|
|
|
|
|
|
|
||
|
|
Building Capacity for Effective Local Governments in Developing Countries: |
||
| Paul Smoke | |||
| Session: Tuesday, April 18, 2000, 8:45 - 10:00a.m. | Author Info | ||
Decentralization and local government reform have been on the rise in developing countries in recent years. Various political, institutional, and capacity constraints often prevent these activities from meeting their intended objectives. Several countries, including Kenya and Cambodia, have managed to make promising progress recently in spite of these obstacles. A number of factors enhance the ability of decentralization programs to strengthen the capacity and improve the performance of local governments. These include: developing broad consensus on a vision of where the intergovernmental system is heading; tailoring reforms to national context and the circumstances of particular local governments; and, defining a strategic implementation strategy to move the system in the right direction at an appropriate pace. (1)
INTRODUCTION
Decentralization and local government reform have become increasingly popular in the developing countries since the mid-1980s. (2)
Rapid economic, political and technological changes and the poor performance of central planning have forced leaders to think beyond conventional top-down approaches to development and to rely more heavily on lower levels of government.
Recent widespread attempts to decentralize and to strengthen local governments have not been particularly successful in many countries, particularly in Africa and Asia. (3)
Local development projects often get constructed as planned, but intended institutionalization of appropriate government structures and procedures has been less successful. In addition, the level of capacity building is rarely adequate to realize and sustain the potential benefits of decentralization.
WEAKNESSES OF LOCAL GOVERNMENT REFORM PROGRAMS
A number of weaknesses are commonly found in programs undertaken to decentralize the public sector and strengthen local governments in developing countries. These include an inappropriate scope of reforms, poor coordination of the often multiple actors involved, homogeneous treatment of very heterogeneous local governments, and an undue preoccupation with overly ambitious or overly simplified approaches to reform.
Scope of Reforms
Decentralization initiatives can generally be classified into two basic types. The first type has often been unworkably comprehensive, overwhelming technical capacity at both the central and local level, and too heavily threatening bureaucratic and political tolerance at the center. (4) The second type has focused on very specific, limited (often technical), rigidly defined activities that are not part of a broader decentralization and intergovernmental fiscal reform agenda. (5) While such limited reforms are often critical to support broader goals and may occasionally serve as a catalyst for decentralization, rarely can they take a system very far on their own.
Weak Coordination
In many developing countries there is poor coordination of the multiple actors typically involved in decentralization and local government reform programs. At a minimum, most developing countries assign local government oversight responsibilities to a Ministry of Local Government or similar agency. Often various sectoral ministries also have some say over local provision of services that fall under their general expertise. Finally, multi-sectoral or coordinating ministries, such as Planning and Finance, typically have some control over local government activities.
These various agencies often have conflicting views of decentralization and their role in it. If coordination is to be effective, the lead agency must be seen as neutral and trusted. In many cases, however, decentralization initiatives are placed under the leadership of a single central agency that is perceived as a rival by other agencies whose cooperation is required for successful decentralization and local government reform. (6)
Lack of Differentiation among Local Governments
Both comprehensive and limited decentralization programs tend to treat all local governments (or classes--large urban, small urban, rural, etc.) as if they were similar in capacity and needs. In fact, there are often great differences among local governments, even those of a particular type. Devolving too many new responsibilities to weak local governments invites failure. Providing heavy technical assistance to capable local governments that don't need it wastes resources.
The "all-local-governments-are-more-or-less-the-same" problem is manifested in highly standardized decentralization reform programs that have similar expectations of all or most local governments and tend to define performance very rigidly in terms of following specific technocratic steps. The circumstances under which local government managers and staff are functioning, however, may differ widely among jurisdictions. Rigid standardization may prevent them from achieving target objectives rather than helping them to do so.
