matter and getting the water ﬂowing through the pipe is another’, International Planning Studies 5(3): 365–91. Riccucci, N. M. (2001), ‘The “Old” Public Management Versus “New” Public Management: where does public administration ﬁt in?’ Public Administration Review, Vol. 61, No. 12, pp. 172–175. Republic of Ghana (1993), ‘Industrial Policy Statement: a strategy for industrial regeneration’, unpublished policy report by the Ministry of Industries, Accra. Ridley, F. (1996), ‘The New Public Management in Europe: comparative perspectives’, Public Policy and Administration, Vol. 11, No. 1, pp. 16–29. Rowat, D. (1990), ‘Comparing Bureaucracies in Developed and Developing Countries: a statistical analysis’, International Review of Administrative Sciences, Vol. 56(2), pp. 211–231. Russell, S. and N. Atanayake (1997), ‘Reforming the Health Sector in Sri Lanka: does government have the capacity?’, The Role of Government in Adjusting Economies Paper 14, School of Public Policy, The University of Birmingham. Russell, S., P. Kwaramba, C. Hongoro and S. Chikandi (1997) ‘Reforming the Health Sector in Zimbabwe: does government have the capacity?’, The Role of Government in Adjusting Economies Paper 20, School of Public Policy, The University of Birmingham. Russell, S., S. Bennett and A. Mills (1999), ‘Reforming the health sector: towards a healthy new public management’, Journal of International Development, Vol. 11, pp. 767–775. Sako, M. (1992), Prices, Quality and Trust, Cambridge, Cambridge University Press.

250 Bibliography Salamon, L. M. (2002), ‘The New Governance and the Tools of Public Action: an introduction’, in Salamon L. M. (ed.), The Tools of Government: a guide to the new governance, Oxford and New York, Oxford University Press. Sandbrook, R. (1993), The Politics of Africa’s Economic Recovery, Cambridge, Cambridge University Press. Savas, E. S. (1987), Privatization: the key to better government, Chatham, NJ, Chatham House. Savas, E. S. (1990), A Typology of Market-Type Mechanisms, PUMA, MTM Series, Paris, OECD. Savas, E. S. (2000), Privatization and public-private partnerships, London and New York, Chatham House. Sen, A. (1988), ‘Sri Lanka’s achievements: how and when’, in Srinivasan, T. N. and P. K. Bardhan (eds), Rural Poverty in South Asia, Delhi, Oxford University Press. Sharma, A. and V. K. Agnihotri (2001), ‘The citizens charter: the Indian experience’, International Review of Administrative Sciences, Vol. 67, No. 4. Shirley, M. M. and C. L. Xu (1997), ‘Information, Incentives, and Commitment: an empirical analysis of contracts between government and state enterprises’. Accessed at http://www.worldbank.org/pub/decweb/workingpapers Shore, K. J. (2002), ‘Who Pays? Municipal Services in South Africa’, available at http://www.idrc.ca/reports/ accessed on 17.01.02. Sifﬁn, W. J. (1976), ‘Two Decades of Public Administration in Developing Countries’, Public Administration Review, Vol. 36(1), pp. 61–71. Silva, T., S. Russell and C. Rakodi (1997), The Opinions of Health and Water Sector Users in Sri Lanka, Role of Government in Adjusting Economies Paper 25, Birmingham: School of Public Policy, University of Birmingham. Slater, R., Ram Khanna, T. K. Bhaumik and R. Basu (1998), ‘The Role of the State in Business Development: the case of the textiles and garments sector in India, Role of Government in Adjusting Economies Paper 27, Birmingham: School of Public Policy, University of Birmingham. Smith, B. C. (1996), Understanding Third World Politics: theories of political change and development, London, Macmillan. Smith, M. and F. Ellis (1997), ‘The Role of the State in Agricultural Marketing in Sri Lanka’, Role of Government in Adjusting Economies Paper 19, Birmingham: School of Public Policy, University of Birmingham. Smithson, P., A. Asamoa-Baah and A. Mills (1997), ‘The Case of the Health Sector in Ghana’, Role of Government in Adjusting Economies Paper 26, Birmingham: School of Public Policy, University of Birmingham. Stanley, W. (2001), ‘The Role of Regulation’, Presentation to the Regional Forum on Regulatory Systems and Networking of Water Utilities and Regulatory Bodies, 26–28 March 2001, Asian Development Bank, Manila. Stiglitz, J. (1987), ‘Principal and Agent’ in Eatwell, J., M. Milgate, and P. Nennan (eds) (1987), The New Palgrave Dictionary of Economics, pp. 241–253, London, Palgrave. Stone, D. C. (1965), ‘Government Machinery Necessary for Development’, in M. Kriegsberg (ed), Public Administration in Developing Countries, The Brookings Institution, Washington DC, pp. 49–67. Stren, R. and R. White (eds) (1989), African Cities in Crisis: managing urban growth, Boulder, Colorado, Westview Press.

