Taking Risks.. Background Papers

Decentralized Development Planning:
Issues and Early Lessons from UNCDF–Supported Local Development Fund Programmes

Leonardo Romeo

The rationale of decentralized planning

Incentives and disincentives to decentralized planning

The local planning process introduced by local development fund programmes

Early lessons from LPP implementation

Some conclusions for future LPP design (in progress)

 

The rationale of decentralized planning

An important element of the policy experiment that local development fund (LDF) programmes are meant to undertake is concerned with the introduction or the improvement of procedures for local-level planning in the public domain. Such procedures are meant to increase the local authorities’ accountability and efficiency in the use of the resources devolved to them in the framework of the decentralization reforms supported by LDF programmes. It may seem obvious that an attempt to improve decentralized planning should be made in the frame of a programme that sets up a decentralized financing facility—the LDF. After all, there is a broad consensus among decentralization analysts that most attempts in the past to decentralize planning have been fundamentally flawed because they were made without decentralizing financing. Thus the devolution of resources to local authorities, particularly through non-categorical, central-to-local grants of the LDF type, is expected to remove one of the most formidable obstacles to meaningful decentralized planning practices. In other words, many agree that the setup of a decentralized financing mechanism does provide a necessary condition and creates an ideal opportunity for improving decentralized planning practices. And the view that there is a need to introduce such improved planning practices, in connection with the devolution of resources and responsibilities to local authorities, has also been expressed by stressing that "devolving project-funding decision-making to local governments is not likely, by itself, to promote increased accountability or efficiency". (1)

However, so many and so diverse are the difficulties associated with the implementation of meaningful, technically sound and participatory planning at the local level, and so vast seem the efforts to build the required local capacity, that some have expressed "grave doubts about whether [local-level "planning from below"] has occurred or can occur in most less developed countries". (2) Whether or not efforts to introduce "planning from below" and improve decentralized planning practices are doomed to fail and should be abandoned altogether depends probably on what is meant by "planning from below" and by different ideological and professional stances. It is in fact the subject of an ongoing and unresolved debate among politicians, administrators and planning professionals. What matters in this discussion of LDF programmes is that when local authorities, because of genuine decentralization reforms, are given new and greater responsibilities and resources for local development, the scope and impact of their decisions is bound to increase accordingly. Then even modest improvements in the local-level decision making process, by making it better informed and more transparent, should be critical enough to justify a commitment to decentralized planning and a concrete investment in the development and extension of improved decentralized planning practices. This is what the local planning processes (LPP), introduced in the framework of LDF programmes, is all about. (3)

Coming to the implementation of the LPP, it is useful to set the framework for a review of the issues and early lessons associated with it by restating the arguments commonly advanced in favour of decentralized planning and which constitute its main rationale. Decentralized planning is expected to: (4)

• Increase popular participation in planning and development.

 

• Make plans more relevant to local needs.

 

• Facilitate coordinated or "integrated" (multi-sector) planning.

 

• Increase the speed and flexibility of decision-making and implementation.

 

• Generate additional resources and encourage more efficient use of existing resources. (Top)

 

Incentives and disincentives to decentralized planning

While decentralized financing is an essential condition for successful decentralized planning, it is certainly not the only factor at play. The political and administrative dimensions of the decentralization reforms also play a major role. Of particular importance for the successful introduction of decentralized planning systems in most less-developed countries are the actual commitment to and investment in them by the state; the level, duration and instruments of external donor support; and the attitude of local political leaders.

State-mandated procedures and related incentives and controls

Participatory and technically sound decentralized planning will not emerge spontaneously out of the granting of local autonomy to some subnational levels. While the argument of lack of local capacity should not be used (as it has been so often) to avoid decentralization reforms, it also cannot be ignored altogether by espousing the principles of local autonomy only to justify the withdrawal of the state from any practical effort to build local capacity. The commitment and ability of the state to institutionalize decentralized planning is, in fact, absolutely critical. The experience of designing and implementing LDF programmes and related LPPs clearly points at the role of the state in the development of the LPP, its extension to local authorities and the monitoring of its implementation. In the LDF programmes the successful adoption of the LPP depends on the ability of deconcentrated state agents (at the provincial or district level and in partnership with nongovernmental organizations) to assist local authorities and monitor their performance. This last aspect is particularly important in the frame of the LDF programmes, because of the need to establish some form of accountability of local authorities, not only to their constituencies, but also to the state—which, after all, is the main source of the financial transfers supporting local choices.