Dominant Models and Development Trends
Many decentralization programs are heavily based on textbook solutions derived from social science models, particularly from the local public finance and fiscal federalism literature (7). These models provide a useful starting point for considering decentralization, but they focus on normatively ideal outcomes and say nothing about how to build towards those outcomes in countries with weak institutional and political capacity. In addition, the models were developed in the context of industrialized countries and are somewhat culture and institution bound. (8)
In addition to the broad adherence to normative models, narrowly focused development trends sometimes dominate decentralization programs. For example, it is currently fashionable to focus heavily in the development of participatory mechanisms that improve communications between local governments and their constituents. Most decentralization analysts would support such initiatives, but the demand-side emphasis sometimes comes at the expense of adequate consideration of supply-side deficiencies in public service provision.
FACTORS UNDERLYING WEAK PERFORMANCE
The weak performance of decentralization programs and the design problems noted above persist for a variety of reasons. Often there is a perceived urgency to reform rapidly, so careful planning is foregone. Local institutional and political capacity is weak, and the capacity at the center is often inadequate to develop sustainable reform programs. Finally, the incentives and objectives of international development agencies may undermine effective reform.
Perceived Urgency for Reform
Many recent decentralization and local government reform programs have occurred during times of crisis, creating a sense of urgency for rapid reform. Sometimes the impetus comes from domestic political and/or economic turmoil. In other cases, the international community mandates rapid action, typically in the form of conditionalities on loans and grants. A key effect of the pressure to undertake decentralization is that supporting programs are often hastily defined and initiated without carefully thinking through their implementation requirements.
Weak Capacity and Incentives
The problem of low capacity at the local level is well known, but central government agencies may also have little understanding of decentralization and inadequate capacity to design reform programs. (9) Where this is the case and there is pressure to move forward, it is easy to end up relying heavily on the simplified models noted above and/or on solutions used elsewhere and/or on external experts and who may not be familiar with the context of the country. In addition, whatever other impetus to decentralize exists, there are rarely strong central government political and bureaucratic incentives to do so. (10) Central ministries generally do not want to lose control over resources or to engage in coordination that may undermine their institutional interests.
Lack of Local Political Development
Local government accountability to a constituency is quite rare in developing countries, although the situation is generally better in Latin America and is slowly changing elsewhere. Even where local government councils are elected, they may not possess genuine political or bureaucratic independence, and they may be dominated by clientelistic local elites. Perhaps most critically, local accountability doesn't mean much if the center can impose its will and override the results of local democratic processes, which is often the case in developing countries.
Incentives and Objectives of International Development Agencies
The interests of international development agencies are not necessarily well served by decentralization and interagency coordination. Such efforts are difficult and cause delays in donor efforts to bring new projects on line and to move funds. Given pressure to keep to expenditure schedules, substantial donor funds continue to flow for local investments, even if recipient governments do not have the capacity to ensure sustainable use of the funds.
Donors often try to get around the difficult problem of sectoral coordination by defining the scope of their projects in terms of particular sectors or local units. Since donors have their own individualized systems and procedures, and central governments rarely coordinate donor efforts, the process of building a consistent decentralized system may be undermined. Moreover, because individual donors often develop client relationships with particular ministries or local governments, they may exacerbate interagency competition.
THE OVERALL SITUATION
As is evident from the above discussion, decentralization in inherently complex and slow. Capacity must be built, and attitudes about the way the public sector works (or could work) do not change overnight. Central government officials are used to making most major decisions unilaterally and controlling local authorities extensively. Local governments themselves are often comfortable with central control as long as it is accompanied by heavy subsidization. Finally, local government constituents are accustomed neither to paying for services nor to expecting much responsiveness from local governments. The mindsets and patterns of behavior that have developed under such circumstances will require years to change. The most important question a decentralization program needs to consider is how to undertake this challenge pragmatically.
TWO CASES OF IMPROVED PERFORMANCE AND INNOVATION
In this section, I contrast two cases in which some of the constraints outlined above are recognized and built into the design of decentralization and local capacity building efforts. Kenya is a case where local governments exist but have performed poorly for many years. Cambodia is a case where autonomous local government has never existed, so a system is being developed from the very beginning.
