issues and challenges in local and regional

2 downloads 0 Views 16MB Size Report
reasons. First, the authors who showed interest in sharing their research findings ... frameworks need to give the incentives for municipalities so that they become effective. .... Despite pro-rural and pro-urban management arguments, however, ...... It will be in the four regional states and will be followed by 31 selected rural.
i



RLDS ISSUES AND CHALLENGES IN LOCAL AND REGIONAL DEVELOPMENT

D~centralisation, Urban Service Delivery, Rural-Urban Linkages, and Inequality in Developing Countries

Edited by Tegegne Gebre Egziabher and Meine Pieter van Dijk IHS Malcln£cititt-"

ISSUES AND CHALLENGES IN LOCAL AND REGIONAL DEVELOPMENT

Decentralisation, Urban Service Delivery and Rural-Urban Linkages, and Inequality in Developing Countries

HS F'.·,rarv

Edited by Tegegne Gebre Egziabher and Meine Pieter van Dijk

Addis Abeba May 2005

ISSUES AND CHALLENGES IN LOCAL AND

REGIONAL DEVELOPMENT Decentralisation, Urban Service Delivery and Rural-Urban Linkages, and Inequality in Developing Countries

Proceedings of the Third International Policy Research Workshop in Regional and Local Development Studies First Published in 2005 by Regional and Local Development Studies (RLDS), Addis Ababa University

Copyright

©

2005

by Regional and Local Development Studies

(RLDS), Addis Ababa University All rights reserved.

NO part of this publication may be reproduced, stored in a database or retrieval system, or any other means, without the prior written permission of the Regional and Local Development Studies, Addis Ababa University

Printed in Addis Ababa, Ethiopia by United printers

Inquiries shall be addressed to: Regional and Local Development Studies

P.O.Box 1176 Tel. :251-1-243684 E-mail:[email protected]

Contents

iii

Forward ..............................................................................................................................................iv 1. Issues and Challenges in Local and Regional Development: Decentralisation, Urban Service· Delivery and Rural-Urban Linkages and Inequality in Developing Countries, Introducing the Themes ....................................................................................................... 1

Meine Pieter van Dijk and Tegegne Gebrre Egziabher

PART I: DECENTRALISATION 2. The Reality of Decentralization Implemehtation in Ghana and Some Emerging Concerns ..................................................................................................................................9 Kwasi Kwafo Adarkwa

3. District level decentralization in Ethiopia, Expenditure Assignments and Fiscal Transfer ................................................................................................29

Worku Yehualashet

4. Decentralized Development Cooperation, Governance and Rural Livelihoods: the Case of Amhara Region, Ethiopia ................................................................................................49 Theodros Bekaffa

PART TWO: URBAN SERVICE DELIVERY 5.

New Public Management Paradigm, Lessons from Africa from the Delivery of Basic Services in Ghanaian Cities ................................................................................................79

Nicholas Awortwi 6. Improving Urban Service Delivery: Water Sector Reform in Ethiopia and its Impact for Addis Ababa .....................................................................................................................105

Meine Pieter van Dijk

PART Ill: RURAL-URBAN LINKAGES AND INEQUALITY 7. Some Indicators of Regional Socio-Economic Well-Being in Ethiopia: Trends and Patters, 1995-2001 ......................................................................................................... 121

Bekure W oldesemait

8. Rural-Urban Linkage in Ethiopia: The Need to Bridge the Divide ............................................................................................................143

Tegegne Gebre Egziabher: 9. The Microfinance Sector in Ethiopia: An Overview ..................................................... 167

Degefe Duressa and Meine Pieter van Dijk

iv

Foreword The need to formulate and implement appropriate policy frameworks and strategies recently become an accepted wisdom in the effort to tackle development problems at local and regional levels. Policies and strategies, however, depend on reliable information and empirical evidence, which could be generated through research undertakings and informed debates. The Institute of Local and Regional Studies in Addis Ababa University, besides its capacity building efforts through training, undertakes research and studies on issues pertinent to local and regional development has been a cardinal element of its mission. Research results and studies are discussed and disseminated in different modalities and avenues. The international policy research workshops, a forum, which bring together scholars from different countries to debate and reflect on diverse issues is one such avenues. In the past few years, the institute has conducted three international workshops on themes focused on: "Environmental Management in Local Development in the Horn and East Africa", "Local Economic Development and Globalization", and "Challenges of Local and Regional Development: Decentralization, Service Delivery and Regional Inequality and Rural-Urban Linkages", which were held on 2000, 2004, and 2004, respectively. The present proceedings present and discuss the results of the Third International Policy Research Workshop. A total of eight papers, which were later edited and published in these· proceedings, were presented in the workshop. These papers fall in three sub thematic categories, namely; Decentralization; Service Delivery and Regional Inequality, and Rural-Urban Linkages, which correspondingly made up the three parts of the proceedings. The introductory part provides the reader with an overview of issues and ideas discussed under the three themes. It is our view that these proceedings will be of interest for wider audience. This is for two major reasons. First, the authors who showed interest in sharing their research findings and reflecting their experiences were comprised of scientists from diverse national origin from Africa and Europe. Their professional composition is also an additional matter of interest, which gives the proceedings diverse perspective, multiplicity of approach and depth of insight. The authors who contributed to the proceedings include two faculties from the Addis Ababa University, one faculty and two PhD candidates from the Institute of Housing and urban development and Institute of Social Studies in the Netherlands, one faculty from the University of Science and Technology in Ghana, and one senior expert and official from the Ministry of Capacity Building in Ethiopia. Secondly, besides the wide range of issues and sectors taken up by the authors, involvement in regional and sub-regional issues and reflections on empirical data and best experiences from different countries make the proceedings comparative in its perspective and wider in its scope. The Institute of Local and Regional Development Studies would like to acknowledge the Institute of Social Studies of the Netherlands which was a partner institution until June 2004 in running the Capacity Building for Training and Policy Research in Regional and Local Development in Ethiopia. The RLDS extends its heart felt appreciation to the SAIL Program in the Netherlands which was the funding agency for the whole capacity program in general and the workshops ir, particular.

Tegegne Gebre Egziabher A/Director Institute of Local and Regional Development Studies, Addis Ababa University

Issues and Challenges in Local and Regional Development

1 Issues and Challenges in Local and Regional Development: Decentralisation, Urban Service Delivery and Rural-Urban Linkages and Inequality in Developing Countries, Introducing the Themes Meine Pieter van Dijk*and Tegegne Gebre Egziabher**

This book addresses issues and challenges of local and regional development in developing countries. It is known that at local and regional levels several issues emerge and several challenges remain to be tackled. Among others, these issues include ctecentralization, service delivery, rural urban linkages and regional inequality which are the three key issues for this workshop.

Decentralization Regional and Local Development has come of age particularly by the twin processes of globalization and decentralization. The former has made regions and localities to become more competitive and they have been forced to restructure their economy. A major feature of this process is· that an increasing amount of economic activities now occur in spaces within nations (Scott cited in Mc Gee and Asfaw, 2001 ). Decentralization, good governance and democratization have also become an essential component of regional development. In fact the drive to decentralization and devolution of power to local and regional entities has strengthened the rational for regional and local development strategies and policies (McGee and Asfaw, 2001 ). Many countries have resorted to decentralization measures as a way of realizing effective public service-delivery and local self-rule. Decentralization has several advantages in terms of smooth and efficient running of public affairs. It provides the means for exercising checks and balances in the different structures and tiers of government, facilitates the participation of non-governmental actors and grassroots organizations in endeavors towards entrenching the governance realm, provides options for individual citizens by prompting government responsiveness, enhances opportunity for local economic activity, and facilitates the taking shape of an active and vibrant civil society (Work, 2001 ). Decentralisation provides space to regional and urban managers, however, the critical role of well-defined inter-governmental frameworks should not be underestimated. Such frameworks need to give the incentives for municipalities so that they become effective.

Institute for Housing and Urban Development Studies (IHS) and Economic Faculty, Erasmus University Rotterdam, P.O. Box 1738 H12-29 3000 DR Rotterdam. The Netherlands Tel. Univ. 31104081870 fax 31104089153 e-mail: [email protected] or [email protected] ** Regional and Local Development Studies, Addis Ababa University

*

2

Meine Pieter van Dijk & Tegegne Gebre Egziabher

Defining tasks and responsibilities to be transferred and how to finance them at the local level is also an essential task. As much as decentralization is now an accepted strategy, it has its own pitfalls arising from design, implementation and impacts. The three papers on decentralization highlight these issues. In chapter II, Kwasi Kwafo Adarkwa discusses the arguments for decentralization using Rondinelli's theoretical framework in reference to Ghanaian experience. The Ghanian case has shown that first there are different types of decentralization that is being carried side by side. These are devolution and de-concentration. Second, the original design and implementation are at variance. For instance governmental agencies continue to retain hierarchy instead of fusing agency. The vertical relationships still exist though there are some signs towards divesting implementation of programs and projects to districts. Generally decentralization in Ghana still continues to face challenges which limit its successful implementation. Problems of conceptual clarity, co'ordination, capacity, budgetary and logistical allocation, constitutional constraints, weak economic base of lower level local governments etc are some of these challenges. The paper by Worku (chapter Ill) on fiscal district level decentralization in Ethiopia has shown that the Ethiopian government has put in place a DLDP program in order to address the capacity building needs of the woredas and the keb/es. Under DLDP, the functional and expenditure assignment of woredas have expanded. Woredas now have some general and sector specific functions. Fiscal decentralization to woredas has particularly been effected as a result of district decentralization. Fiscal decentralization in Ethiopia, however, has its own characteristics in that fiscal expenditure is devolved mostly to payroll and as a result there is a shortage of capital budget in the woredas. With a shortage of capital budget woredas will not be in a position to undertake socio-economic development. The paper by Theodros (chapter IV) gives insight into how decentralized cooperation programs working at local level differ in their development impact. The study has shown that there are inter and intra program differences in livelihood outcome. These differences are attributed to differences in program components and local level governance and resource potential. In terms of program components, those programs which are more centralized and donor driven have proved to be less inclusive and their development outcome to be relatively less satisfactory than those which are more participatory in nature. In the same manner, though all the localities (woredas) are found in the same regional governments there is variation in aspects of governance viz community participation in decision making, responsiveness, transparency and accountability. The close relation between these aspects of governance and livelihood outcome fall in line with the current thinking of development that democratization, decentralization and popular participation are vital for a better economic and political outcomes. Despite the above challenges, all the papers have stressed that the creation of responsive and accountable governance at local level is key for successful development and such effort should be promoted. Government should strengthen and deepen the decentralization process. In this regard:

Issues and Challenges in Local and Regional Development

3

1. A comprehensive programme that addresses the capacities and institutions of the local government is essential; 2. An effective transfer, mobilization and local management of financial resources for socio-economic development is indispensable; 3. A decentralized, recipient driven and locally owned decentralized cooperation program of local development is vital for positive livelihood outcome; and 4. Goood governance and its features: responsiveness, transparency, accountability and community participation in decision making are crucial for local development at grassroots level.

Improving Urban Service Delivery The recent World Development Report underlined the need to make services work for the poor. The poor generally lack access to services in sufficient quantity and quality (World Bank, 2004). As much as poverty is multidimensional and human development is central to human welfare, service provision, in adequate quantity and quality, remains a challenge that needs to be taken up in development in general and regional and local development in particular. In light of this, it is crucial to understand how those services which help the poor escape poverty could be effectively, efficiently and fairly provided. Key words for efficient service delivery are: 1.specificity (Israel, 1989); 2.private sector involvement (Roth, 1989); 3.gaining experience with small projects (the Chinese approach of experimenting reforms); 4.analysing experiences of different countries (this workshop); 5.benchmarking: measure the performance VEWIN (2001 and 2003). Although the literature available on program evaluation is very limited, water sector reform programs in developing countries are being given increasing attention. The elements of such a program include introducing:

• • • • •

institutional changes; policy improvement; financial set-up changes-; establishment of robust sector governance; and, introduction of more efficient and professional management.

A reform program in the water sector usually implies improved quality of the service delivery and better financing structure. The WSP (2002) of the World Bank describes the responsibilities related to water sector reform and the tools vital for these reforms: 1. Institutional changes; 2. Tariff reform; 3. Improvement of sector governance and regulation; 4. Professional managen:,ent of the utility.

4

Meine Pieter van Dijk & Tegegne Gebre Egziabher

The New Public Management Theory (NPM) is a stimulating theoretical framework for studying service delivery. The key terms in the New Public Management (NPM) are autonomy, accountability, customer orientation and market orientation.

Accountability Accountability means rendering explanation of what is going on. Accountability is compulsory in the administration of public utility in general and at the level of local government in particular. The utility is accountable to a variety of actors and groups, including the owner, the .regulator, financing institutions, and NGOs or special interest groups. Two kinds of accountability should be distinguished within the utility. The first relates to the relationship among the service provider and the management oversight agency. The second kind of accountability concerns the accountability of staff within the organization (in this case the service provider), and is also known. as 'managerial' accountability (Rose and Lawton1999).

Autonomy Autonomy refers to the ability to make important decisions at a lower level of government or of a company. A distinction is made between 'internal autonomy' and .. 'external autonomy' (Schwartz and Van Dijk, 2003). The former refers to autonomy within the utility and makes the organization more flexible, effective and innovative, while the latter refers to service providers and shields them from opportunistic political interference and allows them to respond to the needs of their customers.

Customer orientation Customer orientation is related to: attention paid to customers of the utility as measured by considerations to complaints, representation of customers in different bodies, and surveys to be conducted to find out what they think of the service.

Market orientation Market orientation aims at making efforts to reap the benefits of competition. One bestmethod to achieve this is private sector participation. Outcontracting, benchmarking and testing the markets are other indicators of market orientation in a utility. In terms of service delivery, the case studies have shown that reforms are being carried in different countries to improve the efficiency and effectiveness of services. The Ghnaian paper on service delivery by Nicholas Awortwi, (chapter V) examined the application and performance of the New Public Management (NPM) theory and insights in an African city. The author clearly showed the gains to be made from NPM particularly in terms of raising effi9iency and achieving other advantages. The results in Ghanaian cities showed that

Issues and Challenges in Local and Regional Development

5

though the exercise of NPM has resulted in better performance particularly in the collection of solid waste, it did not sufficiently capture all the different gains such as managerial efficiency, getting value for money, shifting responsibility, cost saving, reduction of the public sector, transparency and accountability. The main failure of the Ghanian cities in the application of the NPM has been the failure to get the fundamental rights in particular the failure of the Central and Local Governments. Therefore corollary to the application of the NPM and outsourcing of service delivery, it is important to redefine the roles and objectives of state policies, to put in place third party enforcing regulation, to introduce input and output indicators to measure performance and to build the capacity of the Local Governments who are involved in contracting and monitoring of service and involving citizens. In chapter VI, Meine Pieter van Dijk evaluates the performance of the water and sanitation sector in urban Ethiopia. The Ethiopian case illustrated that the reforms on water provision has not brought the desired results. The study has shown that the Addis Ababa Water Sewerage Authority (AAWSA) has poor performance which, among others, is due to limited degree of autonomy, low tariff structure and other technical and financial problems. Water sector reform needs to be undertaken in light of stable management, effective benchmarking, and performance related rewards, multi-ownership and clear separation of owner and operator, involvement of staff, properly planning the sequence of reforms, promotion of business culture and generally operating under the principles of NPM theory.

Regional Inequality Countries

and

Rural-Urban

Linkages

in

Developing

Regional inequality remains to be the main challenge particularly in developing countries because of its political, economic and social dimensions. That citizens should share the fruits of development in a fair and just manner and social justice be served make regional inequality to be on the top of policy makers' agenda. The division of space into urban and rural has generated heated debate among those who favour urban or rural spaces for resource allocation. Despite pro-rural and pro-urban management arguments, however, the urban and the rural should be seen as a whole and linkages should be strengthened (UNCHS, 2001 ). The strength of rural-urban linkages to a large extent determines the living conditions of people in both urban and rural areas. Problems of rural-urban linkages and inequality are not unique to Africa. China is facing major problems in spreading its successful economic development to the western part of the country (Van Dijk, 2003). Trickling down is out, but trickling plus is the trend. It means that reducing inequality is not automatic, but requires additional government efforts, which can benefit from economic growth. This requires governments to choose and develop relevant programs such as: • Socio-economic development funds; • Improved service delivery for the poor; • Introducing micro finance activities;

6

Meine Pieter van Dijk & Tegegne Gebre Egziabher

Bekure Woldesemait, in chapter VII, takes up the themes of regional inequality and discusses some indicators of regional socio-economic well-being in Ethiopia. He studied the trends and patterns between 1995 and 2001. The Ethiopian Federal states is characterized by such visible inequalities which are difficult to be ignored. The regional patterns of growth has shown that there is no clear narrowing down of growth patterns though the least developed regions have shown dramatic increases in their level of development. Such trend pushes the government to rethink its policies particularly the investment policy to induce investment to the lagging regions. In addition, the capacity constraint of the lagging regions has to be focused if these regions are to register any improvements in their levels of development and narrow the development gap in the country. In chapter VIII, Tegegne Gebre Egzaibher analyses in more depth the rural-urban linkages in Ethiopia. The rural-urban linkage in Ethiopia is found out to be weak. This is manifested in the constraints facing market linkages, limited flow of finance to rural areas, limited road network that hampers mobility and lack of access to public services. Linkages ought to be strengthened and enhanced by activities to be undertaken both in the urban and rural sides. Improved market facilities and information will help strengthen the marketing linkages. Diversified employment opportunities in smaller towns and encouragements for rural-urban labour movement will help improve the linkages. Similarly, urban production should be linked with rural resources and production should be integrated with distribution and marketing. Physical infrastructure will help integrate the rural with the urban and facilitate mobility. In chapter IX, Degefe Duressa and Meine Pieter van Dijk describe the Ethiopian microcredit experience so far and show that there exists a remarkable political support for micro- finance institutions (MFls), although most of the MFls studied have not yet proved to be financially sustainable. The availability of rural finance will definitely help to improve farming productivity, reduce regional inequality and foster linkages.

Conclusions The Millennium Development Goals (MDGs) are very clearly putting a challenge to us to halve the global incidence of poverty and to improve the lives of people living in slums. The number of people not enjoying safe water should be halved just like the number of people not enjoying a sewerage system. The new decentralized structures could be used so that they help governments achieve the Millennium Development Goals and to diminish the rural-urban gaps. The cases presented in this book show that the first steps in the right direction have been taken, but the challenges remain and they require that government and non-government actors need to adopt measures that help overcome the challenges.

Issues and Challenges in Local and Regional Development

7

References Barzelay, M. (1992). Breaking through Bureaucracy: A New Vision for Managing in Government. University of California Press: Berkeley. Burki, S. and Perry, G. (1998). Beyond the Washington Consensus: Institutions Matter, World Bank Latin American and Caribbean Studies. Washington D.C.: World Bank. World Meteorological Organization.(2002). Development Issues for the 21 st Century, International Conference on Water and the Environment (ICWE), 26-31 January, 2002, Dublin, Ireland, Geneva: World Meteorological Organization. Dijk, M. P. van. (2003). Is Nanjing's concentration of IT companies an innovative cluster? In D. Fornahl and T. Brenner (ed.). Co-operation, Networks and Institutions in Regional Innovation Systems. Cheltenham: Edward Elgar, pp. 173-194. _ _ _ _ _ . (2000). Summer in the city, Decentralisation provides new opportunties for urban management in emerging economies. Rotterdam: IHS (english version), (1-36). _ _ _ _ _ _ . (1999). Globalization and economic restructuring: Competitiveness at the regional, country, city and enterprise level. In: M.S.S. EI-Namaki (ed.), Strategic Issues at the Dawn of a New Millennium. Leiderdorp: Lansa. _ _ _ _ _.(1997). The economic activities of the poor in Accra, Ghana. In: D. Bryceson and V. Jamal (eds.), Farewell to Farms, De-agrarianization and Employment in Africa. Hampshire: Ashgate Publishing, (pp. 101-117) Dijk, M. P. van and K. Schwartz. (2003). A series of case studies of well performing drinking water companies, a research project for the World Bank, Paper for the PoWER project at UNESCO IHE in Delft, October 15. Israel, A. (1989). Institutional Development. Washington: IBRD. Lindfield, M. (1998). Preparing Markets for Private Financing of Urban Infrastructure. Rotterdam: Erasmus University, PhQ [ thesis]. Mc Gee Terry and Asfaw Kumsa (2001) 'Introduction'. In: Asfaw Kumsa and Terry Mc Gee (eds.) New Regional Development Paradigm, Volume 1. Westport: Greenwood Press. Osborne, D., and Gaebler, T. (1993). Reinventing Government: How the Entrepreneurial Spirit is Transforming the Public Sector. Addison-Wesley:Reading. Rondinelli, D.A., and K., Ruddle. (1978). Urbanization and Rural Development: A Spatial Policy for Equitable Growth. New York: Praeger. Roth, G. (1989). The Private Provision of Public Services in Developing Countries. World Bank: Washington. Thynne, I. (1998). Government companies as instruments of state action, Public Administration and Development, 18, 217-228.

UN CHS (2001) The sate of the world cities. Nairobi: UNCHS Publication Unit. VEWIN .(2001). Reflections on Performance 2000. Benchmarking in the Dutch Drinking water industry. Rijswijk: VEWIN. VEWIN. (2003): http://www.vewin.nl. Work, R., (2001), 'Decentralization, governance and sustainable regional development' in Sthor et al, (eds.), Decentralization, Governance, and the New Planningfor Local Level Development, Greenwood press, Westport World Bank (2003). World Bank Development Report 2004: Making Services Work for Poor People. New York: Oxford UP. · WSP. (2002). New Designs for Water and Sanitation Transactions. Washington, Water and Sanitation Program.

PARTI

DECENTRALISATION

The Reality of Decentralization

9

2 The Reality of Decentralization Implementation in Ghana and Some Emerging Concerns Kwasi Kwafo Adarkwa *

1.

Some Theoretical Underpinnings 11

Decentralization is viewed as a "word which is surrounded by terminological confusion (Conyers, 1986). In other words, it defies any universally acceptable definition. Consequently, it signifies different meanings in various contexts. As a result, the literature on the usage of the term is full of qualifying adjectives such as "political decentralization", "administrative decentralization", and "financial decentralization".

In spite of the different forms of decentralization, it can succinctly be defined as the transfer of government power, authority and resources to a lower organisation. More specifically, Conyers (1984) suggests that decentralization could be viewed in two contexts; "fundamental decentralization" and "territorial decentralization". To Cot1yers, "fundamental decentralization" refers to the transfer of authority and responsibility from government to the private sector or non-governmental organisations (NGOs), Community Based Organisations (CBOs) and other private enterprises, while "territorial decentralization" refers to the transfer of authority and responsibility to a territorially defined local authority or an identified spatial entity. These two must always go together to ensure the efficient utilisation of local resources for national development and this has characterised many countries' efforts to decentralise. In another instance, Tamakloe (1989) defines decentralization or "decentralised system of governance as empowering various segments of society especially the disadvantaged, including rural poor in developing societies to participate through dialogue, in national and local decisions that affect the quality of their lives. He therefore argues that decentralization is only an instrument for achieving an objective and not an end in itself. Thus, the main objective of decentralization is to promote efficiency of government machinery through dialogue, as well as spatial equity in the distribution of "political and "administrative powers for development decision-making to reduce poverty from the grassroots. One inference from this definition is that decentralization is the system of government which lays down the structures for political, functional and administrative authority to be transferred from a central agency or unit to lower institutions representing identified socio-cultural and political as well as territorial interests. 11

11

11

The foregoing definitions were earlier on reinforced by Manwood (1983) who also defined decentralization to mean "the sharing of part of governmental power by a central ruling

* Kwame Nkrumah University Of Science And TechnologyKumasi, Ghana

10

Kwasi Kwafo Adarkwa

group with other groups, each having authority within a specific area of the state". At the level of "norms", it indicates the existence of formal political structures each covering a defined area, representing local interests as well as the interests of the central rulers with the local share of allocating power protected by formal and normative rules which are accepted by the centre. This definition also highlights the fact that decentralization has to do with laid down legal and institutional framework within which there is power sharing in the process of decision making as well as the election or appointment of local representatives who will promote local interest in the face of central political and administrative interest. At the Commonwealth Secretariat's Workshop on 'Training for Decentralised Administration" held at the Administrative Staff College of Nigeria (ASCON), Badagry, an attempt was made to offer a universally acceptable operational definition of decentralization. At this workshop, the participants agreed that decentralization refers to "any deliberate change in the organisation of government involving the transfer of powers, resources and functions from the centre to units of government and administration at sub-national level". These units of government are mainly, central government field agencies (outside the seat of government), representative of local government institutions ar.id other local bodies involving NGOs and CBOs. Also the transfer of authority or power away from the national capital could be of two main forms; that is either: a)

by "administrative decentralization", delegation or decentralization to field administrative units of the central government; or

b)

by "political decentralization", "democratic decentralization" or "devolution to local authorities" or other local bodies (which are relatively autonomous).

The foregoing definitions indicate that decentralization or decentralised form of government under any circumstance must deal with power transfer (or sharing between) from a central government unit to an institution lower in function and representing an identified spatial unit but which is also based on an accepted legal and institutional structure. In essence, all variants of decentralization include elements of delegation and devolution of power to principal agencies, intermediation and privatisation, together with other centralized practices (Kokor and Kroes, 2000). This has been confirmed by Conyers and Hills (1984) when they argued that decentralization and centralisation should not be seen as separate but rather as a continuum, which always come in various forms of combination of the two systems of government. In terms of institutional arrangement for the transfer of decentralised power, a system of decentralization has been proposed by Kokor and Kroes (2000) along the following lines: a)

a representative system of elected and appointed political bodies and their extensive district committees;

b)

field officers, administrators or managerial staff at the local level;

The Reality of Decentralization

c)

11

private sector and or mixed corporations; and

d) the so-called third sector organisations, such as voluntary not-for-profit associations, community based organisations (CBOs) and activist groups. All these structures have various degrees of autonomy, influence and oversight which are backed by the legal provision that established the decentralised mode of governance. The degree of autonomy which measures the level of decentralization action can be assessed under four main parameters, namely: (i) Political decision making; (ii) Administrative, including staff and personnel management; (iii) Development planning, programming and project execution; and (iv) Financial administration and management. In this paper, my intention is to use the above framework to assess the need for decentralization in Ghana and how far the policy and programme have been implemented as well as difficulties encountered to date. The paper concludes with some indication of what needs to be done to accelerate the implementation of the decentralization policy.

1.2

The Rationale for Decentralization

The major rationale for decentralization is the transfer of significant amounts of power or authority, functlons and capacity (financial and human resource base) from the central government to the local institutions to ensure efficient and effective local decision making based on local knowledge without changing the socio-cultural, political and natural environment of these local areas. Decentralization also aims at promoting high-level community participation, democratic governance and maximum accountability of decision-making. Thus, decentralization or decentralised system of governance seeks to instil in the local people commitment and understanding of their development process so as to make them contribute positively to it. Decentralization is also advocated as a way of improving the management of development by enhancing the governance of development, increasing flexibility and responsiveness (Conyers, 1984). Thus projects and programmes that are implemented at the local level can be adjusted more easily when unanticipated changes are required instead of such matters of a local nature being determined by central government authorities. Development is, therefore, better governed under a decentralised system of governance than other forms of governance. Again, it is an objective of the decentralization process to ensure maximum co-ordination between the various agencies involved in planning and implementation of development programmes at the local level. As a result, under any form of decentralization, local institutions including public, private and community based organisations are encouragad to collaborate in the development process and hence programmes are designed with inputs from all these institutions.

12

Kwasi Kwafo Adarkwa

In addition, decentralization also aims at ensuring spatial equity in the sharing of planning, decision-making and management functions from the central government unit. Once lower units of administration are established, it becomes imperative for central government to channel some level of resources to such units. Such resources are supposed to be used to address the development problems of the local areas as a way of bridging any intra and inter local areas' spatial inequalities in development. It is obvious from the above discussions that another objective of decentralization is to make maximum use of both local, natural and human resources in a sustainable way to ensure rapid development of local areas. There is therefore always a link between decentralization, local resource mobilisation and utilisation, and national development. This is because it is the local people who have an in-depth knowledge about the resource base and will thus be able to utilise them to the maximum benefit of the local area in particular and the nation as a whole. The rationale and objectives of decentralization are also reflected in the features of a decentralised system of government.

1.3

Features of a Decentralised System of Government

From the foregoing definitions of a decentralised system of government, it can be said to exhibit some features that make it functional to achieve its desired objective of efficient and effective local development with maximum participation of the beneficiaries. These characteristics are presented seriatim. •

The existence of a centre (or central government unit) which should transfer power to local bodies/individuals, etc. (The Ghanaian examples are the Ministries, Departments and Agencies as well as the private sector includiflg NGOs and CBOs).



The existence of a legal framework in the form of constitutional provisions that provide the broad institutional framework for local government to take off as well as establishing the various institutions that form the decentralised structure, their functions and inter-relationships.



Local government institutions located at the local levels are equipped (both in human and logistically) to accept and exercise the authority transferred to them. These local institutions are situated in geographically identified locations established with specific political boundaries.



The process and arrangement for power transfer and sharing between the central and local government units as well as non-governmental associations and agencies at the local level.



Local planning, decision-making and financial autonomy have been areas with some amount of controversy as to the extent of autonomy to be given to these local government units. In any case some amount of autonomy must be given to these institutions with some restrictions, where necessary.

The Reality of Decentralization



13

Local political representation is very important because there must always be representatives of the people at the local level forming the base of the local government institutions. This is usually done through either local electoral processes or appointments by central government unit. Here, care must be taken to take into consideration, the existing political (traditional) decision making hierarchy.



Links between the central government institutions and the local government ones (as well as NGOs and CBOs). These are in the form of guidelines as well as providing local data base for national planning and development.



The use of available local resources (natural and human) as much as possible in the development planning and implementation process. There is therefore local resource mobilisation and utilisation under any form of decentralised system of government.



Another important feature of decentralization is that of community participation and dialogue among all stakeholders at the local level. Decentralization provides a broad platform on which all stakeholders make contributions to the development planning process 9f the local area.

These features of decentralization result from the basic principles underlying the concept. These principles have been outlined next.

2.

Some Basic Tenets of Decentralization

Apart from the major characteristics or features of decentralization (or a decentralised system of government), there are some basic principles, which underline any best practice in decentralization. Consequently, these principles should be considered in any institutional or political arrangement for any decentralised system of government. These principles are outlined in this section of the paper. •

Target Group Orientation~ This is where specifically identifiable spatial entities are defined as local government units. These units are given institutional and resource capacity to inspire, enables, guide, facilitate, promote, assist and support the local communities and groups for their own development (Kokor, 2001 ). Thus at the centre of local level development under any decentralised system of government is the people and not the local government institutions (for instance, District Assemblies in Ghana). Again this principle also points to the fact that in the formulation of development goals and targets, the specific needs and demands of the target group (local people) are of paramount importance.



Participation: There should also be a consideration of the total participation of the people in the formulation of local development goals and also in the implementation process. All the necessary structures must be put in place to

14

Kwasi Kwafo Adarkwa encourage grassroots participation in decisions affecting the masses in the localities and thereby creating the awareness in the people that they can, and should, be involved in improving the quality of life in their localities. This is significant because it adheres to the basic planning principle that the people for whom development is meant should not only be the beneficiaries but also the agents of the change so desired. •

Utmost mobilisation of self-help activities of private initiatives and cooperatives and the involvement of the local private sector in development: This principle of decentralization proposes that all agencies (both governmental . and non-governmental), at the local level should be made to get involved in all development programmes. The roles, status and functions of all these agencies and institutions must be spelt out in the locally planned development process, which should be largely based on local resources.



Consideration of local (natural and human) resources and of the endogenous potential: Every decentralization effort should identify and ensure maximum utilisation of local resources without destroying the natural and cultural environment. This therefore calls for a comprehensive data base that will provide the profile of the local units as well as estimating needs and setting goals and targets relative to the local situation.



Rationalisation and more transparency of the decision making process: An important principle that should underline any form of decentralised government should be transparent and rational (logically laid down) decision making structures and processes. These are made possible through the establishment of local agencies/institutions that are supposed to be directly responsible to both the central government units as well as the people in the local communities. NGOs, CBOs and other civil society groups must educate and help the local people to demand more transparency from the local government officials.



Comprehensive information of all actors involved in local development activities: The implementation of any decentralization policy should be underlined by a principle of total information of all actors. That is there should be a broad data base which is accessible to all the actors of development at the local level (district agencies, sectoral departments, Executive Committees, NGOs, CBOs, Civil Society groups, community development committees and local people). Again, the database shoU'ld present structures for all stakeholders to know their roles and functions in the local development process. Such I database can be in the form of planning document, medium term district plans,; etc. I



Local development projects are guided by cost recovery principles: Under this principle, local people are made to understand that facilities provided are theirs and as such their sustenance rests on them. Communities are thus/ motivated to initiate self-help projects which are partly funded and sustained by/ ~em. .

The Reality of Decentralization

15



Creating strong local economic circuits and strengthening local identity: Under decentralization, efforts must be directed towards ensuring the rapid growth of the local economy by particularly ensuring diversification and efficient interaction of the various sectors of the local economy. Again, the local identity in terms of space, tradition and culture as well as economic activities is strongly maintained to ensure socio-economic, political and cultural cohesion.



The economically and socially weaker target groups are given p~iority attention: Decentralization has the aim of empowering the vulnerable, oppressed and poor in society. Thus structures are put in place under every decentrali~ed system of government to improve the lots of the vulnerable in society ·especially those in the local communities. Again, efforts are also made in the provision of basic infrastructural facilities (such as housing, water and sanitation, roads and transport services, education and health facilities), that will enhance the living condition of the people in the local communities.

All these principles were considered before and during the design and implementation of the Ghanaian decentralised system. This is because they had implications for the local development planning process. The major areas of the design which the principles affected were: i. The design of the overall institutional framework for development planning in the country; ii.

The need, scope and content of a legal framework for development planning in the country;

iii.

The role and status of staffing and training for the district development planning and management unit and the sectoral departments; and

iv.

Procedures for inter-dependent (district-region-nation) decision-making.

development

These issues formed the basis of the original concepts of the Ghanaian system discussed below.

3.

The Original Concept

The current system of local government administration was established by PNDC Law 207 enacted in 1988. The law established 11 O District Assemblies as the political and administrative authorities in each district empowered to exercise deliberative, legislative and executive function within their areas of authority. The law was thus designed to "install monolithic institutions at the District, Town and Unit Committee levels to which some functions/responsibilities of government should be transferred and through which access to state power can be created for the entire population of the country to exercise the democratic rights in administrative and development decision making and take up their obligations to contribute, each according to his best ability, to the development of the

16

Kwasi Kwafo Adarkwa

country" (Ahwoi, 1991 ). In the pursuance of this policy, 22 central government departments were decentralised and put under the control of the District Assemblies. The membership of each District Assembly was to consist of a District Secretary (now District Chief Executive) nominated by the Central Government, and voted on by the Assembly, a number of elected members and other persons not exceeding one-third of the total membership of the Assembly appointed by traditional local authorities and organised productive economic groupings in the Districts (PNDCL 207, 1988). Each District Assembly has a Presiding Member (PM) elected from among the membership of the Assembly by at least two-thirds of all the members of the Assembly. The heads of the decentralised departments and other departments established by the District Assembly were ex-officio members of the Assembly without voting rights. Within each Assembly was to be an Executive Committee expected to perform the executive and co-ordinating functions of the District Assembly. It was to be chaired by the District Secretary (currently' District Chief Executive). A diagrammatic representation of the organisational structure of the District Assembly as established by the PNDC Law 207, 1988 is presented in Figure 1. At the regional level, the law provided for the estabHshment of Regional Co-ordinating Councils (there were ten of these) consisting of the Regional Secretary (currently the Regional Minister), appointed by and representing the central government in the region, the Deputy Regional Secretary who had no right to vote on any matter for decision by the Council, all District Secretaries in the region and all Presiding Members of the District Assemblies in the region. The Regional Secretary was the chairman of this Council and another Central Government appointee; the Regional Administrative Officer was the Secretary to the Council. His office provided the secretariat for the Council. With the shift of the country to constitutional rule in 1992, a· new constitution was promulgated in which the Decentralization Policy and local government reform programme were constitutionalized. Essentially the PNDC Law 207, 1988 metamorphosed into the Local Government Act, Act 462 (1993) with some improvements and additions. This Act established the Districts, District Assemblies and their sub-district institutions as well as spelling out their roles and functions within the decentralization process. Other legal provisions backing the current Decentralised System of Government in Ghana include: • • The Civil Service Law 1993; • National Development Planning Commission Act 4 79, 1994; • National Development Planning (System) Act, 1994 Act 480; • Local Government (District Tender Boards Establishment) Regulations, 995 L.I. 1606; • Local Government (Urban, Zonal and Town Councils and Unit Committee Establishment) Instrument, 1994, L.I. 1589; • District Assemblies' Common Fund Act, 1993, Act 455; and • All the various Legislative lnstn:1ments establishing the District Assemblies. With these legal backgrounds (framework), the new decentralization (local government) system policy is aimed at achieving the following;

The Reality of Decentralization

• •

• •



17

Devolving central administration authority to the district level; Fusing governmental agencies in any given region, district or locality into one administrative unit, through the process of institutional integration, manpower absorption, composite budgeting and provision of funds for decentralised services; Divesting the centre of implementation responsibilities and transferring those responsibilities to the districts; Assigning functions and responsibilities to the various levels of government. These, under the current concept are to be as follows; a. Central Government Ministries/Departments to undertake policy planning; monitoring and evaluation of policies and programmes; b.