State commitment to capacity building for decentralized planning varies across the different countries currently involved in LDF programmes. In Mozambique, for example, there is a clear commitment of the central authorities to develop and institutionalize local (district-level) planning. Plans also exist to field district planning officers in all districts. By contrast, the legal status of local (commune-level) planning instruments in Cambodia, Palestine or Vietnam is either unclear or non-existent. Similarly, it is unclear whether governments in these countries will be willing to sustain the effort of LPP extension once the UNCDF external assistance—which currently covers the relatively minor costs of operation of the extension teams—is gone. To a great extent this is also a design problem for LDF programmes. On one hand, the experimental nature of these programmes requires a considerable up-front investment in the development, refinement and extension of the LPP. On the other, the long-term sustainability of the system depends on keeping the operating costs of the LPP extension and facilitation teams low. Experience shows, however, that in most countries some untapped capacity exists within the deconcentrated administration which makes it possible to retrain and reorient existing staff to service local authorities as planning advisers and technical assistants without the need for additional recruitment. Central government recognition of these new professional roles and a minimum of financial encouragement and logistic support could go great distances in the way of improving decentralized planning practices.

Donor and project-driven decentralized planning

A powerful obstacle to sustainable change in local planning practices may be the fact that such practices are driven by donor-funded projects. There are two aspects to this question. First, the more visible and self-contained is the project format, the more likely are the concerned populations—local authorities and state agencies—to adopt attitudes of passivity or opportunistic "rent-seeking". It is the understandable reaction to what they perceive as a temporary exercise to access one-time benefits of external origin. Second, having recognized that some devolution of resources is critical to meaningful local planning, some donor-funded projects (including the UNCDF–supported LDF) also provide access to financial resources for local plan implementation. However, the project format of such assistance implies that targets and deadlines for delivery of physical outputs and financial resources be established. This may try to push too much money too quickly through a still weak locally run system and replace the emphasis on capacity building with a focus on infrastructure delivery. It may also conflict with the long-term time frame needed for systemic and sustainable change in local planning practices.

These problems are obviously at the core of the design of LDF programmes and related LPPs. For example, early lessons from the implementation of the LDF programme in Mozambique suggest that the visibility of the LDF project is a major obstacle to the changes in perceptions and attitudes that are critical for the institutional sustainability of the LPP. A reduction in the project’s visibility, the early buy-in of deconcentrated state agencies and the integration of the project in their regular operations are, therefore, the keys to successful introduction of a sustainable LPP. (5) Similar efforts are being made in Vietnam and Cambodia to ensure that the extension and support of the LPP become an integral part of the mandate and regular workload of the provincial planning agencies. Also, the overall size of the decentralized financing facility (the LDF) is carefully determined to reflect considerations of local-level absorptive capacity and overall financial sustainability. The aim is that the flow of resources in the LDF experiments be compatible with a realistic central-to-local grants system in the medium-term (less than 10 years). Finally, it becomes increasingly clear that the capacity-building effort required by the LPP development and extension calls for a long-term, if not open-ended, frame for UNCDF assistance. All these considerations make the design of UNCDF assistance to the development of decentralized planning systems less and less compatible with the traditional project approach.

Local politics and planning

Difficulties with the introduction and implementation of decentralized and participatory planning procedures do not come only from the state’s unsteady commitment to decentralization reforms. Nor do they originate exclusively in the donors’ lack of instruments to appropriately support them. Local political realities also play a major role. A natural tension exists between local planning and local politics, which has been confirmed in many of the LDF experimental programmes at an early stage of their implementation. To the extent that planning practices attempt to introduce better informed, more transparent and more rational decision-making, they not only specifically conflict with the practice of patronage, but also, and in more general terms, reduce the discretion of local leaders, constrain the mode and the tempo of their decisions and, in one word, limit the "autonomy" of politics. In Palestine, for example, some local political leaders have reluctantly accepted the participatory planning exercises introduced by the LDF programme. They initially felt threatened by such exercises, which they saw as questioning their ability to understand local priorities and manage the solution of local problems. They were eventually fully brought into the participating planning exercises only when it was clear that not only were the exercises a necessary step to access the programme’s resources, but the exericses also helped to manage some local conflicts and make more efficient and equitable use of the programme’s resources. But local politicians’ aversion to planning—and particularly to participatory planning—may also be related to other considerations. When participatory mechanisms are incorporated to planning procedure, the result may be a surge in demand for social infrastructure and welfare subsidies—a strong pressure to spread scarce resources in numerous small projects. Local leaders understandably fear the unleashing of such a local demand, which would then require them to openly exert some form of control by making difficult and potentially unpopular decisions. (Top)