Kenya: Rebuilding Local Government
Kenya is a country with a long history of local government based on the British tradition, but the system was weakened and heavily controlled by the center in the post-independence period, primarily in the name of national unity. As a result, local political mechanisms and incentives to perform were greatly undermined. There have been many efforts during the 1980s and 1990s to re-strengthen local governments. Most have been donor-driven and were not taken particularly seriously by the national authorities. As a result, the performance of local governments in terms of revenue generation and service delivery deteriorated to very low levels. In recent years, however, fueled by political and economic pressures to reform, some genuine attempts to strengthen local governments have been undertaken.
Given the number of local government and the magnitude of their problems, not all of the necessary reforms can be undertaken at the same time or in all places. This reality was well recognized by the first of a new generation of reform programs, the Small Towns Development Project (STDP), a capacity building project for local governments jointly funded by the Kenyan and German governments (GTZ). The STDP has involved experimentation with only a small number (c. fifteen) local authorities, most of them towns rather than the largest cities.
Under this project, there have been some broad common reform goals among the local authorities in the program, but the exact steps each local authority took and the pace at which they took them differed. The strategy in this project was to employ a mobile team of experts to work individually with each participating local government to design a unique package of reforms meeting their needs. The mobile team worked periodically with local governments on site to implement the negotiated program of reforms gradually, and, as necessary, to modify it. Citizen participation mechanisms were required as part of the reform process for certain types of activities, and some training opportunities were provided for local officials. Local governments that met the goals they agreed to were rewarded in various ways, such as greater access to funding and reduced central interference in their decisions. In most of the participating local governments, their performance--in terms of revenue collection and service delivery--improved considerably.
The STDP has also had several other benefits. First, virtually all of the technical assistance personnel have been local rather than foreign, raising acceptance and the credibility of the initiative among Kenyans. Second, STDP efforts generated information that has assisted the Ministry of Local Government to create procedural manuals and training materials that have helped to get central and local officials to accept the use of standards and models in making basic revenue and expenditure decisions. Third, the continual presence of the STDP over a period of years has raised the visibility of local governments among central officials and the general public and demonstrated that, with adequate assistance and incentives, they can improve their long-poor performance.
The Kenyan government, with very substantial World Bank funding and technical assistance, has been defining a broader program of reform--the Kenya Local Government Reform Program (KLGRP)--largely based on the successful experiences of the STDP and lessons learned from less successful past donor efforts. The KLGRP is based on the premise that a variety of interrelated actions are required to improve local government performance, including legal, institutional, technical and procedural reforms at both the central and local levels. Moreover, there has been an explicit realization that, while MLG is clearly the appropriate institution to lead the reform initiative, the close cooperation of a variety of other government agencies is also required. To deal with this reality, an Interministerial Task Force on Local Government Reform governs the effort.
Following the lead of STDP, the KLGRP is being built gradually and systematically in a series of manageable, mutually reinforcing steps. Most of the KLGRP staff includes regular government employees who have continue to perform many of their regular duties in addition to the KLGRP responsibilities. This creates a situation in which KLGRP is not seen as separate from government activities, but an integral and mutually reinforcing part of them. Most of the local consultants being used to conduct investment needs assessments in the local authorities, to develop the simplified accounting mechanisms, etc., are from local firms.
The configuration of KLGRP involves three phases. The first phase, which has essentially been completed, solidified agreement on the most fundamental reforms required. The key tasks included the definition of several critical framework reforms, such as local authority organization and management and intergovernmental fiscal relations. In addition, KLGRP also conducted need, capacity and endowment studies that tentatively identify the most important service requirements of participating local authorities and the technical and financial resources at their disposal to meet these needs. The needs analysis includes the development of simple participatory stakeholder assessment mechanisms that broaden the local decision-making process.
The second phase of KLGRP, which started in 1999, is advancing the process of implementing
basic local authority organization and financial management reforms defined in the first phase in all KLGRP local authorities. In addition, efforts are underway to implement several key intergovernmental and interministerial reforms defined in the first phase, including harmonization of central and local taxes and the establishment of a basic intergovernmental grants program.
The third phase will involve implementation of a physical infrastructure program. Funding for basic infrastructure will begin after minimal reform requirements defined in the previous phases are met. As a precondition to more substantial funding, an increasingly comprehensive set of reforms at both the central and local level must be satisfactorily undertaken.