Bibliography 251 Talbot, C. and J. Caulﬁeld (eds) (2002), ‘Hard Agencies in Soft States? A Study of Agency Creation Programmes in Jamaica, Latvia and Tanzania’, A report for the Department for International Development, London. Talbot, J. (2002), ‘Is the international water business really a water business?’, Water and Sanitation Lecture Series, 13th February, Washington DC, World Bank. The Economist (1997), ‘The Price of Competition’, The Economist, 12 September. Therkildsen, O. (1999), ‘Efﬁciency and Accountability: public sector reform in East and Southern Africa’, mimeo, Geneva: UNRISD. Thynne, I. (1998), Government companies as instruments of state action, Public Administration and Development, Vol. 18, No. 3, pp. 217–228. Todaro, Michael (1994), Economic Development, Fifth Edition, London, Longman. Tritter, J. (1994), ‘The Citizens Charter: opportunities for user perspectives’, Political Quarterly, Vol. 65, No. 4, pp. 397–414. Tullock, G. and K. Eller (1994), Rent Seeking, London: Edward Elgar. Turner, M. and D. Hulme (1997), Governance, Administration and Development: making the state work, London, Macmillan. United Nations Development Programme (UNDP) (2000), Human Development Report, New York, United Nations Development Programme. United Nations Research Institute for Social Development (2000), Visible Hands: taking responsibility for social development, Geneva, UNRISD. Van der Geest, S., M. Macwan’gi, J. Kamwanga, D. Mulikelela, A. Mazimba and M. Mwangela (2000), ‘User Fees and Drugs: what did the health reforms in Zambia achieve?’ Health Policy and Planning, Vol. 15, No. 1, pp. 59–65. Vickers, J. and G. Yarrow (1989), Privatization: an economic analysis, Cambridge, MA, MIT Press. Vining, A. and D. Weimer (1990), ‘Government Supply and Government Production Failures: a framework based on contestability’, Journal of Public Policy, Vol. 10, No. 1, pp. 1–22. Waddington, C. J. and D. A. Enyimayew (1990), ‘A Price to Pay: the impact of user charges in the Volta region of Ghana’, International Journal of Health Planning and Management, Vol. 5, pp. 287–312. Wade, R. (1990), Governing the Market: economic theory and the role of government in East Asian industrialization, Princeton, Princeton University Press. Walsh, K. (1995), Public Services and Market Mechanisms: competition, contracting and the new public management, Basingstoke and London, Macmillan. Weber, M. (1978), Economy and Society, 2 Volumes, Los Angeles, University of California Press. Weimar, D. L. and A. R. Vining (2001), Policy Analysis: concepts and practice, 2nd edition, Prentice-Hall, Eaglewood Cliffs. Wildavsky, A. (1988), Searching for Safety, New Brunswick, Transaction Books. Williamson, O. (1975), Markets and Hierarchies, New York, Free Press. Williamson, O. (1987), The Economic Institutions of Capitalism, New York, Free Press. Wilson, J. Q. (1987), Bureaucracy: what government agencies do and why they do it, New York, Basic Books. World Bank (1983), World Development Report, New York, Oxford University Press. World Bank (1987), Financing Health Services in Developing Countries: agenda for reform, Washington DC, World Bank.