 

The local planning process introduced by local development fund programmes

In the framework of the UNCDF–supported LDF programmes a conscious effort is made to introduce and support improved development planning and management activities at the local level. The complete set of these activities has been synthetically referred to as the local planning process (LPP). The use of the terms "planning" and "local" requires clarification.

"Planning" and "local" in the local planning process

As used in this discussion, the local planning process does not refer only to the activity of planning strictu senso, but also to the knowledge-acquisition activities that precede it and the implementation activities that follow it. (6) As such, the LPP covers all the steps in the cycle of development projects typically undertaken by local authorities. These include diagnostic studies and problem analysis, strategy setting, project identification, selection, formulation, appraisal, negotiation, budgeting, implementation, monitoring and evaluation.

The term "local" also covers different concepts. It refers to a specific (often the lowest level) territorial unit (a district, a sub-district or a commune), that may benefit from budgetary autonomy and financial transfers in the frame of fiscal decentralization. The principle here is that there should be local public sector planning only where local public sector financing can be institutionalized and sustained.

This point deserves some elaboration. Proponents of participatory and community development approaches point to the fact that the spaces with which most rural people immediately identify themselves—and where direct community participation is easier to elicit and organize—are the neighborhood, the village, the terroir, and so on. Programmes supporting community-level planning and the related direct external financing of community-level initiatives have been built on the basis of this observation. But the long-term financial and institutional sustainability of these programmes is questionable. The fact is that village or sub-village spaces are not necessarily viable planning units for the location of most public facilities and the delivery of most public services. And yet, from the point of view of the state and its decentralization policy, the viability of a spatial unit for the delivery of a certain "bundle" of public services is essential to justify fiscal decentralization at that level. And, obviously, the definition of a limited number of devolution levels, corresponding to such viable units, is essential for any manageable decentralization policy. It follows that, in the practice of decentralization reforms, the lowest levels at which public sector responsibilities and resources may be devolved and the conditions for meaningful decentralized planning may be created (the commune, the district, and so on) rarely coincide with socially or functionally homogeneous spaces.

Local planning in the LPP means planning in the public domain by local authorities at the commune or district level. But the LPP focus on local authorities is not an exclusive one. On the contrary it is understood as the entry point in a complex, multi-actor, local space, which requires an approach to planning as dialogue. First a local-local dialogue between authorities, communities and citizens. In this perspective the selection and financing of grassroots initiatives are expected to take place within, rather than outside, the local authorities’ planning and financing processes. The access of communities to resources of their state—more than to those of externally funded nongovernmental programmes—is seen as a key facet of improved local governance. Dialogue between district authorities and higher-level authorities or deconcentrated state administrations—who may be involved in the administration and management of fiscal transfers—may offer an opportunity for additional financing of local development initiatives from the sector programmes they administer, and may play an essential facilitation and technical assistance role.

A generic local planning process

Obviously the LPP will differ from one country to another in response to different institutional environments and degrees of political, administrative and fiscal decentralization. However, in its most generic form the local planning process will be concerned with:

• The assessment of territorial imbalances and the establishment of a formula-driven geographic targeting of central-to-local financial transfers.

 

• The assessment of local development problems and the identification of community needs and preferences, through both conventional diagnostic studies and grassroots participatory appraisal exercises of appropriate scale, format and frequency.

 

• The consolidation of community proposals and priorities at the commune or district level and the selection by a representative local planning authority, through a transparent and informed decision-making process, of the projects or actions to undertake.

 

• The formulation of selected projects with the active participation of representative groups of users or beneficiaries responsible for managing or monitoring the project implementation.