Cambodia: Creating and Developing New Local Governments
During the French colonial period, Cambodia had a deconcentrated system of administration. This system deteriorated during the political and economic turmoil of the past few decades, but the shell of the former French system still serves as the basis for administering government activities in the provinces. After the UN-managed elections in 1992 returned Cambodia to a more stable situation, international agencies began supporting rural development in a major way. Notable among these efforts was CARERE, a UN sponsored effort that transformed an emergency relief operation for returning refugees into a program that supported the development of decentralized governance.
Working from the achievements of CARERE in a number of pilot provinces, the Royal Government of Cambodia drafted decentralization legislation in 1999. This legislation establishes the communes as a separate level of elected government, a dramatic shift away from their previous role as a relatively minor player in the central government's deconcentrated system.
The CARERE experiment and the government's current efforts are largely based on the "Local Development Fund" (LDF) approach developed during the past five years by the United Nations Capital Development Fund (UNCDF). (11)
The LDF approach shares some characteristics with a number of other types of approaches designed to build local governance and to promote local development in very poor countries. Among them are small town/rural development programs, the older family of Community Development Funds, and the more recent Social Investment Funds (SIFs) and Micro-projects programs. These programs typically have a poverty alleviation focus, share a "demand-driven" approach, and target local investments (health and education facilities, roads, water supply, income-generating community projects, etc.) of a similar type and scale. (12)
Under most of these programs, a number of different types of beneficiary groups (local governments, NGOs, community groups, etc.) individually access resources for specific projects though application to the appropriate management unit--centrally located or deconcentrated to the regional level and often partially or fully independent from the formal government structure. In contrast, LDFs provide resources only to local governments, "entitling" them with an objectively defined block allocation of resources (rather than a project-specific grant) from a fund based at a higher level of government. This serves multiple purposes: concentrating aid on the development of a formal system of local government; allocating local budgetary resources in a transparent way; and giving local people an opportunity to channel funding to their own priorities.
Initially the LDF restricts the use of funds to a menu of relatively simple projects that are likely to be within the capacity of local governments to deal with. The LDF also introduces a simple participatory planning process, but local governments use it to plan beyond the use of LDF resources, making requests for larger-scale projects to higher levels of government through a broader intergovernmental planning system and applying for other available sources of funding. The process is demand driven, but technical, institutional and governance standards set by a higher level of government must be met before funds are released. This approach tends to favor stronger local governments, but weaker ones can access resources to obtain technical assistance required to meet the standards. The can choose to obtain assistance through conventional channels (agencies of higher level governments), private firms, or NGOs.
Thus, there are several key differences between LDFs and other approaches to decentralization in underdeveloped areas. These include: the strong LDF emphasis on a contextually tailored local planning process; the provision of LDF resources as "objective" formula-driven grants provided only to local governments; the restriction of initial activities to simple ones that are more likely to be successfully undertaken by nascent local governments; and the existence of supply-side standards that help to ensure reasonably cost-effective use of resources and to encourage inclusive decision making.
It is important to note that the specific context of Cambodia led to making key decisions about how to apply the LDF approach. For example, the program initially bypassed the central level because of the turmoil going on in Phnom Penh at the time the program was initiated. Starting at the provincial level and using mechanisms separate from the government administration allowed the decentralized approach to be tested successfully without much interference from competing ministries at the central level. Now, as the country begins to "mainstream" decentralization and create a system of local government, interagency mechanisms are being developed at the provincial and central levels to bring the system into the formal government structure. In addition, the initial experiment focused on the lowest tier of administration (villages), where villagers elected development committees. This was considered necessary to build the people's trust for government structures at the grassroots level. When the system is formalized and expanded to other provinces, the main local unit will be the communes, which cover larger areas and populations and are more likely to be sustainable as units of government. As capacity is gradually built, new responsibilities will be extended to the commune governments.