252 Bibliography World Bank (1989), Sub-Saharan Africa: from crisis to sustainable growth, Washington DC, World Bank. World Bank (1991), ‘The Reform of Public Sector Management: lessons of experience’, Policy Research Series 18, Washington DC, World Bank. World Bank (1993), World Development Report 1993: investing in Health, New York, Oxford University Press. World Bank (1993a), The East Asian Miracle: economic growth and public policy, World Bank, Oxford University Press. World Bank (1994), World Development Report 1994: infrastructure for development, New York, Oxford University Press. World Bank (1996), World Development Report 1996: from plan to market, New York, Oxford University Press. World Bank (1997), World Development Report 1997: the state in a changing world, New York, Oxford University Press. World Bank (1998), World Development Report 1998: knowledge for development, New York, Oxford University Press. World Bank (1999), Meeting the Health Challenge in Zimbabwe, Washington DC, World Bank. World Bank (2000), The Road to Stability and Prosperity in South-Eastern Europe: a regional strategy paper, Europe and Central Asia Region, Washington DC, World Bank. World Bank (2001), World Development Indicators 2001, Washington DC, World Bank. World Bank (2003a), www.worldbank.org/website/external/news (04.04.03). World Bank (2003b), World Development Report 2004: making services work for poor people, Washington DC, World Bank.

Index accountability 17, 36, 39, 44–47, 71, 80, 102–105, 109, 126, 127, 171, 194, 222, 229, 234–36 advanced countries public sector 3 public sector reform 220, 221 Africa aid 3 civil service 14 ﬁnancial crisis 109 GNP 11 health services 70, 91, 116 hospitals 91 public expenditure cuts 13 public sector decline 29 public sector reform 2, 6 revenue colllection 45 water services 108 see also under names of countries agencies 45, 46, 47, 86–87, 205, 206–11, 214, 216, 223, 232 agricultural marketing build-operate-transfer schemes 151–53 charges 108, 115–16, 121, 229 contracting out 146, 147–48, 161, 162, 169 focus on 22, 23 government intervention 24, 25 IMF and 60 organization 84–85, 94–96 performance 99 poor people and 229 private partners and 141 reform 68–69, 73, 98 types 55 Aguas de Barcelona 157 Aguas de Cartagena 157 aid agencies and donors 2, 6, 52, 57–62, 66, 70, 72, 74, 78, 80, 104, 114, 140, 178, 196, 224 aid programmes 3

Argentina Buenos Aires 69, 154–55 concessions 154–55, 161, 172 economic crisis 155 public service reform 61 as reference country 21 regulation 191 water services 69, 74, 108, 121, 154–55, 172, 192–93, 215–16, 230 Asia economic crisis in 38–39 public expenditure cuts 13 Bolivia contracting out 173 as reference country 21 regulation 191 water services 47, 102, 108, 115, 121, 136, 161, 224 Botswana 229 budget deﬁcits 38 build-operate-transfer schemes nature of 134, 151–54, 233 bureaucracy neutrality 33 politicization of 56 weakness of 31 Weberian model 31, 40 business development focus on 22, 23 public-private partnerships 156–58 business reform: IMF and 60 business, services to 156–57, 204–6, 207 capacity analyzing 19 charges and 124 concepts of 17–20 constraints on 216–18, 224–32

253

254 Index capacity – continued contracting 167–77 development 233–34, 236 external aspects 18–19, 167–75, 203, 214, 217, 227–28, 234–35 internal aspects 18–19, 167–71, 174, 202, 213, 217, 227–28 organization and 136–42 public service reform and 1, 7, 15–20, 102, 105, 106, 232, 235 regulation 186, 193–94, 201–3 charges capacity and 124, 233 controversial 110 developing countries and 51–52 efﬁciency and 108, 121–23, 124 equity and 117–21, 123, 124, 229–30 explaining 106–9 motives for 107, 108, 109 performance and impact 116–23 political aspect 124 poor and 117, 118, 119, 123 problems 52 revenue motive 229 widening 107 Chile 149, 150, 161, 191, 233 civil service: expansion restraint 14 Colombia 157, 161 concessions 134, 154–55, 233 see also under names of countries contracting out administrative failures 176 assessment 158–67 beneﬁts 178 capacity for 167–77 constraints on 169, 179 contractors’ incapacity 177 disadvantage 134 experience of 144–79 extension of 49, 230 external aspects 170, 172 feasibility 163–67 ﬁnancial control 176 forms of 130–39, 144–48 implementation of 161, 179 information systems 170, 176 long-term contracts 164, 176, 178 nature of 133–34

new public management and 49–51 one-off 145–47, 163 opposition to 173, 176–77, 179 performance and 158–62 political aspects 170–1, 173, 175 private sector capacity 173–74 public-private relations 175 service characteristics and 137–38 service and management contracts 147–51 short-term contracts 163, 167, 178, 230 contracts advantages 36–37, 38 difﬁculties 37 see also previous entry corruption 46, 100, 104, 171 countries, selection of 20–21 de-regulation 184–85 developing countries agencies 46 bureaucracy 53 external actors 59 performance appraisal 48–49 public sector 3 public sector reform 27, 53, 54 appropriateness of 27, 29, 38 state, extension of in 2–4 state, role of 5, 220 see also under names of countries development, government and 2–7 East Asia 4 ﬁnancial crisis 4 East European states 5 education privatization 129 reform in 60 electricity generation 185, 191 enablement approaches to 203–4 capacity 211–15, 234 constraints on 216–18 forms of 182 nature of 180, 183–84 regulation and 215–18, 231–32