 

• The adoption of a local development financing plan—including negotiation with potential external sources of local financing, from the government, a nongovernmental organization (NGO) or private sector, in an appropriate provincial-regional consultation and negotiation forum; allocation of internal resources from fiscal transfers; local revenue mobilization and community contributions to selected projects; and linkage of planning and budgeting through the preparation of multi-year and annual development budgets reflecting projected income and capital and recurrent expenditures.

 

• The implementation of local development projects through modalities that maximize the involvement of final users or beneficiaries as managers or monitors in the local procurement of goods, services and works, while securing adequate levels of technical supervision and quality controls.

 

• The design and implementation of monitoring and evaluation systems that provide both tools for local development management and replanning and data for wider-purpose efforts to generate policy lessons on the link between decentralization, improved local governance and local development.

The focus of the LPP is therefore not on strategic or long-term planning but on the preparation of medium-term local investment programmes and annual action plans and budgets. Wherever a multi-year rolling investment programme is prepared at the national level, the medium-term local planning instrument should be synchronized with it to reflect an ultimate policy objective of the LDF experiments—the financing of local investment programmes through appropriate decentralization of the national public investment programme.

Though not a linear process, the LPP is meant to be simple and quick to implement, taking into account the lack of financial and human resources characteristic of most rural areas in developing countries. LPP implementation has also led to one form or another of further simplification with respect to initial designs. In the end the quality and effectiveness of the LPP depends on the gradual buildup of the capacity of local leaders and administrators. This buildup in turn depends on the willingness of national authorities to invest in a long-term effort for the creation of such local capacity. In the frame of LDF programmes such effort is piloted through the appropriate training and reorientation of a selected cadre of deconcentrated state and non-state agents as providers of planning facilitation and technical assistance services to the emerging local authorities.

LPPs have now been introduced in several countries within the framework of UNCDF–supported LDF programmes. Before examining in greater detail some of the issues raised by their implementation, we need to address two more general questions. The first is the introduction of the LPP in association with a specific financial facility—the LDF—and the advantages and risks involved in such association. The second is the links between the LPP and the regional or national planning system, the relevance of such local and regional linkages for local development and the challenges associated with their establishment.

Local planning process and local development fund: advantages and risks of their association

Where LDF programmes are implemented, the LPP is introduced simultaneously with a pilot decentralized financing facility: the LDF itself. Improving local development planning and management practices while providing a modest level of regular financing as an incentive to such improvement is an essential point of the LDF programme design. (7) It allows such programmes to go beyond traditional technical assistance and training approaches to local capacity building. It allows the setup and practical experimentation of real-life systems of local development planning and financing. Beyond training alone, it is this systemic approach and practical experimentation, over an extended period of time, that is key to the changes in institutions, procedures, incentives and attitudes that constitute local capacity. Also, practical system experimentation may provide strong support to policy dialogue, and advise aiming at institutionalizing improved local planning practice through national and local bylaws and guidelines.

On the other hand, the fact that the LPP is introduced in the frame of the LDF carries the risk that the LPP be seen only as a procedure to access LDF funds—an exercise that concerned local authorities would go through to please the donor or the higher-level government agency that sponsors the LDF programme. In the end, the narrow and exclusive association of a planning process (LPP) with a specific financial facility (LDF) induces serious distortions in both the LPP and LDF concepts and implementation. (8)

On one hand, the scope of the LPP is narrowed by the perception that the local initiatives that may receive funding are only those eligible for LDF financial support.(9) What is lost is the opportunity for a broader and more comprehensive LPP in which the initial steps of problem identification and analysis are given adequate attention, and the possibility to apply local public sector capital resources to activities other than infrastructure (such as natural resources management and income-generating activities) remains open.

On the other hand, the LDF is seen as just another donor programme that could finance some locally-relevant project. Its real value as a facility that empowers local authorities with discretionary funds, and therefore with the opportunity and flexibility to leverage both community and other agencies’ additional financing, is not fully grasped. A conventional view of the LDF as another "sectoral" programme may then prevail. As a consequence, little is done to establish horizontal linkages to other programmes and secure co-financing—even when the case for local-level coordination of different assistance programmes is evident and could indeed be facilitated by the flexible application of LDF resources within a broader LPP.

How can the above difficulties be overcome? A national commitment to develop a decentralized planning policy and donor willingness to support it are the most important factors in the adoption of local planning practices independent from specific donor-funded programmes. As for the LPP, its design may contribute to this objective by:

• Broadening its scope.