LESSONS ON DECENTRALIZATION AND CAPACITY BUILDING
There is no single approach to effective decentralization and local government capacity building. Reforms must be determined in a particular case, given available resources and capacity, as well as the institutional, social and political realities that govern an appropriate and realistic scope and pace for decentralization. Most countries, however, face some similar problems and constraints in moving forward. The Kenya and Cambodia cases, among others, suggest a number of basic lessons about approaching decentralization and local government capacity building.
Policy Framework and Process
Some policy reforms at the central level are inevitably necessary to construct the foundation for a sustainable system of local government. A good policy framework articulates a clear vision for decentralization and defines basic rights and responsibilities of different actors. Even a strong framework, however, needs to be operationalized, often in a complex institutional environment. Thus, a process for building consensus about how to proceed should be part of the overall environment for reform. Both Kenya and Cambodia have created inter-agency bodies to define and manage the reform process and to coordinate the multiple actors involved. The likelihood of realizing good coordination and effective results is improved if the decentralization process is properly negotiated and structured from the start.
Strategic Implementation
An effective decentralization program requires a strategic implementation approach, as in both Kenya and Cambodia. Even though a clear vision of the eventual system is needed, those in charge of reform efforts must be careful not to do too much too quickly, and they should try to phase in steps in a logical way. Reforms that have the greatest possibility of succeeding in a relatively short time frame should be undertaken first. This requires prioritizing reforms, focusing on fairly simple tasks that don't immediately threaten in a significant way the central power base or overwhelm local capacity. Both the Kenya and Cambodia cases profited from such an approach, although the particular steps they took reflected differences in the level of development of the intergovernmental system. It is also useful to differentiate among local governments, a particular highlight of the Kenya approach. Some are likely to be relatively capable and can be given greater responsibility, while others will require substantial technical assistance. Starting carefully and modestly should raise the prospects for initial success, thus creating a stronger base on which to build further momentum in the future.
Balancing the Roles of Different Levels of Government
Decentralization and local government reform need to be framed in broader institutional terms than just shifting power from central governments to local governments. In fact, decentralization may sometimes involve a modified and/or increased role for centralized agencies, at least at early stages. Both Kenya and Cambodia recognized this need in the design of their decentralization efforts. In both cases, there is an appropriate balance between decentralization reforms on the supply and demand side of service provision. Each recognized the importance of establishing political credibility by consulting citizens, but neither ignored the importance of using standards set at higher levels to ensure adherence to minimum technical and cost standards.
Building Capacity and Governance
It is necessary to build the commitment and capacity of local governments to assume greater responsibility on behalf of their constituents. Building on the type of appropriate starting point discussed above, decentralization reforms should be progressively phased in with incentives, such that it is clear to local governments what they have to accomplish before they will be assigned additional responsibility or resources, as in the Kenya case. In addition, the importance of building the credibility of local governments with their constituents cannot be overemphasized. Without increased local input and accountability through political and procedural reform, decentralization is ultimately a meaningless exercise. There are various ways of realizing this important goal, but often-disillusioned local residents have to be gradually convinced that their local governments can and will respond to them, and this requires some concrete results in the form of improved performance of service delivery functions.
CONCLUDING STATEMENT
Decentralization and local government reform are part of a broader shift in the practice of development that is likely to persevere for the foreseeable future. As economic and political pressures continue to escalate, many countries will feel an increasing urgency to formulate policies that produce good results. Following some of the basic principles outlined here can help, but there is much more to learn about how to approach decentralization and local government reform under different circumstances.
BIOGRAPHY
Paul Smoke is Associate Professor in the International Development and Regional Planning Program of the Department of Urban Studies and Planning, MIT, and Senior Institute Associate at the Institute of Public Administration, NYU. His interests include urban and regional development, public finance and public sector decentralization. He has worked in numerous countries and for various international development agencies, and he has published widely on these topics. He can be reached at: smoke@mit.edu
1. The ideas discussed here are based on a variety of projects and programs financed by the World Bank, the United Nations Capital Development Fund (UNCDF), the United Nations Development Program (UNDP), the Deutsche Gesellschaft fur Technische Zusammenarbeit (GTZ), the United States Agency for International Development (USAID), and various governments. The perspectives expressed here are those of the author and should not be attributed to any of these organizations.