Regions, through the Regional Co-ordinating Councils and their respective Regional Planning Co-ordinating Units, to play the important role of co-ordination, not in a regulatory manner but to ensure consistency, compatibility and coherence of district level development, facilitate joint ventures among districts and monitor the activities of District Assemblies within regions; and

c.

District Assemblies to be primarily responsible for the implementation of development policies and programmes coordinated by the National Development Planning Commission (NDPC) which is the highest development planning institution in the country providing advisory roles to the government and the sectoral Ministries (see Figure 2);

Promoting popular grassroots participation in the administration of the various areas concerned from the start point of planning, implementation, monitoring and delivery of those services which go to improve the living conditions of the people and the orderly, fair and balanced development of the whole country (Government of Ghana, 1996).

Conceptually, the new Local Government of Ghana can be represented diagrammatically as shown in Figure 2. The composition and functions of the structures/institutions shown in Figure 2 are not very different from those established by the PNDC Law 207, 1998. The characteristics of the component have however been summarised in Figure 3. From Figure 3, it can be seen that the three Metropolitan Assemblies were not mentioned in the description. There are however four Metropolitan Assemblies; Accra Metropolitan Assembly, Kumasi Metropolitan Assembly, Sekondi/Takoradi Metr:opolitan Assembly and Tamale Metropolitan Assembly. These are cities with over 250,000 inhabitants. They

18

Kwasi Kwafo Adarkwa

have sub-metropolitan offices as well as Town and Zonal Councils and Unit Committees at community levels. The structure of the new local government system is conceptually operationalised through the new planning system also presented diagrammatically in Figure 4. From the diagram it can be seen that the inputs for local level planning and development must evolve from the local communities with consultations and advisory inputs from the DPCU, RPCU and the NDPC. The RCC however has a co-ordinating function to ensure that plans at the local levels conform to broad national goals and aspirations to ensure that local development does not take place at the expense of national development. Thus far, the foregoing presents the broad Conceptual Framework of Ghana's Decentralization and Local Government Policy. The implementation of the policy took effect from 1992 under a constitutional government. After about twelve years of its implementation, the policy implementation process has presented some realities bordering on successes, challenges and prospects which constitute the thrust of subsequent sections of this paper.

4.

The Reality (Current Situation in Ghana)

In order to present a more comprehensive picture of the reality as far as the implementation of the decentralization policy is concerned, two broad areas have been chosen. These are the understanding of the concept of decentralization and the empirical reality, which exist in the implementation of the policy with and among the Ministries, Departments and Agencies. There appears to be two variants of the concept of decentralization that are currently being implemented in Ghana. One is the devolution of major political and administrative responsibilities from central government to District Assemblies, which comprise of partially elected members and bodies with the mandate for local, and or community level governance and general development. The other concept, which operates alongside the first, is the process of administrative and technical de-concentration of the planning and delivery or provision of specific services such as water and sanitation, health, education and agriculture from Ministries, Departments and Agencies. Devolution in Ghana has been to restructure technical set-up, political and administrative institutions and mandates to deliver services particularly at the local level. On the other hand, de-concentration has involved the re-arrangement of the location of key personnel and where their functions are performed. Thus far, decentralization policies of most African countries are generally confined to "decentralization" as opposed to "devolution" - Ghana, Cameroon, Kenya and Zambia all practice decentralization. It is important to note that even though Ghana's Decentralization Policy is underpinned by Article 35 (60) (d) of the 1992 Constitution, which provides for making "democracy a reality by decentralising the administrative and financial machinery of government to the regions and districts ... ", the reality has been a growing concentration of power and

The Reality of Decentralization

19

resources in key sector Ministries, Departments and Agencies (MDAs) that plan, implement, monitor and evaluate essential community services. Consequently, in the implementation of the Decentralization Policy, there has been some variation between the original design and what is being implemented. Some of these variations are identified seriatim. •

The policy design on Political Decentralization is represented by partial election of Assembly Members into the District Assemblies in 11 O Local Government areas of jurisdiction and devolution of central administrative authority at District Assembly's level. The reality is that, Political Decentralization is still on-going and not yet completed. The associated challenges which have been identified by government are (i) the election of District Chief Executives and Assembly members to make the system both representative and democratic and (ii) only partial devolution of administrative authority to the District Assemblies.



The policy also aims at the fusion of governmental agencies in all the regions, districts and locality into one administrative unit. The truth is that governmental agencies continue to retain their "hierarchy" from national to region and to district offices. Those institutions which are considered to be decentralised are the ones in the "District Assembly Secretariat". In fact, real fusion has not yet been achieved.



The policy intended to divest from the centre of implementation responsibilities and transfer these responsibilities in 22 functional areas of Central Government to the District Assemblies. What has been achieved is that, there has been some divestiture of implementation authority for programmes and projects to the district level. These are however still occurring vertically from national to regional and district offices of the same department. The policy objective which intends to shrink the 22 departments into 11 has not yet occurred in reality.



Another intent of the decentralization policy is that through the Regional Coordinating Councils (RCCs) and the Regional Planning and Co-ordinating Units (RPCUs), the regions will formulate and harmonize district level development interventions, monitor implementation activities of the District Assemblies; develop implementation monitoring indicators and reporting to the centre. The realisation of this has been faced with slow process of incorporation of the RCCs and the RPCUs into the decentralization system. This is partly due to lack of capacity for these institutions to play their roles as provided under the law. It is also because many of the functions to be delegated to them are still being performed by regional offices of the MDAs. It is interesting to note that the MDAs continue to obtain the needed funding from central government, through the Medium Term Expenditure Framework (MTEF) budget arrangements to perpetuate their role as regional policy makers and co-ordinators (but on sectoral level). This notwithstanding, some progress has been made in the Volta and Upper West Regions (where DANIDA support is providing needed skills and capacity for RPCUs) and in the Upper East and Northern Regions (where CIDA

20

Kwasi Kwafo Adarkwa project is undertaking similar capacity building efforts as those in the Volta and Upper West Regions). •

District Assemblies are to be primarily responsible for the general development of local level policy based on, and informed by, national policies/programmes and local circumstances, preparation of district development plans based on or informed by national and regional development policies, plans and programmes that are co-ordinated by the National Development Planning Commission (NDPC).ln this area, substantial progress has been made. This is aided by the availability of the District Assemblies Common Fund (DACF) to support district develoi:,ment policy, planning and programme implementation. This is however occurring in tandem with equally strong "district-level" policy, planning and programme implementation by competing MDAs (for example, Health, Education, Agriculture) which continue to receive funding and priority-setting directives from their national and regional offices. Occasionally, this could be at variance with the District's priorities.



A major achievement in the implementation of the decentralization policy is the development and an on-going implementation of a National Decentralization Action Plan (NDAP). Among the several objectives of the NDAP is to reinvigorate national policy and institutions for decentralization, guide the strategic targeting of resources to poverty reduction efforts at the local level and promote greater participation of key private sector, civil society and traditional authorities in the process of governance and development at the district level.



A Presidential Advisory Committee on Decentralization (PACO) has also been set up with the mandate to advise the President on how to get decentralization implementation on smooth active course and how to remove obstacles and constraints to enhance the achievement of the objectives of the decentralization policy.

Clearly, these realities reveal some challenges which constrain the implementation of the decentralization policy. They include the following: •

Conceptual differences in the interpretation of the decentralization policy by various sectors. There has been no clarity at the various levels of government and within different sectors, m1nistries and departments regarding the intended shifts in power functions and resources and how the processes should be ·managed. This is manifested in the creation of sectoral services in education, health and forestry. Decentralization implementation in Ghana is characterized by two parallel systems of devolution and de-concentration both supported by the Constitution and appropriate Acts of Parliament. This obviously has a lot of significant implications in trying to increase the pace of decentralization.

The Reality of Decentralization

5.

21



Effective collaboration is required because of the wide-ranging groups of stakeholders involved in the process. This has however been weak owing in part to the absence of a central body with sufficient clout to oversee and guide the implementation process. lntersectoral collaboration therefore remains weak.



Inadequate managerial and technical capacity to implement the policy: The implementation of the policy has suffered generally from the lack of adequate qualified personnel at all levels - national, regional and district.



There is also the element of weak co-ordination of donor-support, NGOs/CBOs programmes. This has resulted in some cases of competition for scarce financial and human resources, conflicting programmes and duplication of effort. Regular consultation and systematic co-ordination of activities should be institutionalised so as to minimize duplication of efforts and to attain complementarity and synergy in the activities of the various stakeholders.



Inadequate budgetary allocations.



Inadequate logistic support including office and residential accommodation, vehicles and equipment.



Weak capacity to effectively monitor the implementation process and evaluate the performance of local government institutions.



Weak economic base of some District Assemblies constraining their ability to mobilise adequate resources for the provision of services.



Constitutional constraints, including the election of DCEs and some Assembly members.



Personnel and attitudinal problems which are manifested generally in the reluctance by top-level bureaucrats to relinquish power and responsibility to regional and district level operatives.

The Way Forward

The national vision of decentralization as contained in the Ghana Poverty Reduction Strategy (GPRS) is to: "Promote responsive and accountable governance at the local levels that allows effective participation, equity in resource allocation and effective delivery of services, especially for the poor". The GPRS has guided the formulation of government policy, medium-term sector plans, district plans and budget processes. The weak capacity of local government institutions will however pose a great challenge to the effective implementation of the GPRS. Government intends to strengthen and deepen the decentralization process and to make the sub-district structures serve as conduits for service delivery and instruments for local democracy. The commitment of government to good governance makes implementation

22

Kwasi Kwafo Adarkwa

of the decentralization policy a priority as decentralised government machinery promotes good governance. There is the need, therefore, to develop a comprehensive programme to strengthen the capacities and institutions of the local government system to enhance planning, implementation and monitoring of programmes at all levels - national, regional, district and sub-districts to support the implementation of GPRS and other development programmes. The Ministry of Local Government and Rural Development is the lead agency responsible for co-ordinating the implementation of the decentralization policy and the local government reform programme. Guided by the above stated vision, the overall policy objective of the Ministry is to promote the establishment and development of a vibrant and well-resourced local government system to ensure good governance and balanced development. This is to be done by: •

Formulating appropriate policies and programmes to establish and operationalise a decentralised government machinery;



Strengthening human resources for all levels of the decentralised government machinery;



Improving the capacity of communities and local government institutions to mobilize and manage resources for accelerated rural and urban development; and



Monitoring and evaluation of the effectiveness of local government institutions for improved management performance.

The priorities of the Ministry with respect to decentralization in the medium term, i.e., (2003 - 2005) include: • •

Strengthening political leadership and institutional decentralization implementation; Strengthening of the Inspectorate Division in MLGRD;

arrangements

for



Legal clarification regarding the division of roles and responsibilities between the different layers of government and within the Local Government Service (LGS);



Establishment and operationalisation of LGS Council;



Review of local government investment funding practices and design of a harmonised District Development Fund (DDF) Facility;



Harmonisation of local government capacity building; and



Strengthening of local governance structures and accountability mechanisms.

The Reality of Decentralization

6.

23

Conclusion

It is clear that the decentralization policy design in Ghana had good intentions. The dynamics and complexities of its implementation require comprehensive approach to meeting the challenges which are evident in the contemporary Ghanaian situation. To move the process forward, it is essential that efforts are focused on how to bring about the following: •

Essential shifts in power, functions and resources from the Central Government (MDAs) to the Local Authorities (RCCs and DAs);



Capacity building and integration of decentralised departments within the RCCs and District Assemblies;



Effective transfer, mobilisation and local management of financial resources for local development;



The Decentralization Action Plan be put in public domain for public sensitisation, acceptance and support; All future legislation emanating from Ministries, Departments and Agencies must be consistent with national policy of decentralization. Parliamentarians should be sensitised about this;





All Donor Support projects should have implementation management arrangements that are consistent with national policy of decentralization;



A Decentralization Secretariat is established and located under the Presidency with the responsibility to co-ordinate the implementation of the Decentralization Policy; and,



The need for effective monitoring and continuous evaluation of official policy on decentralization.

Democracy and good governance has now become a permanent feature of the Ghanaian political landscape. There is therefore the need to enhance the implementation of government's decentralization policy because it is complementary to democracy and good governance and are mutually reinforcing. It is also clear that for the policy to be successfully implemented, it needs more than simply passing legislation and legal instruments. Such legal instruments are necessary but do not guarantee the successful implementation of decentralization.

24

Kwasi Kwafo Adarkwa

References Ahwoi, Kwamena. (1991).Decentra/ised Administration; Progress, Problems and Prospects, From Centre to the Grassroots. Ministry of Local Government: Accra. Conyers, Diana. (1984). Local Government in Developing Countries. Joe Heitey: London. Conyers, Diana and P. Hills. (1984). An Introduction to Development Planning in the Third World. John Wiley and Sons: New York. Conyers, Diana. (1986). The Management and Implementation of Decentralised Development Administration:.. Commonwealth Secretariat: London. Government of Ghana. (1992). Constitution of the Republic of Ghana. Government of Ghana: Accra. Government of Ghana. (1996).Ghana - The New Local Government System:.. Ministry of Local Government and Rural Development: Accra. Kokor, J.Y. and G. Kroes. (2000). Central grants for local development in a decentralised system, Ghana, Research Series, 23, SPRING Centre, Dortmund. Mawhood, P. (1983). De centralization: The concept and practice. In Mawhood, P. (ed.), Local Government in the Third World: The Experiences of Tropical Africa. John Wiley and Sons: New York. MLGRD.(2003). National Decentralization Action Plan: Towards a Sector Wide Approach for Decentralization Implementation in Ghana (2003-2005). Ministry of Local Government and Rural Development: Accra. Tamakloe, E. K. A. (1986). The Prospects and Problems of Promoting Regional Development Planning as a Dialogue, in Ghana. TRIALOG: Zeitsehrift tor das Planen und Bauen in der Dritten Welt. 2. Quartal.

25

· The Reality of Decentralization

Fig 1 :District Administrative Machinery (Under PNDC Law 207)

I

- - - - - - - - - - - Command Link

'

I

- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - Consultative Link

26'

Fig 2: The New Local Gov·ernment System (Under Local Government Act, 1993, Act 462)

Source: Ministry of Local Government and Rural Development, 1996

27!

Fig 3: Basic Characteristics of the New Local Government Structure

Source:Ministry of Local Government and Rural Development, 1996

28

Fig 4: The New Planning System RPCU-Regional Planning Co-ordinator Unit DPCU-District Planning Coordinator Unit

Source: Min. of local Government and Rural Development, 1996

District Level Decentralization

29

3 District Level Decentralization in Ethiopia: Expenditure Assignments and Fiscal Transfer Wo~lrn Yehualeshet* 1. Introduction Recently decentralizing local governments has become a vital concern in a number of developing countries. The task, indeed, has also become a complex process as it involves resource mobilization and allocation and service delivery. It has also become an issue of poverty reduction, good governance and capacity building (Litvack et al, 1998). These complexities made designing decentralization and measuring its impact a difficult task if not impossible. Th~ extent of influence between down-streaming technical support and up-streaming policy advice in local level decentralization are not yet clear. This made·the interventions to be dispersed and less focused. This is why stakeholders have been worried much about decentralization. Like any other country, empowerment, governance, capacity building, poverty reduction and sustainable development are interwoven aspects in the development of decentralization in Ethiopia. It is therefore imperative to understand the challenges in the design and execution of the program to contribute to better outcomes of the decentralization process. The present paper, therefore, attempts to provide information and describe the status and challenges of District Level Decentralization (OLD). Section 1, 2 and 3 describe the feature and performance of OLD. Section 4, 5, and 6, respectively, deal with expenditure assignments, fiscal transfer issues, and problems/challenges and issues of decentralization. In addition to the review of available literature and secondary .data, primary data were collected through interviews and observation from three woredas.

2. Features, Objectives and Components of District Level Decentralization Program (DLDP) -PSCAP DLDP is one of the 14 programs that have been undertaken by the Ministry of Capacity Building (MCB). It is one of the six sub-programs incorporated in the Larger Public Sector Capacity Building Program (PSCAP) implemented by MCB-DLDP in the nine regional states. At present, there are eight major intervention areas or components that address the capacity building needs at woreda and kebele (community unit) levels. These include Capacity Building (CB) on: • institutional/organization arrangement; • manning/staffing and training;

* Director, District Level Decentralization, Ministry of Capacity of Building

30

Worku Yehualeshet

• fiscal transfer and revenue enhancement; • local planning and control; • grassroots participation system; • minimum standard service; • equipment support; • strengthening institutions. The objective of the District-Level Decentralization Program -DLDP-PSCAP under the larger program is scaling up capacity of rural woredas. In other words, it aims at improving decentralized public service delivery (by deepening district-level decentralization),promoting good governance, and enhancing organizational effectiveness, manpower capability; institutionalizing decision-making processes (for participation at the grassroots level), and improving rural woreda working systems by way of contributing to Poverty Reduction and Sustainable Development.

Specific objectives and missions to be carried out during the medium-term phase include the following components: Component I: Institutional/ organizational rearrangement Refining the functional assignment of woredas, improving local governance, building efficient organizational structure of frontline public sector offices and defining the roles of kebeles in 441 rural woredas in Tigray, Amhara, Oriomia and SNNP Regional States; and,

Creating legal and institutional environment to devolve power to woredas, improving local governance and introducing efficient organizational structure of public sector offices and defining the roles of kebelel lower level units in 111 rural woredas found in Afar, Somali, Benshangul-Gumuz, Gambela, and Harari regional states. Component II: Manning and training Increasing the availability of sufficient manpower; enhancing professional and leadership competence; and, introducing human resource development plan for rural woreda. Component Ill: Woreda planning and financial control systems Speeding up the overall development and resource administration of rural woreda by introducing year-framed local planning and fiscal framework mechanism; improving the intervention of donors in local development, introducing financial rules, reviews, reporting mechanisms that decrease delays in corresponding and help facilitating assessments. Component IV: Grassroots participation Empowerment of the rural woreda people at grassroots level through increased participation in the planning and decision-making· processes of development programs; Component V: Region-Woreda fiscal transfer and own revenue enhancement Allocating financial resources required for local development on efficacy and equity basis (by introducing credible, transparent and fair Region- woreda transfer formula) and enhancing the capacity of and incentives for Woreda to mobilize own revenues and use of funds.

District Level Decentralization

31

Component VI: Minimum standard service indicators and performance mechanisms Improving planning and implementation capacities of public sectors in rural woredas by setting minimum standard systems (minimum standard service norms, benchmarks, performance tracking and reporting mechanisms and related guidelines). Component VII: Program development Strengthening the development of District Level Decentralization and facilitate its efficient and effective implementation by streamlining regipnal decentralization strategies, building capacity for coordination and execution of the progl'am, creating capacity for Monitoring and Evaluation (on going- monitoring and assessment of impacts), sensitization of DLDP, and identification of linkages and gaps for CB interventions at rural woredas. Component VIII: Office equipment Creating enabling-working environment by strengthening disadvantaged rural woredas and woreda offices with minimum basic office equipment or ICT related office facilities. Component IX: Strengthening capacities of lead institutions Maximizing the supply side intervention of affiliated lead public sector organizations at regional level which are directly involved in building the ~apacity of rural woradas in the area of training provisi2_n, human resource development, data base and system development.

The priorities of DLDP depends on long term strategies of national capacity building, on the phases of intervention envisaged by each regional government and other stakeholders as well as on scaling -up the on- going activities. Therefore, the activities planned for the fiveyear period are categorized under enabling, deepening and consolidating stages. The implementation of these stages varies in the nine regional governments. Four regional governments, namely; Tigray, Oromia, Amhara, and SNNP, concentrate on deepening and consolidating human capital, reinforcing grassroots participation and planning systems, and refining functional assignments and restructuring of some of the public institutions, etc. The other five regional governments, however, are still in the process of creating enabling environments in a credible way. They are assumed to begin establishing functional assignments, building-up institutions, creating financial management system and devolving decision-making on financial resource, etc. The various areas of intervention in these regions are given in the following table: Table 1: Areas of Intervention in Different Regions Four Regional Governments Capacity building for manning and training

Five Regional Governments Capacity building for Woreda Institutions/Organizations

Grassroots participation

Capacity building for manning and training

Capacity Building for Woreda

Reporting and supervision system for woreda

Institutions/Organizations

budgeting

Capacity Building for program Development

Region/woreda block grant credible model formula

Reporting and supervision system for woreda

Grass-roots participation

and budget allocation system budgeting Region/woreda block grant credible model

Capacity building for proqram development

32

Worku Yehualeshet formula and budget allocation system Capacity gaps on own revenue collection,

Capacity gaps on own revenue collection, utilization

utilization and revenue sharinQ.

and revenue sharing.

Local (Woreda) multi-year planning system

Local (Woreda) multi-year planning system and

and fiscal framework

fiscal framework financial

Minimum standard service

Minimum Standard Service

Implementation was categorized into two phases of three and four year durations: • First phase (2002-2004): This phase comprises initial intervention in 433 woredas. It will be in the four regional states and will be followed by 31 selected rural woredas in the five affirmative-seeking regional states •

Second phase (2005-2008): During this phase, deepening and consolidation of Capacity Building will be done in the 443 woredas in the four regional states. The initial intervention as well as deepening and consolidating of 108 woredas in the five affirmative seeking regional states will continue.

Rural woredas which were given priority in the decentralization and capacity building interventions are shown below. Table 2: Envisaged number of Woradas for Decentralization and Capacity Building Regional

Number of

States

Kebles

rural

Envisaged number of woredas for decentralization & CB intervention

woredas Enabling

Enabling

Deepening

Consolidation

stage

stage

Stage

Stage

Phase 1

Phase 2

Tigray

34

572

34

34

Amhara

106

3229

106

106

Oromia

197

NA

SNNP

96

3700

Afar

26

326

17

Somali

51

NA

22

Benushangul-

20

504

20

Gambela

9

209

Harari

2

17

197

197

96

96

9

26

26

29

51

51

20

20

9

9

9

2

2

2

541

541

Gumz

Total Source:

541

70

38

PSCAP,DLDP program Document Revised, MCB (2004)

2.1. Linkages with other programs and integration of crosscutting issues Decentralization program is invariably related to arrays of Capacity Building programs and sectoral activities at the local level. It requires the launching, facilitating and coordinating of the pertinent sectoral activities and programs effectively at the required speed. In some cases, there

District Level Decentralization

33

is a need to adopt systems tailored to local needs. Decentralization can be effective if only it is implemented in conjunction with capacity building interventions, namely, human resource development, institutional transformation and changes in working systems and processes. The civil service reform, which was initiated in 1996 aimed at creating a smaller, efficient and effectively performing public service, is amongst the major means of attaining the capacity building needs of rural districts. So are other initiatives such as urban management (in the transformation and urban-rural linkages), justice system reform, information and communications technologies, etc .• Information and communications technologies for fast, efficient and dependable exchange of information across sectors and regions one important in a large country such as Ethiopia with underdeveloped infrastructure. DLDP, therefore, fills the various gaps of the key initiatives through assessments and bottom-up stakeholder opinions. Developing and building capacities will create opportunities to address crosscutting issues effectively. Among the various crosscutting issues, the focus of the program depends on how far decentralization is harmonized with national development priorities and fiscal policies of the macro-economy. At the local level, empowerment is not realized without active participation of the rural women particularly in decision-making. It is hoped that utmost efforts will be made to advance the participation of the disadvantaged sections of the population. Improving the delivery of social services, ensuring property ownership and enhancing asset creation capabilities indeed, lessens the problems of women. Appropriate exploitation of the environment should be embodied in the plans of all the institutions at local level. To achieve sustainability, it is necessary to place the woreda structure and the regulatory system so as to ensure the responsiveness of every institution. At local level, institutions working against HIV/AIDS should be strengthened; planners will be involved in creating awareness along with the enhancement of assets, and creation of employment for the citizens.

3. Brief Review of DLD Performance The decentralization process at district level is a far-reaching and ambitious program (World Bank: PSCAP Ethiopia, 2004:23). The decentralization process is already operational in the four regional states with varying degrees of success.

3.1. Fiscal Decentralization The experience of bilateral organizations (Dutch, Italian, Swedish and Irish) shows that ,following the regional institutional structure transferring 90% of the budget through nondiscretionary means (block grant), there was a considerable improvement in the activity of woreda. This has given full autonomy over the use of fund. Decentralization is an important element of the federal system of government in Ethiopia. The application of fiscal federalism provides budgetary subvention to the woredas in the form of federal block grants and has given them autonomy in budgetary process and actual expenditures.

3.2. Capacity for Planning and Budgeting Although strategic plans have been prepared in most of the woredas, the planning system bear similarity with that of the pre-decentralization level and did not involve the community. Budgeting processes have been expanded at woredas. However, most of them were not able

34

Worku Yehualeshet

to exercise decision on all choices and preferences of development needs due to shortage of budget. Budget utilization was modest compared to the available resource at the early stage of managing the system.

3.3.Training Training on planning and budgeting were given to all woredas. Thus currently a sense of awareness and understanding has developed among woreda officials in the various procedures such as reporting, procurement and accounting. However, the community that was assumed to be major beneficiaries was gain little from these opportunities.

3.4. Service Delivery During the process of the pre-decentralization to woreda, the focus of delivering service was mainly on primary education, health care, and agriculture including rural water supply. Despite financial constraint and lack of 5killed manpower, the woredas were able to manage the bulk of services. To support recurrent and capital expenditure united block grants brings the resources closer to the target group for service provision. This, in turn, provides a basis for participation of the people in local development programs. This includes areas such as primary education, primary health care, rural water supply, rural roads and agricultural extension.

3.5. Manpower Manpower transfer was made from the zone and regional states along with logistic supply to woreda. Regional bureaus retained about 40% of their projected requirement and deployed the rest to the woredas. Attempt had been made to fill the key posts, but niost of the posts which require skilled manpower, were vacant or held by semi-professionals. In 2002, woredas had about 50% of the required manpower. At the time, the number of the civil servants in 430 woredas of four regional states was estimated to be 65, 172. The demand for professionals was assumed to be twice as much as the existing one. (PSCAP Ethiopia, 2004: 41). Salary scale, which was low during pre-decentralization, was raised in line with the civil service commission policy.

4. Functional and Expenditure Assignments The functional assignments of the tiers of governments in Ethiopia have their roots in the realization of the right for the self-determination of nations, nationalities and peoples of Ethiopia. The constitution of FORE of 1994 lays down the basis for building a political community that is founded on the prevalence of the rule of law, capable of enhancing a lasting peace; guaranteeing a democratic order; and achieving the country's social and economic development. The Constitution also enables the consolidation of the structure of regional states, their assignments and building up the decentralization process across the country. Major developments such as implementation of sectoral development policies and strategies have been taken place in Ethiopia following political and economic reforms during the last 10

District Level Decentralization

35

years. Along this, restructuring government has been the agenda both at Federal, Regional and District levels. To meet the intended missions institutions and ministries such as the Ministry of Rural Development, Capacity Building, Federal Affairs, Domestic Revenue, Finance and Economic Development were reorganized or introduced. In line with this development, functional assignments of electoral bodies including the role of the executive, cabinets, spokespersons and civil servants have been streamlined to promote transparency, avoid duplications and improve efficiency in public service provision and sub-national governments/ administrations. The second development was the reorientation of capacity building through sector development programs and reforms to achieve sustainable and accelerated development reduce poverty, promote responsive governance structure, efficient civil service arrangement, and justice system, etc. The third intervention was related to enhancing political and economic interaction processes for the common goal of attaining Sustainable Development and Poverty Reduction (SDPR). Towards this end, capacity building, civil service reform program, justice reform programs including decentralization, governance and empowerment were considered the building blocks of SDPR. In implementing the second wave of decentralization and empowerment it was found that the ultimate beneficiaries, i.e., the woredas had not been reached as intended. The key government administration bodies at district levels were not furnished with sufficient power and capacity that enable them discharge their responsibilities. The zones at the third tier superseded the decision-making powers of districts. Therefore, as of 2001, major change has been made by way of reorganizing functional assignments and shifting powers from zones to woredas in matters pertinent to financial affairs, i.e., administering their own income and expenditure. Based on the aforementioned developments, the functional activities related to the assignments of the four regional states and the respective districts are now being refined. The expenditure assignments of the federal government, regional State and Woreda administration are given in Annex 2. Tables 3 provide the overall and sector-specific expenditure assignment of federal government and regional state and woreda administration.

36 fSect

C SECTOR

FEDERAL

Expenditure

Consolidated budget

planning and financial control

REGIONAL STATE /SPECIAL ZONE

DISTRICT (WOREDA)

Federal budget

Regional budget

District budget

Transfer to regional states

Transfer to districts

Civil service reform (CSR) Agriculture

Worku Yehualeshet ·fie exoenditure assi!mment

Food security, water harvesting strategy and guideline Extension package strategies

M&E, reporting

M&E, reporting

CSR

SCR

Regional strategy and

District based

guideline Regional specific extension package strategies Medium and small

Large scale irrigation

scale irrigation Regional research

strategy planning & implementation Implementation of extensions Small scale irrigation Adaptive research

Research Natural resource

Administration

Administration

Disaster prevention and

OPP

Preparedness (OPP)

Land use administration OPP Natural resource protection development

Water supply Drinking)

Road

Water supply standard

Urban water supply

Hand dup wells,

Procurement

Deep wells

spring and

inter-regional road

River diversion

pond development

Inter woredakone

Rural travel and

road

transport integrated development (rural road, pavement)

District Level Decentralization SECTOR

FEDERAL

REGIONAL STATE /SPECIAL ZONE

Air transport

Federal arrangement

Air fields and

T elecomm unicatio

Federal assignment

Branch offices management

Education

Natural SOP policies and

Regional SOP policies

37 DISTRICT (WOREDA)

Facilities manaqement

strategies

strategies and

Standards of service

primary education

Curriculum

Curriculum

Higher education

Colleges

Federal management training

Regional management

institution TVET, IC/ITC Large scale procurement

1st and 2nd cycles

institution, TTI Secondary school cost sharing

Primary school management and cost sharing Allocation of district fund

Publication of education method resource mobilization and

Allocation of Regional fund

coordination Health

Natural HSOP policies and strategies Standard of services (4 hierarchy of service delivery,

Regional HSOP policies, strategies and actions Referral hospital, IC/ITC Allocation of district

medical training institutions IC/ITC Large scale procurement Resource mobilization and coordination

ports Health extension package

drugs etc) Referral Hospital and higher

Health centers, health

Allocation of regional fund

fund

38

Worku Yehualeshet

5. Fiscal Transfer 5.1. Revenue Assignments to woreda The revenues assigned to woreda are mostly location specific. In principle, woredas are expected to collect and utilize their own revenue. The utilization of the revenue with the block grant varies from region to region. The following are revenue sources assigned to woredas. 1. Personal Income tax from woreda employee and small traders, (TOT below birr 500,000 and above birr 9000 at a rate of 2%); 2. Rental income tax from individuals; 3. Rural land use fee; 4. Agricultural income tax; and 5. License, fees and charges from service rendered by woreda offices.

5.2. Rationale for Inter-Governmental Fiscal Transfer and Types and Mechanisms of Transfer The literature on fiscal federalism suggests several economic rationales for inter-governmental transfer. Addressing vertical fiscal imbalance is one of the reasons for the national government to retain the major tax bases leaving aside insufficient fiscal resource to cover the expenditure needs of the sub-national governments. In terms of horizontal fiscal imbalances, some jurisdictions have better access to natural resources and other tax bases that are not available in others. And they may have higher income, better fiscal capacity and expenditure need. The national government has the obligation to maintain a minimum standard of public service in all sub-national units. These are mostly caused by uncontrollable factors and should be addressed by national government transfer. The names given to the types of transfer differ from country to country. But they generally fall within the categories of conditional and unconditional grants. Conditional grants include matching open-ended and closed-ended grants, and nonmatching grants. Unconditional grants include basic need or expenditure need and fiscal capacity/equalization approaches, expenditure need and fiscal capacity/equalization approach considering standard and standardized expenditure and revenue (in per capita relativity terms and using disabilities), and specific purposes grants calculated by considering sectoral priorities, expenditure priorities projects, time and economic space (area) dimensions.

The basic objective of fiscal transfer to woreda level is to empower the woreda and grassroots population to decide on development activities affecting their lives/locally consistent with national and regional priorities. The type of .transfer is block grant to woreda. The kind of formula to be used takes into account efficiency and equity criteria, which would be credible and transparent. The allocation formula is decided by Regional states.

District Level Decentralization

39

5.3. Prototype Fiscal Transfer from Regional States to Woredas In order to assist regional states, prototype formula and allocation system was designed by MCB, DLDP in May 2002. The formula was, at first, simulated for 77 woredas of SNNP and provided 0.4 to 2.2 percentage share allocations to each. The model was a replica of the Federal Budget Subsidy Formula. A basic need approach incorporating revenue raising effort has been suggested considering the experience of other countries, availability of data and its presentation to the public for decision-making. In addition to the indicators that were used for Federal Budget Subsidy Formula in health, education and water sectors (excluding performance indicators), the following were included to reflect the realities of woredas. Agriculture: Ratio of development agents to household, percent of households covered by extension service, and ratio of livestock to number of people. Recurrent cost implication: Area of woreda and ratio of revenue collected to recurrent cost of the woreda.

Prototype allocation/guidance 1. For Regional States whereby data could be available on expenditure practices the use of indicators was recommended. 2. Treasury, revenue collected .from and dividends relegated to woreda were assumed to be sources of the financial pool. The mechanism and proportion of the woreda revenue collection, and revenue sharing was assumed to be improved. 3. External funded projects and programs shall not be considered in computation of the allocation. 4. At least 50% of the envelope was suggested for woreda to cover recurrent expenditure requirements and to make devolution practical. 5. Dividends from sales and trade taxes should be included in the allocation formula. 6. The financial pools between region. and woreda (region including zone) shall be administered by each tier and decided based on expenditure ahead of allocation. 7. Urban centers with high level population size and revenue shall obtain proportionate share. 8. Subject to the approval of regional Councils and communicated to woreda representatives the formula components to be used in the allocation at region level shall be designed by a technical committee. 9. The pay out to woredas should be carried out in accordance to the disbursement and reporting procedures developed in advance. 10. In 2003, working papers were distributed as additional source of information to make the resource allocation and prioritization process more realistic and consistent with national priorities and grassroots level needs. Efforts were made to communicate the draft grassroots participation framework, which also contains participatory budgeting process (DLDP: grassroots participation ensuring mechanism, manual).

5.4. Performance: Formula, Indicators and Weights Tigray, Amhara, Oromia and SNNP were pilot regional states whereby a block grant system that closely replicates the Federal approach has been implemented.