Index 255 Ghana agricultural marketing 68, 85, 95, 100, 115, 229 business development agencies 156 business services 210–11, 212, 213, 214 charges 51, 52, 109, 110, 111, 115, 117–18, 121, 122 civil service 14 Civil Service Performance Improvement Programme (CSPIP) 47 Cocoa Marketing Board (Cocobod) 85, 95, 100, 102, 115, 224 cocoa sector 60, 62, 68, 73, 100 contracting out 139, 147, 157, 158, 163, 168, 171, 173, 174, 175 Economic Recovery Programme 10 Finance Ministry 104 ﬁnancial crisis 9, 10 Food Distribution Corporation 85, 95, 100 GNP per capita 11, 12 Health Ministry 111 health service 65, 77, 82–83, 88–89, 90, 91, 98, 99, 104, 109, 111, 112, 117–18, 121, 122, 147, 157, 158, 163, 168, 171, 196, 197, 198, 200, 202, 203 hospitals 47, 90 industrial sector 84, 97, 98 Industry and Commerce Ministry 72, 140, 212, 213 Investment Promotion Centre 97 leasing 154, 171 new public management 47 perfomance agreements 48 private partners 141 Produce Buying Company 85, 95, 100 public expenditure cuts 13 public opinion 222 public service reform 61, 62, 220, 223 Public Utilities Regulatory Commission 113 regulation 191, 196, 197, 198, 200, 202, 203, 216 revenue collection 46

structural adjustment 10, 21 as study choice 21 water services 65, 66, 69, 74, 77, 83, 92–93, 94, 98, 101, 102, 104, 112–13, 121, 122, 141, 154, 191, 216, 235 World Bank 92 government intervention public choice theory 34–35 reasons for 32 theories on organization of 33–38 government, role of capacity and 15–17 changing views on 1–30 development and 2–7 effectiveness of 6–7 intervention model 8 liberalizing 7 reform and 16–17 scepticism about 4 social services and 8 health services capacity issues 168–71, 233, 234 charges 106–7, 108, 109, 110–12, 117–18, 123, 124, 229 contracting out 147, 149–51, 157–58, 159–60, 162, 169, 170, 177, 178 focus on 21, 22, 23 government intervention 25, 27 hospitals, autonomous 90 human resource capacity 103 insurance schemes 107, 123 organization 82–83, 88–92, 105 performance 99 private partners 142 privatization 129 reform 60, 61–62, 70–71, 73–76, 79, 98, 223–24 types 55 regulation 195–203, 215, 216, 231 users’ opinions 222–23 IMF (International Monetary Fund) economic crisis and 5 public service reform and 59 structural adjustment and 10, 14, 51, 59, 220

256 Index IMF (International Monetary Fund) – continued UK and 4–5 Zimbabwe and 10 India agencies 216 agricultural marketing 1, 60, 68, 85, 116, 141, 147–48, 229 build-operate-transfer schemes 171, 233 business development agencies 156 business services 208, 209, 210, 211, 212, 213, 214 charges 109, 110, 116, 119, 122 concessions 154, 171 contracting out 147–49, 151, 163, 173, 175 Finance Ministry 73 Food Corporation 85 GNP per capita 11, 12 health care 71 health service 13, 65, 71, 82, 83, 91, 104, 149, 151, 163, 168–70, 196, 198–200, 202, 203 industrial sector 72–73, 84, 140 private partners 141 public service reform 56, 61 public expenditure cuts 13 public opinion 222–23 regulation 190, 196, 198–200, 202, 203 staff cuts 14 structural adjustment 10, 21 as study choice 20–21 water services 65, 83, 94, 101, 114, 119, 141, 147, 148–49, 154, 172 World Bank and 75 Indonesia 191 industrial sector charges 108 development 22, 23 organization 84 private partners and 140 privatization 72, 144 reform 55, 72, 96–98 regulation 204–5 Jamaica 45, 46 joint venture 135–36, 156–58