 

• Conceptualizing its initial steps as problem analysis or goal setting rather than simply project identification.

 

• Focusing on the project selection, negotiation steps and the related responsibility of the local planning authorities to act and establish links and dialogue with many potential sources of financing in the wider regional and national areas.

Relation with regional and national planning

The establishment of the LDF and the linking of decentralized planning with decentralized financing also bring a different perspective to the relations between local-level planning and the regional-national planning system. Under decentralized financing conditions, local planning is no longer the preparation of a wish list of projects that are meant to be fed bottom-up to regional-national agencies for their selection, financing and implementation as part of sector programmes they administer. Instead local resource availability and discretion over their use allows a more meaningful local planning process—one in which local choices can and must be made under clear budget constraints.

But opening a space for local choice does not entirely eliminate the need or resolve the problem of articulating local and regional-national planning. For one, many problems identified as bottlenecks of local development in the course of the local planning process are often common to several local authorities at the same time and do require a supra-local solution. They may require the set up of ad hoc consortia of concerned authorities, or may be tackled in the frame of regional-national programmes. Even if substantial resources are devolved to local authorities by decentralization reforms, an important part of public investment financing will continue to flow through national sector channels and be managed through national or regional programmes. Two problems then arise. The first is how to promote interjurisdictional cooperation between local authorities to take advantage of economies of scale in service provision and pool resources to tackle common problems more efficiently. The second is how to associate local authorities to the planning and implementation of regional-national programmes and make such programmes more relevant to local needs and more apt to mobilize additional local resources. Many UNCDF–supported LDF programmes have not directly addressed these problems. However, early lessons from those that have indicate that neither interjurisdictional cooperation nor a greater involvement of local authorities in the design and management of regional-national programmes will emerge spontaneously from the initiative of the local authorities themselves, even if explicitly instructed that they could use LDF resources for those purposes. Horizontal cooperation between local authorities must be, at least initially, actively fostered from above through the creation of appropriate multijurisdictional consultative platforms and the setup of related incentives for participation. Also incentives for local authorities and communities to be associated and contribute to the financing, planning and implementation of regional-national programmes must be created by offering appropriate contractual (principal-agent) arrangements. In summary, an appropriate regional development policy environment is what could make the difference between emphasizing local choice and accepting a high degree of inefficient localism. (Top)

 

Early lessons from LPP implementation

The implementation of the LPP in a number of countries has provided some preliminary and limited lessons directly related to the rationale of decentralized local-level planning.

Increased popular participation in planning and development?

The extent of feasible popular participation in local-level planning and development, in any given country at any given time, is determined first and foremost by sociopolitical and cultural factors. These factors have a scope and depth that far exceed the domain of administrative re-engineering, in which the improvement of local planning procedures takes place. This is not to say that—given the appropriate design and the right timing—such improvements cannot also contribute to increased community participation and even wider social, political and cultural changes. But formal changes in administrative processes (for example the introduction of decentralized local-level planning) are no guarantee of actual improvements in broad-based popular participation and may have different outcomes in different countries.

Studies of decentralization reforms in the highly stratified societies of South Asia have shown that the local elite may take advantage of the opportunities and capture the benefits associated with decentralized planning, leaving the poor in a state of identical if not increased marginalization. (10) This leads to the somehow obvious need for a better understanding of the local power structures and dynamics in order to maximize the chances of increasing popular participation in the LPP. An even more useful insight from these studies is that the ill-preparedness and the cultural resistance of the deconcentrated bureaucracy might have been decisive factors to allow the local elite to capture the decentralization process. The unholy alliance of local political patronage and persisting centralist bureaucratic practices have betrayed the promises of decentralization for popular participation in local planning and decision-making. The ultimate cause of such betrayal can be traced to the lack of central commitment and adequate incentives for the professional reorientation of deconcentrated state agents as both supervisors and facilitators of local planning. This lack of real commitment—disguised under the rhetoric of decentralization and local autonomy—has proved fatal to decentralization reforms in general and to their participatory development goals in particular. While decentralization reforms may be a necessary condition, they are not sufficient for broad-based participation in local-level planning. Central governments embarking on decentralization reforms must understand their role as one of "change agent" to open spaces for popular participation in local-level planning. The key is adequate investment in the extension of participatory planning practices and accountability of local authorities for their implementation. The role of the deconcentrated, front line state agents is critical.