Worku Yehualeshet

40

In general, the financial pool (envelope) transfer to woreda consists of federal grant to regional states, own regional state revenue, sub-regional own revenue (excluding municipal revenue), and external fund (assistance and loan).The envelope, however, varies among regions. The Amhara Regional State includes specific percentage of external fund, while the other regional ·states opted to consider external fund in the calculation process. Similarly, woreda own revenues, which were excluded from the formula had been reconsidered by regional states. Thus, external loan and assistance including woreda own revenue were offset. The main indicators and weights assigned in the formula are given in Table 4. Tabl e 4 In d'1cat ors an d we1g . hts .m gran t a II oca f10n formu a Region

Population (CSA

Development

1994 census)

level/expenditure

Own revenue effort

Area

need Tigray Amhara Oromia SNNP Average

70 55 55 60

20 20 30 13.3

10 15 15 13.3

5

The number of specific expenditure related indicators was smaller than the proposed one, while weights assigned to the indictors were much higher than those envisaged. Budget allocations in the first year resulted in deficits in some woredas than expected by the regional states. Irrespective of adjustments, reallocations were made for urban centers in Amhara and SNNP. Requests from woredas of Oromyia were considered while other regional states replenished the gap through reallocations from the contingency.

5.5. First year (2002/03) Division of Financial Pool The first year of woreda level decentralization was implemented with strong enthusiasm from regional states. It was, however, entangled with a number of activities, especially the introduction of budget reform, disbursement, accounting, reporting and supervision systems, structuring of offices, and transfer of personnel and logistics. Since the allocation system required better and integrated performance of these activities, rigorous assessment of woreda was carried out. The fund for on-going capital projects was deducted from the pool based on the general picture of expenditure at woreda and regional state (including zone) levels, i.e., according to the importance of the budget prior to the devolution of assignment to the two tiers. Table 5 shows the shares of envelope applied between a regional state and their respective woredas.

41

District Level Decentralization Tabi e 5 : Resource a II ocat,on b etween Regional and Woreda governments Region

Regional State

Woreda

Zone

Contingency

%

% Amhara Oromia SNNP AveraQe

38 35 15

62

Earmarked

57 85

Earmarked & reallocated

29.3

68

-

8

Source: BoFED, BCB Amhara, Oromya & SNNP Field report On Fiscal relationships, 2003, DLDP Field review notes.

The amount reallocated to each woreda's municipality (higher order) and to the respective zones was not more than Birr 5 million in Amhara and 15% in SNNP.

5.6. Budgeting and Prioritization Budgeting and prioritization involves various steps. Each woreda sector office should submit budget requests to the woreda cabinet. Reviewing against the identified priorities of the woreda, the council may suggest some reconsideration to be made on. the submitted budget plan. Then sector offices will resubmit the revised budget request to the cabinet. Based on the revised request, a joint summit involving administrators, spokesperson, deputy, and sectoral offices negotiate differences of interests in line with envisaged problems and priorities of the woreda. Finally, agreement will be reached with referral discussion and approval of the woreda council. This process, however, had not been practiced in all woredas. Although one can say that the needs of the population are implicitly considered, participatory budgeting process has not yet been in place. Disbursement to woredas is made upon submission of capital and recurrent expenditure plans to regional state. Almost all the four states had attempted to adopt budget reform and account reform practices with varying degrees.

5.7. Expenditure Allocation In the first year (2002/03), on-going capital investment projects in basic sectors and other operating expenditure were covered through transfer and specific purpose resources allocated by regional states to the woreda. At this early stage, the expenditure was devoted mostly to financing payrolls. Similar practices continued during the second year (2003/04). The emolument for teachers and health personnel including for additionally deployed staff, at woreda level was unexpectedly high. As a result of a number of factors, the budget allocated to finance expenditure was found to be inadequate. Salary and operating cost consumed about 90% of the budget. Prior to the allocation, the patterns of the woreda expenditure were not properly assessed in the previous year. Structures parallel to the regional level were established at woreda level in line with the budget organizational arrengement at regional and federal levels. Number of desks and offices were organized. The personnel at woreda had not been in line with the projected requirement. It is only recently that skilled manpower has been assigned to woredas by way of replacing the existing semi-skilled personnel. Moreover, services such as health and education, which were managed by the zones, have been transferred to woredas. The woredas, have now larger assignments compared with their pre-decentralization status.

Worku Yehualeshet

42

In some instances, for example, in SNNP expenditure need approach formula was developed using major cost drives in 2003/04 allocation. In this approach factors related to expenses in the education sector include number of students, teachers' salary, and operational expenses such as teaching materials, stationary, maintenance, textbooks, other expense, etc. The allocation of capital grant and recurrent expenditures for the other basic sectors had been also treated in the expenditure need approach formula. External funds were taken into account during the calculation for distribution of financial resource to woredas. In Oromia, the allocation purposefully excluded on- going capital projects. The regional state issued memorandum and encouraged woredas to spend 8-10% of their allocation on capital investment and increase the share overtime.

6.Problems/Challenges and Issues of Decentralization 6.1. Problems The major problems of the decentralization process in Ethiopia are the following: 1. Acute shortage of budget for capital investment; 2. Extensive institutional structure with a large number of semi-skilled personnel; but with acute shortage of skilled manpower; 3. High cost of decentralization for operating; expenditure; and "non-economic cost" such as office furniture and equipment; 4. Delay in accounting reports (lack of transparency in reporting); 5. Lack of clarity on the division of assignments between woreda, region and zone; 6. Less compliance to orders and accountability, and slow pace of civil service reform practices; 7. Lack of vibrant grassroots participation, especially of women, in decision-making, development and democratic process (for example in prioritization for ensuing selfreliance, and in promoting pro-poor market development); 8. Lack of systematic integration and coordination of civic societies and NGOs, government and executing agencies ; 9. Lack of multi- year local planning system, coherent strategic planning and management; 10. Absence of transparency (for example on accounts, accruing information); 11. Lack of credible fiscal transfer methodology;

6.2.

Issues of Deepening Decentralization

The question how best to design decentralization and bring about success in developing countries has remained central in the ongoing debates among academicians, professionals and politicians. Major financiers and partners especially the World Bank sector studies stress the following important points to help rethink decentralization in developing countries. 1. Designing decentralization by incorporating institutions: This scheme emphasizes the need for separation of finance and capital market from inter-governmental issue and for allowing local governments to finance their own investments through fundamental approach of removing unsuitable local borrowing (reassigning revenue, revising transfer, arranging reliable reporting mechanism, etc.).

District Level Decentralization 2.

3.

4.

5. 6.

43

Synchronizing local policy: for example, expenditure assignments should match the availability and flexibility of budget and the ability to influence wage and employment polices and wage. Organizing service delivery at local level in alternative ways to produce or provide local public services (through a number of arrangements among public, private and tiers of government). Adopting asymmetric decentralization such as rural districts, tiered municipalities, and service delivery units like water boards, health districts; as one-size fits all is not appropriate for decentralization. Establishing competitive government and information system by making them more accountable. Measuring fiscal decentralization :Litwack et al., (1998) agrees on the need for more case studies, data and research for country specific /actively specific/ orientations, while the World Bank Decentralization and Sub-National Regional Economies Thematic Group (2004) contend the data available are imperfect. Thus, following this, measuring fiscal decentralization is the key issue to assess the impact of sub-national expenditure on service delivery in comparison to the national and overall expenditure effect.

In Ethiopia, the success of decentralization depends on the nature of its design, intermediate and final outcomes sought to bring a state of change for achieving poverty reduction and sustainable development. Despite lack of sufficient data, the following issues influence the degree of .decentralization and its positive contribution to Ethiopia. These are; 1. Consistency of macro-economic policies to woreda level decentralization: The expenditure -budget or GDP ratio at sub-national level basic sectors (health, education, water) the increase of expenditure in basic sectors aligned to the PRSP, Millennium Development Goals; 2. Cost of decentralization: Estimating the cost and efficiency of pursuing decentralization for adjusting strategies; 3. Empowerment: The extent of transferring decision-making (to local governments, disadvantaged section of the population, and women) at local level in the delivery and usage of services; 4. Local governance: The existence of good governance, and enhancement of democratic participation in a sustainable way; 5. Fiscal decentralization: There is a need to design improved, transparent, and credible formula based on standardized expenditure (meeting minimum standard and standardized revenue, exploiting own fiscal capacity to finance minimum standards) .The formula to be incorporated in the allocation mechanisms should encourage revenue raising effort. It should also take into account performances carried out for ensuring minimum standard service provision, and efforts made for participatory budgeting processes at the grassroots level. 6. Integration of programs, projects, and actors in the development and democratization process.

Annex 1 Government structure Ethiopia had a four-tier system of government until 2001. Below the Federal government there are nine Regional States, one city administration (the capital Addis Ababa) and one city council (Diredawa).The antecedent to the Constitution - the 1991 Proclamation that establishes national regional self-government-

44

Worku Yehualeshet

recognizes 63 different ethic groups, 48 of which are large enough to establish their own self-government structures. Previously, the Regional States were divided into 66 zones, which in turn were divided into 550 local administrations (woredas). Therefore, up to 2001, the FORE consisted of four government tiers, namely;

+

The Federal Government;



9 Regional States including one city administration and one city council;

+ +

66 Zones Administration which also comprise of municipals; and Over 550 Districts administration consisting of 541 rural woredas and municipalities of highest and lower orders.

Recently, the Regional States have been divided into local administrations (woredas). Under these, there are a number of smaller electoral units, the communities. To implement the decentralization in four Regional States, the government structure follows the structure of Regional state with the exception of a few special zones having higher states. Apart from this, urban local governments especially at zone administration level are being restructured to function in line with the jurisdiction. The woredas and kebe/es of the government state structure are provided in the following Table (Table annex 1). Table Annex 1: Number of Woredas and Kebe/es in Regional States Regional States

Number of rural woredas

Kebles

Tigray

34

572

Amhara

106

3229

Oromia

197

NA

SNNP

96

3700

Afar

26

326

Somali

51

NA

Benushangul -Gumz

20

504

Gambela

9

209

Harari

2

17

Total

541

Source: PSCAP Ethiopia 2004,MCB Legislative and executive functions Like the Federal

government, ,Regional States elect councils,

administrator/president,

cabinets,

spokesperson, and deputy spokesperson who work in different branches of the governance, i.e., the legislative, judiciary and executive. The States also have organized administrative, legal and audit bodies to backup their functions. HIV secretariat, for example has been established within administrative office of the administration. Institutional structure At least, 24 sectoral public bodies with specialized functions and responsibilities for public service delivery are organized under the regional states. By 2001, similar structure has been adopted at 430 districts of the four regional states. Between 20 and 30 offices/desks are organized in the varies woredas. The functional assignments between district and regional state have by and large been made distinct.

District Level Decentralization

45

In SNNP at district (woreda) level, the following public bodies have been established. 1. Capacity building office (Pool No.1) A. Capacity Building Office B. Health Office C. Education Office 2. Rural Development Office (Pool No. 2) A. Rural Development Office B. Rural Roads Desk C. Rural Water Development Office D. Rural Mine and Energy Desk E. Loan and Product Promotion Desk F. Disaster Prevention and Preparedness and Food Security Desk G. Irrigation Desk H. Natural Resources and Environmental Protection Desk 3. Finance and Economic Development Office (Pool No.3) A. Finance and Economic Development Office B. Public Mobilization Office C. Public information and Public Relations Desk D. Youth and Sport Desk

E. Women's Affairs Desk F . Cooperatives Office G. Social Affairs Desk 4. Woreda Administration Office (Pool No.4) A. Woreda Administration Office B. Legal and Administration Office C. The Office of the Prosecutor D. Police Office E. Militia Office F. Public Complaints Affairs Office G. Neighboring Regions Affairs Office

Structural issues regarding number of woredas and roles of zones Zones in Amhara and Tigray Regional state conduct disbursement for the woredas. In SNNP, there are 13 zones with constituency and greater legislative role than the zones in Tigray and Oromia Regional States. Nationality zones have special role over woredas. In principle, zones are not considered as different layer of the Regional state following the deepening of decentralization to woreda. In fact, zones are not points of decision-making and not center of control for woreda. In some cases, however, the legacy of control has not been ceased entirely. Still, there are signs of reinstitution by reason of inaccessibility of woredas for reporting, bank transaction, and information flow. Exception to this prevails in Ormia and partly in SNNP. The number of woredas in the four Regional State has marginally increased from the original 430. Currently, there are some measures taken in the refinement of the structures of woredas. The steps include demarcation and amalgamation of woreda based on the requests of the communities while maintaining their entity as electoral units, linking rural and urban woredas in service provision and expenditure division.

46

Worku Yehualeshet

Annex 2. Expenditure assignments of the Federal Government, Regional State and Woreda administration 4.3.1 Federal Government Assignments + Defense, public security, federal police; + Foreign affairs; • Overall economic and social policies; + Country's fiscal, monetary and foreign investment policies commerce; 111 Regulate foreign affairs 111 Federal budget 111 Establish uniform standards of measurement and calendar 111 Conferring of citizenship and administering immigration related issues 111 Deployment of army in situations beyond capacity of regional governments Declaration of state of emergency Levy taxes and collect duties on revenue sources reserved Administer national bank printing of currency, foreign exchange, domestic borrowing. Sector specific expenditure assignments • Enact laws for the utilization and conservation of land and other natural resources, historical sites and objects; • Establish and administer air, rail, waterways and sea transport and major roads linking two or more states, as well as postal and telecommunication services; • Regulate inter-state trade; • Administer federally funded institutions that provide services to two or more states; • Establish national standard and basic criteria for public health, education, science and technology, protection and preservation of cultural historical legacies; • Utilization of waters, rivers and lakes linkage or cross up boundary of national territory.

For example, the Federal Ministry of Health (FMOH) responsibilities include creating enabling environment and capacity building of regional states. This includes, among others, drafting and implementing overall policies, strategies development of guidelines, manuals to facilitate the implementation of family health services, child development, integrated RH services, HIV/AIDS Case Management, medical services, etc. Similarly, the Federal Ministry of Education is responsible especially for capacity building of the nation's education system focusing on the creation of enabling environment for universal primary education, vocational and technical training expansion and licensing of higher education, curriculum development, etc.

Regional state expenditure assignment

+ State administration; + Developing policies, strategies and plans of the state; + Enacting and executing state constitutions and other laws;

District Level Decentralization + Levying and collecting taxes and duties on revenues reserved for the state; +

+ +

47

Drawing up and administering budget; Enacting laws on the state civil service and implementing appropriate to national standards; Administering police, maintain peace and order. Sector specific expenditure assignments • Administering land and natural resource in accordance with Federal law; • Drawing policy and strategy (culture, social security, curriculum, information, public relation, capacity building, agriculture service delivery, etc.); • Administering Civil servant of the state, implementation of CSR; • Regional agricultural research and technology; • Human capital development; • Regional water resource development and protection policy; • Building Inter- woreda and regional road; • Vocational and technical training, TTI, medium level college; • Hospital and referral hospital; • Building regional health service standard setting; and • Disaster prevention and preparedness.

Woreda (District) administration expenditure assignments

+ Initiating and implementing of their own administration, rules, and procedures; + Implementing of policies, and laws by the regional state; + Preparing and approving district development plans, programs and budgets; + Setting specific tax rates and collection of taxes and levies (such as agricultural income tax, land use fee) and utilization of the same not the under administration of regional state;

+ Administering of functional resources (own resource and transfers); + Administration and management of civil servants in the districts, implementation of civil service reform;

+ Security and peace (police, militia). Sector specific expenditure assignments nd 1 • Capacity building and primary schools (15 and 2 cycles); • Coordinating school management and cost sharing activities; • Primary health care preventive and curative activities (health posts, health centers); • HIV aids secretariat ; • Financing health care; • Rural road (lower grades); • Agricultural development activities (extension packages, food security, water; harvesting, disaster preparedness, prevention, small irrigation); • Land administration and natural resource development and protection; • Supply of drinking water (hand- dug wells, springs, etc.); and • Rural energy

48

Worku Yehualeshet

References DLDP. (2002). World Bank Project, MCB Framework of Local Government (Woreda) Fiscal system, intra regional fiscal relations and grant formula, DLDP Ethiopia

working

document.

MoFE (2003).lmpact of decentralization on program implementation, Experience of Dutch, Italian, Swedish and Irish Development Program : Irish and Ethiopia, a paper presented at MoFE, 2003. Litwack,

Jannie, Junaid Ahmed and Richard Bird. (1998). Rethinking Decentralization. World Bank, PREM,

Sector Studies Series. Ma, Jun. (1995). Intergovernmental fiscal transfer: A comparison of nine Countries, abstracts

on overview of

fiscal transfer mechanism MCB (2003) Framework of local government (woreda) fiscal systems, intra regional fiscal relations and grant formula, MCB, World Bank project ID NO P 050938, MCB. (2003).Medium Tem National Capacity Building, MCB: Ethiopia. MCB, DLDP (2002). Study for allocation of block grant to woreda, task force for financial collection and utilization. Addis Ababa: Ethiopia (In Amharic). MCB, DLDP. (2002). Working papers, advice notes on the design of formula, allocation and woreda budgeting, working document, Ethiopia.

Decentralized development Cooperation

49

4 Decentralized Development Cooperation, Governance and Rural Livelihoods: The Case of Amhara Region, Ethiopia Tewodros Bakaffa*

1.

Introduction

Ethiopia has decentralized (devolved) significant power to regional states since the Transitional Government of Ethiopia (TGE) came to power in 1991. The regional states which are established by the Federal Democratic Republic of Ethiopia (FORE, 1995) are provided with substantial degree of autonomy and accorded with wide range of power and authority. The regional states are "autonomous governments in that they have elected bodies and have the power to make legislation and policy decision regarding many activities" (Tegegne, 2001 :22). The woredas, which are the lowest and basic units in the hierarchy of the government structure, are also provided with more power than it had been before to play a much greater role. The Amhara National Regional State (here after ANRS) is one of the regional states established under FORE. It is located in the Northwest part of the country. The Region has an area of approximately 160,553 km2. In terms of area, the Region stands third next to Oromiya and Somali Regions and covers 15% of the country's total physical area (BoPED, 1998/99). In 2002 the population of Amhara National Regional State (ANRS) is estimated to be 16.8 Million with an annual growth rate of 2.9 percent. This makes the region the second most populous region in the country next to Oromiya. About 90 percent of the region's population live in rural areas mainly depending on agriculture. On the average, agriculture accounts for 69.2 percent of the regional GDP. Amhara region has a four-tier system of government: the Region, Zone, Woreda and Kebele. It is divided into eleven administrative zones and 105 woredas. There are councils at regional, woreda and kebe/e levels having power and authority within their own capacity and respective territory. Thus, the administrative structure goes down to the vast majority of the rural farming society and urban dwellers which are organized into 5300 kebe/e administrations. Zones, being representative of the region at sub-region level and meant to facilitate administrative functions, do not have their own council led by an Executive Committee elected by and from members of the regional council. The legislative organ and sector departments are also in place at zone level. The woreda government has a Council, an Executive Committee and sector offices and legislative organs. The Executive Committee having 9 to 15 members are elected from and by the Council for the purpose of managing the day-to-day

* PhD Candidate, Institute of Social Studies

Tewodros Bakaffa activities of the council. The fundamental powers of the woreda government "range from preparing development plans and budgets and levying taxes to constructing, maintaining and administering lower level infrastructure such as rural roads, primary schools, junior health institutions" (Tegegne, 2001: 25). According to the ANRS constitution (ANRS, 1995), the "Kebele administration is part of the state administration and as the lowest form of government receives instructions and guidance from the woreda government" (Tegegne, 2001 :26). It has a Council, Executive Committee consisting of 9 members, sector offices and a social court (the judiciary committee) comprising 3 members elected from and by the Council. Some of, functions of the kebele administration are collection of land tax, enforcement of decisions and instruction, including mobilization of residents for community campaign work and the mediation of minor conflicts which arise among members. In line with the government structure, there are Development Committees established at zone, woreda and kebele levels that are meant for coordinating and overseeing the development effort in their respective territory and areas of jurisdiction. With the intention of reaching the community at the grassroots levels so as to easily 'mobilize' the people for the purpose of community work campaign and other activities-- the government structures below the kebe/e level - often know as Got (sub-kebe/e), Mengistawi Budin (Government Team) and Yelimat Budin (Development Team) are established in each and every woreda by the directives of the regional state.

2. Statement Methodology

of the

Problem,

Research

Question

and

The woredas being the lowest (basic) hierarchy of any nation-al regionaJ state in the country are provided with powers and authorities to play a much greater role in local development programs. Decentralization cooperation programs whereby donors relate directly with regional and local governments and other local development actors are accompanying this decentralized development effort. How do these decentralized development programs relate to and interact with multiple of local development actors? What has happened when donors, which have been linked to and closely working with central government authorities for decades, start to deal witt, multiple of local development actors? What are the opportunities and challenges experienced by decentralized cooperation programs in working closely with multiple of local development actors? Did these programs succeed or fail in terms of attaining the intended objective of improving the livelihood of the community at the grassroots level? These, among others, are some of the critical questions that will be raised while dealing with decentralized development cooperation programs. Three decentralized local development cooperation programs- Sida supported "Swedish-Amhara Rural Development Program (SARDP)", Finnida supported "Rural Water Supply and Environmental Program (RWSEP)" and UNICEF

Decentralized development Cooperation

51

supported "Woreda Integrated Basic Services (WIBS)" which are operating at different woredas in the Amhara Region are subject of this research. These different programs are operating differently while they are operating down on the ground and dealing with local development actors particularly the community. The then available data suggests that the difference among the programs is not limited to their modus operandi, but also in their performances. The main research questions, therefore, are: 1. Are there significant differences among the programs particularly in terms of improving the livelihood of the community? 2. What are the major factors that attribute to such difference in livelihood outcome? The objective of this research is, exploring whether or not there are significant differences among the programs under consideration in their performance and identifying the major factors that attribute to such difference. The major hypotheses established for the purpose of this research are: 1. There are significant differences among the woredas participating in different programs in terms of improvement in the livelihood of the community. Particularly, woredas participating in WIBS program are less performing than woredas participating in SARDP and RWSEP. 2. Difference in the livelihood outcome of woredas participating in these programs arises mainly from a number of factors that can generally fall under two broad categories- program-specific factors and woreda (local conditioning) factors. 2.1. Program-specific factors include: Problem identification needs reflection and responsiveness; Planning and decision-making; Financial management; Follow-up, monitoring and Evaluation; Tendering, contracting and procurement; Backstopping managerial and technical services; Participation. 2.2. Woreda (local conditioning) factors are: Economic factor (agriculture potential); Woreda enablement policies and strategies; Woreda Governance; Election and community leadership; Responsiveness; Transparency and accountability; The woreda is the unit of analysis for this research and Comparative study is the major methodological thrust. Both primary and secondary sources of data are collected. The field research went through two-stages- the primary rapid appraisal study and a more in-depth study. Since there was no much a priori knowledge particularly as to the enabling policies and strategies and system of governance established in the woredas, structured and semi-structured questionnaires have been developed. Of the total 40 woredas, the first stage study was conducted in 22 woredas. This stage also helps to capture and gain

52

Tewodros Bakaffa

better picture as to different factors that could determine the performance of the programs and select the woredas that need to be incorporated in the secondstage in-depth study. Based on preliminary findings of the first-stage RRA study and some other criteria, 10 woredas are selected for the in-depth study. Dembecha- a high agriculture potential woreda where there is no donor supported program and known as one of the best performing is included both in the first and second stage studies as counterfactual woreda. Different approaches and methods have been employed in the undertaking of the second-stage (in-depth study) data collection. Much of the data collected covers the years 1994-2001.

3.

Theoretical and Conceptual Issues

The theoretical and conceptual framework of this research deals with myriad of issues which are intricately linked to each other and that would help capture as what is meant by performance of programs. Concepts and theories in relation to livelihood, local development, decentralization,(good) governance, participation and decentralized development cooperation are taken into account. With the failure of the traditional top-down and centrist approaches to development and giving way to bottom-up and participatory approaches to development, the notion of 'decentralized development cooperation' gained prominence in the development arena (Wunsch and Olowu, 1990; Bossuyt, 1995). The concept of decentralized development cooperation unlike the traditional 'project approach' to development insists on the importance of collaborative and partnership approach to development which is inherently process oriented (Amanor and Annan, 1999). Decentralized development cooperation, if designed and implemented based on realities on the ground could provide much wider opportunity in reaching the community at the grassroots level and in bringing about improvement in the livelihoods of the community and sustainable local development (Hertogs, 1999). The concept of livelihood refers to "the activities, entitlements and assets by which people make a living" (Wanmali, 1998:4). Improvement in the livelihood of the community refers to improvement in activities, assets and entitlement of the community. Though income is an important component of livelihood, the concept of livelihood is broader. It covers quite wide-range of dimensions. Education and health, for instance, are crucial components of capability dimension of livelihood. The fundamental objective of decentralized development cooperation is "to ensure 'better' development prospects, by taking greater account of the needs and priorities expressed by the population .... It consists on the one hand in bringing together and ensuring collaboration at different intervention levels of the potential economic and social actors ... On the other, it consists of the active and determining participation of direct beneficiaries in decision-making and the different stages of actions that concern them" (Hertogs, 1999:2). Local actors are seldom involved in project identification, appraisal and selection. Program management remains highly centralized (Bossuyt, 1995:4).

Decentralized development Cooperation

53

The milestone for successful decentralized development cooperation programs is participation. It is imperative to create enabling environments whereby various development actors discuss, dialogue, identify, design, implement, monitor and evaluate programs. This condition will create strong sense of ownership of the programs that, in turn, is critical in attaining the objective of improvement in the livelihood of the community. Difference among the programs in the participation of various stakeholders particularly the primary stakeholders- the community is one of the factors that could attribute to difference in livelihood outcome of the people participating in various programs. Albeit the varying views, there are strong justifiqations for the increasing and widening acceptance of participation. Among others, the ability to make use of the potential immense local resourceslocal knowledge, experience, skills, technology, labor, etc.; increased efficiency and effectiveness of development programs; enhanced local capacity; better reach of the poor and women; the ability to achieve the objectives of improved and sustainable livelihoods, and (local) development are some of the major reasons justifying the need for and importance of adopting and pursuing participatory approach to development (OECD, 1997a,b; UNDP, 1999a). Though there is a general consensus that 'participation' lies at the heart of development, the concept is understood, defined and interpreted varyingly by different development thinkers and international agencies. The conceptualization of participation could generally be categorized into two: 'popular participation' and 'participatory development'. 'Popular participation' over-emphasizes the participation of the people at the grassroots level particularly of the poor and the marginalized by way of considering them the most critical and decisive one that shapes and brings about sustainable development. Program studies are mainly designed in such a way that local governments and the community take prime position and play leading role in various aspects of the programs. From the context of these programs and the localities, the concept of government enablement to the community and the arrangement of local government-community partnership, inter alia, are critical and relevant. Therefore, in localities where enabling environments for the participation of the community and CBOs in particular are created, ceteris paribus, it is likely to find the performances of local development programs in a better position than localities where this is not the case. Three definitions of governance are widely observed in the literature. International development agencies such as the World Bank and United Nations institutions have tried to define what governance is all about. The United Nations Development Program (UNDP) defines governance as "the exercise of political, economic and administrative authority to manage a nation's affairs. It is the complex mechanisms, processes, relationships and institutions through which citizens and groups articulate their interests, exercise their rights and obligations and mediate their interests" (UNDP, 1997b: 1). Governance, as conceived by these multilateral organizations, emphasizes leadership- the manner in which political (state) leaders manage, use, or misuse power- to promote social and economic development or to pursue agendas that undermine such goals. Good governance is conceived from a process perspective with emphasis on rule of

54

Tewodros Bakaffa

law, accountability, participation, transparency, and human and civil rights (Olowu and Sako, 2002: 4). The second definition is related to sharing of authority and responsibility between state and non-state actors. The co-arrangement system of governance where by the state and non-state actors are not operating separately but together and in conjunction is emphasized in this definition of governance (Kooiman, 1993; 2003). "The school therefore views governance as forms of multi-organizational action rather than exclusively state actions. An important difference from the first approach is that governance is judged as good or bad by both processes as well as outcomes: the use of state and non-state institutional resources to solve social problems. This approach is also referred to as the partnership approach to governance" (Ibid: 5). The third characterization of governance, gives especial prominence to "the fundamental rules that regulate the relationships between rulers and the ruled, the rules in use, or constitutive choice rules and operating at deeper levels of analysis" (Ibid). According to this approach both process and outcomes are crucial. "While democratic governance promotes rules that ensure the fundamental equality between the rulers and the ruled, autocratic forms assume and reify inequality between them. But the primary criteria for distinguishing between good and bad governance is the outcome of policies promoted by public organs. Good governance results from the activities of the public sector institutions as they work with other societal organizations to formulate public policies and programs, which are implemented to improve the people's welfare, reduce poverty and realize other public and societal goals" (ibid.). The ultimate goal of good governance is, of course, achieving sustainable human development. Sustainable human development (SHD) "is development that not only generates economic growth but distributes its benefits equitably, that regenerates the environment rather than destroys it, that empowers people rather than marginalizes them. It gives priority to the poor, enlarging their choices and opportunities and providing for their participation in decisions affecting them. It is development that is pro-poor, pro-jobs, pro-women and pro-children" (UNDP, 1998b: 1).

4. Livelihood Outcomes Are there differences among the woredas in their livelihood outcomes participating in different programs and within the same decentralized local (woreda) development cooperation programs that are being carried out in the Amhara Region? Given the overall situation of the woredas in the Amhara region, income and basic social services are the most basic foundation on which livelihood rests on. Information generated from the household survey, focus group discussions and interviews made with leaders of CBOs made clear that income and provision of basic social services (health, potable water, and education) are critical components of livelihood.

Decentralized development Cooperation

55

In the Amhara region more than 85% of the livelihood of the population depends on agriculture. Crop production followed by livestock husbandry is one of the major factors that explain the increase or decrease in livelihood in of the rural population. Social and health services are at the top of the list of priorities of the rural population of the region followed by potable water supply and education. The results of the livelihood analysis based on information extracted from statistical data are presented on Table 1. Except Hulet lju Enessie Woreda (a woreda participating in Finnida supported RWSEP), which stood 4th, Sida supported SARDP woredas took the rank ranging from the 1st to the 6th. The two sample woredas from Fihnida supported RWSEP - Hulet lju Enessie and Dera stood 4th and i\ respectively. From the ten in-depth study woredas last raking two are Ebinat and Belessa, which are UNICEF supported WIBS programs. Dembecha, which is a control woreda, stood 8th. The results are quite interesting from the point of view of the woredas participating in different programs. Changes in livelihood registered in woredas participating in Sida supported SADRP, generally speaking, are substantially higher than those of Finnida supported RWSEP and UNICEF supported WIBS program woredas. Finnida supported RWSEP woredas are found in a better position than UNICEF supported WIBS program woredas. The counterfactual woreda, Dembecha which is a high-agriculture potential woreda ranked 8th which is still better performing compared to UNICEF supported woredas, Ebinat and Belessa which are, of course, low agriculture potential woredas The difference in livelihood outcomes among the woredas participating in different programs becomes much more evident by investigating the livelihood outcomes of woredas participating in different programs but having the same or different agriculture potential. It is found that two low agriculture potential woredas participating in SARDP have performed better than the high agriculture potential woreda participating in Finnida supported RWSEP - Dera, the two low agriculture potential woredas -Ebinat and Belessa which are participating in UNICEF supported WIBS program and the high agriculture potential woreda which is taken as a control group- Dembecha. In addition, the three high agriculture potential woredas participating in Sida supported SARDP are found to be better performing in terms of changes they have registered in livelihood outcomes than the two high agriculture potential woredas participating Finnida supported RWSEP- Hulet lju Enessie and Dera. It is important to underline that woredas participating in UNICEF supported WIBS program are the lowest ranking woredas compared to woredas participating in SARDP and RWSEP whatever their agriculture potential is. To better figure out the livelihood outcomes registered by different woredas in the programs period, a household survey is carried out in addition to the analysis of the livelihood conditions based on woreda level statistical data. The household survey on issues related to livelihood is believed to shed light as to the perceptions of the community on changes brought by the intervention programs on their livelihood conditions in the years under consideration. Whether or not , there were improvements in their livelihoods, income, health condition, and there

56

Tewodros Bakaffa

were increases on their children school attendance and potable water supply and water-borne diseases decreased were some of the major areas of livelihood components which the household survey covers. The results of the household survey by program show that the share of households reporting improvement in the livelihood conditions and increase in income in the years under consideration (1994/95 - 2000/1) is greater in SARDP woredas followed by RWSEP, WISS and the government program woredas in their order. This suggests how community's perception of livelihood is strongly related to that of income. None of the households from SARDP woredas were reported decline in livelihood condition or income. In the case of RWSEP and WISS program woredas 16.67% and 38.33% of the households, respectively, claim that their income has decreased and livelihood condition has declined. The substantial difference in the claim of improvement in livelihood and income condition among the programs has been validated by undertaking statistical significance test. Statistically significant difference in the livelihood outcome and income observed between Sida supported SARDP on one hand and Finnida supported RWSEP, UNICEF supported WISS and government on the other hand. Impact on improvement in livelihood and income condition is found to be higher than the other programs. There is also a statistically significant difference in livelihood condition and income between RWSEP on one hand and SARDP and WISS on the other. Livelihood outcome and income improvements by RWSEP is found to be less than SARDP, it is higher than WISS. Moreover, the difference in the livelihood condition and income registered between WISS and the other programs- SARDP and RWSEP is statistically significant. This clearly shows how WIBS is performing poorly in improving both the livelihood condition and income of the community in the face of SARDP and WISS. Moreover, the health condition of households covered by SARDP is found to be in a better position than households from RWSEP WI BS program and government program woredas. Besides, as of conditions of livelihood and income, the share of households asserting improvement in health condition in RWSEP is found to be higher than those households covered by WISS and government programs. Statistical significance tests in relation to difference registered among the prograrris in health condition shows that there is statistically significant difference between SARDP on one hand and RWSEP, WISS and government on the other. Though not with the government program, RWSEP has statistically significant difference with SARDP and WISS programs. Thus, RWSEP is found to be less performing in its impact on health condition compared to SARDP and better performing compared to WISS. The difference in this respect between the government program on one hand and RWSEP and WISS on the other, however, is not statistically significant.

Decentralized development Cooperation

57

When it comes to provision of potable water and prevalence of water-borne diseases, it is found that Finnida supported RWSEP and Sida supported SARDP are in a better position than UNICEF supported WIBS and government programs. It is worth mentioning that all the households requested in all programs woredas witnessed the accessibility of school in the nearby. However, there are differences among the households in different woredas in sending (enrolling) their children to school. The statistical significance test discloses that there is a statistically significant difference between SARDP and WIBS in terms of school enrollment of children. The difference among other programs in this respect, however, is not statistically significant. Based on the above discussions it can be concluded that the livelihood condition of households from Sida supported SARDP woredas has been showing betterment in comparison with households from other programs woredas. Following from SARDP the Finnida supported RWSEP has brought significant impact on the livelihood of households than communities under WIBS and government programs. Apparently, the livelihood condition of households from UNICEF supported WIBS program woredas is found at the lowest level-even lower than the government program woreda, Dembecha in most of the critical elements of livelihood. The above inter-program comparison clearly reflects the existing disparity among programs in impacting livelihood outcomes. Generally speaking, woredas participating in Sida supported SARDP are found to be better performing than others participating in RWSEP, WIBS and government programs. Interestingly, it is observed that the low-agriculture potential woredas under SARDP, namely Wogedi and Legambo were found in a better position than Dera and Dembecha, the high-agricultural potential woredas participating in RWSEP and governments program, respectively. Next to SARDP, woredas participating in Finnida supported RWSEP were found to be better in improved livelihood outcomes of their respective communities than both the UNICEF supported WIBS program and counterfactual woredaDembecha. The intra-program assessment is one of the crucial dimensions that need further investigation so as to figure-out whether there are differences among woradas participating in same program in terms of livelihood outcomes. With-in-program comparison not only helps to see the magnitude of disparity among woredas in terms of livelihood but also calls for further exploration of the specific woreda (local) conditioning factors that led to such difference. While Sida supported SARDP has been operating in both high and low agriculture potential woredas, Finnida supported RWSEP has been focusing mainly in high-agriculture potential woredas. UNICEF's WIBS program is concerned with low-agriculture potential woredas. The counterfactual government program woreda - Dembecha is a high-agriculture potential woreda.