Kenya agricultural marketing 76, 85, 96, 98, 115, 122, 229 charges 51, 115 contracting out 146, 174, 175 as reference country 21 regulation 189–90 revenue collection 46 Latin America authoritarian rule 4 charges 52 civil service 14 public expenditure cuts 13 public sector reform 5–6, 9 state in 4 water services 108 see also under names of countries leasing 134, 154–55, 233 licensed competition, producers 134–35 Malaysia 154, 161 market failure 23, 24, 32, 181, 183, 203, 204, 215 market mechanisms bias towards 5 contracting and charging 50, 51 new public management 41–44, 220 privatizing management 125–28 replacing hierarchies 40 state, need of 221 Menem, President Carlos 74 monopolies concession 134 Namibia 46, 229 Neo-classical theory 32, 33 neo-liberalism 35 economic crisis and 4–5 economic policies 31 new public management adoption of 40 decentralizing 44–47 deﬁnition 40, 41 evolution of 40–44 ﬁnancial control 46 key elements 41, 42–43, 44 mixed experiences 53

Index 257 performance 47–49 review of 44–52 New Zealand 39, 45, 47, 48 NGOs 129 OECD (Organization for Economic Co-operation and Development) 3, 41, 185 oil prices 4, 54 Ondeo 154–55 organization capacity and 136–42 pre-reform 81–85 private-public 129–32 reform of 88–98 partnership 135–36 performance agreements 48, 87–88 performance auditing 47, 48–49 Gulf War, ﬁrst 10 Philippines 74, 155, 191 poverty re-distribution of 10–13 reduction 6, 9, 29, 221 Principal-agent theory 35–36, 54, 58–65, 67, 125–27 private partners case for 125–28 categories of 133–36 functions split 129 public interest and 127 privatization 16, 38, 49, 128, 143 Property rights theory 37–38 Public choice theory 34–35 public ownership 33 public sector: reducing 5, 108 public sector reform adjustment of 6–7 administrators 71–72 agents 35, 58–66, 71–72, 125, 127 appropriateness of 27, 29, 38, 106, 219, 220, 221, 235, 236 autonomy 102, 105, 224–29 background 7–15 central tendencies 222 citizens and 63, 65–66, 71, 222, 235

countries studied 7 crises and 9–10, 38–39, 58, 75, 108, 220, 221 demanding nature of 29 donors and 59, 61, 62 economic crises and 38 emergency constituencies 76–77 expenditure reduced 13–14 history of 220–21 implementation of 98–99, 105, 235 institutional background 8–9 ofﬁcials and 223 ‘ownership’ of 5 performance and 98–102, 105, 225–28 policy-makers 67 political aspects 9, 14, 54–80, 222–23, 234–35, 236 political leaders 62, 66–67, 79 poor countries and 5–6, 221–22 pragmatic rationales 38–39 private partners 125–43 problems with 235–37 provision-production distinction 15–16 questioning of 220 sector differences 67–71, 72–76 staff cuts 14–15 success of 98–102 see also following entry public service management changing approaches to 31–53 costs 34 criticism of 39 decentralization 99, 102, 103, 224–29 deconcentration 86 government intervention 32–33 growth of 34 neo-classical rationalism 32–33 organizational arrangements 81–105 policy-makers 36 reform 1, 5 resistance to 57, 62, 66, 80, 161, 223 service providers 36 theoretical basis 31–38

258 Index public service management – continued traditional public administration 39–40 see also preceding entry and new public service management re-regulation 181, 185–86 reference countries 21 regulation capacity for 186, 193–95, 201–3, 234 case for 185 constraints 193–95 deﬁnition 181 economic 181–83 enablement and 215–18, 231–32 failures 232 forms of 181–84 independent 191–93 information and 200, 202, 232 nature of 180 necessity of 222 new approaches 185, 186 performance 198–201 price 188, 189 privatization and 190–91 problems 193–95 professional councils 216 review 185 research framework 27–28 Saur International 155 Singapore 46 South Africa charges 52 contracting out 151, 152 health service 151, 152 public service reform 61 as reference country 21 revenue collection 46 water services 119 South Asia aid 3 GNP 11 public sector reform 6 South-East Asia: aid 3 Sri Lanka agencies 216