This lesson is being relearned in the early stages of implementation of many LDF programmes. These experimental programmes rely on a national infrastructure of planning facilitators and technical staff (at the district-provincial-regional level) to support and supervise adoption of the local planning process by lower level local authorities. These include provincial and district staff in the planning support groups and technical support staff in Cambodia, district planners in Vietnam and Mozambique and planning and technical staff of the micro-regional planning councils in Palestine. Their performance is critical to the success of the LDF experiments and they are therefore the targets of intensive capacity-building efforts. The pace at which capacity can be built in this national infrastructure by and large dictates both the pace of possible expansion of the LDF programmes and the quality of their implementation. International experience has shown that building local planning capacity is a long-term endeavor and a particularly challenging one, given the mix of technical, managerial and social facilitation skills required from local planners. The limited experience in LDF programmes so far, seems to indicate that:

• There is great variation in the quality of trainable human resources in different countries, but this does not seem to pose insurmountable problems to the introduction of the local planning technology under the LDF programmes, and relatively quick results have been achieved in this respect.

 

• Greater difficulties are encountered in changing attitudes than in teaching techniques, particularly when local planners are selected from the technical staff of local administrations. Actual on-the-job experience and the related development of confidence and sophistication cannot be substituted by any amount of formal training.

 

• The association of nongovernmental organizations to the deconcentrated state agents as part of the above national infrastructure of planning facilitators is most valuable in the early stages of the effort to extend the LPP, because of their greater familiarity with participatory development attitudes and techniques. However, it poses problems of its own since most NGOs are unaccustomed and sometimes reluctant to work within the formal public sector planning system.

 

• Together with adequate remuneration, incentives in terms of central support and professional recognition are critical to secure a good performance. The long-term sustainability of the local planning support infrastructure depends on the ability of central and local authorities to provide such remuneration and incentives.

While the central commitment to create an infrastructure to support and supervise local-level participatory planning is essential, equally important is the design of the LPP itself to create extensive and meaningful opportunities for participation. This is a matter of institutional design—that is designing both planning procedures and structures for participatory consultation, decision-making, implementation and evaluation of local development activities. This institutional design must strive for a balance between direct participation and political representation as it can be achieved in any specific country situation.

Whether the local planning authority is an elected council (Nicaragua, Palestine, Vietnam) or a decentralized administration (Mozambique), there is need for a network or hierarchy of specific structures that provide a vehicle for systematic consultation between many local development actors and for community participation in the identification, selection and implementation of development projects. Such structures may take different forms and names (commune development boards in Vietnam, commune development committees in Cambodia, district development consultative councils in Mozambique and municipal development committees in Nicaragua). LDF programmes support the establishment of such institutions as platforms for dialogue between the local government authorities, the local state administration, existing civil society organizations (associations of farmers, traders, fishermen, churches and affiliated groups and locally active national and international NGOs) and traditional authorities and other community structures emerging in the process of local participatory planning exercises. However, the functions and limits of these structures in promoting popular participation must be clearly understood.

First, in those situations where local councils have been elected and genuine local governments operate, these structures should not be conceived as alternative planning authorities, but rather as platforms set up by the local planning authority (the local government) to reach out to the community and solicit a dialogue with respect to major decisions and actions for local development. This point seems obvious, but it is worth making if one considers the widespread practice of donors and NGOs to set up and finance directly all sorts of community, village or municipal development committees, bypassing and disempowering locally-elected councils. In extreme cases (Nicaragua) this has created additional cause of local political tension. Nevertheless, if properly conceived, these structures could provide a valuable consultative forum influencing and facilitating major planning decisions. These include the establishment of strategic objectives, the selection of areas for priority interventions and the intra-area targeting of LDF resources, down to the selection of projects to be included in the local plans and the monitoring of all project implementation.