58

Tewodros Bakaffa

While Machakel, Awabal and Gozamin are the high agriculture potential SARDP woredas taken as sample woredas for the in-depth study, the remaining two sample woredas from the same program- Wogedi and Legambo are lowagriculture potential woredas. It goes without saying that the sample woredas for the in-depth study from Finnida supported RWSEP and UNICEF supported WIBS program are high and low agriculture potential woredas respectively. The two low agriculture potential woredas participating in SARDP- Wogedi and Legambo have registered less performance in terms of changes in livelihood over the years under consideration than the high agriculture potential woredas participating in the same program- Machakel, Awabal and Gozamin (see Table 1). The lower performance of Wogedi and Legambo in comparison to the highagriculture potential woredas participating in SARDP indicates that difference in agriculture potential contributes to difference in livelihood outcomes. Interestingly, however, Wogedi and Legambo which are low-agriculture potential woredas participating in SARDP are found in a better position in livelihood outcome than Dera and Dembecha which are high-agriculture potential woredas participating in RWSEP and government programs, respectively. However, differences in agriculture potential matters among the woredas led to difference in livelihood outcomes, but it is not necessarily the only local (woreda) conditioning factor to make difference. Though the high-agriculture potential woredas participating in SARDP are all categorized as high level performing, there are with-in-group variations among themselves in their livelihood outcomes. Gozamin is found at the top of Machakel and Awabal which are all high-agriculture potential woredas participating in the same program, SARDP. This brings into attention the need for considering one or more woreda conditioning factor(s) that attribute(s) to differences in the livelihood outcomes of the woredas. Of the woredas participating in SARDP Gozamin, Awabal, Machakel, Legambo and Wogedi stood from 1st to the 5th rank in their share of households that state improvement in their livelihood conditions in the years under consideration, respectively. From the household survey statistically significant differences were observed between high and low agriculture potential woredas in terms of percentage of households claiming improvement in their livelihood condition. The high-agriculture potential woredas are found to be better performing in this respect than the low-agriculture potential woredas. The share of households in high-agriculture potential SARDP woredas claiming that their livelihood condition has been improved is 83.33% in Gozamin, 70% in Awabal and 63.33% in Machakel. On the other hand, the share of households reporting improvement in their livelihood condition in two low-agriculture potential SARDP woredas is 56.67% in Legambo and 53.33% in Wogedi. The results reached lead to the conclusion that differences among the woredas in agriculture potential is one of the crucial factors that attribute to difference in livelihood outcome of the woredas. The higher or the lower the agriculture potential of the woreda is the higher or the lower the respective livelihood outcome of the woreda is. One of the crucial components used for livelihood assessment during the household survey was income. Interestingly, the results revealed more or less

Decentralized development Cooperation

59

similar to that of livelihood. This condition suggests how strongly was households' (communities') perception of livelihood related to income. Income improvement between high and low agriculture potential woredas shows statistically significant difference. Obviously, the high agriculture potential woredas are found in a better position than the low agriculture potential woredas regarding improved income. In addition, it becomes apparent that the share of households that report increase in income is found to be higher in highagriculture potential woredas under SARDP- Machakel, Awabal, Gozamin than low-agriculture potential woredas participating in the same program: Wogedi and Legambo. High and low agriculture potential woredas participating in SARDP were also found to be significantly differing in terms of improved income. This implies that agriculture potential is one of the critical factors that attribute to income variation, which is one of the crucial components of livelihood. Marked differences were observed among percentages of households claiming increased in income in SARDP woredas. While the share of households claiming increased income accounts for 86.67% of the households in Gozamin, it is only 63.33% and 66.67% for Machakel and Awabal, respectively. In this respect, the difference in between Gozamin and Machakel in particular is found to be statistically significant. This suggests that besides differences in agriculture potential, there could be one or more local conditioning factor(s) that attribute(s) to such difference in the income among the woredas. Health is the other crucial component of livelihood. The share of households claiming improved health condition between low and high agriculture potential woredas (irrespective of the programs they are participating in) was found to be statistically significant. Likewise livelihood and income, statistically significant differences in share of households claiming improved health condition were observed between low and high agriculture potential woredas participating in SARDP. While 90%, 83.33% and 76.67% of the households from high agriculture woredas participating in SARDP, i.e., Gozamin, Awabal and Machakel respectively claim improved health condition, 73.33% and 76.67%of the households from Legambo and Wogedi woredas assert improvement in their health condition, respectively. More statistically significant difference in health condition was observed between Hulet lju Enessie and Dera, the two woredas participating in RWSEP. While 90% of the households from Hulet lju Enessie allege that their health condition has improved, it is only 40% of the households from Dera that confirm the same. Similar statistically significant results were observed between the two WIBS program woredas, Ebinat and Belessa. While 60% of the households from Ebinat respond positively, only 26. 7% of the households witness improved health condition.

60

Tewodros Bakaffa

Access to potable water supply and decreased prevalence of water-borne diseases were other crucial components taken into account in the analysis of livelihood condition of the woredas. Statistically high agriculture potential woredas participating in SARDP were found to be more benefited from the provision of clean water and relieved from the incidence of water-borne diseases than low agriculture potential woredas participating in the same program. Of the woredas under SARDP, Gozamin is showing outstanding results both in provision of potable water and low prevalence of water-borne diseases. 93.33% of the households from Gozamin woreda claim to benefit from provision of clean water and decreased water-borne diseases. Excluding Gozamin, the share of households from SARDP woredas, claiming safe water provision and lower incidence of water-born diseases ranges from 56.67% to 70% and from 60% to 80%, respectively. It is worth mentioning that while the difference in potable water supply between Gozamin on one hand and Machakel and Awabal woredas (which are all high agriculture potential woredas participating in SARDP) on the other is statistically significant, Gozamin also has a statistically significant difference in water-borne diseases with Machakel. Moreover, considerable difference in the share of households claiming provision of clean water supply and lower prevalence of water-borne diseases is observed within RWSEP and WIBS program woredas. While 96. 7% of the households from Hulet lju Enessie participating in RWSEP claim benefiting from provision of potable water supply and decrease in waterborne diseases, it is only 50% and 53.3% of the households from Dera woreda who report benefiting from the same, respectively. The difference in these aspects between the two woredas is statistically significant. Similar results were observed between the two WIBS woredas- Ebinat and Belessa. 50% and 70% of the households from Ebinat and only 23.3% of the households from Belessa claim access to potable water and decreased waterborne diseases, respectively. Generally speaking, although differences in incidence of water-borne diseases were not high differences iri access to and provision of potable water between high and low agriculture potential woredas participating within the various programs comes to be statistically significant. High · agriculture potential woredas are found in a better position than low agriculture potential woredas in terms of potable water supply. In addition, the difference SARDP high agriculture potential woredas have with SARDP low agriculture potential woredas in both potable water supply and prevalence rate of waterborne diseases is statistically significant. Moreover, the share of households claiming benefiting from potable water supply and decreased water bornediseases between Gozamin and other SARDP woredas, between Hulet lju Enessie and Dera participating in RWSEP and between the two WIBS program woredas- Ebinat and Belessa, appears to be statistically significant. The other crucial component of livelihood included in the survey was education. The number of households reporting sending their children to school shows significant difference between the high and low agriculture potential woredas. The difference in education among the high and low agriculture potential woredas

Decentralized development Cooperation

61

participating in SARDP is also found to be statistically significant. Hence, from among the woradas under SARDP households from higher high agriculture potential are found to have enjoyed better access and opportunity for schooling than households from lower agriculture potential woredas. Inter-program livelihood analysis has provided interesting insights on difference in livelihood outcome among the woredas that emanate from program specific factors. Woredas participating in Sida supported SARDP, are performing better than woredas participating in RWSEP, WISS and government program in terms of change in livelihood outcome they have registered in the years under consideration. The strong impact of program(s) has/have on livelihood becomes crystal clear when observing that the low-agriculture potential woredas under SARDP, Wogedi and Legambo, perform better than Dara, a high agriculture potential woreda participating in RWSEP and Dembecha, another high agriculture potential woreda taken as a control. In addition, these two low agriculture potential SARDP woredas are found performing far better than the two low agriculture potential woredas participating in WISS program - Ebinat and Selessa. Following woredas under Sida supported SARDP, woredas participating in Finnida supported RWSEP are found in a better position in terms of improved livelihood outcomes than UNICEF supported WISS program woredas. It is striking that the performance of woredas under WISS were not only found well below woredas participating in SARDP and RWSEP but also below the counterfactual government program woreda- Dembecha. Livelihood analysis made among woredas participating within same program (intra-program livelihood analysis) also provides interesting results. Observed significant differences in the livelihood outcome (condition) in general and in some of the critical components of livelihood such as income, health, potable water supply and prevalence of water-borne diseases, and education, is among the major findings this study revels. Marked differences in livelihood outcomes in general and income in particular, which is the most critical components of livelihood, become evident between the high and the low agriculture potential woredas participating in the same program. A case in point is the statistically significant difference income between Machakel, Awabal and Gozamin which are high agriculture potential and Legambo and Wogedi which are low agriculture potential woredas participating in SARDP. Study results suggest also the existence of one or more local conditioning factor(s) that attribute(s) to or explain(s) difference in livelihood and income among the woredas. Considerable differences in livelihood outcome improvement was observed, for instance, between Gozamin and other high agriculture potential woredas (Machakel and Awaba) participating in SARDP. Similarly, significant differences in various domains of livelihood were also observed among the woredas participating within RWSEP and WISS program. Considerable differences observed between the two high agriculture potential woredas

62

Tewodros Bakaffa

participating in RWSEP namely: Hulet lju Enessie and Dera attest this fact. Hulet lju Enessie found in better position than Dera woreda in terms of improved income, health, provision of potable water supply, and decreased prevalence of water-borne diseases. Though not as magnificent as SARDP and RWSEP woredas, variations were also common among low-agriculture potential woredas participating in WIBS program. Ebinat, generally speaking, is found to be in a better position in comparison to Belessa woreda in improved income, provision of health and potable water supply as well as in low incidence of water-borne diseases.

5.

Factors Attributing to Difference in Livelihood Outcome

The discussion above was focused on differences observed among the woreda in their agriculture potential. Other program and local conditioning factors responsible for such a wide difference in livelihood outcome have not been specified and illustrated yet. The purpose of this section, thus, is to identify and specify factors that led to such difference. Factors that contribute to livelihood outcomes differences among the woredas may generally be categorized as program specific and woreda (local conditioning) factors. Difference among the programs in capturing and reflecting community needs and preferences, responsiveness, financial management, tendering, contracting-out and procurement, monitoring and evaluation, training and backstopping managerial and technical services and rate and type of community participation of various stakeholders particularly the community are, among others, some of the program specific factors that stimulate differences in the livelihood of the community. Differences in the program factors such as planning and decision-making; problem identification, needs reflection and responsiveness; financial management; tendering, contracting-out and procurement; follow-up, monitoring and evaluation; training and backstopping technical and managerial services happened to be pervasive among the programs. Such dissimilarities in practices among the programs ultimately boil down to differences in local governanceparticipation, one of the critical dimensions. In regards to program factors taken into account SARDP followed by RWSEP have statistically significant difference with WIBS and government programs. There is also statistically significant difference between SARDP and RWSEP. SARDP followed by RWSEP are found in much better position compared to WIBS and government programs. These program-specific factors have prompted differences among the programs in terms of participation and sense of ownership of the programs by the community. Differences in reflection needs and preferences, in participation as well as in responding to and solving community problems, seems to affect the performance of programs and ultimately caused variations among programs in improving the livelihood outcome of the community. This in turn stimulates

Decentralized development Cooperation

63

differences in degree of satisfaction and happiness among the communities participating in various programs. Statistical analysis of household survey and secondary source of data provides strong woreda rank correlation coefficient between participation and improved livelihood outcome that validates how strongly associated participation and livelihood outcome are. The rank correlation between participation in terms of its specific variables and livelihood outcome based is over 90% and 83%, respectively. This suggests that participation and livelihood outcome go hand in hand and are strongly correlated. In addition to program-specific factors, difference in woreda (local conditioning) factors is another critical factor that could cause differences in the livelihood outcome among woredas. Among the various possible factors, enablement and system of governance aspects calls for further attention as they contribute to difference in livelihood outcomes among the woredas under various programs. Despite the fact that "as a political orientation, the government has launched decentralization that is assumed to devolve power and resources to lower government unit, thereby creating autonomy and smooth functioning of the government", decentralization does not seem to go deeper to the lowest units of government, the woredas. As Tegegne (2001: 22) notes that "the most prominent feature of decentralization in Ethiopia is the extensive power which regional governments are accorded", and woredas are provided with some limited powers which spillover from the extensive powers of regional states. Thus, "for these reasons, in fact decentralization in Ethiopia could be termed as mid level decentralization" (Ibid). Due to the limited power they have by their own, the woredas under the various programs were not in a position to formulate and exercise policies, rules and regulations that would create enabling environment for the community, CBOs, the private sector and NGOs to participate in development endeavor. This makes the woredas to be upward-looking and passive recipients of rules, regulations and guidelines from the regional state. Interviews made with the various development actors at different levels particularly at woreda level have made clear that there is no as such meaningful difference among the woredas in formulating and exercising their own rules, regulations and guidelines. What the household survey and discussions made with various development actors has actually been disclosed is that is that the woredas do differ in exercising and translating into to action the already given rules, regulations and guidelines by the regional state. Information acquired from various actors at different levels and from the household survey, however, reveal prevailing differences among woredas in terms of practices related to enhanced participatory development, heartening democratic election and leadership, responding to the felt needs and preferences of the community, in nurturing and fostering transparency and accountability to the public, as well as in efforts made to integrate these in the system of governance.

64

Tewodros Bakaffa

Against this backdrop, attempts were made to look into the various woreda governance factors that are likely to attribute to differences in the livelihood outcomes of woredas. The existing structure intends to ensure the participation of the communities in economic, social and political affairs that affect their lives by way of making governance and decision-making closer to the people. The existing guideline stipulates leaders of community-based structures should be elected by democratic mean. There were also strong claims by regional, zonal and woreda authorities on its implementation. However, information obtained through the household survey did not verify this claim to be true across the board. It, rather, uncovers that only in some woredas the leaders of community-based structures were elected democratically. Survey interviews were structured in a way that they inform the study whether community members have knowledge of the community-based structures and whether or not they know the leaders in person. The survey clearly demonstrates that the overwhelming majority of the population has knowledge of these structures and knows the leaders in person. Despite this fact, however, there is a statistically significant difference between woredas as to the degree of direct participation of the people in the process of delegating community leaders. In Gozamin while 53.3 % of the respondents state that community-based structure leaders were elected through the direct participation of the community another 46. 7% affirm that their participation in the election process was limited to approval of the nominees proposed by the higher authorities. The situation in other woredas is quite different. The share of community members who report direct participation in election of communitybases structure leaders range from 3.3% in Wogedi, Belessa and Ebinat Woredas to 16.7% in Hulet lju Enessie woredas. The claim for electing the leaders based on the proposals of higher officials ranges from 63.33% in Belessa to 90% in Machakel, Wogedi and Ebinat woredas. However, none of the respondents from Machakel, Gozamin and Legambo woredas confirm assignments or appointments of leaders of the community-based structures from the top, without the consent of the respective communities, while those argue for this from other woredas vary from 3.33% in Hulet lju to 33.33% in Belessa woreda. The feeling the communities have whether or not they have elected their leaders has far reaching implications in a community the development program. The higher the feeling of the community members that their leaders are elected by themselves, the higher the trust will be on the leadership and the higher the willingness of the members' to participate in economic, social and political affairs that affect their lives. This, in turn, will greatly contribute to betterment of the livelihood of the community. The difference among the woredas in the way their leaders are positioned is, therefore, one of the critical factors that prompt differences in magnitude of livelihood outcome improvement among the woredas. The household survey conducted on different dimensions of community participation unfolds some differences among the woredas in this respect. Majority of respondents from all the woredas witness the existence of discussion

Decentralized development Cooperation

65

forums whereby leaders discuss vital problems and issues of common interest with their respective constituencies. However, in this respect, statistically significant difference in degree and magnitude was also observed between Belessa and other woredas. While 50% of the community in Gozamin asserts high participation in the discussion forums, the participation rate for other woredas ranges from 0% in Belessa to 26. 7% in Wogedi woreda. Surprisingly, Dara was found to be in a better position than Belessa in terms of the rate of community participation in the discussion forums. Likewise statistically significant differences were observed between Gozamin and other woredas in terms of the rate and magnitude of public involvement in making- decisions on community development programs and activities. While 50% of the community in Gozamin claims active involvement in the discussion and decision-making processes related to community development programs and activities, community members respond positively for the same from other woredas vary from 0% in Belessa to 33.33% in Legambo and Wogedi woredas. Once again, the situation in Dera happened to be much better than Belessa. Generally speaking, community's involvement in discussions and decisionmaking processes related to development activities was found to be superior in Gozamin than the other woredas. Contrary to this community participation in Belessa and Dera woredas seems to be non-existent or in its infantile level. Highly associated and intricately linked to the above discussed woreda political environment factors, i.e., election in community leadership, participation and decision-making, are responsiveness, transparency and accountability. The difference among the woredas in transparency and accountability to the public as well as in responding to the needs and preferences of the community are likely among the key factors that stimulate difference in terms of improved livelihood outcome among the woredas under study. Asked as to the level to which community leaders were enable to identify and opt . for development plans that capture, reflect and address common needs and problems of the community, respondents from various woredas report differently. Statistically significant differences were observed between Gozamin and the other woredas. In this connection, while 50% of the community in Gozamin values the development programs high in terms of their ability of identification and reflection of community problems, responses in favor of development programs from other woredas range from 0% in Belessa to 26. 7% in Wogedi woreda. 86.7% of community members from Belessa assert that the performance of community leaders in this respect was low. What is surprising, rather, is Dera, the far less performing compared to other woredas, was found to be better performing compared to Belessa woreda. In connection to differences observed in terms of the effectiveness and accommodativeness of their development plans responses elicited from community members from Gozamin and the other woredas suggest statistically significant differences in terms of communities scaling of their respective leaders' ability for inventing practical responses that address the needs and problems of

66

Tewodros Bakaffa

the community. 36.7% and 56.7% of the households from Gozamin, rank their community leaders high and medium respectively for the vigilance they show in practically addressing needs and preferences of their respective constituencies. Belessa woreda was found to be the bottom in this aspect too. The rank correlation between livelihood outcomes based on household survey and secondary data on one hand and responsiveness in terms of its specific variables on the other is over 83% and 85%, respectively. In the same analogy, the higher a given woreda is placed in the ladder of responsiveness, the higher the likelihood that woreda would rank higher in terms of its livelihood outcome. For instance, like that of rank correlation between participation and livelihood outcome, Belessa and Gozamin which are found to be the lowest and highest ranking woredas in responsiveness and also found to be lowest and highest ranking in terms of livelihood outcome, respectively Among others, difference among the woredas in practical responses to the needs and preferences of the community is, therefore, one of the crucial factors that can attribute to and make difference in the livelihood outcome of the woredas. Intricately linked to the above discussed woreda political environment factors are transparency and accountability. The higher the level of transparency and accountability to the public, the higher the good governance is established, the higher the trust between leadership and the public is cemented, and the higher the participation of the community will be in the economic, social and political affairs that affect their lives. These are critical factors which ultimately engender to improvement in the livelihood of the community at large. Whether the public is informed by the leadership as to the resource available and the intended use of it is one feature of transparency and accountability on which an attempt was made to probe into. In this respect, thus, data from the household survey has brought to light interesting results. Except Legambo, Wogedi and Hulet lju Enessie woredas, there is statistically significant difference between Gozamin on one hand and the other woredas on the other hand. While 83.3% of the households from Gozamin assert that they are informed about the resource available and allocated for the development effort by the leadership, this condition for the other woredas rages from 23.3% in Belessa to 63.3% in Hulet lju Enessie. The fact that the public in Gozamin is better informed than the public in other woredas uncovers how the leadership in Gozamin is more transparent to the public than the leadership in the other woredas under review. The least performing in this respect are Belessa and Dera. There is statistically significant difference between Belessa on one hand and Legambo, Gozamin, Wogedi, Hulet lju Enessie and Ebinat woredas. The former woredas were found to be better placed in their level of transparency to the public than Belessa. Excluding Ebinat, there is also statistically significant difference between Dera woreda on one hand and Legambo, Gozamin, Wogedi and Hulet lju Enessie woredas on the other. What is interesting is that there is statistically significant difference in this factor between Gozamin on one hand and Awabal and Machakel on the other which are all high agriculture potential woredas participating in SARDP. Likewise, there is statistically significant difference

Decentralized development Cooperation

67

between Hulet lju Enessie and Dera and between Ebinat and Belessa woredas which are participating in RWSEP and WIBS programs, respectively. Whether or not the community scrutinizes the performance of community leaders is the other area on which an attempt has been made to look into in relation to one of the critical factors of governance, accountability to the public. Along this line, the household survey unfolds interesting results. Aside from Legambo, Wogedi and Hulet lju Enessie, there is statistically significant difference between Gozamin and the rest of the woredas under consideration. Likewise· the transparency, while the share of households claiming inquiry of the performances of community leaders by the community in Gozamin accounts for 83.3% of the households, this condition for the other woredas ranges from 26.67% of the household in Belessa to 63.3% of the household in Legambo woreda. Belessa is again found to be the least performing in terms of poring over the performances of community leaders by the community. Except Machakel, Awabal and Dera, there is statistically significant difference between Belessa and the other woredas in this respect. Following Belessa, Dera is observed to be the next poor performing woredas. The difference between Dera on one hand and Legambo, Gozamin, Wogedi and Hulet lju Enessie woredas on the other is statistically significant in terms of putting community leaders under scrutiny by the community for their actions and performances. Interestingly, like that of the transparency factor there is also statistically significant difference in putting community leaders under scrutiny for their actions and performances between the woredas which are characterized by similar agriculture potential and participating in the same program. To this effect, the difference between Gozamin on one hand and Machakel and Awabal on the other which are both high agriculture potential woredas and participating in ·sARDP is statistically significant. In the same manner, the difference between the two high agriculture potential woredas participating in RWSEP, Hulet lju Enessie and Dera and between the two low agriculture potential woredas participating in WIBS program, Ebinat and Belessa is found to be statistically significant. While Gozamin is found to be better performing than Machakel and Awabal, Hulet lju Enessie, and Ebinat, respectively is found in a better position than Dera and Belessa. There is a positive and significant woreda rank correlation between the two specific variables operationalzing transparency and accountability: provision of information to the community and scrutiny of community leaders by the community with livelihood outcome. This clearly shows the higher the ranking of woreda in transparency and accountability, the higher the ranking of woreda is in terms of livelihood outcome.

68

Tewodros Bakaffa

6. Theory and Empirical Findings: A Synthesis Decades of development experience, which was essentially centralized and donor-driven, resulted in failure particularly in terms of bringing about meaningful improvement in the livelihoods of the community. Besides the paternalistic and donor-driven approach to development, the heavy emphasis given to economic growth with little attention provided to other critical dimensions of development contributed to this frustrating experience. This condition paved the way for the nurturing and fostering of a new development thinking and approach which, essentially, bases itself on the principles of peoples-centered development, decentralization of authority to local development actors, creation and strengthening of good governance, promotion of and respect for local development initiatives and participatory development. This development approach, besides paying due and appropriate attention to economic growth, extends its spectrum as to encompass wide range of critical components of development. This, essentially, is livelihood approach to development. For instance, though there is a general tendency to explain poverty in terms of economic variables, it cannot, however, be fully captured and explained by economic variables per se. "Poverty is much more than income or consumption . . . . longevity, access to health and education facilities, and perhaps security of consumption levels from extreme shocks' are equally relevant in the analysis of poverty" (Chambers, 1995:7). Pronk, in agreement with this view, said: "Poverty cannot be captured in terms of money and income alone. If poverty is seen as a lack of opportunity to acquire lasting control of resources in order to strengthen one's capacity to acquire basic necessities of life .... that requires more than money, more than income" (Pronk, 2003: 10). Thus, though the necessity and imperatives of economic growth for development is not disputable, economic growth per se does not suffice to explain development. For instance, livelihood or human development do notions constitute wide-range dimensions of development than mere economic growth. "Human development consists primarily in the strengthening of fundamental sectors, i.e. education and health and to increase employment. This entails an investment in people, to enable them to participate in society, to improve their productivity and to increase their 'space' within society, thus leading to further participation within and democratization of the society to which they belong. Human development must aim at development of, for and by people" (Pronk, 1991: 13). Moreover, "Livelihood comprises the capabilities, assets (stores, resources, claims and access) and activities required for a means of living" (Chambers and Conway, 1992:7). In addition to assets, which are critical component of livelihood, capabilities being the other important components of livelihood are "both an end and a means of livelihood: a livelihood provides the support for the enhancement and exercise of capabilities (an end); and capabilities (a means) enable a livelihood to be gained" (lbid:6). Education and health, in particular, are among the crucial components of capabilities. (Ibid) It is against this reflection an attempt is made to see whether or not there have been improvements in the livelihoods of the community living in different woredas and participating in the same or different programs. Moreover, there has been an attempt to probe into and identify the factors that attribute to such difference in

Decentralized development Cooperation

69

the livelihood outcome of the communities. Given the complexity and difficulty of analyzing livelihood in an exhaustive manner this research probes into some of the major dimensions and components of livelihood. "Exhaustive analysis may not be appropriate in all cases. optimal ignorance must always be applied" (Scoones, 1998: 13) Agriculture being the mainstay of rural livelihood of the population of Amhara Region, it is of critical importance to give due emphasis to agriculture in the analysis of rural livelihoods. Thus, crop production and livestock husbandry- the two key activities on which rural livelihood depends on in the region, are taken into account in this research by way of the exploration and analysis of rural livelihoods of the community in the woredas under consideration. Moreover, provision of education, health and potable water which, by and large, are related to capabilities dimension of livelihood were also paid due consideration. Taking into account these components of livelihood unfolds that there have been differences in improvements of the livelihood among communities in different woredas participating in same or different programs in the years under consideration. Household survey and secondary source data suggest more or less similar in terms of improved livelihoods of the community. The woreda rank correlation of 96% in terms of livelihood outcome based the primary and secondary sources of data clearly show how the two sources of data are compatible. Differences among woredas in their agriculture potential, among others, are found to be one of the critical factors attributing to differences in the livelihood outcome. Generally speaking, high agriculture potential woredas are found to be better performing than low agriculture potential woredas in terms of improved livelihood outcome. Such difference among the woredas is one of the basic reasons that calls for the need to pay due attention to local development. The diversities and differences among localities such as economic situation, culture, religion, ethnicity, etc., (Singh, 1999) could influence the performance of development programs including livelihood outcome (Romoe, 1996). This general premise is asserted by and found to be valid by the present empirical study since difference among the woredas in their agriculture potential is found to be fundamental in attributing to difference in the livelihood outcome of woredas. In this relation, Scoones (1998: 11) states that "Livelihood ... can be expected to differ along ... natural resource endowment gradients. . .. This variation in resource .... has implication for development... in different areas." However, findings of the present research also makes clear that difference in agriculture potential is not the only factor accounting for difference in the livelihood outcome of woredas. What has become evident is that there is difference in livelihood outcome among the woredas participating in same program but having more or less similar agriculture potential. This suggests the intervention of other factors related to woreda (local) conditioning factors that attribute to such difference in the livelihood outcome of woredas in the programs under review.

70

Tewodros Bakaffa

Apart from decentralized development cooperation, among others, is about (1) promoting the active participation of the community in particular and local development actors in general, (2) supporting development initiatives coming from the grassroots levels, (3) ensuring better development by taking care of and responding to the needs and preferences of various development actors at the grassroots levels particularly the community; experience witnesses that the implementation of decentralized development cooperation programs is not that easy and smooth. Rather it is fraught with myriad of challenges and problems (Hertogs, 1999). The threat of central governments that they lose control of aid resources, the entrenched centrist attitude of some of the donor agencies, lack of experience of donor agencies in dealing with multiple local development actors, etc., are some of the challenges encountering decentralized development cooperation programs (Bossuyt, 1995). Though operating in line with locally-owned strategies, respecting and promoting local initiatives and participation are the basic tenets of decentralized development cooperation; in reality some donors are still entrenched with their centrist attitude and unable to do away with their rigid systems and procedures of development cooperation which essentially violates the basic principles of decentralized development cooperation and undermines local initiatives and ownership of programs (Bossuyt, 1995; Regumamu, 1997; OECD, 1996; OECD, 1998). The difference among donor agencies in this respect is, therefore, is one of the fundamental factors attributing to difference in the effectiveness and impact of the programs on the livelihood outcome of the community. This view has been found valid in the present empirical study. Taking into account program-specific factors such as: (1) problem identification and needs reflection, (2) planning and decision-making, (3) financial management, (4) tendering, contracting and procurement, (5) monitoring and evaluation, and (6) training and backstopping technical and managerial services, an attempt was made to explore whether or not there were differences among programs and/or differences in the regulations and procedures of the donor agencies supporting the respective programs. To this effect, the present study brings to light that UNICEF supported WIBS program, generally speaking, is found to be more centralized and donor-driven than Finnida supported RWSEP and Sida supported SARDP. Some of the program-specific factors also revel RWSEP is more centralized and donor-driven than SARDP. Such differences in regulations and procedures among the programs supported by different donor agencies, among others, boil-down to programs variations in terms of rate and type of participation of various development actors par,ticularly of the principal stakeholders, the community. Empirical data drown from field also attested this view. Elements of participation taken into account (community knowledge of the programs, rate and type of community participation), unfold that there were considerable differences among· SARDP, RWSEP, WIBS and government programs in this respect. The comparison made between SARDP on one hand and RWSEP, WISS, and government programs in terms of these domains of participation is statistically significant differences. Moreover, difference in these elements of participation between RWSEP on one hand and

Decentralized development Cooperation

71

WIBS and government programs on the other were also statistically significant. Such a significant difference among the programs as to the knowledge of the community about the programs, the rate and type of community participation are some of the critical factors, among others, attributing to difference in the livelihood outcome of the community. Communities participating in SARDP and RWSEP are found to have better knowledge of the programs, better degree of participation and higher level of involvement in priority setting and decisionmaking processes than the communities participating in WIBS and government programs. This is in line with the view which sees participation as a process that nurtures and fosters along a continuum that may start from lowest level of participation - manipulation to the highest level- self management. According to this view, the other levels in between manipulation and self-management include: information, consultation, consensus building, decision-making and partnership (World Bank, 1996). However, it is worth mentioning that it is hard to draw welldefined and clear boundary among the different levels of participation along the continuum and to make detail specification as to what they constitute. Thus, taking into consideration the overall trend and broader components of participation, a modest, but, safe conclusion could be drawn that SARDP and RWSEP are found in a better level along the continuum of participation than WIBS and government programs whereby the community under the programs witness no or little knowledge (information) of the programs which is at the lowerend-levels along the continuum of participation. Whether or not and the extent to which various stakeholders at different levels participate in the programs under consideration is the other crucial indicator of participation that has been dealt with. It is understood that difference among the programs is not only limited to the participation of the community but also of the participation of other stakeholders at different levels. Though the participation of primary stakeholders, the community is of critical importance, it is also imperative that other stakeholders at different levels participate in the programs at different capacities if the intended objectives of attaining improved livelihood of the community in particular and development at large are to be achieved. Accordingly, the present empirical study unfolds that various stakeholders from SARDP and RWSEP (at different levels, capacities and decision-making powers) participate better starting from the woreda to the regional levels in than WIBS and government programs. Attempts that have been made by SARDP and RWSEP to reflect the aspirations and interests of various stakeholders at different levels including the community by way of involving them in program coordination and management and by creating various fora were promising. This is not the case when it comes to WIBS program. Except the limited involvement of the program woredas' administrators and BoPED at regional level the participation of other stakeholders including the community in WIBS program was non-existent. The donor- UNICEF and the Federal Government authority, MEDaC, are found to be the key actors both in the planning and decision-making processes of the program. Though much is said about the critical importance of participation of the community, woreda development actors and other stakeholders at different levels for the successful program outcome, this was not found to be surpassing beyond rhetoric. In the case of government sponsored woreda development program, the participation of both the community and other woreda development actors is

72

Tewodros Bakaffa

constrained by the limited authority provided to the woredas. The woredas under various development programs were not authorized to exercise formulating their own enabling policies and regulations that promote participation of various actors in the woreda development. Limited expenditure authority and high fiscal vertical imbalance, among others, are some of the factors hindering the performance of government sponsored woreda development programs. Much of the power in relation to planning and decision-making of government development program remains in the hands of regional and zonal level authorities. To this effect, the participation of the community in government development programs has been found to be essentially limited to labor and material contribution. Such differences in establishing effective system that facilitate optimal opportunities for the involvement of various development actors at different levels is one of the crucial dimensions of participation that attributes to difference in the livelihood outcomes among the programs. This finding is in line with the notion of 'participatory development', which heavily pronounces the vital importance of partaking of various stakeholders at different levels over and above, if the livelihood of the people is to be improved in particular, and sustainable development is to be realized in general (OECD, 1995). The populist view, 'popular participation, albeit important, is not sufficient to bring about improved livelihood and sustainable development in the community since other development actors who have strong hold in the political, economic and social arena are likely to obliterate the whole development effort if their voices are not heard and their interests are not captured and reflected by the programs (OECD, 1995; World Bank, 1996; UNDP, 1999). Such a difference in various dimensions of participation, therefore, is one of the critical factors that attributes to differences observed among woredas under various programs in livelihood outcome. The woreda rank correlation between participation and livelihood outcome based on secondary data and household survey which is over 83% and 90%, respectively clearly indicates that participation and livelihood outcome go hand in hand. The difference between SARDP on one hand and RWSEP and WISS program on the other in their livelihood outcome is statistically significant. There is also statistically significant difference between RWSEP and WISS program in their impact on livelihood outcome. SARDP followed by RWSEP are found to be better performing in terms of livelihood outcome than WISS program. It can, thus, be concluded that difference program-specific factors attribute to difference in livelihood outcome through difference in various dimension of participation of stakeholders. To conclude, the more decentralized, recipient-driven and locally-owned a given program is, the better will be community participation which in turn results in better livelihood outcome. In the same analogy, the more centralized and donordriven a given program is the lower, will be community participation in the program which in turn results in lower livelihood outcome. Besides difference in agriculture potential (economic factor), difference in enablement and practices of governance among the woredas are hypothesized to be the other woreda (local) conditioning factors that have stimulated differences in the livelihood outcome of the woredas under consideration.

Decentralized development Cooperation

73

Though complex and hard to discern its impact on livelihood, "decentralization does seem to have a positive impact on livelihood (Goldman, 1998: 48) To this end, in a condition where significant power and authority is entrusted to local governments (woreda enablement), the prospect of local governments to formulate and exercise their own rules and regulations, that are meant to promote the participation of various local development actors (the community, CBOs, private sector, etc.), in the local development endeavor would be promising and higher. Such a condition contributes to improvement in the livelihood of the community in particular and various stakeholders at large. What the present study has come across, however, is that the powers and authorities the woredas were endowed with in formulating and implementing of enabling policies and strategies were inadequate to stimulate any meaningful improvement in livelihood outcome. Apparently, the woredas were found to be passive recipients of instructions, guidelines, rules and regulations from the respective regional state rather than being active agents of change in matters decisive for their own livelihood. Therefore, differences in existing practices related to woreda enablement were not found to be one of crucial factors accounting for difference among the woredas in their livelihood outcome. Differences in system of governance were observed among the woredas. Improved livelihood of the community and sustainable human development can be attained, among others, through the prevalence of good governance. Sustainable (human) development and improved livelihood of the community can be achieved, among others, where good governance is which firmly stands on the principles of responsiveness, and transparency and accountability, prevailing (Edward and Hulme, 1997; UNDP, 1998). It is in this relation, as Pronk 1991: 18) observes that in "both economics and politics" currently, "[t]he emphasis has now shifted to good, participatory governance, accountability, democracy and institutional pluralism". Against this reflection, it is found that there is statistically strong and significant rank correlation between factors explaining good governance and the state of livelihood outcomes of woredas under study. Based on household survey and secondary data woreda rank correlation between responsiveness and livelihood outcome is above 83% and 85%, respectively. This clearly shows how positively and strongly responsiveness is related to livelihood outcome. In a similar vain, rank correlation between transparency and accountability, and livelihood outcome is found to be positive and statistically significant. The woreda rank correlation between livelihood outcome on one hand and provision of information (transparency) on the other is slightly more than 46% and 47%, respectively. Besides, woreda rank correlation between livelihood outcome on one hand and scrutiny of community leaders by the community (accountability) on the other is above 44% and 41 %, respectively. Such a positive and statistically significant woreda rank correlation of responsiveness and transparency and accountability with livelihood outcome validate strong positive correlation of good

74

Tewodros Bakaffa

woreda governance to livelihood outcome. Thus, differences in governance were found to be one of the fundamental factors attributing to difference among the woredas in terms of achieved livelihood outcome improvement.