agricultural marketing 68, 76, 85, 100, 116, 121, 141, 147–48, 151–53, 229, 235 Agriculture Ministry 116 build-operate-transfer schemes 233 business development agencies 156 business services 208, 210, 212, 213, 214 charges 109, 110, 116, 119, 121, 122 citizens 62 concessions 154, 171 contracting out 147–48, 149, 158, 163, 173, 174, 175 economic liberalization 10 ﬁnancial crisis and 9–10 foreign direct investment 76 GNP per capita 11, 12 health service 65, 83, 89, 91, 92, 99, 104, 149, 163, 168–70, 196, 197, 199, 202, 203 industrial sector 84, 97, 140 National Water Supply and Sewerage Board 75 new public management 47 Paddy Marketing Board 85, 100, 116 public expenditure cuts 13 public opinion 222 public service reform 61, 62, 220, 223 regulation 190–91, 196, 197, 199, 202, 203 staff cuts 14 structural adjustment 21 as study choice 21 Urban Programme Unit 47 water services 65, 83, 93–94, 98, 101, 114, 119, 154 state 2–8 structural adjustment efﬁcacy of 220 failures in 7 study country focus 20–21 overview of 20–28 sector focus 21–27

Index 259 Sub-Saharan Africa: public sector reform 9 Tanzania 45, 46, 47 textiles industry intervention and 26 privatization 140 reform and 61 Thailand contracting out 149, 151 health service 121, 149, 151, 168–70, 196, 216 private partners 142 public service reform 61 as reference country 21 regulation 216 Transaction cost theory 36–37 Trinidad and Tobago 142, 161, 173 Uganda 14, 47 charges 51 revenue collection 46 United Kingdom economic recession 4–5 National Health Service 90 new public management 47 Next Step Agencies 45, 46, 48 public expenditure cuts 5 United States of America: economic recession 4–5 urban water supply agencies 47 autonomy 105 build-operate-transfer schemes 153–54 capacity issues 171–74 charges 51, 107, 108–9, 112–15, 116–19, 123, 124, 229 contracting out 146–47, 148–49, 157, 158–61, 162, 169, 171–74, 178 donors 60 focus on 22, 23 government intervention 24–27 IMF and 60 joint ventures 157 leasing 154–55 organization 83–84, 92–94, 105 performance 100–2

private partners and 141–42, 157 reform 60, 69–70, 73–75, 98 types 55 regulation 186–88, 215, 231–32, 234 users’ opinions 222–23 ‘Washington consensus’ 5 welfare state 31, 34 World Bank accountability and 126 capacity and 235 citations 32 economic crisis and 5 performance contracts 48 poor and 107, 123 private participation and 137 public service reform and 59 structural adjustment 10, 14, 51, 59, 220 water services and 113, 114, 115, 116–17 World Trade Organization 61 Zambia charges 52 regulation 191 revenue collection 46 Zimbabwe agencies 216 agricultural marketing 68–69, 95, 96, 98, 99, 100, 116, 122, 141, 163, 235 build-operate-transfer schemes 153, 174 business services 156–57, 211–12, 213, 214 charges 109, 110, 111–12, 116, 117, 118, 122 civil service 14 contracting out 146, 149, 158, 161, 163, 168, 171, 175 crisis 111 Dairy Board of Zimbabwe (DZL) 100 Finance Ministry 72 ﬁnancial crisis 75 GNP per capita 11, 12 Grain Marketing Board 100

260 Index Zimbabwe – continued Health Ministry 90, 109, 149 health services 65, 77, 82, 83, 88, 89, 90, 98, 99, 109, 111–12, 117, 118, 122, 149, 158, 163, 168, 196, 199, 200, 201, 202, 203 Industrial Development Corporation 206, 208–9 industrial sector 84, 97, 98 Investment Centre 72, 140, 211 Investment Promotion Centre 97 Lands and Agriculture Ministry 73, 96, 99 Local Government Ministry 113, 114, 190

performance agreements 48 private partners 141 privatization 10 pubic service reform 56, 61, 62 public expenditure cuts 13 public opinion 222 Public Service Commission 149 regulation 190, 196, 199, 200, 201, 202, 203 structural adjustment 10 as study choice 20–21 water services 65, 69–70, 74–75, 83, 94, 100–1, 102, 113, 120, 121, 122, 141, 153, 224 Zimtrade 47, 102, 156–57, 211, 212, 214