2. World Bank, Entering the 21 Century: World Development Report 1999/2000. Washington, DC: The World Bank.
3. Much of this literature is reviewed in Smoke, P. (1994) Local Government Finance in Developing Countries: The Case of Kenya. Nairobi: Oxford University Press. Other references include: Dillinger, W. (1995) Better Urban Services: Finding the Right Incentive. Washington, D.C., The World Bank; Tendler, J. (1997) Good Government in the Tropics. Baltimore, MD: Johns Hopkins University Press; Litvack, J., J. Ahmad and R. Bird (1998) Rethinking Decentralization in Developing Countries. Washington, DC: The World Bank; Cohen, J. and S. Peterson (1999) Beyond Administrative Decentralization: Strategies for Developing Countries. West Hartford, CT: Kumarian Press.
4. See, for example, Smoke (1994); Khellaf, A. (1992) "Decentralization and Centralization of Local Public Services in Tunisia," Cambridge, MA: Department of Urban Studies and Planning, Massachusetts Institute of Technology.
5. There have been, for example, at least five different and entirely separate attempts to improve budgeting and accounting systems in Kenya since the mid-1980s. Until recently, none of these were linked to a broader reform effort.
6. A good example of the level of competition among central agencies is shown in the development of the World Bank's Fourth Urban Project in Kenya. Efforts to start this project began in 1986, with protracted battles among several ministries for control over it continuing until the project was essentially abandoned in 1994.
7. These models originated with a few seminal works, including: Tiebout, C.M. (1956) "A Pure Theory of Local Expenditure, Journal of Political Economy, Vol. 64, No. 5, pp. 416-424; Oates, W. (1972) Fiscal Federalism. New York, NY: Harcourt, Brace, Jovanovich. Recent discussions of how these models have been applied, expanded and empirically tested is found in: Bahl, R. and J. Linn (1992) Urban Public Finance in Developing Countries. Baltimore, MD: Johns Hopkins University Press; Raimondo, H.J. (1992) Economics of State and Local Government. New York, NY: Praeger; Fisher, R.C. (1996) State and Local Public Finance. Chicago, IL: Irwin.
8. Smoke, P. (1989) "Is Local Public Finance Theory Relevant for Developing Countries?" Development Discussion Papers, No. 316. Cambridge, MA: Harvard Institute for International Development; Bahl and Linn (1992); Smoke (1994); Bird, R. and F. Vaillancourt (1998) "Fiscal Decentralization in Developing Countries: An Overview." Discussion Papers, No. 11. Toronto, ON: International Center for Tax Studies, University of Toronto; Litvack, Ahmad and Bird (1998).
9. Many of these weaknesses are elaborated in: Smoke (1994); Dillinger (1995); Tendler (1997); Bird and Vaillancourt (1998); Litvack, Ahmad and Bird (1998); Cohen and Peterson (1999).
10. There are some cases in which there are significant political motives for decentralization. In Ethiopia, for example, decentralization has been largely driven by ethnic concerns. See: Cohen, J. (1995) "Ethnic Federalism in Ethiopia" Development Discussion Papers, No. 519. Cambridge, MA: Harvard Institute for International Development. Similarly, South Africa's attempts to decentralize and strengthen local governments is more politically than fiscally motivated. See: Manche, J.M. (1994) "Restructuring Urban Government in South Africa" Cambridge, MA: Department of Urban Studies and Planning, Massachusetts Institute of Technology.
11. The LDF approach is discussed in considerable detail in UNCDF's 1996 publication Local Development Funds. UNCDF's 1999 publication, Taking Risks, reviews experiences with the LDFs to date.
12. Smoke, P. and Romeo, L.
(1997) "Designing Intergovernmental Fiscal Relations and International
Finance Institutions Allocations for Rural Development," Decentralization
for Rural Development. Rome: Food and Agriculture Organization of the
United Nations.
Copyright 2000 By Author
Paul Smoke
Massachusetts Institute of Technology