To conclude, differences in agriculture potential (economic factors) and system of governance, more specifically differences in community participation, responsiveness and accountability and transparency among the woredas are found to be the fundamental factors attributing to differences in earned livelihood outcomes of the woredas under consideration. This clearly shows that the findings of the present study are in line with the theoretical and conceptual framework that guides this research.

References Amhara National Regional State (ANRS). (1994). The Rural Water Supply and Environmental Programme: Programme Document (1994-1998). _ _ _ _ _ _ _ _ .(1995a). Amhara National Region Five Year Programme on Development, Peace and Democracy (Amharic Version), Bahir Dar: Ethiopia. _ _ _ _ _ _ _ _ .(1995b) "A proclamation to Provide for the Establishment of the st Constitution of the Am hara National Region", Zikre Hig, Bahir Dar, 1 year, No.2, Bahir Dar: Ethiopia. _ _ _ _ _ _ _ _ .(1996). ANRS and Sida Cooperation in Rural Development: Programme Document, Bahir Dar, Ethiopia _ _ _ _ _ _ _ _ .(1997) Amhara National Regional State (ANRS) and Sida Cooperation in Rural Development: Plan of Operation and Bugdget 1997-1999, Bahir Dar: Ethiopia. _ _ _ _ _ _ _ _ .(1998a). ANRS and Sida Cooperation in Rural Development: Physical and Financial Report (1 June 1997 - 30 June 1998), Bahir Dar: Ethiopia. _ _ _ _ _ _ _ _ .(1998b). ANRS and Sida Cooperation in Rural Development: 1998/99 Plan of Operation and Budget, Bahir Dar: Ethiopia. _ _ _ _ _ _ _ _ .(1999). ANRS and Sida Cooperation in Rural Development: Physical and Financial Report, First Quarter (1 July - 30 September 1998), Bahir Dar: Ethiopia. Bebbington, A. (1999). "Capitals and Capabilities: A Framework for Analyzing Peasant Viability, Rural Livelihoods and Poverty" in World Development, 27/12. PP. 20212044, Elsevier Science Ltd. Bossuyt, J. (1994). Decentralized Cooperation and the African Public Sector: Several players in search of a dramatist, ECDPM. _ _ _ _ _ _ _ . (1995). Decentralized cooperation: Lessons from EU experiences under the Lame convention, Maastricht: ECDPM. Bureau of Planning and Economic Development (BoPED) (December 1998). Statistical Bulletin for the Year 1993/94 - 1996/97), Bahir Dar, Ethiopia. _ _ _ _ _ _ _ _ . (1999a). Atlas of the Am hara National Region, Bahir Dar: Ethiopia. _ _ _ _ _ _ _ _ . (1999b). Statistical Bulletin, Bahir Dar: Ethiopia. _ _ _ _ _ _ _ _ . (1999c). Estimation of Regional Gross Domestic Product of the Amhara National Region: Aggregate Report of Economy and its Structure (1993/94-1996/97). Bahir Dar: Ethiopia. _ _ _ _ _ _ _ _ . (1999d). The 1994 Population and Housing Census of Ethiopia: Results for Amhara Region Vol. I Analytical Report, Bahir Dar, Ethiopia. Burgess, R., M. Carmona and T. Kolstee. (1994). Position Paper on Urban Strategies and Urban Design, prepared for the International Seminar The Hidden Assignment, Delft: Universiteitsdrukkerij.

Decentralized development Cooperation

75

_ _ _ _ _ _ _ _ (eds.). (1997). The Challenge of Sustainable Cities, Neoliberalism and Urban Strategies in Developing Countries. London: Zed Books. Central Statistical Authority (CSA). (1998). The 1994 Population and Housing Census of Ethiopia: Results at country level, Addis Ababa. _ _ _ _ _ _ _ _ _ . (1991). Putting People First: Sociological Variables in Rural Development, New York: Oxford University Press, The World Bank. Chambers, R. (1983). Rural Development: Putting the Last First, Longmans: London. _ _ _ _ _ _ _ _ _ . (1987). Sustainable Rural Livelihoods: A Strategy for People, Environment and Development, Brighton, Sussex: Institute of Development Studies (IDS). (1988). Sustainable Livelihoods, Environment and Development: Putting Poor Rural People First, IDS Discussion Paper, Brighton, Sussex: Institute of Development Studies (IDS). _ _ _ _ _ _ _ _ _ . (1995). Poverty and Livelihoods: Whose Reality Counts?, IDS Discussion Paper, Brighton, Sussex: Institute of Development Studies (IDS). Chambers, R. and Conway, G. (1992). Sustainable Rural Livelihoods: Practical Concepts for the 21 st Century, IDS Discussion Paper, Brighton, Sussex: Institute of Development Studies (IDS). ECDPM. (1993). Democratization in Sub-Saharan Africa: The Search for Institutional Renewal. Maastricht: ECDPM _ _ _ _ _ _ _ _ . (1994). A Governance Approach to Civil Service Reform: Operational Considerations, Maastricht. _ _ _ _ _ _ _ _ . (1995). Decentralized Cooperation in Africa: Finding A New Balance. _ _ _ _ _ _ _ _ . (1996). Decentralization: Acid Test for NGOs. Economic Commission for Africa (ECA). (1999). Progress in Good Governance Since the Summit, (ECA): East and Southern Africa Sub regional Follow-up Conference to the World Social Summit, 15-17 March, 1999. Nairobi: Kenya Hertogs, E.J. (1999). What Role for Local Authorities in Decentralized Cooperation under Convention of Lame, ECDPM Discussion Paper No.8, Maastricht: ECDPM. Ministry of Economic Development and Cooperation (MEDaC). (1993). An Economic Development Strategy for Ethiopia: A Comprehensive Guidance and a Development Strategy for the Future, Addis Ababa _ _ _ _ _ _ _ _ . (1994). Capacity Building for Regional Administrations in Ethiopia, Strategic Action Plans (Final draft), Addis Ababa. _ _ _ _ _ _ _ _ . (1995). Inter-regional Data on Some Aspects of Economic and Infrastructure Conditions in Ethiopia, Addis Ababa. _ _ _ _ _ _ _ _ . (1996). Federal Democratic Republic of Ethiopia's Five Year Program on Development, Peace and Democracy (Abridged unofficial translation), Addis Ababa. _ _ _ _ _ _ _ _ . (1997). Criteria for Allocating Subsidies to the Regional Governments from the Federal Government, Amharic Version, Addis Ababa. _ _ _ _ _ _ _ _ . (1997). Federal Democratic Republic of Ethiopia and UNICEF Program of Cooperation (1994-1999) Amhara National Regional State Mid-term Review, Addis Ababa. _ _ _ _ _ _ _ _ _ . (1998). Regional Development in Ethiopia (Overview), Addis Ababa, _ _ _ _ _ _ _ _ _ . (1999). Woreda. Integrated Basic Services (WIBS) Program Plan of Operations (July 1999-December 2001), Addis Ababa. _ _ _ _ _ _ _ _ _ . (1999). Povery Situation in Ethiopia, Addis Ababa. Morgan, P. (1997). The Design and Use of Capacity Development Indicators, CIDA OECD. (1993). Good Governance and Development Assistance: A Background Paper, Paris. _ _ _ _ _ _ _ _ _ . (1995). Participatory Development and Good Governance, Paris

76

Tewodros Bakaffa st

(1996a). Shaping the 21 Century: The Contribution of Development Cooperation, Paris _ _ _ _ _ _ _ _ _ . (1996b). DAG Seminar on Technical Cooperation and Capacity Development: Synthesis Reporl, Paris _ _ _ _ _ _ _ _ _ . (1997a). Evaluation of Programs Promoting Parlicipatory Development and Good Governance: Synthesis Reporl, Paris _ _ _ _ _ _ _ _ _ . (1997b). Final Reporl of the Ad hoe Working Group on Parlicipatory Development and Good Governance, Part I, Paris _ _ _ _ _ _ _ _ _ . (1997c). Final Reporl of the Ad hoe Working Group on Parlicipatory Development and Good Governance, Part 11, Paris _ _ _ _ _ _ _ _ _ . (1998). DAG Informal Network on Institutional and Capacity Development, Paris Olowu, D. (1993). African Local Governments as Instruments of Economic and Social Development, The International Union of Local Authorities (IULA), The Hague, The Netherlands. _ _ _ _ _ _ _ _ _ . ( 1998). African Local Governments as Instruments of Economic and Social Development, The Hague: The International Union of Local Authorities - - - - , - - - - - - · (1999). Governance Indicators for African Smar States of the st 21 Century, ECA Romeo, Leonardo. (1996). Promoting Decentralized Planning, UNCDF Rugumamu, S.M. (1997). The Green Paper: Myths, Facts and Neglected Details, ECDPM Working Paper No. 24, Maastricht, ECDPM SIDA. (1993). Shifting the Balance? Towards Sustainable Local Government, Decentralisation and District Development In Botswana, Stockholm. Singh, Kashi N. (1999). Locality and Local Economic Development, Working Paper No.3, Regional and Local Development Studies, Addis Ababa University, Addis Ababa. Tegegne Gebre Egziabher. (1998). ''The Influence of Decentralization on Some Aspects of Local and Regional Development Planning in Ethiopia" in East African Social Science Research Review, Vol. XIV, pp.32-63. _ _ _ _ _ _ _ _ _ .(2001). Institutional setting for local development planning in Ethiopia: An assessment and evidences from Amhara region, P.R.F. Series No 37, Institute of Development Economics: Japan External Trade Organization. The Transitional Government of Ethiopia (TGE). (1991 ). Ethiopia's Economic Policy During the Transition Period, Addis Ababa. _ _ _ _ _ _ _ _ _ . (1992). Proclamation No.7/1992, A Proclamation to Provide for the Establishment of National Regional Self-government, Addis Ababa _ _ _ _ _ _ _ _ _ . (1992). Proclamation No.33/1992, A Proclamation of Revenue Sharing Between Federal and Regional Governments, Addis Ababa UNDP. (1990). Foreign Aid Accountability: Perspectives of Donors and Recipients (1993). Human Development Reporl 1993: Oxford: Oxford University Press (1995). Public Sector Management, Governance and Sustainable Human development. _ _ _ _ _ _ _ _ _ . (1995). Sustainable Human Development, From Concept to Operation: A Guide for the Practitioner. _ _ _ _ _ _ _ _ _ . (1997a). Capacity Development, Technical Advisory Paper 2 _ _ _ _ _ _ _ _ _ . (1997b). Governance for Sustainable Human Development: A UNDP Policy Document _ _ _ _ _ _ _ _ _ . (1997c). "Civil Society Organization Dialogue" in International Conference on Governance for Sustainable Growth and Equity, United Nations, 28-30 July 1997, http://magnet.undp.org/icg97/part5.htm _ _ _ _ _ _ _ _ _ . (1997d). Capacity Development for Sustainable Human Development: Conceptual and Operational Signposts

Decentralized development Cooperation

77

_ _ _ _ _ _ _ _ _ . (1997e). Sustainable Livelihoods: Concepts, principles and Approaches to indicator development _ _ _ _ _ _ _ _ _ . (1998a). Capacity Assessment and Development in a System and Strategic Management Context, Technical Advisory Paper No.3, _ _ _ _ _ _ _ _ _ . (1998b). Decentralized Governance Country Thematic Assessment: Framework and Guidelines _ _ _ _ _ _ _ _ _ . (1999a). UNDP's Guidebook of participation _ _ _ _ _ _ _ _ _ . ( 1999b). Governance for Sustainable Livelihoods: Operational Issues USAID. (1992). Concept of Governance and its Implications for A/D's Development Assistance Programme in Africa World Bank. (1992). Governance and development. Washington DC: World Bank _ _ _ _ _ _ _ _ _ . (1996). World Bank Participation Sourcebook, Washington DC _ _ _ _ _ _ _ _ _ . (1997). "Bringing the State Closer to the People" in World Development Report 1997, Oxford University Press, Inc. New York. Wunch, James S. and Olowu, Dele. (1995). The Failure of Centralized State: Institutions and Self Governance in Africa, Institute of Contemporary Studies, USA Young, John. (1998). "Regionalism and Democracy in Ethiopia" Third World Quarterly, Vol. 19, No.2., PP. 191-204

Tewodros Bakaffa

78

TABLE 1

IMPROVEMENT LIVELIHOOD OUTCOMES I

Woreda Machakel Awabal Gozamin Hulet lju Enessie Legambo Wogedi Belessa Ebinat Dera Dembecha

Total

N C O M

INCOME TOTAL (60%)

CROP TOTAL (40%)

L I V E LIVESTOCK TOTAL (20%)

7.1185 8.9185 13.6898 5.1404 5.4781 5.3644 1.2298 1.9914 5.4809 5.5881 60

4.0749 6.1952 10.0683 2.9403 3.4605 3.3476 0.8942 1.4475 3.7385 3.8330 40

3.0436 2.7233 3.6215 2.2001 2.0177 2.0168 0.3356 0.5440 1.7424 1.7551 20

s

T

E

SOCIAL SERVICES

C~ttle

0 C K Sheep and Goat

Equine

{8%)

(6%)

(2%)

1.1608 1.0273 1.0181 0.8293 1.3344 1.1764 0.8465 0.7388 0.8746 0.6078 0.9454 0.5535 0.1045 0.0002 0.2152 0.0819 0.7776 0.5215 0.7231 0.4632 8 6

0.4439 0.2692 0.5058 0.1964 0.0904 0.0978 0.0832 0.1059 0.0712 0.1363 2

Poultry (4%) 0.4116 0.6066 0.6049 0.4184 0.4449 0.4201 0.1477 0.1409 0.3722 0.4326

4

Social services TQtal

Health

Water (10%)

(40%)

{iO%)

4}369 4.7541 5.~492 5.7329 3.6218 4.8984 1.5213 3.1290 3.4902 2.1663 40

3.2002 0.6182 2.8862 0.7100 3.8935 q.8631 2.6729 1.8109 1.1828 1.4771 2.8420 1.0289 0.1359 0.6112 1.3486 0.7325 0.9006 1.9972 0.9372 0.1510 20 10

Education

GRAND TOTAL

(10%)

(100%)

0.9186 1.1579 1.1926 1.2491 0.9620 1.0274 ~.7742 1.0478 0.5924 1.0781 10

11.8555 13.6725 19.6390 10.8733 9.0999 10.2628 _ 2.7511 5.1204 8.9711 7.7545 100

RANK 3 2 1 4 6 5

10 9

7 8

PART TWO

URBAN SERVICE DELIVERY

New Public Management Paradigm

79

5 New Public Management Paradigm: Lessons for Africa from the Delivery of Basic Services in Ghanaian Cities Nicholas Awortwi 1

1. Introduction In the 1980s African cities were described as being 'in crisis' following unprecedented urbanization of poverty, increasing failure of governments to provide public services, the virtual collapse of local institutions, persistent public sector mismanagement, inability of the cities to attract new investments, lack of maintenance of existing public infrastructure, and general deterioration of the living environment (Stren and White, 1989; Amin and Lloyd, 1990). 2 Since inadequate provision of public infrastructure services was one of the most visible features of the crisis, their continuous decline attracted the most attention. Although economic downturn and rapid urbanization also contributed to the crisis, one factor that attracted major criticism was the fact that in many African countries, only the government was responsible for service provision. This was based on the premise that the ability of a government to legitimately tax and govern people depends on its capacity to deliver a range of services required by the population, services that no other player will provide (Olowu, 1999: 123). The public sector provided services with its labour, materials and equipment. Apart from providing physical infrastructure, politicians used the sector to fulfil multiple objectives, including creating jobs, satisfying political expediency, and delivering a wide array of non-commercialized services and patronage. Many analysts argue that without the discipline of competition and focus on objectives, public bureaux workers had little incentive to innovate, control costs and deliver services effectively and efficiently. Even where governments found the resources to operate services; delivery of the services was poor and large segments of the population were left out. The best way to improve, according to these analysts, was to change the way in which public infrastructure and services were administered in Africa. For those on the ideological right, privatization was the only solution. For instance, according to the World Bank, the key task is to "manage infrastructure like a business, and not like a bureaucracy" (World Bank, 1994:2), which means the private sector can do a better job. Private sector participation in public service provision is expected to foster government downsizing, reform governmental management and functions, and improve delivery. Nee-liberals call for 'reinventing government' by inspiring· a new entrepreneurial spirit (Hammer and Champy, 1993; Osborne and Gaebler, 1992; Barzelay, 1992). Osborne and Gaebler (1993:xvii) demonstrate in their ten 1 The Author is a postdoctoral fellow at the Institute of Social Studies, The Hague, The Netherlands. This article is an extract from his PhD thesis entitled 'Getting the Fundamentals Wrong: Governance of Multiple Modalities of Basic Services Delivery in three Ghanaian Cities.' Correspondence: P.O. Box 29776, 2502 LT, The Hague, The Netherlands, Email: [email protected]. 2 These authors vividly depict the crisis situation in African cities by tracing the decline of urban resources over the critical years of the 1980s and 1990s. They show how the problem was compounded by rapid urbanization so that per capita levels of urban expenditure on services and infrastructure fell dramatically.

80

Nicholas Awortwi

principles for entrepreneur government that the basic problem is neither too much nor too little government, but rather the wrong kind of government. Many of these analysts proposed to shrink government, remove it from as many infrastructure programmes as possible, and turn the functions over to the private sector. Programmes that have to remain in the government's hands should be contracted out, they argue (Farnham and Horton, 1993; Starks, 1991; Diiulio Jr, 1993; Savas, 1982). The consensus among many of these analysts is that the market would make government infrastructure provision self-correcting in three ways: efficiency, cost minimization and quality delivery. Some of these market assumptions have become a very powerful ideological belief within the Right, to the extent that Kettl (1993) argues that, in the United States, this presumed superiority of the private sector had led to contracting out everything possible with relatively less concern about how to structure and manage it. Many of the persuasive arguments and ideologies have influenced the way many African states have administered infrastructure services re9ently. This includes acknowledging the vital role of other actors (formal and informal) and subsequently decentralizing responsibilities to them. This practice has further been bolstered by advances in technology, which have made possible 'unbundling' of the components in the service delivery process to several institutions.

2. New Public Management Paradigm According to the World Bank, the public sector lies at the core of the stagnation and decline in growth in Africa. The sector is considered to be over-extended, bloated, inefficient, unproductive and incapable of delivering efficient services (World Bank, 1981, 1989).The solution, according to the World Bank, lies in taking a market-friendly approach to development, in which public sector activities are transferred to the private sector through a process of privatization and corporatization. The argument for giving public infrastructure services a commercial orientation is that it ensures efficiency. This is to be achieved through three major means: management incentives, competitive forces and firm objectives. According to the World (1995a: 12) "experience shows that competitive markets - mainly involving private actors - are the most efficient way to supply goods and services and the most accountable for users' needs. Government role usually can be limited to policy-making, regulation, ownership or financing, leaving actual investment, operation and maintenance to nongovernmental entities." In addition, two fundamental notions have changed thinking on the delivery of services. First, 'provision' can be separated from 3 'production, and second, governments have to recognize that they do not need to dominate the provision of services; they only need to provide the enabling environment (Helmsing, 2000). The rationale for introducing the NPM approaches especially contracting out of 3

The primary responsibility of a 'provider' is to aggregate and articulate the dema~d of its constituents and to raise funds using its coercive governmental powers over the citizens, to finance public goods. The 'producer' on the other hand, performs the wholly technical function of transforming inputs into outputs.

New Public Management Paradigm

81

service delivery to agents it is argued in the literature that its leads to cost savings and better value for money by removing the production of such services from inefficient public bureaucracies that are more intent on satisfying the wishes of producer groups than of consumers (Hope, 1997: 165). Contracting out reduces costs in three principal ways. First, private contractors are free from Civil Service requirements, they use an incentive pay system and have greater freedom to hire and fire workers, they employ part-time workers, have less absenteeism and therefore are more flexible than the public 'Sector in reducing costs. Second, they pay lower wages than government agencies. Third, and most importantly, contractors tend to pay their workers substantially lower fringe benefits (Kettl, 1993:161; Helmsing, 1997a:72). There are four other fiscal arguments: if services are provided by the private sector, governments will be able to reduce subsidies to loss-making public agencies and reduce expenditure risk (for example, having to buy new trucks); increase tax revenue from private operators; reduce the public debt by selling assets; and reduce public borrowing by encouraging the private financing of capital expenditure on infrastructure (Rakodi, 2003:3). Furthermore, it is suggested that private provision will result in improved equity, since charging everyone for services will generate sufficient revenue to expand services to unserved residents and areas. The application of NPM in Africa has taken may forms including divestiture of public enterprises to private hands; creation of joint stock companies; the longterm leasing of public infrastructure to private enterprises; the use of decisionmaking processes such as contracting and bidding; internal replacement of coordination and authority with market mechanisms; and incorporation of market incentives into the reward system for public officials. Also, some public bureaux that had till then enjoyed monopolistic positions were being split up in order to separate their commercial, social, developmental and regulatory roles. By the end of 1998, there were 3,387 privatization transactions worth U.S. $7,352 m. across Africa, which according to the World Bank had brought about the following fundamental changes (World Bank, 2001 :259): 1. The fiscal burden of public enterprises had been reduced or eliminated. 2. Privatitation receipts had contributed to a reduction in fiscal deficits. 3. Privatization had attracted foreign direct investment, both in acquiring enterprises and for post-privatization investment in those businesses. 4. The process had stimulated private sector development and contributed to a more competitive business environment. Recent experience in public infrastructure services development in Africa shows that changes are occurring in terms of who manages the services in the cities. Governments in many African countries are not only encouraging private sector in public service delivery, but are also addressing power, authority and resource imbalances between the state and city authorities by providing an enabling environment ii') trie form of decentralization, democratization and community selfhelp (Rakodi, 2003: 13).The NPM approach of contracting out has now been extended to all areas of public infrastructure such as water supply, sanitation and solid waste services, roads and telecommunications, and other utility services. For instance, in the Cote d'Ivoire, the government has given a concessionary contract to the Societe d'eaux de Cote d'Ivoire (SODECI) to supply water to

82

Nicholas Awortwi

Abidjan and other towns; in Dar es Salaam, Accra, Conakry and Lagos, water vendors get their supply from the government water reservoir; and in Malawi, 20 per cent of the 770 health clinics are operated by private voluntary organizations (Olowu, 1999: 132; Rakodi, 2003: 10). In Nigeria, the National Electric Power Authority has contracted out maintenance of power stations and transmission facilities to private companies (Hope, 1997: 165). Also through the ·support of donor agencies, many governments in Africa are interested to involve citizens more directly in the managemel')t of services. For instance, donor support for undertaking service delivery surveys (user assessment) is being intensified in Uganda, Zambia, Tanzania and more recently Ethiopia (Olowu, 1999: 136). Although the degree of success is unclear in many African countries, governments' roles have fundamentally changed from providers to enablers, with emphasis on the ability to act as regulators (monitoring service quality and ensuring equitable access), partners (contributing to project finance directly or through incentives and credits) and catalysts (providing incentives and streamlining procedures and regulations). To understand the present trend in Africa better, there is a need for further research to assess the effectiveness of the new approach.

2.1. The Fundamentals that Need to be Right To achieve success in this new approach, lessons from organisational models (public choice theory, agency theory, and collective action theory) and countries that have used the NPM approaches (New Zealand, Canada, Australia, etc.) suggest some basic fundamentals, which need to be right if the promises of reforms and contracting-out are to be achieved (Awortwi, 2003: 105-110). 2.1.1. Incentives to Share out Responsibilities

Governance through contracting means passing government responsibility and risk to multiple agents. However, unless compensated by sufficient potential reward, commercial private agents cannot be expected to assume that role. A key task for LG and public service regulators is to find a way to guarantee adequate incentives to agents without exhausting public funds or taking all the risks. This means not only that incentives should be favourable to agents, but also that the government should be seen as obtaining the best alternative in terms of efficiency and effectiveness of service delivery. Attempts at a modality that do not recognize these, or that have unrealistic expectations of any of the actors, are certain to encounter severe problems. 2.1.2.Creating Competition Among Multiple Agents and Modalities Whether the aim is to reduce government or just to make it work better, public choice theory indicates that it is the introduction of competition, rather than awarding contracts to agents, which is the critical factor in lowering costs and improving efficiency (Damberger and Hall, 1996; Damberger et al., 1986). The influence of ownership in competitive and non-competitive market conditions has been extensively inyestigated (see Johnson, 1990; Hartley et al., 1992; Audit

New Public Management Paradigm

83

Commission, 1984, 1987, 1989). While the much-praised self-discipline of the market exists only when competition can reward success and punish failures, the government can influence any market through its regulative, facilitative and monitoring functions. If market imperfections hinder self-discipline and the government is unable to act intelligently, problems ranging from conflicts of interest to fraud can ruin contracting out modality. 2.1.3. Writing and Managing Contractual Agreements Agency theory suggests that it is imperative to distinguish the identity of the principal and the agent. This enables the development of arm's-length relations between the two. This means that the principal and the agent should both be seen as independent partners and nothing in the relationship should be construed as creating a merger. Even where the number of suppliers in the market is small, competition can still be created through short-term contracts. This ensures that although there may be only a few suppliers of SSW services, competition occurs before the contract is signed and, in principle, when the contract expires and is due for renewal. Thus, there is competition for the market even though there is no direct competition in the market. Secondly, the adoption of competitive tendering in governance through contracting essentially represents a switch towards market discipline in service delivery. It involves a greater separation between purchaser and provider (LGs' and contractors' functions) regardless of whether a contract was won in-house or externally. The implementation of competitive tendering and competition at the agency level involves four steps: specification of the service to be provided; request for tenders; evaluation and selection of the contractor; and managing the contract after the tender has been awarded. In that case, highly specific contract terms that establish duties, performance targets, and rules for changing prices hold the key to contractual exchange. Also, as agents perversely find it in their interest not to reveal all their information, there is a tendency for the principal to select wrongly. In such cases, informal constraints and personality traits in the form of honesty, integrity and a good reputation can be used to make the right selection, but care should be taken to ensure that the process does not ignore potential new entrants, as these actors may not have past business records. Research suggest that efficiency gains stemming from the application of competitive tendering and competition are substantial and widespread (Damberger and Hall, 1996; Prager, 1994; Damberger et al., 1986).

2.1.4. Performance Monitoring and Enforcement Agency theory suggests that monitoring and reporting becomes a natural corollary, whether the services are delivered by a public bureau or through the market or collaborative action. The reason for monitoring is that all agents cannot be trusted to keep to the contract. Actors do not give up shirking and self-interest just because they have entered into contracts. If the risk of getting caught is nonexistent, opportunistic agents will have a strong incentive not to make any effort to fulfil their obligations. The need for monitoring at both the agents' and the principal's levels arises from the risk of free-riding. However, one cannot take contractual enforcement for granted. It is (and always has been) the critical issue in agency theory. There should be provisions in the contract to check pre- and post- contractual opportunism by actors. The implication is that LGs must have

84

Nicholas Awortwi

an effective capacity to monitor, among other things, achievement of performance indicators set out in the agreement and the resources devoted for this purpose. Being the principal, the LG should reserve the right of supervision and appoint a monitor with or without the consent of the agent. Monitoring carried out by an independent regulatory body rather than the LG is often suggested as a way of increasing the credibility and effectiveness of regulation. Although third party enforcement is never ideal, never perfect, and a clientelistic relationship can develop, neither self-enforcement by parties nor trust can be completely successful. When trust becomes the criterion for selecting and monitoring agents, it is likely to be undermined.

2.1.5. Enhanced Autonomy and Competition at LG Level In order to foster competition in practice at the Local government (LG) level, it is necessary for the central government (CG) to permit substantial local autonomy. LGs should have the autonomy to regulate, facilitate and monitor agents they have contracted with. With greater autonomy and development of multiple ties, LGs can create competition within the bureau by encouraging different departments to compete for the public sector budget. In addition, LGs cannot be truly competitive if the costs of their decisions can be externalized through grantsin-aid. Boyne (1996) argues that there is an inverse relationship between central funding and local competition. In a competitive market, inefficiency is reflected in sharp price differences between alternative suppliers. The payment of large CG subsidies not only provides inefficient authorities with a buffer against competitive pressures, but it also distorts the price signals sent to the local residents. Nevertheless, there may be an argument for CG funding in the form of 'equalization grants' to neutralize local variations in service needs and tax base resources. If public choice arguments concerning competition are valid, then central policies will have substantial adverse effects on LG performance.

2.1.6. Capacity of Public and Private Actors As a result of imperfect information, there will be inevitable shirking in governance through contracting. Agents with better resources can hold back much information from the LG and twist many benefits out of contracting. Therefore, it is important for LGs to have the capacity to leverage information from agents. LGs can build their capacity by learning from some of their own delivery experience. This means that LGs should keep a level of in-house capacity to deliver services so as to leverage information, especially on costs. Reforms put pressure on government managers, who not only must do more with less public money, but also must build new capacity to manage agents in ways they have never done before. Building LG competence and effectiveness is, therefore, a vital prerequisite for successful contracting. Only LGs with "competent professional staff' would be fully able to develop, negotiate, manage, monitor and enforce a competent contract instrument (Cointreau-Levine, 1999; Helmsing, 1997; Kettl, 1993). The opposite is also true. Many LG officers are corrupt and have the tendency to collude with private agents to defraud the public purse if there are no higher-level controls, checks and balances. These crucial elements are too easily overlooked in the rush to adopt new public management models. The fundamental issues that need to be borne in mind are that reform strategies do not manage themselves. They require aggressive and thoughtful

New Public Management Paradigm

85

oversight by highly skilled public managers and not the usual traditional bureaucracy. 2.1.7. Basic Ethics for Politicians and Bureaucrats Several public administrative models (classical organization theory, Weberian bureaucracy theory, and public law notions) provide some basic rules to guide politicians and bureaucrats as they undertake their duties. Weberian ethics are conducive to safeguarding against abuse of power and misuse of office and funds by bureaucrats. The rule of law requires room for complaint and remedy. The public administrative framework advocates openness in relation to procedures and close scrutiny of leadership. 2.1.8. Dialogue and Stakeholders' Involvement In the new urban management, it is imperative to create a viable and central role for residents of the affected area through their own neighbourhood organizations (Spiegel, 1981; Cunningham, 1981; Langton, 1983). The literature shows that in many community partnerships for public service delivery, it is impossible for agents to create value or deliver a service unless users/clients actively participate in its production. In public services, the question is not whether the role of users is important, but rather how it is best elicited. Effective community participation requires strong community organization and the support of NGOs (UNCHS, 1993). The absence of such third-sector organization leads to unbalanced power relations, and, as CBOs often lacks adequate information to bargain, this influences negotiation over resources, terms and conditions, and roles and responsibilities. Through community organization there could be enhanced accountability and legitimacy. This is because CBOs may be used to question the extent to which service providers are meeting their contractual obligations and by doing so they act in the public interest (Barns, 1999).

2.1.9.Strengthening Institutional Regulation The assumption is that markets grow according to the rules in the environment in which they exist. Changing the rules shifts the balance of interests among the actors and creates different winners and losers. Regulatory institutions have to be structured in such a way that they are fairly robust against opportunistic tactics. If new urban management modalities are to be successful, the government has to develop and manage regulatory problems. If the government is involved in direct delivery of services, there should be an autonomous and independent commission to regulate all agents. However, enforcement requires an effective and impartial judicial system as well as existence of a transparent regulatory framework. In addition, there should be political will to let institutions work.

3.Governance of Basic Urban Services in Ghanaian Cities Since the colonial and post-colonial periods, the development of public infrastructure services in Ghana has been centred ori .state provision. Infrastructure services such as roads and railw·ay lines were built to link mineral and other industrial raw material areas, while electricity, piped water, housing and telecommunications, among others, were provided in the regional capitals

86

Nicholas Awortwi

(notably Accra, Kumasi and Takoradi). After independence in 1957, many parastatal organizations, such as the Ghana Water and Sewerage Corporation (GWSC), Electricity Corporation, State Housing Corporation, etc., were established to expand infrastructure services to satisfy social and political expediency. In the late 1970s, when the Ghanaian economy took a nosedive, there was deficiency in public infrastructure service provision. The situation deteriorated badly in the cities, and the CG could hardly manage the wide range of services in the neighbourhoods. It therefore became politically, economically and institutionally expedient to decentralize the provision, delivery and management of some of the basic services to lower levels of government. These services included SSW management, market infrastructure, parks and gardens, streetlights, minor roads/street repairs, etc., while education, water, health, telecommunications, postal services, electricity and public transport remained the responsibility of central ministries or parastatals. Following decentralization of basic services, the LGs set up production units, defined production technologies and offered the services virtually free of charge, in the same way that the central parastatals used to do in the past. With worsened economic conditions in the early 1980s, there were insufficient funds for acquisition of capital equipment, resulting in regular breakdown of vehicles, plant and equipment. Coupled with unmotivated waste management personnel, inappropriate management organization and poor planning, improper attitudes of residents towards waste management, and lack of enforcement of waste and sanitation by-laws, the ability of LGs to deliver SSW deteriorated sharply. The situation reached a crisis level in 1985 (Benneh et al., 1993:38; AMNColan Consult, 1995). In solid waste collection (SWC), about 82 percent of the population in the cities depended on a few communal container disposal sites from which garbage was not collected for weeks, while only 10 percent of the households had home collection. The rest of the population had no access to waste collection services at all and dumped their waste anywhere they wanted. Frequency of waste collection worsened to the extent where it was not uncommon to observe residents creating their own refuse dumps, sometimes forming layers of garbage stretching to the street. The conditions of public latrines (PL) were abysmal. Since about 60 percent of urban houses had no inside toilet, 4 the majority of residents had to resort to public latrines. As a result of poor management, Ghanaian cities were engulfed in filth, with heaps of uncollected refuse, choked gutters and overflowing public septic tanks. The most alarming 5 aspect of sanitation was the prevalence of bucket latrines, which was a threat to public hygiene and safety (Frantzen and Post, 2001 ). With the establishment of Waste Management Departments in all the major cities in 1985, equipped and funded by the German Agency for Technical Co-operation (Deutsche Gesellschaft fur Technische Zusammenarbeit GmbH - GTZ), the 4

In 1985 only about 30% of urban residents (medium to wealthy households) had access to flush latrines, while the majority used pit latrines, bucket latrines or other means. Three out of four households shared their private latrines (Benneh, 1985). 5 There were over 33,000 bucket latrines in Accra, and the bulk of them were in the low-income, highdensity residential areas.

New Public Management Paradigm

87

problem of waste management began to be addressed more seriousJy. To further improve services, there was a need for regular replacement of worn-out vehicles and equipment. Funds were required for capital and recurrent costs, but beneficiaries did not pay appropriate service charges to the LGs. The city authorities could not sustain the momentum to equip and enable the waste management departments to operate satisfactory. Collection of service charges from few wealthy and medium-income neighbourhoods did not cover even half of the operational costs (AMA/Colan Consult, 1995:35). In Accra, service coverage, which as a result of GTZ assistance had increased to about 80 percent, began to fall again. By early 1990 it had fallen to 45 percent because of limited financial resources for maintaining the trucks (AMA/Colan Consult, 1995). Furthermore, there was managerial inefficiency and no effective monitoring of activities and cost. Solid waste collection and sanitation management became headaches for the city authorities as well as residents. Generally, the public saw the waste management departments as inefficient and ineffective. In the early 1990s, following the CG policy towards private sector-led development in the economy, critics began to question the delivery role of LGwaste management department. One alternative considered was the employment of private contractors. Although provision of the service remained a public responsibility, the delivery would be private. In 1990 some LGs took the initiative and subsequently started contracting and franchising the collection and management of SSW services to the private sector. In the plans to involve the private sector, it was predicted that in the short to medium term, LGs would provide about 60 percent of the basic services, with 40 percent being in the hands of the private sector. This proposition required LGs to change the regulatory framework and tendering procedures, specialize in the management of standard contracts and implement strong and effective monitoring, supervision and performance measurement systems. LGs were also required to set tariffs at realistic and economically viable levels, with due allowance for cost recovery and depreciation of capital investments, and to build private sector capacity through access to investment funds. In other words, the move from direct provision to contracting implied that LGs had to completely reorganize and reorient their management and regulatory apparatus and strengthen the capacity of their staff. In 1999, it was further projected by the Ministry of LG that 80 percent of SSW services were to be provided by the private sector on a competitive basis. LGs were to maintain a residual capacity to manage only 20 percent and make emergency intervention in the event of failures by the priv9te sector (Zanu, 1999). In another dimension, LGs have also collaborated with CBOs in areas where both public and private sectors have been ineffective in delivering SSW services. In Ghana, the policy shift changed solid waste collection (SWC) in the cities from a waste management department monopoly to multiple modalities. Figures 1.1 and 1.2 show the pattern in Accra and Terna, respectively.