Second, the potential of these structures for participation of the poorer strata of the rural population should not be overstated and should be seen in an evolutionary perspective. In practice the full potential of such structures to represent all strata of the rural society and offer a strong institutional conduit for broad-based community participation will be realized gradually and over an extended period, since it depends on structural changes in the sociopolitical distribution of local power. Such changes cannot be induced only by improving local planning practices and institutions. Nevertheless, one could expect that the composition and activity of the consultative platforms set up as part of an improved LPP should evolve to reflect the emergence of new forms of organization and leadership at the community level induced inter-alia by the more direct community-level participatory planning exercises (participatory surveys and rural appraisals) also advocated by the improved LPP. Therefore, while the need for these platforms, their institutional nature and a core set of their functions is clear, the process of their establishment, their composition and rules of operation remain to be defined through an intensive local-level information and consultation process. LDF programmes may not have paid sufficient attention to this information and consultation process and in the future should allocate appropriate resources to support it.

As for the appropriate techniques for local participatory planning, the basic problem in the design of LDF programmes and related local planning processes is achieving the "maximum feasible degree of grassroots participation" within the formal systems of local public sector planning. This has two main implications. First, the very fact that popular participation is embedded in a specific administrative system and therefore must somehow be functional to its workings has a direct bearing on its nature, format, intensity and timing.

Second, extensive limitations and constraints on participation typically characterize these systems. (11)

• First, there is a lack of clearly defined and legally mandated modalities for incorporating popular participation in local-level planning.
• Second, at all levels in the system there is a dearth of trained staff with skills for involving the community in local-level planning.
• Third, there is a lack of financial resources to cover the cost of organizing and implementing participation (workshops, meetings, travel) often compounded by poor access to and between communities in sparsely populated rural areas.

In trying to address these constraints, the design of the local planning processes within the LDF programme has been forced to evolve. An initial emphasis on extensive use of PRA techniques for problem analysis and project identification at the grassroots level was replaced by the use of more rapid and focused rural appraisal and participatory survey techniques. (12) Simultaneously, more importance was placed on the participation of final beneficiaries (through project holders committees) in the actual formulation and implementation (including procurement responsibilities) of selected projects. It was also recognized that the process through which the extended local planning authority analyses community priorities, selects projects and agrees on the strategy for their financing, was the core of the LPP and had to be better supported. While it was recognized that the local decision-making process was ultimately a political one, it was also felt that such a process could be a better informed and more transparent one by making available to local decision-makers some basic planning and technical information and simple techniques to evaluate projects against commonly agreed multiple criteria.

Make plans more relevant to local needs

To the extent that community participation in needs assessment and resources allocation has been enhanced by the adoption of a LPP, plans and actions generated through the LPP are more responsive to locally perceived needs and priorities. Although a comprehensive, comparative evaluation has not yet been made, numerous LPP monitoring reports provide strong, anecdotal evidence to this effect. They also point to three main areas of difficulty that prevent a fuller realization of the potential of decentralized planning to generate plans and actions of greater strategic relevance to local needs. These difficulties apply to decentralized planning in general, but may be worsened by donor interventions in support of decentralized financing.

The first is related to local politics and the lack of planning capacity. When empowered with resources from donor-supported financing facilities—given relative discretion in allocation and asked to implement a participatory planning approach (as in the UNCDF–supported LDF programmes)—local authorities may end up spreading such resources throughout their jurisdiction in the politically neutral form. The result is the one-village-one project approach followed at the outset of the LDF programmes by commune-level authorities in Cambodia and Vietnam. This is the same phenomenon of saupoudrage (dusting) of resources in numerous low-quality civil works projects experienced in Senegal in the late 1970s. Clearly, a long-term investment in capacity building is required to overcome this bias of local leaders and build a local capacity to make difficult allocative decisions and optimize the use of scarce resources. But it is encouraging to see that this capacity can be increased with the introduction of appropriate planning techniques and allowance for the required learning and experiencing time. In Cambodia and Vietnam implementation of a second round of annual planning exercises—better support to the project selection step in the LPP—showed that local decision-makers can move from the initial political concerns with equal spreading of what they perceive as donor-funded one shot windfalls to a more strategic use of regular fiscal transfers.