88

Nicholas Awortwi Fig.1.1 Modality of solid waste collection in Accra, 1995-2000

100% 80% 60% Vol. ofSWC

40% 20% 0% 1995

1996

1997 1998

1999

2000 CJ NGO/CEO/others 1111 Private Ill Public

Years

Figure 1.1 shows that the volume of SWC by waste management department (public delivery) fell from about 78 percent in 1995 to less than 0.5 percent in 2000, while private delivery rose from 10 percent in 1995 to about 97 percent by the year 2000. A small amount of waste is also collected by organizations such as CBOs, NGOs, prisons services and others, including unorganized groups. However, in Terna (Fig. 1.2), public delivery by waste management department has persistently been not less than 40 per cent over the past five years, and that by the non-private and non-public sector not below 20 percent. Fig.1.2: Modality of solid waste collection in Terna, 1996-

2000

100% 80%

Vol. ofSWC

60% 40% 20% 0%

c.o

r--

0) 0)

0) 0)

~

~

co 0) 0)

0) 0) 0)

~

~

Years

0 0 0

N

D Org/others II Private Ill Public

Generally, when it is possible to make productivity comparisons between public and private programmes for similar goods and services, the overall conclusion from a number of studies in various countries appears to be that private provision is more efficient than public provision, all other things being equal (Mueller, 1989). In solid waste, one of the aims of inviting the private sector was to increase the volume collected. Data from the LGs shows remarkable improvements in all the cities (Figs. 1.3 and 1.4).

New Public Management Paradigm

89

Fig.1.3: Volume of solid waste collection in Accra, 1990-2000 400,000 350,000 300,000 Cl)

250,000

C:

.8

200,000

0

~

150,000

~

100,000 50,000 0

Years

Figure 1.3 shows a consistent increase in SWC with private sector participation in Accra. In 1997, when the biggest private company was contracted, collection almost doubled and has shown continuous improvement since then. Fig. 1.4: Solid waste collection in Kumasi (2000-01) 60000

50000

40000 en

C:

.9 (,)

EQ)

30000

~

20000

10000

0

90

Nicholas Awortwi

In Kumasi, under full capacity (when all equipment was in operation) in July 2000, the waste management department was able to collect at least 12,000 tonnes of 6 · waste for disposal (KMA- waste management department, 2000). When the private sector was contracted in August, SWC took a dramatic turn (more than quadrupled, Fig. 1.4). However, it fell sharply a month later and achieved stability at a slightly higher level than when SWC was delivered directly by the waste management d~partment. The sharp fall was due to the fact that accumulated waste had been cleared. If the NPM that introduced multiple modalities of private, community and public sector services delivery did produce cheaper, more efficient government, with high-quality services and more effective programs, and if it simultaneously enhanced political control, freed managers to manage, and made government more transparent, it would deserve its current widespread promotion. Unfortunately, matters are not so simple. There is a plenty of evidence that expectations of the NPM modalities may be elusive if the fundamentals necessary for successful implementation are not first put in place (Awortwi, 2002 & 2003). The NPM practice of contracting-out the delivery of basic services to multiple agents involving the private sector, CBOs and government waste management department has been in existence for at least a decade in Ghana, but their results or the preconditions for success have yet to be analysed carefully. NPM is theoretically and conceptually thought of as a means to an end, not an end in itself (to be precise, a potential means to multiple ends). The ends include: making savings in public expenditure, improving the quality of services, making government operations more efficient and increasing the chances that the policies that are chosen and implemented will be effective (Pollitt and Bouckaert, 2000: 1). While many policy-makers and academics have expressed profound optimism about the new approach, there are many questions that demand further exploration and answers. Considering the remarkable improvement in the volume of SWC, would it be correct to conclude that governance through contracting has been successful in Ghana? Some more questions arise: Did LGs save money through contracting to other actors? What happened to value for users' money? Has there been technology improvement in service delivery? To what extent has private capital been introduced into LG services? What happened to LG labour management problems? To what extent did LGs transfer/share risks with the private sector? Has quality of service delivery improved? Did the contractual processes follow laid-down procedures and were they transparent? These are questions that demand answers but cannot be deduced by merely looking at the volume of waste collection.

6

Until 2000, there was no proper documentation of the volume of SWC. Officials based their assessment on hypothetical parameters of a total population of one million and a per capita generation rate of 0.6 kg. which gives a daily generation of about 600 tonnes. An estimated additional 250 tonnes is generated from the two main markets (central market and Asafo market) to bring the total city-wide daily generation to about 850 tonnes (KMA-WASTE MANAGEMENT DEPARTMENT, 2000).

New Public Management Paradigm

91

3.1. NPM Expectations and Outcomes NPM movements have dominated governments around the world in the last two decades, in both the developed and developing worlds. From the energetic Westminster style of reform in New Zealand to re-engineering government in the United States, we are confronted with a plethora of conflicting results. Whether these reforms produced solid results has been a controversial issue in the literature. Comparative analysis of administrative reforms suggests that there have been impressive gains in productivity (Keating, 1998). However, Kettl (2000:51) cautions that "analysis was incomplete at best, but anecdotal evidence, coupled with cross-national comparisons, indeed suggests that many succeeded in improving government's ability to produce more and better government services at lower cost. However, no good, reliable data are available in any country regarding the savings that the reforms produced". The Ghanaian research provides rather surprising results that add to the controversy surrounding NPM outcomes.

3.2.lnfusion of Private Capital and Technology Before the introduction of NPM and especially the private sector participation, there was no desire on the pwt of the LGs to collect fees and they bore almost all the costs of service delivery. Given these difficulties, the waste management department could hardly mobilize adequate financial resources to ensure sustained service delivery. With the introduction of private delivery, LGs are gradually shifting to cost recovery, though at a slower pace in solid waste collection (SWC) than in public sanitation. In fact, private sector participation in public sanitation in Ghana shows many forms of recent development including build operate and own (BOO), build operate and transfer (BOT), rehabilitate operate and transfer (ROT) and afterimage which have improved LG finances to the extent that in the sub-metro units, surcharges form 60-70 percent of their revenue. In addition, there has been remarkable improvement in technology infusion in the services. The use of compaction trucks in SWC and water closet in public sanitation has replaced tipper trucks and bucket latrines. In Terna, where the LG continues to directly deliver services, the old technology still prevails. The difference between the public technology of service delivery and that used by private agents makes technology improvement one of the most significant benefits of private sector participation in sanitation and solid waste services delivery in Ghanaian cities.

3.3. Manageria1 Efficiency The core presumption from NPM is that the skills needed to find and operate in new markets, enhance productivity and stay ahead of the competition can also improve the way the LG works. LG managers learn by emulating their private partners, as they adapt to the rigours of competition. If successful, they become more entrepreneurial and flexible, occupied with deal-making and attracting capital rather than with administrative procedure (Linder, 1999). So, to a large extent, the success of improving LG service delivery depends heavily on that subjective concept of market and management principles. However, in the three cities there is little to show that there has been any significant improvement in efficiency of LG personnel and managers. It has been business as usual as LG

92

Nicholas Awortwi

officers do not sign any performance contract, nor are their salaries dependent on service improvement. Their regulatory, facilitative and monitoring capacities demonstrate little or no significantchange from the pre-reform period. Labour and attitudinal problems continue to bedevil the LGs because LGs do not have the powers to hire high-quality staff and fire incompetent staff. The claimed efficiency gains of contracted services are unsupported by evidence. Furthermore, introduction of private delivery in many instances would require raising user fees for some user groups so as to guarantee the financial viability of the service in other areas in order to expand coverage. However, this is based on the assumption that LGs will improve their management efficiency to collect the revenue. The findings on this in Accra are not encouraging. As the franchise system was abandoned and local contractors were made sub-contractors to the city and country waste limited (CCWL), the LG assumed responsibility for collecting user fees. In the year 2000,. only 2 percent of the revenue was collected. This means that as long as LGs remain inefficient in fee collection, raising tariffs will not improve their financial balance so as to guarantee expansion of service coverage. In Kumasi, the LG is paying a very high contract fee to the Kumasi Waste management Limited (KWML), yet it has failed to bill users for the services. In addition, the LG would have been expected to improve its handling of complaints from residents and operators in prompt, proper and responsible fashion now that its responsibilities have been streamlined, but it has been business as usual in this area, too. There has not been a single assessment survey in any of the cities to incorporate users' opinions on how to improve service delivery. In addition, the NPM approach requires LGs to invest a fair amount of time and resources in learning to speak different business languages to commercial agents, CBOs and the bureaux. But, when LGs continue to practise the old ways of doing things and agents exert corrupt influence on corporate governance then LGs can expect to lose many of their expected benefits. Certainly, a decade of implementing the NPM modalities has· not contributed to the development of more professional. LGs that can efficiently and effectively facilitate new contract management, performance monitoring, and the quest to improve LG finances, as well as achieve value for money. The LGs do not regulate better and intelligently, partly because.of inadequate CG enabling policies but substantially because the LGs are not made accountable to the citizens and continue to live in the past when public. services and civil servants were to be courted and worshipped. Service delivery has not been made a major issue on political platforms. Politicians are elected not on the basis of what users see as improvements in public service delivery, but on party affiliation, personal relationships, ethnicity, monetary influence and slogans. Besides, mayors are appointed and not elected, and as long as they can praise the President they care little about service delivery in localities. The role of users and cJvil societies in influencing service delivery is minimal, and the transparent corporate governance necessary in NPM is absent in Ghana.

New Public Management Paradigm

93

3.4. Shifting the Responsibility and Value for Money A consensus is gradually emerging in the literature that contracting to the private sector enables government to transfer financial and managerial risks to the private realm, which can better assess, bear and manage them. In undertaking multiple modalities, the LG is concerned with two main risks: First, private agents will do a bad job, and second, they will not comply with regulations of the LGs. In Ghana, the bureaucrats purport to design strategies to ensuring that these risks are managed before contractual modalities are entered into. The three main strategies th~y claim to adopt inclu'de (i) scrutinizing the background of the prospective agents, (ii) asking them to sign performance agreements, and (iii) engaging them in a continuous dialogue to extract more information from them. However,- if the cases of the CCWL, KWML and Terriwhite contracts in SWC can be used to generalize about what goes on in LGs, then scepticism can be expressed.· Little ,or no risk was actually transferred from the LGs to these companies. The companies did not bring in any substantial capital or managerial expertise. The CCWL and the KWML took over the assets of the LGs at the lowest risk level. They did not pay cash but used the abnormal profit from their dubious service contracts to defray the cost. In Accra, the contract also made it mandatory for the LG to transfer the ownership of the machines to the CCWL after five years. Records from the LG show that the CCWL did not make any payment during the three years of its operation until the contract was abrogated. Turning over service delivery to agents is expected to give both the LGs and users their money's worth in terms of the services they receive. 'Value for money' is measured by comparing the quality of services received and the amount paid in return in relation to what existed in the past or what would have been. However, considering the lowest risk that the LGs passed on to the agents in Accra and Kumasi, and the fact that the amount paid to the CCWL (Accra) and the KWML (Kumasi) is disproportionate to the level of services delivered, all contracting out of services is no more cost effective than traditional LG delivery. The LGs and users are not getting good value for their money. This shows that unless a private operator is engaged with a clear link between the levels of risk, quality of performance, and payment received the high expectations in the literature about efficiency gains and value for money will not be met. Nevertheless, passing delivery responsibilities to the waste management department (in-service delivery) has proved to be necessary as users in start-up areas and poorer communities in Terna are getting services that would not have been possible through franchising. This solidifies the argument that public sector participation allows the LG to concentrate its direct delivery on vulnerable segments of society. However, even here, since the same public budget is used and the LG delivery is rated the poorest in terms of quality, there is still no value for money. The waste management department is not conscious about the cost of its delivery. In fact, NPM suggests that when service delivery is retained inhouse, it should also be competitive. In Scotland, authorities made the decision to retain work in-house and made conscious efforts to cut labour and other variable costs by entering into contracts with the bureau. Kerley and Wynn (1991 :48) indicate that "contract-based services, even where LGs have won contracts with no alternative bids logged, has of itself forced a reduction of cost and/or an increase in productivity". In Terna, the LG contracted-in SWC to the waste

94

Nicholas Awortwi

management department without reforming the latter's mode of operation. It did not change the internal organization of the department to adopt efficient or commercial principles. The LG did not separate the monitoring role of the department from its delivery, thereby creating the ground for failure in both. However, despite the argument for efficiency, value for money, and advantages of shifting responsibility, it should be borne in mind that direct public delivery may be necessary in some circumstances, especially in vulnerable groups. This is because if rich citizens are served by private enterprises, and they pay directly for service provision, there will be less solidarity with poor citizens who are unable to pay the full cost of a service. It may even be possible that high-class area residents may resist cross-subsidization and/or paying higher taxes for that purpose.

3.5. Contracting out Service Delivery and Cost Savings McDavid (1985) made a cross-sectional analysis with data from a survey of 126 Canadian municipalities conducted in 1981 and 1982, and showed public collection was 50 percent more expensive than private collection. The substantial cost difference was confirmed when a regression analysis was used to allow for service level, scale of output, collection method and environmental conditions. Public collection was still 41 percent more costly than private collection. The Audit Commission (1984) looked at refuse collection in England and Wales and concluded that although privatized services produced on average better results than the average local authorities' direct labour organization, privatization was not necessary to secure efficient performance as a quarter of LGs had costs lower than the average privatized service. Damberger et al.'s (1986) results suggest that the introduction of private contracting has yielded substantial cost savings a conclusion confirmed by analysis of the trend in costs in authorities where private contracting has been introduced. The ICMA survey also revealed that 40 percent of LGs saved more than 20 per cent from previous outlay and another 40 per cent saved 1O percent to 20 percent (Kettl, 1993: 160) in all services that were contracted out to agents. The OECD survey of experiences in Australia, Denmark, Iceland, Sweden and the UK showed that savings from contracting out ranged from 5 per cent to 50 percent, although Keating's report a year later showed 20 percent (OECD, 1997; Keating, 1998). · In the Ghanaian case, in general it is the opposite. LGs have not saved but rather spent more than before. The LGs are now paying more than twice the cost per tonne of SWC than they used to before contracting out. This has come about as a result of collusion and corruption affecting contract prices. In fact, in Accra and Kumasi, it made little economic sense for the LGs to have contracted out SWC. And because their revenues are not enough to cover this huge undertaking, the CG has in the past five years come to their aid. However, since one of the main priorities of LGs in adapting NPM approaches was to shrink the annual deficit, in Accra the practice did not continue to get political support; so, by early 2001, the government abrogated the main private contra'ct. Indeed, the results regarding cost savings in LGs' contracting out services in Ghana confirm Prager's . (1994: 176) assertion that at times, instead of stemming the flow of budgetary red ink, the modality can intensify the haemorrhage. In the countries that realized improved savings, the evidence showed that who performed the service delivery

New Public Management Paradigm

95

was less important than how it was done. In other words, it was not about public versus private delivery, but about competition versus monopoly. This indicates that it is the introduction of competition, rather than awarding of contracts to private firms, which is the critical factor in achieving lower costs (see also Kay and Thompson, 1986). The threat of substitution creates incentives for agents, both 'in-house' and private, to minimize costs. In addition, if there is competition for different levels of service, the bidding process generates information for policy-makers on the cost of alternative supply options. That is, information on marginal costs and marginal benefits (Hartley, 1984). Elizabeth Proust (1997: 133) in her report on Victoria State indicated that contracting out enabled Victoria to save 16 percent. However, in Victoria's case there was substantial 'inhouse' contracting, meaning that private contractors did not necessarily always win the contract. The Citizen' Charter market testing in the UK also reported 27.8 percent savings where 'in- house' bids were included and only 14.4 percent savings where no 'inhouse' bids were involved. Damberger et al. (1986:83) also show that 'cost savings was achieved where tendering results in the contract being awarded to the 'in-house' units'. In the Ghanaian case, as shown in the analysis, there was no proper 'in-house' contracting. The waste management department operates, using the public budget, with less emphasis on value for money. In fact, apart from the accounting department, the rest of the waste management department have no idea how much their own delivery is costing the LG and do not bother to find out. The NPM approach, which focuses on commercial and client-oriented principles when services are contracted-in, was absent in the waste management department because the LG did not change its policy towards them. The department needs internal reorganization to achieve the benefits of 'in-house' delivery. The waste management department should also be made to deliver competitively even if services are delivered free of charge. The findings also show that in SSW services it is preferable to implement a low-cost solution that is affordable by users. As CBOs seek the welfare of their residents, their fees are cheaper and affordable by many users, which cannot be said of fees charged by commercial agents. Although CBO services cost less and deliver better, they have rarely been given the necessary attention by LGs in Ghana.

3.6.Reduction in Size of Government The reform movement in many nations was to reduce the size of the public sector in order to reduce public spending. Government spending as a percentage of gross domestic products (GDP) fell in many countries that implemented NPM. For instance, between 1990 and 2000, government expenditure in New Zealand fell by 18 percent, and in the United States by 8 percent (Kettl, 2000:52). Kettl's report also showed that during the same period, government payrolls in the UK decreased by more than one-third, and those in New Zealand by 14 percent. In Ghana, following the policy shift towards contracting governance to multiple modalities, it was expected that the workload and consequently the size of the labour force of the waste management department would be reduced. Keeping core staff to monitor the activities of agents was to be the focus. The findings show that although contracting out did succeed in trimming the department, the reduction was minimal and affected only some labourers, leaving the

96

Nicholas Awortwi

administrative personnel untouched. For example, in Kumasi, contracting out total waste collection to the KWML succeeded only in transferring 55 out of a total of 268 personnel (20. 7 percent). In Accra, too, the LGs still keep a large number of personnel whose services are no longer needed. Instead of dispensing with their services and recruiting new officers to perform new responsibilities, the LGs are restricted by Civil Service rules on employment and dismissal; hence, labour management problems continue to bog down progress towards LG effectiveness and efficiency.

3.7.Transparency and Accountability Transparency and accountability are complements to responsibility, and they raise concerns regarding the manner in which LGs operate as regulators and how they explain, answer for, and bear the consequences of the way in which they discharge their duties, fulfil functions and utilize resources in the interest of the public (Ferris, 2001 ). As technical personnel in charge of sanitation and solid waste management, many of the bureaucrats have little control over the issuance of contracts and openly express their discomfort. Secondly, not only do the CG transfers facilitate LG inefficiencies, but also the way in which some of the transfers are made raises governance and accountability concerns. Transfers that are not clearly publicized to local users can undermine LG accountability and jeopardize the improvement in incentives sought from NPM. For instance, the research revealed that between October and December 2000 a total amount of ~10.3 b (US $ 1.4million) was transferred from the CG in three separate cheques to the LG in Kumasi for the purposes of SWC. This amount was deposited in an account outside the control of the LG, ostensibly to avoid misuse. Considering that the amount was transferred during the heat of the political campaign and that the government at that time was losing power, some bureaucrats at the Ministry of LG allege that part of the money was used for party politics. The confirmation by the KMA finance officers that they have no control over the use of the money shows lack of transparency and possible corruption. Yet this cost is to be included in the total cost of SWM in the city. In addition, when the LG in Accra was winding up direct delivery of SWC, many of the waste management department's machines were sold to a private agent at ridiculously low prices. There was no competitive bidding or auction sale. l_n what is described among the private agents as 'inner circle corruption', it is alleged that engines and other vital parts of the machines were removed or written o.ff as unusable. This paved the way for many of the old LG vehicles to be sold as scrap to some agents. A Ghanaian weekly, The Statesman, reported on 30 May 1999: The CCWL service contract is one of the most ridiculous 'rape cases' against the people of Ghana by a government agency. The stench of the whole contract assumes an empowering dimension with the revelation that the costs of the Madvac 101 D which is priced in the contract at US$ 41,300 each cost only US$ 28,000 in the market.

Evidence of shady deals in government contracting-out services and the NPM is widespread not only in Ghana but also in many countries in the developed world (Kettl, 1993; Schneider, 1992; Martin 1993).

New Public Management Paradigm

97

4. Policy Implications In the light of what happened in Ghana and comparable and conflicting results in other countries, the following policy implications are presented for critical assessment, basically advising governments with similar problems to go back to the basics and get them right. Re-engineering government using the NPM approaches is a comprehensive activity backed by theories and principles and not just an art inspired by ideology.

4.1.Redefinition of State Policies and Objectives in Public Services The first task required to improve public services using the NPM approach is redefinition of the state's role. The new role should draw clear policy boundaries between the CG, LGs and agents (public, private and CBOs). When government services are contracted to agents including LG, the state has a responsibility to regulate, facilitate and monitor them. The regulative role will require the state to set overall rules of the game that take into consideration local potential and challenges. Since public service is a political decision, it is the responsibility of the state to set public objectives to be achieved. The facilitative role of the state is to provide space for multiple actors to perform to their utmost potential. This can be achieved through political, market and community enablement strategies that ensure that there is competition in the three sectors. Enabling these sectors will require substantial decentralization of powers, financial resources and human capacity. Serious commitment to decentralization, and not just empty promises, is a key issue here. For instance, the CG may support information diffusion by organizing regular courses and training workshops for LG personnel on new and better ways of regulating. It is important for the CG to first strengthen the bureaucracy in its right sense before initiating a new model. The monitoring role requires bureaucrats who can set objective verifiable indicators to judge service outcomes that -were transferred to multiple agents. When the state is able to differentiate its role from that of agents by clearly establishing regulatory, facilitative and monitoring policies and defining the objectives to be achieved by the three policies, efficiency in service delivery by the public sector, private sector and community can be made more similar than we have assumed in the past. This does not suggest that modality does not matter. It matters when efficiency is considered along with effectiveness, equity, capacity of actors to act, and local circumstances of users. This suggests that setting the right policy objective to be attained is very crucial in public service provision.

4.2. Need for Independent Third Party to Enforce Regulation of Private Agents When LGs, which are supposed to take care of the public interest, engage in direct service delivery as agents and at the same time act as regulators of agents, there is a risk of overlooking their own deficiencies. In addition, there is a risk of partiality and favouritism as LGs themselves are interested ·agents. Separating the roles of LGs as agents of service delivery and regulators helps to resolve many ambiguities. While it is not appropriate for LGs to perform this dual function, it is also suicidal for regulation to be left to agents. Therefore, under no circumstances should an agent be made a supervisor of its own activities. NPM

98

Nicholas Awortwi

modalities of service delivery should ensure that there is arm's length relationship between the government as a purchaser and agents as suppliers. The analysis of Ghanaian case has shown that implementation of NPM modalities has many benefits as well as governance problems (institutions, management, accountability, ethical, informational, and attitudinal). This means tbat institutional development to regulate private agents effectively and efficiently is a prerequisite for success. In the short to medium term it may be acceptable for LGs to perform a regulatory role as long as they hope to gain some experience and information in the new area. In the long run, what is needed is a statutory or an independent regulatory body that clearly defines the rules for both the government and agents and openly enforces them. Involving an independent third party in designing contracts that protect all parties' interests, and in setting performance targets, monitoring and sanctions has the potential not only to improve arm's-length relationship but also to increase the credibility of the agreement. Nevertheless, involving an independent third party may also not totally remove the risk of regulatory capture as this body may also be embroiled in rent-seeking and private interest. So, as third party regulators become stronger, regulating and monitoring regulators may be desirable as well. This role, to be performed by the CG, needs to be insulated from short-term political expediency and other improper influences and to be based on competent analysis of costs and benefits to society. For the agency to be truly independent, its staff, source of revenue and powers ought to be autonomous of the state.

4.3. Setting Clear Input and Output Indicators to Monitor and Penalize Poor Performance The NPM modality of contracting agents to deliver LG services has a number of problems emanating from selecting the right agent and ensuring that targets are met. For this to be successful, clear input and output indicators should be set. Input indicators are relevant as far as using them as criteria to select agents is concerned. The moment a contract is awarded, monitoring should shift towards output indicators. The two should therefore be separated in the contractual agreement. Monitoring provides evidence to check compliance and noncompliance among actors, which means that the LG should provide sufficient incentives for monitoring officers to perform. Monitoring s~ould then be supported by a clear threat of sanctions. In fact, monitoring may not be worthwhile if there is no sanction to penalize poor performance. This is because, if non-compliance is costless, whereas compliance is costly, actors will have a strong incentive not to comply. Sanctions in the form of fines should be structured in such a way that they cover the cost of remedying the situation, including the administrative cost of investigation. In addition, sanctioning for non-compliance should not only be limited to liquidation (monetary payment or fines). Since good reputation is often important to firms and LGs, the risk of being exposed to public contempt as a result of poor performance or dishonesty may provide an additional incentive to comply. Therefore non-compliance by actors should be exposed to the public and subsequently penalized.

New Public Management Paradigm

99

4.4. Building Capacity for Local Governance of Service Delivery As responsibilities between public, private and community sectors are shared, the actors must have the technical, organizational, managerial and financial capacity for the selected functions. NPM modalities do not guarantee efficiency, effectiveness and lower costs to LGs. Problems arise when the approach is poorly conceived, unregulated and, in particular when competition between agents is lacking and the LG is unable to shape the service market. Governance through contracting may also degenerate into private rent-seeking and other corrupt practices if no effort is made to hold bureaucrats and politicians accountable. Findings from LG corrupt practices suggest that efforts at capacity building should also support moral values such as personal responsibilities and accountability. This means that ethics inherent in administrative bureaucracy must not be downplayed for the sake of quick decision-making. Finally, public departments delivering services should be structured to permit closer budgetary oversights, which ensure that their activities provide value for money. Public departments should change the focus of their employees from pure administrative bureaucracy to business orientation, time consciousness and respect for the public budget. These get enforced by cost recovery as a basis for defining the bureau's budget. For this to be successful there should be motivation for civil servants. There should be commensurable remuneration to bureaucrats in order to obtain their commitment and loyalty. However, motivation is more than just salary improvement, but also respect, recognition, autonomy and opportunities for skills training. This means that public service motivation should be based on a carrot-and-stick policy of incentives and sanctions.

4.5. Establishment of A Citizens' Charter The public sector in the past was related to users of public services on a basis of authority because it tended to have monopoly traits. However, with the application of new forms of public management, the distinction between users as consumers and/or clients suggests that consumers have the exit option when they are not satisfied with the performance of agents or have a strong voice. Therefore, the pressure for improving quality of services is through empowering citizens to be the market-testing power. Though in many public services this may be difficult, in the absence of choice there are things that can be done to commence consumer power to the service. These are the sort of measures to be implanted firmly in citizens' mind. Public servants need to tell citizens what level of service they are entitled to expect (including indicators of performance), and a means of redress should be in place if they are not provided. There is a need for many African countries to introduce a citizens' charter like the one in the UK and elsewhere if politicians and bureaucrats are to be held accountable. It may involve putting someone in place to represent the citizen, like a regulator, or creating and strengthening public complaint commissions and giving them broad powers to ensure compliance and or obtaining redress for the citizen, like an ombudsman or a small claims court. The charter should be simplified and involve regular contacts with users. Users should be able to hold politicians accountable for the services that they provide to the public. The corollary is that if the service provided is unsatisfactory, politicians could end up being voted out during elections. In countries where civil society and other consumer groups are strong,

100

Nicholas Awortwi

users as clients have looked at service delivery and (where they could not vote with their feet) cast their votes to remove politicians and governments. Politicians who cannot be accountable to the public through the services they provide are in the wrong business.

5. Conclusion: Is there a Need for an African Way of Public Administratjon and Management? I will conclude this paper by using the aftermath of Ghana's application of the NPM to rekindle the debate on the relevance of western management concepts in Africa. Despite many persuasive arguments by 'environmentalist' and 'organizational' theorists that Africa's socio-political and management culture does not permit acclimatization and application of some of the new ideas in public management, the findings of the present study, while not conclusively rejecting their views, provide interesting new insights. The outcomes of the implementation of NPM elements in the three Ghanaian cities provide reason to question some stereotypical assertions in contemporary literature. Many countries in Africa have implemented different types of market reforms, including NPM and partnerships modalities, but the ,results have been disappointing. Comparison of the results in Ghana with those in New Zealand, the UK, Australia and the U.S. clearly shows that competition and regulation, which are the fundamental theoretical principles underlying success in cost saving, efficiency gains, quality of service delivery, etc., are universal and applicable to Africa. In general, adoption of new approaches would have many unexpected results, as happened and continues· to happen in Europe, the U.S. and elsewhere in the developed world. Socio-cultural, political and institutional factors may have some influence in getting many of the fundamentals right in many countries; hence, reform outcomes may not be satisfactory. But to what extent can they be attributed to culture? Are socio-cultural and political differences between countries that achieved significant success in NPM and those that failed rise to such a level to suggest that they cannot be undertaken in African public administration, or that they will automatically produce poor results? Can cultural difference alone be a sufficient reason to agitate for an African way of public administration and management? I am aware that many authors have suggested the use of indigenous institutions as a substitute for western administration and management concepts that seem to be failing in Africa. Yes, there is a need to recognize the role of indigenous institutions and to integrate them into management, but to what extent can these institutions solve the overwhelming problems in many African cities? Can these local institutions not be reformed to make them more efficient? And does integration of local institutions into public administration mean that basic principles of competition, regulation and motivation which have become the foundation of western management concept become incompatible with the African administrative set-up? These are issues requiring further reflection, nuance and research. That would be much more productive than the continual use of cultural stereotypes to explain all that has gone wrong with reforms in African countries. For instance, Haruna (2003:348) argues that the American type of bureaucratic reform is unsuitable for Ghana

New Public Management Paradigm

101

because the Ghanaian social structure is predominantly rural, with low literacy, multicultural, with little infrastructure, and marked by ethnic loyalty! Future research may help us to establish the influence of each of these socio-cultural factors on getting the fundamentals right. In other words, what is the minimum degree of socio-cultural change that countries must bring about to enable them to adapt and undertake new ways of solving public problems? The author of the present study certainly has no knowledge of the types of reform that would be suitable only in Africa. However, we can be reasonably certain that a combination of universal organizational theories (such as competition, regulation and motivation), taking into account the socio-cultural, political and economic circumstances of Africa, will be necessary; but the latter need not be overstretched and repeated over and over again, thereby making culture a black box . This is because we also know that some African public organizations (in South Africa, Mauritius and Botswana) are performing well without explicit reference to an African way. The World Development Report (1997) indicates that these are the only countries in sub-Saharan Africa that have improved institutional capacity to perform basic tasks of governance satisfactorily. In fact, it is disappointing to say that NPM have done little to improve the governance of public services in the three Ghanaian cities, not because the Ghanaian socio-cultural and political environment cannot adapt but because the CG/LGs failed to get the fundamentals right. The approaches the LGs took in terms of introducing competition, creating arm's-length relations between themselves and agents, building regulatory, facilitative and monitoring capacities of bureaucrats and motivating them to undertake their new responsibilities were implemented unsystematically and half-heartedly (Awortwi, 2003). Getting the fundamental rights is a necessary condition for achieving results in the NPM modalities.

References AMA/Colan Consult. (1995). Public & Private Sector Management of Solid Waste in Accra: An Evaluation and Prospects for Sustainable Services Delivery. Accra: AMA. Amin, P., and P., Lloyd (eds.). (1990). Housing Africa's Urban Poor. Manchester: Manchester University Press. Audit Commission.(1984). Securing Further Improvements in Refuse Gol/ection: A Review by the Audit Commission. London: HMSO. Audit Commission for Local Authorities in England and Wales.(1987). Competitiveness and contracting-out of local authorities' services', Occasional Papers, 7. Audit Commission for Local Authorities in England and Wales.(1989). Preparing for compulsory competition', Occasional Papers, 3. Awortwi, N. (2002). Getting the fundamentals wrong: The woes of partnerships in solid waste collection in Ghanaian Cities. Paper presented at ISS Lustrum Conference, The Hague, 7-10 October. (Forthcoming in), Public Administration and Development, 2004. Awortwi, N. (2003). Getting the Fundamentals Wrong: Governance of Multiple Modalities of Basic Services Delivery in three Ghanaian Cities. Maastricht: Shaker Publishing. Barns, M. (1999). Users as citizens: Collective action and the local governance of welfare', Social Policy & Administration, 33(1): 73-90. Barzelay, M. (1992). Breaking through Bureaucracy: A New Vision for Managing in Government. Berkeley and Los Angeles: University of California Press. Benneh, E., J. Songsore, J.S. Nabila, A.T. Amuzu, K.A. Tutu, Y. Yangyuoru and G. McGranahan. (1993). Environmental Problems and the Urban Household in the Greater Accra Metropolitan Area (GAMA)- Ghana. Stockholm: Stockholm

102

Nicholas Awortwi

Environment Institute. Boyne, A. G. (1996). Competition and local Government: A public choice approach, Urban Studies, 33(4-5): 703-21. Cointreau-Levine, S. (1999). Private Sector Participation in Municipal Waste Services in Developing Countries (Vol.1), Urban Management Programme Discussion Paper 13, Washington: The World Bank. Cunningham, J.V. (1981). Assessing the urban partnership: Do community forces fit? National Civic Review, 70 (1): 521-526. Dilulio, J.D. Jr. (ed.) (1993). Deregulating the Public Service: Can Government Be Improved? Washington: Brookings Institute. Damberger, S. and C. Hall. (1996). Contracting for public services: A review of Antipodean experience, Public Administration, 74: 129-47. Damberger, S., S. Meadowcroft and D. Thompson. (1986). Competitive tendering and efficiency: The case of refuse collection, Fiscal Studies, 7(4): 69-87. Farnham, D. and S. Horton (eds.). (1993). Managing the New Public Services. Basingstoke: Macmillan. Frantzen, A. and J. Post. (2001). Public toilets in Kumasi: Burden or boon? In K.K. Adarkwa and J. Post (eds.), The Fate of the Tree: Planning and Managing the Development of Kumasi, Ghana, pp. 123-38. Accra: Woeli Publishing Services. Hammer, M. and J. Champy.(1993). Reengineering the Corporation: A Manifesto for Business Revolution. New York: Hamper Collins. Hartley, K. (1984). Why Contract Out: Contracting Out in the Public Sector. London: RIPA. Hartley, K., D. Parker and S. Martin.(1992). Organisational status, ownership and productivity', Fiscal Studies, (May). Haruna, F.P. (2003). Reforming Ghana's public service: Issues and experiences in comparative perspectives, Public Administration Review, 63(3): 343-54. Helmsing, A.H.J. (1997). Local government and finance: An introduction, mimeo, Institute of Social Studies: The Hague. Helmsing, A.H.J. (2000). Decentralisation and Enablement: Issues in the Local Governance Debate. Utrecht: University of Utrecht. Hope, K.R. Sr. (1997). African Political Economy: Contemporary Issues in Development. Armonk, NY: M.E Sharpe. Johnson, E. (1990). Effects of competition within local government services, mimeo, Institute of Local Government Economics, Stockholm: University, pp. 1-10. Kay, J.A. and D.J. Thompson. (1986). Privatisation: A policy in search of a rationale, The Economic Journal ,96 (March): 18-32. Keating, M. (1998). Public Management Reform and Economic and Social Development. Paris: OECD. Kerley, R. and D. Wynn. (1991). Competitive tendering - The transition to contracting in Scottish local authorities, Local Government Studies, Sep/Oct, 33-51. Kettl, D.F. (1993).Sharing Power: Public Governance and Private Markets. Washington, D.C.: Brookings Institution. Kettl, D.F. (2000). The Global Public Management Revolution: A Report on the Transformation of Governance. Washington, D.C: Brookings Institution Press. KMA-WMD .(2000). Waste Management Department Annual Report, 2000, KMA, Kumasi. Langton, S. (1983). Public-private partnerships: Hope or hoax?' National Civic Review , 72(5): 256-61. Linder, S.H. (1999). Coming to terms with public-private partnership: A grammar of multiple meanings, American Behavioural Scientist, 43(1): 35-51. Martin, B. (1993). In the Public Interest? Privatisation and Public Sector Reform. London: Zed Books. McDavid, J.C. (1985). 'The Canadian experience with privatising residential solid waste collection services, Public Administration Review, (Sep/Oct). Mueller, D.C. (1989). Public Choice II. Cambridge: Cambridge University Press. OECD .(1997). Managing Across Levels of Government. Paris: OECD.

New Public Management Paradigm

103

Olowu, D. (1999). Public service Delivery. In Ladipo Adamolekun (ed.), Public Administration in Africa. Oxford: Westview Press. Osborne, D. and T. Gaebler.(1992). Reinventing Government: How the Entrepreneurial Spirit Is Transforming the Public Sector. Reading, MA: Addison-Wesley. Pollitt, C. and G. Bouckaert.(2000). Public Management Reform: A Comparative Analysis. Oxford: Oxford University Press. Prager, J.(1994). Contracting-out government services: Lessons from the private sector, Public Administration Review, 54(2): 176-84. Proust, E. (1997). Implementing the contract state, Australian Journal of Public Administration, 56(3): 132-34. Rakodi, C. (2003). Beyond public failure and private success: Disentangling theory, practice and outcomes in the provision of urban environmental services. Paper presented at the N-AERUS conference, Paris, 15-17, May. Savas, E.A. (1982). Privatising the Public Sec.tor: How to Shrink Government. Chatham, N.J: Chatham House Publishers. Schneider, K. (1992). US cities waste in its contracts, New York Times, 2 December 1992. Spiegel, H.B.C. (1981). The neighbourhood partnership: Who's in it? Why?, National Civic Review, 71(10): 513-21. Starks, M. (1991). Not for Profit, Not For Sale: The Challenge of Public Sector Management. New Brunswick, NJ: Transaction Books. Stren R.E. and R. White (eds.). (1989). African Cities in Crisis: Managing Rapid Urban Growth. Boulder and London: Westview Press. UNCHS .(1993). Public/Private Partnership in Enabling Shelter Strategies. Nairobi: UNCHS. World Bank.(1981). Accelerated Development in sub-Saharan Africa: An Agenda for Action. Washington, D.C.: World Bank. World Bank. (1989).Sub-Saharan Africa: From Crisis to Sustainable Growth. Washington, D.C.: World Bank. World Bank. (1991 ). Urban Policy and Economic Development: An Agenda for the 1990s. Washington, D.C.: World Bank. World Bank. (1994). World Development Report 1994 Infrastructure for Development. Oxford and New York: Oxford University Press. World Bank. (1995). A Strategy for Restoring Urban Infrastructure in Nigeria. Washington, D.C.: World Bank. World Bank.(1997). The World Development Report 1997: The State in the Changing World. Washington, D.C.: World Bank. World Bank .(2001). Human Development Indicators, 2001: Drawn from World Bank Africa Database, Washington D C.: World Bank. Zanu, S.Y.M .(1999). Implementation of the environmental sanitation policy: A programme of action, Selected Speeches and Papers: Sixth Annual Conference of District Chief executives, Ho, Volta Region: Min. of Local Government, Accra (Sept 5-10, 1999).

Improving Urban Service Delivery

105

6 Improving Urban Service Delivery: Water Sector Reform In Ethiopia And Its Impact For Addis Ababa Meine Pieter van Dijk*

1. Introduction Poor people often have no access to key services such as water and sanitation (watsan) in acceptable quantity and quality. Urban service delivery is a task for the urban manager, who can sometimes delegate it to a municipal water, electricity or infrastructure company. Even in that case, the overall responsibility and the supervision and regulation will be part of the job of the local urban managers. This chapter deals with urb?n service delivery in Ethiopia in general and in particular in the capital Addis Ababa. The main thrust is: Public service delivery in general the water supply and sanitation for urban, peri-urban and rural communities in particular has been in poor shape in developing countries. The expectations were that decentralization and regional development would change the situation for better. However, despite the efforts made towards decentralization, our analysis of the situation in Ethiopia shows that there have not been improvements in the water supply and sanitation sector. In this case, experiences with public service delivery in other African countries and South Asian countries are not different.

2. Assessing the Quality of Urban Service Delivery Public urban service delivery of water and sanitation of most developing countries has been weakening over years. Quality of urban service delivery could be understood in terms key concepts such as: 1. Specificity; 2. Private sector involvement; 3. Gaining experience with small projects; 4. Analyzing existing experiences; 5. Benchmarking: measure the performance. Before we proceed to discuss the situation in Ethiopia, a brief explanation of these terms may facilitate the ground for common understanding. These terms will now be explained briefly, since they are crucial for a discussion about the

* Institute for Housing and Urban Development Studies (IHS) and Economic Faculty, Erasmus University Rotterdam, P.O. Box 1738 H12-29 3000 DR Rotterdam. The Netherlands Tel. Univ. 31104081870 fax 31104089153 e-mail: [email protected] or [email protected].

106

Meine Pieter van Dijk

quality of urban services and will lead to an indication of the need to undertake a water sector reform program in Ethiopia.

2.1.

Specificity

The implementation of urban service delivery varies from one service category to another and hence knowledge and experience of the specific sector is necessary. Although, the practice is taking place at the management level, cross-sector knowledge and experience in deed would help to analyze the problems under discussion and lends hand to come up with appropriate solutions. In the literature, the most common examples of urban service delivery are: a. Water and drainage; b. Electricity; c. Roads; d. Waste collection and treatment; and e. Other urban infrastructure and services.

2.2.

Private Sector Involvement

There are strong arguments to involve the private commercial or non-commercial (NGO) sector. A distinction should be made between economic arguments, ideological arguments, and historical practices in different countries. It is also important to define privatization in the broad and in the narrow sense of the word. Privatization in the broad sense implies promoting private sector involvement in service delivery but not necessarily changing the ownership of the utility. Outsourcing, management contracts, and concessions are ways to increase private sector involvement without changing the ownership. In the narrow sense of the word, privatization refers to ownership transfer by selling the shares (divestiture) or selling the utility to a private party. We like to use the term divestiture for outright sales of public property and privatization for other forms of private sector involvement or participation. Roth (1989) provides an overview of the historical practices and theoretical arguments (also Van Dijk and Schulte Nordholt, 1996).

2.3.Gaining Experience with Small Service Delivery Improvement Projects Countries often conduct experiments on a small scale, but the results have not always been analyzed systematically (Van Dijk, 2004), with the intention to apply the outcomes on a larger scale in other cities. However, the decision China has made to adopt and test a different approach to service delivery improvements on a small scale seems interesting. In most countries there are experiments on a small scale, but the results have not always been analyzed systematically (Van Dijk, 2004), with the intention to then apply the insights on a larger scale in other cities.

Improving Urban Service Delivery

107

2.4. Analyzing Existing Experiences with a Comparative Studies Framework We have always encouraged our students to look for the existing cases of urban service delivery. i The New Public Management theory provides a better framework for such a comparison. For an application of this framework to a number of successful water utilities in different countries see Van Dijk and Schwartz (2003). The results are presented in Schwartz and Van Dijk (2004) and partially used in this chapter.

2.5.

Benchmarking: Measure the Performance

Benchmarkit]g is a technique of comparing companies or utilities on a number of key indicators. It is good to use different measures of performance, because the results of reforms should be clear and the comparative performance of different utilities within the same reform program provides an interesting perspective. What explains the differences in performance? Involving the private sector does not assure unlimited supply of the service at the desired price and quality, thus, monitoring and benchmarking are good ways to assess the results. In the Netherlands, benchmarking is undertaken by the association of water companies on a voluntary basis (VEWIN, 2001 and 2003). The analysis of these data shows significant differences in performance between different actors (Van Dijk, 2003).

3. The Ethiopian Administrative System and Decentralization In December 1994, a new Ethiopian Constitution was promulgated. National and regional elections were held in May and June 1995. The Constitution has brought the establishment of a decentralized federal system of government. It provides for the existence of fairly hornogenous ethnically and linguistically defined groups of people under separate federal units with their own political and administrative setup. The Constitution· retains state ownership of the land, while recognizing user rights and opening the possibility of leasing land. The Regional State Council is the highest organ at the various regional states of Ethiopian region. The Executive body, which is also called the Executive committee, comprises of 24 Cabinet members and led by the Chief Administrator of the regional state. The major professional institutions are the social sector, the economic sector, the legal affairs and the Peoples militla and security sections. This description does not tell us exactly which responsibilities have been put at which level In fact, the distribution of responsibilities may differ from state to state, as in the case of India. A systematic study could help to clarify this. We found, for example that in the SNNP water is the responsibility of the state authorities, while in the east (Harar) it is a municipal affair. Similarly, in most regions municipalities are not allowed to issue bonds, while most state banks can do so.

108

Meine Pieter van Dij k

Is Ethiopia really a federal system with decentralization to lower levels of government? The Reporter (28-1-2004) mentions the de facto non-existence of federalism in Ethiopia, given the repeated interventions of central government in regional states affairs. For example, the government is controlling the major towns and their financial activities. The Prime Minister of Ethiopia keeps time and again telling his ministers that: the utility companies of Ethiopia should bear some semblance of effectiveness following a sweeping administrative reform" (Capital: 25-1-2004). In particular, six ministries (including the Ministry of Infrastructure) have been selected for reforming dysfunctional institutions. The reform process is designed to introduce efficient and speedy service from a government that is to be reengineered. 11

Figure 1: Finance and management dimensions of different forms of PSI in the water and sanitation sector

~ C cc C iI

-~

::0 :::,

CL

Private

Public Management

Source: Janssens and Baietti, (2002).

Improving Urban Service Delivery

109

The objective of the reform is to bring a managerial transformation that help service provision companies become more efficient, competitive and customer friendly. However, the problem with public policy in Ethiopia is that there is no space for private sector in the public services delivery. Capital (25-1-2004) concludes that, "many companies are still under state control and management, despite their being out of tune with the times, incurring losses or simply failing to be profitable ... [The reform] has failed to give them a real managerial shake-up to make them competitive or efficient". If we look to public service delivery in terms of private or public ways of managing and financing utilities (see Figure: 1), Ethiopia is very much on the left of the continuum. Reform would mean moving from the left to the right and for the World Bank it also means moving from the bottom to the top of the figure. In terms of type of utility it would mean, going from a municipal department, through ring-fenced and water boards to the water PLC. These are different ways of achieving better performance in the utility sector. Another argument for such a trend of moving to the right and the top is that the reforms on progress cannot so easily be reversed or stalled, if you are more on the right in this figure.

4. The Watsan Sector in the Capital: Addis Ababa Addis Ababa came into existence in 1886 and become the capital it:J 1892 by Emperor Menilek (r. 1889-1913) and became gradually the largest city of Ethiopia. The near recent practice indicates that the city was given the right to collect and use income derived from leasing of land within its jurisdiction through proclamation 80/1993, and has subsequently obtained a chartered status with all its benefits through proclamation 87/1997. According to the Central Statistical Authority, it has generated about one quarter of Ethiopia's GDP in 1994. The city is situated in the centre of the country and sprawls over an area of 28,000 hectares. The central part of the city, the so-called metropolis is about 7,500 km2, and is considered crucial in all aspects. The current population is estimated about 3 million people though some put the figure at much higher level. Although the Addis Ababa Water and Sewerage Authority (AAWSA), has often said to have the status of an authority and hence an example of what Rondinelli and Ruddle (1978) would call 'administrative decentralization through de!egation', the AAWSA is not really autonomous. As it has no power to fix its own tariffs, for, example, it gives the service with loss and, thus, depends on other bodies to cover the costs, although customer orientation has improved. Tariffs are extremely important, just like customer orientation. In fact, one may lead to the other. Successful utilities have generally been able to increase their tariffs and pay more attention to the needs and problems of their customers. Asset management is important for the water sector, because the assets are investments with a horizon between 50 and 100 year. AAWSA depends on the municipality and foreign donors for financing its investments. If they are not

110

Meine Pieter van Dij k

properly maintained, they may generate no return at all. If there is a need for an extension this may be a quantum jump. There is often a grant element in the. financing. Cost benefit analysis usually doesn't take into account a period longer than 30 or 40 years. In the UK, Coopers and Lybrand developed a procedure for the valuation of such assets.

5. Water Sector Reform in Ethiopia AAWSA is an 'autonomous' government body in charge of providing water and sewerage services for the city of Addis Ababa. In the old days, springs were the only sources of drinking water. In 1960, the Gefersa Dam was built and subsequently the Legedadi Dam and treatment plant were put in operation. Two additional resources are the Dire Dam and the Akaki well fields. A program of spring rehabilitation and borehole drilling has enabled the city to supply water to outlying areas (Desta, 2003). AAWSA's poor performance can be reflected by the fact that it can satisfy only 60 · percent of the current demand. It is also shown by its Unaccounted For Water (UFW at least 35 percent) and the low coverage (only 2% of the urban population is linked to) of the sewerage system. It is also shown by the lack of transparency for customers or researchers, who need to find out what is really going on in the current water sector reform process in Ethiopia. To improve urban water supply a clear reform program is desired, with changes at the policy and the utility level. The formulation and implementation of such a program need to be supported by the municipal and local government authorities. To improve urban water supply, a reform program was started. Desta (2003) has made attempts to see the nature and assess the progress of the Water sector reform in Ethiopia. Although the literature available on program evaluation is very limited, water sector reform programs in developing countries are being given increasing attention. The elements of such a program include introducing: • institutional changes; • policy improvement; • financial set-up changes; • establishment of robust sector governance; and, • introduction of more efficient and professional management. A reform program in the water sector usually implies improved quality of the service delivery and better financing structure. The WSP (2002) of the World Bank describes the responsibilities related to water sector reform and the tools vital for these reforms: 1. Institutional changes; 2. Tariff reform; 3. Improvement of sector governance and regulation; 4. Professional management of the utility.

Improving Urban Service Delivery

111

We want to determine the elements used in the water sector reform programs and try to assess the results. It is important to analyze which factors have hindered or facilitated the formulation and implementation of the program and which factors explain the eventual success or failure. An evaluation should be made in terms of effectiveness, efficiency and impact. Sustainability of the effort can be added as an additional factor. It should be realized that different objectives imply criteria for the effectiveness (the degree to which these objectives have been achieved). Desta (2003) notes a number of structural and tariff reforms and concludes that the current tariff does not allow AAWSA make the necessary investments that help improve its performance substantially. AAWSA cannot increase the tariffs sufficiently, but the new organizational setup gives more emphasis to water management and managing public relations. Describing the state of current organizational setup and tariffs and listing performance indicators, Desta (2003) shows that the degree of autonomy is very limited. The Board of Directors entrusted with powers of tariff setting and investment planning as well as the hiring and firing, and incentives of workers. These are clearly elements, which according to the NPM should be the responsibility of the management. Finally, she notes that a combination of technical and financial problems has affected the scope and quality of the service delivered by AAWSA. This, in turn, had a negative impact on the willingness of the consumers to pay for its services.

6. Water Sector Reform what it should be in Ethiopia The major research question for our inquiry on urban service delivery in general and drinking water supply in particular was: Why are some public utilities function well whilst other utilities perform poorly (Van Dijk and Schwartz, 2003)? The perspective taken in this project is to analyze this question in the framework of different approaches to the way in which public sector organizations are managed. The analytical framework focuses on the distinction between the bureaucratic paradigm of management in public sector organizations and the 'New Public Management' approach. The latter can be separated in NPM principles and NPM instruments, as illustrated in Table 1 and Figure 2, respectively.

112

Meine Pieter van Dij k

• Ies an d correspon d"mg NPM.ms t rumens T abi e 1 NPM prmcIp t

NPM principles

NPM instruments

1.

6. 7. 8. 9.

2. 3. 4. 5.

Autonomy Accountability Customer orientation Corporate culture Market orientation

10.

Influencing tariffs Using contracts A service oriented company Incentive programs Benchmarking/inventory PSP

The hypothesis formulated is that well-performing utilities are likely to be subjects to considerable accountability mechanisms that involve explicit performance targets. The study also looks at the corporate culture of the utility. Finally, introducing a stronger market-orientation, which aims at reaping the benefits of (quasi) competition between either suppliers working outside the utility, or between outside suppliers and internal departments, or between internal departments of a utility is recommended and promoted. A range of contracts, (quasi-) competition between (public) agencies, inter-agency fee-charging, and outsourcing enhance increased market-orientation at a public sector organization (Burki and Perry 1998). An increased market orientation promotes efficiency and stimulates innovation in service provision. The assessment of market-orientation of the utilities is carried out by taking into account: • The use of outcontracting/outsourcing in service provision; • The use of bench-marking practices in the utility; and, • The use of other mechanisms to introduce (quasi)-competition. Dependency on government agencies rather than the people they intend to serve to secure the funds needed to carry out their activities, and 'captive' customers with no range of options (Osborne and Gaebler 1993), are common characteristics of most public utilities. The fact that their being dependent on government agencies and their having a 'captive' clientele, make utilities more concerned to their financiers than their customers. Customers are also not 'empowered' to hold accountable the service provider, because they are not 'directly paying' fo'r the services they received (Barzelay 1992). Figure 2: Framework for analysing the relationship between the utility and its environment

Autonomy of the utility

~

Performance

~ targets

~

Accountabilit

~ y for results

Improving Urban Service Delivery

113

Figure 3: Conceptual Overview of Accountability between Utility and its Environment Government

Financing Institutions

Regulator

Owner

Utility

Customers and Customer Organizations

Non-governmental Special Interest Groups

7. Ideas about what the Water Sector Reform in Ethiopia should be Like The World Bank has been looking for a strategy to engage again in a larger scale with the public water sector in developing countries. That means working with national and local governments and assuming that the utilities have implemented some reforms in the past, finding out: 1. What prompted the process in the first place and what were the critical moments? 2. What was the sequence in which the reforms were implemented? 3. Was the sequence of the implementation knowingly planned? 4. If so, what was the reas0n for this sequence of implementing the reforms? 5. What can we learn from the earlier efforts to reform? Commencement of water sector reform, for example after macroeconomic stabilization or restructuring programs, usually intends to bring changes at three levels: at the level of the sector, in ownership, and finally, the internal reforms in the utility. Our research took place at the entity level. The internal reforms need to be spelled out in a lot of detail for the public sector reform, if you don't want the private sector to do the job for you in a private sector participation framework. It means you have to follow in the foot steps of the public sector to a large extent

114

Meine Pieter van Dij k

rather than the measures the private sector would take, when the private sector would be asked to step in. Ad I: There need to be opportunities to reform. What can we say about triggers for change? In Uganda, the threat of privatization was important. Management expected a lease contract and wanted to show that they could also perform well. In Singapore, the shortage of water led to considering the option for wastewater treatment in an innovative way. Similarly, in Scotland, the price cap forced them to improve efficiency and develop alternative sources of income. So the first conclusion concerns the importance of external pressure: 1. A water shortage crisis may force utilities to look for other approaches to the problem; 2. A serious case of pollution or health problems can play this role; 3. The threat of privatization can bring about a change; 4. Too high prices trigger political unrest, which may force politicians to change current policies. Ad Ill: Sequencing could for example take the following forms: start with new management, training of staff, increase the tariffs, and build up continuing pressure to perform. Subsequently, more customer orientation, the development of internal procedures and systems to measure performance and, finally, more market orientation will do the job. There is, however, the problem of conflicting time horizons: the new management, the politicians and the donor all have different time horizons.

It is also important to look at the capacity requirements for reform: a certain size of the utility makes it easier to find the necessary capacity. Also important is the need for funds to train the staff. Capacity building is also needed at different levels. Sometimes the utilities need at the town level trained low skill people, with manuals to do the work. There is significant difference between the single municipal water utility and the multi-utilities and multi-municipal cases, often concerning the whole country (for example in Burkina Faso, Tunisia, etc.). Ad Ill. A key message in the reform process is: separate the different roles in a utility. The main roles are ownership, oversight and service provision. After separation, it is necessary to define the relations between the different units. NGOs can help to monitor the process and play a role in the web of accountability that needs to be woven. To be able to turn around utilities, the director and the board need some autonomy. Below, a list of functions of the board and the managing directors (Table 2; use functions of the Board and directors column as checklist: does the board deal with ... and do the directors deal with ... ), is provided.

Improving Urban Service Delivery

115

Table 2 Functions of the board and the managing directors Functions of Board

Directors

Hire and fire executive managers Fiscal matters Health and safety of personnel Legal matters: to comply with the law Monitor performance and advise management Approves business plan

Implement the company policy Manage the institution Hire and fire lower level managers Prepare corporate plans and budgets Prepare financial accounts Assure the necessary finance

Corporate governance is the right term and the detailed picture of the functioning of the Board and of the management will be based on local legislation and the articles of association of the utility or company. It involves issues such as: a. Who nominates, candidates are suggested by ... , b. For how long c. Members and the Managing Director d. With which mandate? e. How many politicians, how many specialists f. Investments above a certain level have to go to the Board g. Who signs-off on price increases? Ad IV. The internal reform process would follow the following sequence: a. Find out what the public sector considers to be the priorities in drinking water supply; b. Identify the necessary reforms and link them to what these reform packages would cost; c. Find out how much money is available; d. Link different reform packages to different price tags and choose what you can afford, given available means; e. The package should be concentrated, follow the 20/80 rule, with 20 percent of the means you can achieve 80 percent of the objectives; f. Make a business plan that.will eventually become .a three year rolling annual plan g. That requires physical concentration of your activities, choosing: going for standpipes, more connections or more pressure (choose one that is urgent) and refuse exceptions, which jeopardize your planning (the connection of the sister-in-law of the mayor); h. Start higher tariffs as soon as you have something to show; i. For public companies it may be necessary to re-concentrate activities before you decide on outsourcing certain activities to private or public

116

Meine Pieter van Dijk partners (the South African case of metering, billing and complaints which are still in the municipality); j. Major Instruments in the internal reform process are: k. Internal benchmarking' I. Internal contracts m. Agreements linked to easily measurable indicators n. Enforcement of contracts and agreements o. lncentivy the agreements and contracts p. The challenge is to sequence these measures and then visualize reform tracks, which may be different, depending on the point of departure, the identified bottlenecks and priorities; q. Then locking in the reforms is extremely important.

Ad V: We found an enormous variety of cases and that there is not one solution for all water companies. Much depends on the past achievement, where they are in the reform process and what political leadership and commitment there is improve public water supply. Examples as far as success are influenced by such accidental factors: 1. 2. 3.

In Uganda, there was the threat of privatization and a strong managing director. In Scotland, the importance of having a price cap forced the water utility to go for efficiency improvements. In Singapore, the lack of resources stimulated experimenting with reuse of wastewater. Hence, an impulse to innovate.

Another factor that surfaces a number of times is what we would call the importance of external pressure. Such external factors can bring about the necessary turnaround because people are ready for the radical reforms that need to be introduced. We also found that the utilities should get some autonomy. Finally, the commitment to reforms is vital. The change can come from inside or from outside, but the commitment to change should be inside and outside the utility and here urban managers play an indispensable role. Examples of turnaround utilities are found in Uganda and in Haiphong (Vietnam). Sometimes a lack of interference by local politicians in the business is enough to allow for the necessary reforms. For example, Singapore functioned well for a long time, although it was only a municipal department. It did not even corporatize to separate its finance from the regular municipal finance. However, the authorities did not interfere! Linked to this is the need to eventually push autonomy down the line, which boils down to further decentralization. For example, the dealing with customer's complaints can be pushed down the line and there are examples where this worked. The people who need the feedback would get it and will be able to deal with it. This brings out an important point that is supported by a lot of research, namely if management introduces a change it

Improving Urban Service Delivery

117

is important to explain it, to introduce it properly to the personnel and to allow them to make it their point. In certain utilities (Haiphong, for example) the management is very centralized, which may work with a director who knows the company for a long time, but it may be a constraint for his successor. It is interesting to note that in Sansa (Brazil) a lot of the decisions were suggested by working groups. This seems a good way to involve different layers in the organization. Autonomy allows a utility introducing incentives and competitive salary scales. Therefore, it is certain that utilities set up as companies do better than the others. These are, of course, companies and start behaving like ones. Another way of formulating this point is that there should be a weak link between the owners and the management. Preferably there is a business plan, with a vision, a strategy and concrete objectives, which guides the management. Leadership and the ability to recruit qualified people is also an important factor contributing to the success of the utilities studied. This applies for Uganda since 1997, for Haiphong from 1993-2003 and also for the case studies in Scotland and even Philadelphia (USA). The role of the donors can be very positive. They put a lot of money in the water sector, within a water sector reform framework put in place by the World Bank. If 80 percent of the investments are donor financed the question is whether this will be sustainable in the future. The reform process often takes a lot more time than is expected. In Haiphong it took ten years. It is certainly not going to happen (and become sustainable) in 3 or 4 years, the normal period of a project. This leads to the importance of the phase where the utility is in and the sequencing of reforms. A sequencing of the proposed recommendations is necessary. For example, tariff increases have to come before more private sector involvement (PSI or PSP). PSP is compatible with the public model if forms like outsourcing, concession contracts and build operate and transfer (BOT) formulae are used. We need to come to a road map where to go with public utilities. We would like to provide practical tools through a comparative analysis of the cases. This involves a prioritization and indications of what to do. Finally, the stability of management is important. In particular, the Latin American success shows the importance of the stability of management. In Mexico and Brazil, most senior staff is replaced when there is a change in government. When the new mayor has been elected, the board and senior management is sacked! The sustainability of reforms is a worrying issue. This is not just in developing countries. Even in the Philadelphian case a lot of senior managers threatened to leave the company and the question is whether the utility will be able to attract qualified new people. The question becomes how you can lock in reforms: a. Multi-ownership; b. Certification procedure for management; c. Transparent recruitment procedures;

118

Meine Pieter van Dij k d. e.

f.

Link loan to performance criteria (World Bank in Haiphong); Keep politics away from the water entity, use the Board as a buffer. It would create a weak link with the owners, namely only through the business or annual plan: with general objectives and specific targets); Use working groups (Sanasa) to prepare decisions on for example tariffs. · This is one type of decentralization and employee involvement

8. Some Recommendations The analysis can lead to some recommendations for sector reform in the drinking water sector in Ethiopia: 1. Stability of management can be improved by going for multi-ownership. If several public bodies own the shares and are represented on the board it becomes more difficult to change management after an election. In Mexico, they are even thinking about a certification procedure for management in the water sector, to keep political nominations from spoiling the reform perspective. Along this line, the State Commission for water would certify certain people for such jobs. 2. Introduce effective benchmarking, sometimes sometimes at the level of the technical variables.

informal

benchmarking,

3. Develop more performance related rewards for the utility and the people working in the utility. Given the current very low levels of salary, every possible improvement may lead to more effort to achieve better service delivery. 4. Promote multi-ownership and a clear separation of the owner and the operator to avoid political interference and changes in the policy after elections. 5. Use the principles of the NPM theory and translate those into instruments and reform actions. It is not necessary to insist on implementing all elements of the theory in an un-integrated way. The NPM is a powerful way of convincing utilities of the need for change, and also suggest tools to achieve the objectives. 6. Involve the staff of the utility in the price fixing process, because they are the only ones knowing the real cost. 7. Plan the sequence of reforms carefully and improve stakeholder participation. 8. Solve the governance issues by separating ownership, oversight and the service provider and by clearly defining the role of the Board and of management. 9. Promote a business like culture with business plans, a vision and mission, focus on what you try to do, tell your staff where you want to go, devolve decision

Improving Urban Service Delivery

119

making as far as possible, etc. If it is set up as a company it can also be mapped against a normal business: is a Business plan produced, what are the standards you aim for? Are the directors appointed in a normal way, what is the level of services, etc? 10. Use the board with a few hard controls: strategy, business plans, financial supervision and performance monitoring.

9. Conclusions Urban service delivery for the poor gets more attention since the Millennium development Goals have been formulated. The relevant ones are listed in Box 1. Box 1: Relevant Millennium Development Goals 1. 2. 3. 4.

Improve the situation of slum inhabitats Decrease urban poverty Halve the number of people without access to clean water by 2015 Halve the number of people with no access tosewers by 2015

The Millennium Development Goals (MDGs) are very clearly putting a challenge for us to halve the global incidence of poverty and they specify what this means for the water and sanitation sector by 2015. Ethiopia has a long way to go to achieve these goals. Urban service delivery can be improved by putting poor people at the centre of service provision, by enabling the poor to monitor and discipline service providers, by amplifying their voice in policy making and by strengthening the incentives for providers to serve the poor (World Bank, 2003). Designing proper water sector reform programs is an important step in the right direction. An analysis of experiences in different countries helps to bring out the critical factors to look at.

References Barzelay, M. (1992). Breaking through Bureaucracy: A New Vision for Managing in Government. University of California Press: Berkeley. Burki, S. and Perry, G. (1998). Beyond the Washington Consensus: Institutions Matter. World Bank Latin American and Caribbean Studies, Washington D.C.: World Bank. World Meteorological Organization. International Conference on Water and the Environment (ICWE), Development Issues for the 21st Century, 26-31 January, 2002, Dublin, Ireland, Geneva. Desta M.K. (2003). Assessment of the Water sector reform for AAWSA in Ethiopia. Elfdt: UNESCO-I HE Institute for Water education. Dijk, M. P. van. (1998). Effects of government policies on small enterprises in Zimbabwe, A case of good governance? African Development Perspectives, 6, 1-13.

120

Meine Pieter van Dij k

Dijk, M. P. van. (2003). Is Nanjing's concentration of IT companies an innovative cluster? In D. Fornahl and T. Brenner (eds.), Co-operation, Networks and Institutions in Regional Innovation Systems. Cheltenham: Edward Elgar, (173-194). Dijk, M.P. and N.G. Schulte Nordholt (ed.).(1994). Privatization Experiences in Africa and Asia. SISWO: Amsterdam. Dijk, M.P. van and R. Rabellotti (eds.). ( 1999). Clusters and Networks. London: F. Cass. Dijk, M. P. van and H. Sandee (eds.). (2002).lnnovation and Small Enterprises in the Third World. Cheltenham: Edward Elgar. Dijk, M. P. van and K. Schwartz. (2003). A series of case studies of well performing drinking water companies, A research project for the World Bank. A paper for the PoWER project at UNESCO IHE in Delft, October 15. English, P. and G. Henault. (eds.). ( 1995). Agents of change. Ottawa: IDRC. Janssens,.J. and A. Baietti. (2002) Internal World Bank publication. Washington: IBRD. Ndimo, D.N. (1998). The potential for private provision in urban water supply and solid waste in Mbarara Municipality in Uganda, Rotterdam: UMC Thesis. Rondinelli, D.A. and K. Ruddle. (1978). Urbanization and Rural Development: A Spatial Policy for Equitable Growth. New York: Praeger. Roth, G. (1989). The Private Provision of Public Services in Developing Countries. Washington: World Bank. Thynne, I. (1998). Government companies as instruments of state action, Public Administration and Development, 18, 217-228. VEWIN. (2001). Reflections on Performance 2000. Benchmarking in the Dutch Drinking Water Industry. Rijswijk: VEWIN. VEWIN, (2003), http://www.vewin.nl. World Bank .(2003). World Bank Development Report 2004: Making Services Work for Poor People. New York: Oxford UP. WSP. (2002). New designs for water and sanitation transactions. Washington, Water and Sanitation Program. _ _ _ .(2002a). Uganda Rural. Washington, Water and Sanitation Program. _ _ _ .(2002b). Second Country. Washington, Water and Sanitation Program.

Notes i A number of theses at ECSC, RLDS and the UMC in Rotterdam have analyzed experiences in different African cities with urban service delivery (Ndimo, 1998).

PART III

RURAL-URBAN LINKAGES AND INEQUALITY

Regional Socioeconomic Wellbeing in Ethiopia

121

7 Some Indicators of Regional Socioeconomic Wellbeing in Ethiopia: -Trends and Patterns (1995 2001) Bekure Woldesemait*

1. Introduction Development efforts undertaken at the national level do not necessarily ensure regional equity in the distribution and share of development opportunities. While a national economy as a whole prospers it is possible that decline and/or stagnation occur in some regions. Public intervention with appropriate regional policy can help avoid unjustifiable regional inequalities. On the other hand, citizens claim, and are justified to do so, equitable share of the benefits of the development taking place in their country. When their claim is not satisfied, enormous negative political and social consequences may take place. Based on this premise, the paper attempts to assess the regional development processes and levels in Ethiopia .using . some indicators of socioeconomic development. The socioeconomic indicators applied are demographic, agricultural, manufacturing, and social data th;3t are generally accepted by development scholars as manifestations of level of development. The regional framework used is the federal regional structure of the country (Fig.1 ): After briefly discussing the concept of development, the paper attempts to: • Examine the regional policies that explain the proc'ess and pattern of development attained; • Assess the level of development achieved by the federal regions; and, • Review the process _of development as indicated by growth indices of each federal region during the last six or seven years.

• Department of Geography &

Environmental Studies, AAU.

122

Bekure W oldesemait

Ethiopia: Regional States AA = Addis Abab DD = Dire Dawa HR = Harari

Fig. 1

2. The Concept of Development Regional socioeconomic wellbeing is a result of development or underdevelopment. Hence, a brief discussion of what is development is in order. Development students distinguish between two interdependent, nonetheless, slightly different concepts: economic growth and development. 2.1. Economic growth is commonly defined as the percentage annual change in the national income of a country, generally expressed in per capita ·income. It is the steady process by which the productive capacity of the economy increases over time bringing about rising level of national income. Hence, a rise in real per capita income from 1,000 birr five years ago to 1,500 birr in the current year represents economic growth. The implication is, there is economic growth without, necessarily, being accompanied by development. 2.2. Development, in addition to economic growth, creates conditions conducive to the growth of people's self-esteem, and freedom of choice. Societies establish social, political, and economic systems to regulate their affairs, and in principle have confidence in the impartial functioning of the systems they have created and are in control of. They have reasonably wide latitude of choices in appointing managers of the affairs of the State, in the consumer and intermediate goods

Regional Socioeconomic Wellbeing in Ethiopia

123

work and residence. Thus, development is more than economic growth as it brings human dignity, respect, and freedom (Todaro, 1986). Accordingly, an individual. citizen believes that he is neither superior nor inferior to anyone else in front of the law and expects equal treatment in any part of the country. The institutes in which establishment he has participated, he believes, render their services impartially.

3. The Policy Framework Befekadu and Berhanu (2000) describe EPRDF's economic strategy as difficult to categorize compared with that of the PMAC/PDRE - its predecessor. They continue to say, while EPRDF has grudgingly accepted capitalism, it by no means believes that pure market allocation of resources without government intervention is a viable option to Ethiopia (Befekadu and Berhanu 2000). While accepting the difficulties of identifying clearly stated policy framework of development, we propose that objectives, areas, minimum capital requirement, as well as participants in investment stated in the relevant proclamations, are determining factors of the pace, pattern, and regional distribution of development, and could be cited as policy framework. We shall briefly discuss each of them in the following few paragraphs.

3.1. Objectives of Investment Initially, the objectives of investment were stated as peace/stability, democracy, and rehabilitation. To a government that had just taken power by force after a long civil war that resulted in considerable destruction of infrastructure and trying to establish its rule, the stated objectives were, I think, realistic. The objectives of investment stated above, changed to 'accelerated and balanced development' in subsequent proclamations. It was not clear in Proclamation No. 15 of 1992 weather 'balanced' referred to sectors, regions, or both. In proclamations No. 37 and No. 280 - the expression 'balanced and integrated regional development' clarified the cloudy statement in proclamation No. 15. This is particularly relevant to our discussion of regional socioeconomic wellbeing. However, what the tools, instruments, and strategies for attaining balanced and integrated regional development are is not clear from the proclamations we have cited above. Befekadu and Berhanu (2000) mention budgetary allocation as an instrument for achieving regionally balanced and integrated development. The regional budget allocation criteria appear to have been biased towards some regions at the beginning as one could conclude from the relevant session of the federal parliament. Representatives from 'lagging' regions complain that they are not provided with sufficient support to enable them catch up with relatively advanced regional states (e.g., discussion during the report to the parliament by the Minister of Finance of the 1997 EC proposed budget).

3.2. Participants in Investment PMAC/PDRII: made the government the sole promoter of development - regional or national - and discouraged private participation. TGE/EPRDF, on the other hand, recognizes three participants in investment - the government, private domestic, and private foreign. Each can invest on its own or jointly with the other

124

Bekure Woldesemait

in sectors of the economy variously specified at different times. Thus, there are certain areas of investment that are reserved exclusively to the government or domestic private and foreign investors. However, the boundaries separating the three participants have been changing or modified since TGE/EPRDF came to power. The major reasons for the changes seem to be the fact that the policies have not been strong enough to attract private investors either of domestic or foreign origin as promulgated, and that international institutions like the World Bank put pressure on the government to give more room for the private sector. It is noticeable that the government is· gradually reducing its ,exclusive ·area of investment in favor of private domestic and foreign investors.

3.3. Areas of Investment The areas of investment assigned to the participants have been experi.encing change with time as mentioned above. Hence, the boundary has been rather fluid. A wide range of activities have been exclusively reserved to the government at the initial period. In subsequent documents, the areas reserved exclusively for government investment have considerably diminished. For instance, the 1992 · proclamation reserved defense industries, large scale electric production and supply, postal and telecommunication services, large-scale air, rail and marine transport, insurance and banking and import and export trade of strategic materials for government. In the 1996 proclamation, insurance and banking, import and export trade of strategic materials were completely scrapped out of the list of activities exclusively reserved for governm.ent. Air transport up to a ceiling capacity of 20 passengers and a cargo capacity of 2,700 kg and electric production and supply with a ceiling capacity installed capacity of 25 megawatts were opened to domestic private investors. Though the 2002 proclamation left many areas of investment to be open for private domestic and foreign investors in accordance with regulations to be issued by the Council of Ministers at later dates, only electric production and supply through the integrated system remained in the list of activities reserved for government. The activities relinquished by the government have gone either to domestic or foreign private investors.

3.4. Capital Requirement Similarly, the capital requirement for domestic and foreign private investors has gradually changed. For instance, the 1992 proclamation set a minimum amount of capital of US $ 500,000 for a foreign investor undertaking a project on its own or jointly with a domestic investor. The 1996 proclamation reduced the minimum capital requirement for joint investment to US $ 300,000. The 2002 proclamation further reduced the minimum amount of capital to US $ 100,000 for a foreign investor investing on its own and to US $ 60,000 if investing jointly with a domestic investor. The minimum amount of capital for a private domestic investor was specified as 250,000 birr in the 1992 proclamation. Subsequent proclamations mentioned no minimum requirement.

Regional SocioeconomicWellbeing in Ethiopia

125

3.5. Agricultural Development-Led Industrialization A strategy accompanying the policy framework we have discussed above is Agricultural Development-Led Industrialization (ADU). It stresses that the country's industrial development must lean heavily on processing agricultural raw materials. Implicitly or explicitly, the agricultural sector must lead the industrial sector. This has generated, as witnessed in many forums of socio-economic development conferences, considera_ble controversies. The government defends that this is the only realistic way to promote industrialization while others insist that a sector as weak as the Ethiopian agriculture cannot play a leading role in the development of the country. In conclusion, we can state, the objectives of investment, areas of investment, minimum capital requirement, and privatization listed in the relevant proclamations serve as the policy framework of development in general and regional development in particular. These elements of development policy have changed from severely limiting factors initially to relatively less limiting factors later on. This process of change took place through the gradual reduction of the areas of exclusive government investment in favor of the private investment both domestic and foreign, and diminishing of the minimum capital requirement for private investors. Accompanying the policy pillars is the ADU strategy, which has its opponents and exponents.

4. Socio-economic Indicators of Development Certain patterns (phenomena) are observable in association with levels of development because they trigger or/and are reflections of development. The most important ones are: (a) demographic; (b) industrial structure; (c) social; and, (d) types and patterns of energy consumption. We will examine some of these indicators, at times,· the changes they have undergone during the last six or seven years, and at other times as indicators of the level of development attained in the year 2001. Our choice of certain indicators among many, and the way we use them - as indicators of the process or level of development achieved is determined by the availability of data. However, it is hoped that this does not involve a significant compromise on the analysis and conclusion of the paper. The following sections will review processes and levels of regional development.

126

Bekure Woldesemait

5. Regional Development: Patterns and Process As stated earlier, we shall examine the level of development attained ·and. the process followed by the federal regions of Ethiopia.

5.1. Levels of Development of Federal Regions We shall attempt to assess the levels of development of the federal regions on the bases of demographic, agricultural, manufacturing and social indicators of development. 5.1.1. Demographic Indicators of Regional Development Demographic characteristics expected to vary with level of development are: age structure, birth and death rates, expectation of life, and rural-urban proportion. In general, less developed areas have youth population. With regard to birth and death, they are characterized by either high rates in both or high birth rate and relatively reduced death rate. Expectation of life is low compared with developed countries. The rate of urbanization is also low.

Since the differences in level of development between the federal regions are small, and are of recent origin, the first three demographic variables (age structure, birth and death rates, and expectation of life) do not manifest readily observable variations. In general, changes in these demographic aspects are slow to come by. What is readily observable is difference in level of urbanization by federal regions. Here, one needs to distinguish between city-based regions/units such as Addis Ababa, Dire Dawa, and Harari on one hand, and noncity based regions (eight) on the other hand. For each region the percentage of urban population is calculated and the regions ranked from 1 to 11 in order of level of urbanization. The standard deviation of the

rank values is calculated applying the model: a-

=

_i=_I_ _ _ •

=3.2. Then,

N

Z-scores, which indicate the number of standard deviations of a given rank are calculated; for instance, Addis Ababa, which ranks first is -1.7 on the z-scale. The

model for calculating z-scores is: Z

x-p

= - - . The

regions are then grouped on

(J"

the bases of the z-scores with the objective of minimizing intra-group differences (Table 1).

127

Regional Socioeconomic Wellbeing in Ethiopia . f I0n b1y regional state (2001) T a bi e 1 Leves of ur bamza Total Population

l

Urbanization

N C 0

~ u

(I)

·5, , .c: Q..

~

C:

C:

~

ci.

- - .c: C:

0

.c: 0

0

::0 (D

(/)

co



C:

:::J

~ ::i

~ ::i

'+-

(1)

.C)

~ ::i

(1)

0

·O

0

E ::J

u (J)

C:

0

Q..

0:::

D..

AA AF AM BG DO GM

2 7 10 6 3 4 1 9 8 11

HR OR

SM SN TG Source:

131

Q..

D..

3 7 11 4 2 5 1 10 8 9 5 6 CSA. (2000, 2002).

Q..

D..

3

0

en

I...

0

I

N

e

0

(.)

'{> N 'O

e Cl>

'E

0

-1.6 -1.1 HR -1.2 DO 0.7 -1.1 11 f.5 AA -0.3 5 -0.3 BG -0.1 2 -1.2 TG 0.5 4 0.5 GM 0.7 -1.6 1 AF 0.7 SM 7 0.9 0.9 8 0.7 OR 1.2 SN 9 1.2 1.5 -0.1 6 AM Statistical Abstracts ,(compiled & computed by the author).

10

8 24 32 15 7 13 3 26 24 29 17

The distribution of health facilities reveals both expected and unexp·ected results. Naturally, the three urban-based regions of Harari, Addis Ababa, and Dire Dawa are favorably provided. The fact that Benishangul-Gumuz, along with Tigray, makes group two is unexpected. That Gambella, Afar, and Somali regions along with Ormia belong to the third group also defies expectation. It is even more surprising that SNNP and Amhara regions, regions that are expected to fare well, form the last group. 5.1.5. Rating of Regions under Aggregated Development Indicators So far we have grouped the level of evolution of the regional states under each major economic sectors: urbanization, agriculture, industry, and social services. We present hereunder how the regions fare under all of them taken together.

132

Bekure Woldesemait

Table 6. Re ional develo ment under a ~

0

(.)

"'N I

:::0

('I)

cc

o· :::,

C:

0

~ C:

0

·5, Q)

0:::

AA

AF AM BG OD GM HR OR

SM SN TG

N

·c("O ..0

Q)

.....

C:

::i (.)

~ (.)

.....:::,

0

-~

..... ::::>

< Cl) "C 300 s:: 250 .s:: 200 i0 I.. 150 (!) 100 50 0 1994

--11--AF · ·AM BG

-¾-00 -e-GM -+--HR --OR SM 1996

1998 Year

2000

2002

SN TG ET

Table 12 and figure 6 show the following: (1) The process of development is visibly erratic especially for Afar and Somali regions. (2) Afar, Tigray and Somali regions have recorded the highest growth rates. For Somali and Afar regions, one of the reasons for high growth rates is the low of level of development at the initial year. (3) Intermediate growth indices are recorded by Amhara, Oromia, Gambella, SNNP, and Benishangul-Gumuz regions. Inclusion of BenishangulGumuz and Gambella in this group is due to low initial level of development. (4) Addis Ababa, Dire Dawa, and Harari - the three urban-based regions have achieved least growth rates. What has reduced their growth rates is the already high level of development attained at the initial year - 1995.

6. Summary and Conclusion An attempt has been made to assess the levels and patterns of development of the federal regions of Ethiopia using socioeconomic development indicators. The examination of the socioeconomic indicators by region between 1995 and 2001 reveals unequal levels and patterns of regional development. In terms of levels attained, the general low level of development of all regions recognized, Harari, Addis Ababa, Tigray, Dire Dawa, and Oromia, form the first group. Group two embraces Amhara, Benishangul-Gumuz, Gambella, Southern Nations, Nationalities, and Peoples Regions. The third group is made up of the Afar and Somali regions. Yearly oscillations characterize the process and pattern of development. The oscillations are more pronounced in the case of the less developed regions of Afar and Somali. Excepting the urbanization process, there is no visible narrowing down of development gaps for the rest of the. indicators.

140

Bekure Woldesemait

The overall development level is determined by the government policies regarding the choice of investment participants and minimum value of capital required to secure investment permit. The policies have changed overtime as and when the government recognized that the provisions in the policies formulated are after all inadequate. The disadvantage of such procedure is that considerable time elapses before a policy is replaced by a more attractive policy, delaying development progress. One of the important bottlenecks in this regard is the land policy - land remaining under government ownership. In addition to policies, there is the issue of capacity of the regions to manage development. The developmentally peripheral regions: Somali, Afar, BenishangulGumuz, and Gambella suffer from low capacity. Therefore, they are disadvantaged in the first instance, to mobilize resources, and in the second instance, to effectively use resources that are allotted to their regions. Also, the developmentally peripheral regions have less accessibility and low level of security. The low level of development of the infrastructure is a real handicap. They also have relatively limited influence on the central government to obtain federal resources. The regions that have better understanding of central government policies and confidence in their representatives in the central government move more steadily forward. It is this author's belief that the fast development of Tigray is largely based on these factors. As stated earlier, citizens show legitimate dissatisfaction when their region does not get its appropriate share of the development benefits. Such dissatisfactions lead to protests, some times to violent protests. The Federal Government needs to do more to level out unjustifiable regional differences. The present regional inequalities are too visible to be overlooked.

References Befekadu Oegefe & Berhanu Nega (eds.). (1999/2000). Annual Report on the Ethiopian Economy,(Vol.l). The Ethiopian Economic Association: Addis Ababa. CSA. ( 2003). Report on Large and Medium Scale Manufacturing and Electricity Industries Survey. Addis Ababa. CSA/FORE. (2002). Report on Large and Medium Scale Manufacturing and Electricity Industries Survey. Addis Ababa. _ _ _ _ .(2000).Report on Large and Medium Scale Manufacturing and Electricity Industries Survey. Addis Ababa. _ _ _ _ .(1998).Report on Large and Medium Scale Manufacturing and Electricity Industries Survey. Addis Ababa. _ _ _ _ .(1997).Report on Large and Medium Scale Manufacturing and Electricity Industries Survey. Addis Ababa. CSA/FROE. (2002). Ethiopia: Statistical Abstract. Addis Ababa. CSA/FROE. (2000). Ethiopia: Statistical Abstract. Addis Ababa. CSA/FROE. (1998). Ethiopia: Statistical Abstract. Addis Ababa. CSA/FROE. (1997). Ethiopia: Statistical Abstract. Addis Ababa. CSA/FROE. (1995). Ethiopia: Statistical Abstract. Addis Ababa. FORE. (2002). Proclamation No. 280/2002: Investment Proclamation, Addis Ababa. FORE. (1996). Proclamation No. 37/96: Investment Proclamation, Addis Ababa.

Regional Socioeconomic Well being in Ethiopia

141

Gilbert, A. (ed.). (1976). Development Planning and Spatial Structure. John Wiley & Sons: London. Hoyle, B., S. (ed.). ( 1978). Spatial Aspects of Development. John Wiley & Sons: London. Ministry of Education. (2003).Educational Statistics Annual Abstract, 2002-2003, Addis Ababa. rd Todaro, M.,P. (1986). Economic Development in the Third World ( 3 ed). Longman: New York and London. Transitional Government of Ethiopia (TGE). (1996). Proclamation No.3711996. Investment Proclamation, Addis Ababa. (1992). Proclamation No. 15192: Proclamation to Provide for Encouragement, Expansion & Coordination of Investment, Addis Ababa.

Rural-Urban Linkages in Ethiopia

143

8 Rural-Urban Linkages in Ethiopia: The Need to Bridge the Divide Tegegne Gebre Egziabher*

1. Introduction Development theories are based on the dichotomy between rural and urban areas. Consequently, the urban and the rural have received different attention in practice and planning. Such approaches have undermined the potential of ruralurban relation for poverty reduction and mutual development. It also has created disparity between the two spatial units in levels of living. In reality, however, the urban and the rural need each other to perform their functions well and develop symbiotically. Agricultural growth depends on access to urban market and industrial inputs, while urban industries require raw materials from the rural sector and use the latter as sources of demand and market. In the same way, households' livelihood in both rural and urban areas transcends the regional boundaries of a town and its hinterlands as it is becoming multi-locational and multi-dimensional (Haan and Ufford, 2001 ). In both rural and urban areas, a significant portion of households relies on income diversification and on the combination of rural and non-agricultural income sources (Tacoli, 1998). Rural-urban interactions are now considered as manifestations of people's livelihood strategies (Haan and Ufford, 2001 ). As a result, rural-urban interactions in many· places have expanded. For instance, a Club du Sahel's West Africa study by Cour and Snrech (cited in Haan and Ufford, 2001) found that:(1) population growth; (2) changes in the natural environment; and, (3) brutal exposure to world markets brought a rapid increase in rural-urban interactions. In many places evidences related to modernization of agriculture, increased export crop production, increased rural-urban migration, expansion of cities to hinterlands abound and they underlie increased rural-urban interactions. However, the complementarity between urban and rural areas for mutual development is yet to be achieved particularly in developing countries. In these countries, urban areas are not capable of producing goods and services in sufficient quantity for rural areas (Haan and Ufford, 2001 ). Rural areas in turn are not producing enough food and raw material, and serve as market centers for urban areas. In addition, rural-urban interactions may be marked by competition as is evidenced in land market in rural-urban contact zones (Ibid). The situation in Ethiopia attests to this fact as urban and rural, agriculture and industry have evolved separately. For instance, urban centers in Ethiopia have evolved as political, military.and administrative centers rather than as centers of

* Regional and Local Development Studies, Addis Ababa University

144

Tegegne Gebre Egziabher

economic activities. The rural hinterland is cut off physically from the main urban centers as infrastructures such as road and air transport, telecommunication and others are poorly developed. Therefore, the major objective of the present study is to review the existing ruralurban relations and identify some possible policy measures to foster strengthened linkages between the urban and the rural sectors in Ethiopia. The paper has six sections including the introduction. Section two discusses the different views on rural-urban in development · theory, while section three examines the nature of rural-urban linkages. Section four brings into attention factors associated with the observed linkages, while section five sketches out possible areas of intervention that help strengthen rural-urban linkages. Section six is a concluding remark.

2. The Rural and Urban in Development Theory During the immediate post war-period, most governments, informed by modernization theory, encouraged urban development. According to the modernization. perspective, the change from traditional to modern was to occur through the diffusion of capital, technology, values, institutional arrangements and political beliefs from the west to the traditional societies. The industrialized Waste was believed to be a model that should be emulated. The notion of urban-rural duality was derived from the conformist literature on urban development that considers urban areas as centers of modernization and development and rural areas as zones of economic backwardness and cultural traditionalism. In this perspective, development should come through the positive influence of the city acting on the countryside as a catalyst of socio-economic transformation (Slater, 1986). Industrialization, which is coincidental with urban growth, was regarded as the engine of social and economic development while rural development was believed to be achieved through trickle down effects of the urban sector. The modernization theory is severely criticized by the dependency school of the 1960s and 1970s.The dependency school rejects the notion that western contacts and influences were necessarily beneficial for the third world. In fact, it has been noted that relations between the advanced capitalist and the peripheral countries resulted in the underdevelopment of the latter (Slater, 1986). Along these lines, it is argued that the growth of urban centers was based on the exploitation of rural areas, which prevented rural areas from taking advantage of their own development potential. Besides the above theories, Lipton (1977) presented an 'urban bias' thesis in which he showed that tariff, trade, taxation, and sector investment policies pursued by most governments deprived rural areas of resources and infrastructure. As a reaction to 'urban bias' thesis, emphasis was shifted towards

Rural-Urban Linkages in Ethiopia

145

rural development particularly in the 1980s. Two paradigms along this line were the rights-based approach to human welfare and the role of agriculture in economic development. The rights-based approach puts emphasis on access to basic needs and the provision of universal coverage of primary education. It implies that the needs of the rural poor should be met. On the other hand, the agricultural growth thesis argues that agriculture is the engine of growth. The growth of agriculture provides not only food to the urban sector, but also stimulates demand for industrial growth and urban development. According to this view, urban development is believed to be dependent on rural development. The modernization theory that argues for urban centers to bring rural development, and the agricultural growth linkage theory which argues for urban development to be contingent on rural development underlie the dynamics of rural-urban link. Both theories specify the type of linkages the two can have. This led researchers to focus more on the interdependency and symbiosis of the rural and urban spheres. The virtuous circle model of rural-urban development envIsIons a mutually reinforcing pattern of linkages between rural and urban centers. The model describes the economic linkages between agriculture and urban activities as inter-reliant behaviors in which an increase in agricultural income prompts diversification of urban activities which in turn absorbs rural surplus labor, raises demand for rural produces, and boosts agricultural productivity and incomes (Evans, 1992). Despite the fact that, the virtuous circle model put emphasis on the symbiotic development of the rural and urban, some researchers have noted that, it has not yet been considered by most developing countries (Dougals, 1998)

3. The Nature of Rural-Urban Linkages in Ethiopia Various forms of rural-urban linkages exist in Ethiopia. These linkages are, however, constrained by a number of problems and are poorly developed. The linkages and their problems could be discerned by looking at factors related to flow of agricultural and industrial goods, labor, and finance. In addition, sectoral linkages between agriculture and industry and public service linkages are other forms of linkages though they also have their own problems.

3.1. Flows of Agricultural Goods Grain and livestock marketing represent the major form of agricultural goods flow between rural and urban areas. The market for grain is the largest of all markets in the country. It involves mtllions of farmers and consumers as well as a number of private marketing agents providing diverse marketing services - buying, selling, transporting, storing, processing, retailing etc (Girma, cited in Wolday and Elleni, 2003). Grain trade links rural areas and small towns with major regional towns and terminal markets or the capital city. There are three grain market

146

Tegegne Gebre Egziabher

types in Ethiopia: ruml, urban and export. Export grain market, which was banned during the Derg period, has become functional since 1996. Each of these markets attracts a number of actors including assemblers, wholesalers, processors, brokers and retailers, who are involved in complex marketing channels. Studies indicate that producers directly sell 31.4% of their grain to consumers; about 35. 7 % to inter-regional traders; about 19.8 % to retailers, and about 12.1 % and 1% to assemblers and processors, respectively (Gebre Meske! cited in Wolday and Elleni, 2003). Inter-regional traders sell the bulk of their grain in deficit areas and in the central market. A number of factors imposed serious constraints on grain marketing. First, small holder farmers in Ethiopia produce for household consumption. As a result, the proportion of food grain production to be marketed is relatively small. For example, in 2001/02 the average marketed grain was only 17 percent of the total production in the country (Table 1). During the same year, the marketed surplus ranged from 25 percent for wheat to 8 percent for sorghum. In addition, the marketed grain varies among regions and between seasons. Much of the marketing take place few months after harvest and seasonal peaks creating difficulty to mobilize finance and transport (Wolday and Eleni, 2003). Secondly, the fact that almost all agricultural produces are transported by road implies access to transport infrastructure is a crucial aspect of marketing and thereby rural- urban linkage in Ethiopia. Ethiopia with an average of only 21 km per 1000 km 2, or 0.44 km per 1,000 persons has one of the lowest road networks in Africa. Limited road network increases consumer prices and also penalizes producers. Market level survey reveals that rural markets are connected with the central markets by poorly paved roads and also many of the roads to the villages and rural markets are not motorable during the rainy season (Wolday and Eleni, 2003). For instance, in Tigray the proportion of population that is cut-off during impassable months is 22 %; while the proportion is28 % and 29% in Amhara and Oromia, respectively (Wolday and Eleni, 2003). Table 1: Marketable Surplus of Grains in 2001/2002 Production Season 000qts) Type of crop Teff Wheat Barley Sorghum Maize Other cereals Total

Production

17,730.0 9,739.4 10,249.1 24,044.7 18,200.3 4,055.1 84,018.5

Source: Wolday and Eleni (2003, p.8).

Marketable surplus

3,779.1 2,444.4 1,148.6 1,855.6 4,467.36 337.0 14,031.9

Percentage

21.3 25.1 1·1.2 7.7 24.6 8.3 16.7

Rural-Urban Linkages in Ethiopia

147

Third, an important element of a marketing system is market information. Market information systems are primarily concerned with price formation, supply and demand. Lack of accurate and timely information leads to higher risks in production and marketing. It also limits the opportunity for interaction between places. Studies show that personal observation is more important to get information on prices, supply and demand than other sources in Ethiopia (Eleni and Wolday, 2003). Significant number of traders did not have information on the supply and demand of grain in other markets (Wolday and Elleni, 2003). Other factors which affect the performance of agricultural marketing in Ethiopia include limited storage capacity of traders ~md lack of access to finance. The latter is due to high collateral requirement, complex procedures and limited information on accessing finance from various sources (Wolday and Elleni, 2003). In one study, about 58 %, 25 % and 36% of respondents in Oromiya, Amhara and Tigray regions, respectively, said that financial institutions in market towns failed to deliver financial services to grain traders (Wolday and Elleni, 2003). The problem of storage could be seen within the broader problem of market facilities in many towns of Ethiopia. Warehouses built for storage and retaining of goods are not common in Ethiopian towns. Grain marketing, therefore, has a number of constraints that have implication for rural-urban link. For instance, poor road conditions lead to lack of exchange between surplus and deficit areas. It also leads to higher prices hampering trade between rural and urban areas. Similarly the lack of marketing facilities leads to weak storage capacity of traders. 1

The pastoral area of Ethiopia depends on subsistence livestock production. In eastern and southern pastoral areas, the marketing options include domestic and 2 external markets . The domestic market includes rural-rural and rural-urban. The intra-rural are intra-pastroal trade and trade with adjacent highland farmers while the rural-urban is trade with urban consumers, feedlot owners, and butchers. The marketing channel for livestock is of different types. They are classified as local, primary, secondary and terminal markets or export markets. The type of participants and the volume of livestock traded become more complex as one moves from the local to the terminal market. The distance from pastoralist areas also gets longer following the hierarchy of markets. In livestock marketing transport cost is the most important marketing cost livestock traders have to face. The poor transportation system also results in loss of animals due to death, theft and disappearance. In addition, the absence of market information and seasonality of livestock production and demand also affect domestic livestock marketing.

1 The pastoral area of the country are low-lying areas covering about 60 % of the country land mass and about 12 % of the country's population. 2 The external market include trade with official exporters and unofficial cross-border trade with neighboring countries.

148

Tegegne Gebre Egziabher

The fact that domestic livestock marketing is not well developed and that there is an alternative livestock trading such as cross-border trade makes rural-urban integration in the pastoral areas weak.

3.2 Flows of Industrial Goods Industrial goods and merchandise, which are imported or domestically manufactured, flow from urban to rural areas. It is difficult to document the types and varieties of these goods. Wholesaling, retailing and petty trading activities are involved in the marketing of industrial goods. Wholesaling in the past was entirely under parastatal agency called EDDC. The institution was restructured in 1992 under the name of Merchandise Wholesale an.d Import Trade Enterprise. The two objectives of the enterprise were to stabilize consumer market and become profitable. The sales and purchases of the enterprises have fallen drastically as a result of competitive environment it faces from the private sector. Similarly, the retail -trade was liberalized and the former retailing enterprises were privatized. The participation of the private sector in retail activities has grown significantly since then. Rural-based service cooperatives were also active in retailing in the past since they used to obtain goods from government owned factories and distributing centers (Tegegne and Tilahun, 1996). Many of the recently existing service cooperatives in rural areas however discontinued this function. While these are major developments in the institutional setups of wholesale and retail activities, currently towns are the major suppliers of consumer and manufactured goods to rural areas. Retail shops in small towns are the major sources where rural demand for industrial goods is met. But, these retail shops do not have the capacity to engage in wholesale transactions and are involved in trading consumer goods of low value. Other industrial goods which may be demanded by rural people such as irrigation pumps, veterinary medicines, spare parts for grain mills and other small industries are often not available in small towns. People have to travel to bigger urban centers to get access to these types of goods. The size distribution of towns will therefore be an indicator of the extent of the opportunity for the flows of industrial goods to rural areas. Only few regions have relatively bigger towns. Tigray, Oromia, Amhara and Diredawa contain one town each in the size category of 100,000- 500,000. Yet, these towns are found in the lower end of the size bracket. Some regional centers in SNNP and Harari are found within the size bracket of 50,000-100,000. Other regions lack such towns and still others such as Benishangul suffer from absence of towns even in the size category beyond 20,000 people. The unbalanced spatial and size distribution of towns at national and regional levels thus means that there are places which are not served by urban centers and those which are served suffer from lack of bigger centers. Size of towns has a positive correlation with the type, nature and

Rural-Urban Linkages in Ethiopia

149

diversity of functions that can be supported. The smaller the size, the more limited the number of functions. In this regard, all the small towns do not seem to have the necessary capacity to promote the flows of industrial goods to rural areas.

3.3 Flows of Factors of Production: ·Labor and Capital Labor is the major resource that flows between rural and urban areas. Generally, it is the mismatch between demand and supply for labor at a given time and space which drives the movement of labor from one part of the country to another. There are four main directions of movements: intra-rural or from one rural area to another; rural to urban; urban to rural; and intra-urban or from one urban area to another. The level of development is one of the crucial factors that determine the direction of movements. In countries where most of the population is still rural, the largest movement of labor is intra-rural while in countries at a more advanced stage of urbanization, the predominant flows are more likely to be inter-urban or from one city to another (Evans, 1990). An in-depth analysis of the 1994 census and the 1999 labor force survey shows that the major flow of people in Ethiopia is between one rural area and another. It is found that nearly 37 % of the migrants are intra-rural. This is to be expected since the majority of the Ethiopian people (nearly 85 %) live in rural areas and the country has a very low level of urbanization. There is, however, a substantial flow of rural-urban migration as well. The proportion of rural-urban migrants in the country is 23% while 16 % moved from urban to rural areas. The two flows together account for 39 % (Table 2).

Table 2: Proportion of Migrant Population by Area of Origin and Destination '%)

Destination Origin Rural Urban Total

Rural

Urban

Total

22.9 36.8 59.7 24.2 40.3 16.1 47.1 52.9 100 Source: Golini et al. (2001 ). According to the in-depth analysis of 1994 census, the major reason for urbanward migration was economic while non-economic reasons dominate the ruralward migration. Nearly 31 % of the migrants to urban areas have stated job search while 25 % of the migrants who went to rural areas did so for the purpose of accompanying family. This implies if towns do not have enough absorptive capacity, the flows of people from rural to urban areas will be constrained and will have serious implications for the towns themselves. Capital flows between rural and urban areas is channeled through financial intermediaries. Types of financial institution prevailing, their geographical distribution, and the actual flow of finance determine the extent of financial linkages.

150

Tegegne Gebre Egziabher

In the Ethiopian context, informal financial institutions involving money lenders, friends and relatives, and traditional arrangements such as iqqub and idir are the predominant types. These institutions are believed to be important in Ethiopia because of the thinness of the formal financial institutions and the difficulty of getting credit (Befkadu and Berhanu, 1999/2000).The volume of the transaction of these institutions exceeds that of the formal financial institutions. lqqub is particularly important in mobilizing savings. Generally, the size of funds mobilized by iqqub is large and is estimated at 8-1 O % of the GDP (Befakdu and Berhanu, 1999/2000). Since informal institutions are local, the circulation of capital is mainly within the local community and there is little exchange with outside places and has limited contribution for rural-urban link. Most of the financial transaction between urban and rural areas is therefore undertaken by formal financial institutions. As of October 2002, the Commercial Bank of Ethiopia (CBE) and the Development Bank of Ethiopia (DBE) had about 171 and 32 branches, respectively. In addition, the private banks all put together had 108 branches. Most of these branches, however, are found in major urban centers far from rural areas. The possibility for rural people to be served by these institutions is therefore limited. In 1998, banks were found only in 21 % of the woredas or districts (Mulat, 2001 ). A look at sectoral distribution of bank loans indicates the extent to which finance has flown to agricultural and rural sector. The amount of credit disbursed to agricultural sector is small compared to trade (both domestic and international). In the period 1992/93-2001 /02 agriculture received a loan of 556.4 million while trade received 1.6 billion on average (Jonse, 2003).The share of agricultural credit disbursement as percent of total credit and GDP was low (16.5 % of total credit disbursement and 1.3 % of GDP for the period 1992/93-2001/02) compared to trade which accounted for 50 % of total credit disbursement and 3.7% of GDP (Jonse, 2003). Agricultural bank credit disbursement was not only small but declining as well. The declining rate was worse for Development Bank of Ethiopia which is the main supplier of agricultural credit.In addition, agricultural disbursement per capita was almost nil with an average of birr 3 for the period 1992/93-2001 /02 (Jonse, 2003). The poor which are mostly found in the rural areas are regarded as credit risky by the commercial banks. As a result, even if there are banks, their chance of getting bank loan is very low. Micro-finance is considered as a way out of this situation since micro-finances are geared to serving the poor in particular. A microfinancing proclamation was issued in 1996.The objective of the Micro-Finance Institution (MIF) is to enhance the delivery of the financial services (credit, saving and insurance, etc) to the productive but resource poor people in rural and urban areas (Wolday , 2002).Currently there are nineteen institutions registered under the National Bank of Ethiopia. The profile of these. institutions shows that the majority of the clients are rural people (see Table 3).Though the size of the

Rural-Urban Linkages in Ethiopia

151

average loan is not big; this is an indication of the flow of finance from urban to rural areas. The saving products of the MFls are generally categorized into two: agricultural loans and micro business loans. Agricultural loans are loans given to inputs, livestock, beekeeping, etc. while micro-business loans are specified to petty trading, handicraft, and services. In the year 2000 about 51 % of the loan of Dedebit Credit and Saving Institutions S.C (DESCI) disrupted among 61 % of the beneficiaries for the agricultural activities (Table 4). Of the agricultural loan, 56 % was for buying oxen, about 19 % and 17 % were for buying small ruminants and cattle raising, respectively. The same pattern was also noticed for Amhara Credit and Saving Institution S.C (ACSI), Oromia Credit and Saving Institution S.C (OCSI), Omo Micro- Finance Institution S.C (OMFI), and Sidama Micro- Finance Institutions S.C (SMFI) (Wolday, 2002). Despite the micro-finance institutions being important intermediaries in channeling resources to rural areas, they meet only less than nine percent of the demand for financial services of the poor (Wolday, 2002). There are quite vast rural areas which have been not served by micro-finance.

t f Eth·1op1an MFI s, 2001 Tabi e 3 0 utreac h Average Loan an d Int eres t Raeo Microfinance Institutions

'

Active Clients

Women

Rural

Urban

(%)

(%)

(%)

Average Loan (Birr)

Interest on Credit (%) per annum

Interest on Deposits &

Savings Amhara Credit Savings & Institutions S.C. Debit Credit & Saving Institutions

140,000

47

75

25

900

12.5

6

187,550

41

80

20

600

12.5

6

37,342

12

99

1

1,000

12.5

8

39,342

35

95

5

600

12.5

6

3,700

80

1

99

1,000

16

7

3,217

85

0

100

800

13

6

8,535

30

85

15

755

14.5

6

4,286

60

90

10

1,800 ·

15

6

2,300

85

100

0

400

25

6

s.c. Oromia Credit Saving & Institution S.C. (OCSI) Omo Microfinance Institution S.C. Specialized Financial & Promotional Institution Gasha Microfinancing Institutions Wisdom Microfinancing Institution Sidama Microfinancing Institution S.C. Mekket Microfinance Institution S.C.

152

Tegegne Gebre Egziabher

PEACE 974 Microfinance Institution S.C. Addis Credit & 7,000 Saving lnstitutipn S.C. 516 Eshet Microfinance Institution S.C. Wasasa 562 Microfinance Institution S.C. Asser Micro- 3,100 financing S.C. Africa Village 450 Financial. Service S.C Bussa-Gonofa 2758 Microfinance S.C. Mekit 1,001 Microfinance Institution S.C Benishangul 425 Microfinance Institution S.C. Shashemene 200 iddir Yelimat Agar MFI Source: AEMFI various data

62

100

0

682

12.5

6

70

-

100

1,300

12.5

7

54

70

30

500

24

6

31

69

31

498

24

6

-

73

27

750

16

7

60

-

100

1500

16

-

85

87

13

-

24

8

73

-

100

700

16

6

60

100

-

-

12.5

6

58

-

-

1,275

16

6

Table 4: Loan Products of Dedebit Credit and Saving Institutions S.C. (DECSI), October 2000 Activities

Amount of

No of

Loans {10,000)

clients {000)

Agricultural

Male

Female

rrotal

Male

Female

Total

%

105.4

53.0

158.4

164,73.8

62,43.2

227,17.0

50.9

4.5

15.5

120.0

13,26.6

27,52.2

40,78.8

9.1

Loan Handicraft & Processing Petty trading

46.3

31.9

178.2

112,92.6

61,51.0

174,43.7

39.1

Service

0.9

.3

1.2

2,60.2

1,23.1

3,83.3

0.86

Grant total

157.2

100.0

~58.2

293,53.3

152,69.7

446,23.0

100.00

Source: DECSI. (2001 ). (unpublished data)

Rural-Urban Linkages in Ethiopia

153

3.4 Public Service Linkage Among the manifestations of rural-urban linkage are the public service linkages. Rural areas derive various kinds of services: administrative, health, education, police, postal, etc., from urban centers. Higher order services such as high schools, colleges, hospitals, etc., in Ethiopia are found mainly in urban centers. The rural people have to come to the urban centers to obtain these services. Access to public services at national level shows the strength of service linkages in the country. Physical proximity as a measure of access to public services could be a useful indicator for a large country like Ethiopia where the transport network and efficiency is quite low. A study on poverty profile of Ethiopia for the year 1999/2000 provides information on distance between various facilities and the residence of households. The mean distance to reach the secondary school, a higher order services and mostly found in urban areas, is nearly 22 kilometers (Table 5).This distance is not affordable to make a daily commuting to and from the service center. Since primary schools as a lower order good are found in most rural centers, the distance traveled by students become less. Though the study does not differentiate between the types of health centers, the average distance traveled for rural areas is nearly 8 kms in contrast to the 1 km distance in urban areas.

Table 5: Mean Distance to Reach Public Services (1999/2000) Public Service type Primary school Secondarv school Health Center

Rural 3.38 21.90 7.98

Urban 0.73 3.47 1.17

Ethiopia 3.00 19.25 7.01

Source: Tassew and Teklie. ( 2002). The distribution of the population by distance from the nearest public service shows that nearly 76 %, 34 %, 76 %, 74 % of the population live more than 10 kilometers away from a secondary school, hospital/health center, post office, and 3 telephone service, respectively (Table 6).The implication of this observation is that those urban centers which house higher order public services are found far from the rural population thereby limiting rural people's access to these services. A study in two woredas of East Shewa has shown that the main reason for households not sending their children to school was unavailability of secondary 4 school in the nearby small towns (Demeke, 1998) .

3

The percentage of rural population living more than 10 kms is lower for health center since it also includes lower order services such as health post, clinics, etc. 4 It is important to note that while availability is one factor that affects utilization, the latter is also affected by a number of other factors such as attitude, affordability, communication, etc.

154

Ta bi e 6

Tegegne Gebre Egziabher

o·IS tn"b uf10n ofP opu Iaf10n biy o·1st ance f rom p u bi"IC serv1ces