The second difficulty is related to the concentration of decentralized planning (including most LPPs implemented so far) on projects as the main agents of change and on capital budgets as the only planning tool. (13) Important as capital programmingby projects is, the concentration on projects excludes from the planning focus a number of other actions with potentially great impact on local service delivery.(14) The concentration on capital budgets excludes from the planning focus other tools like the recurrent expenditure budget and the regulatory, organizational changes often necessary to achieve the desired levels of service delivery. The large role played by donor assistance in local development financing and the fact that it privileges project-led development and change is a big part of the problem. Because of the place external assistance has in financing development activities in Mozambique, this problem was perceived as acute by programme designers involved in the setup of the UNCDF–supported LDF experiment in Nampula province. Fortunately, they could count on a central commitment to introduce district-level planning procedures that were open to but independent from specific external funding programmes or facilities (including the UNCDF–supported LDF). The district planning process (as the LPP is called there) was designed not only to support local capital programmingbut to integrate such programminginto the preparation of an annual District Economic and Social Plan (Plano Economico e Social (PES) do Distrito). The PES is a tool for short-term development management driven by service delivery objectives. It includes all activities and investments to be carried out by multiple institutional actors and is linked to the preparation and execution of the annual budget. Early lessons from the LDF programme implementation indicate that this is indeed the right direction to follow.

The third difficulty, which stands on the way of a greater relevance of decentralized planning to local needs, has to do with the restricted range of objects of financing allowed under most central-to-local development grants systems (including the UNCDF–supported LDF). This refers first to the strict exclusion of all recurrent expenditure financing, in spite of an obvious need for local-level flexibility in this respect. Second, it refers to the lack of policies and instruments enabling local authorities to go beyond the provision of local infrastructure and services and be proactive in protecting the environment and supporting the community management of natural resources or promoting local economic development.

As for the objects of financing supported by the LDF, these are classified into three broad areas:

• Provision of infrastructure and services.

 

• Protection of the environment and management of natural resources.

 

• Promotion of local economic development.

In practice, either by design or default, the objects of financing have been restricted to small-scale economic and social infrastructure facilities. In this area local authorities have a clear delivery comparative advantage with respect to other local institutional actors. In fact, if they wanted to deliver the goods in the two remaining categories, in most cases they would have to go through some form of intermediary institution. Practical, organizational and even legal difficulties in structuring such inter-mediation, cumulate with the reluctance of local politicians and administrators to further devolve responsibilities and public funds to civil society organizations and produce a strong bias in favor of the exclusive allocation of LDF resources to the building of physical infrastructure.

If we turn then to the potential objects of financing as they emerge at the project identification stage and travel through the local planning process, in a number of LDF programmes (Cambodia, Malawi, Mozambique, Palestine) we may observe a symmetric bias in favor of local infrastructures. Project proposals other than infrastructures do emerge at the community level in the participatory problem analysis and project identification exercises. They are often supported by poorer and disadvantaged groups, as in the case of food subsidies for the landless and poorest in Cambodia and income-generating activities for women in Cambodia, Malawi, and Mozambique. However, if such proposals are forced to compete with public infrastructure projects in the same project selection forum they will inevitably be overrun and put aside. The reasons are multiple and differ in the case of safety nets or income-generating projects. The income-generating are usually concerned with a limited sub-sector of the local community; application of simple criteria of number of beneficiaries or cost-effectiveness may be fatal to them at the project selection stage. Second, these proposals are seldom off-the-shelf projects. Even assuming that templates for their analysis could be developed, they would require more time and effort to be filled in than in the case of simple infrastructure projects. And most important, they are seen as ways of favoring a small subset of the local community or putting public funds in private pockets—something that cannot be undertaken directly by a local political forum but that requires the mediation of a neutral and transparent allocation mechanism.

Facilitate coordinated or "integrated" (multi-sector) planning

The assumption that decentralized planning facilitates inter-sector coordination and leads to locally integrated programmes and related development synergies has long been part of the rationale of decentralized planning. Experience has shown that such an assumption is verified only to the extent that the local planning authority responsible for horizontal, multi-sector planning is strengthened by financial resources and technical capacity. In many countries local-level representatives of national ministries have often been invited or mandated to cooperate among themselves, usually under the direction of the chief state representative (provincial governor, "préfet"or district administrator) to put together a local, integrated development plan. Most of these plans have been technically weak (such as the collating of sector-specific wish lists) and have seldom been implemented. On one hand, the absence or technical-institutional weakness of local planners has made it difficult to identify and assess alternative local development strategies, obtain community views and priorities and prepare integrated plans. On the other hand, the fact that decision-making and financing remained centralized and individual deconcentrated line agents had to seek approval from their headquarters for each sector’s activities made it impossible to approve and implement integrated plans.

In the LDF programmes both the need for devolution and financial empowerment of local planning authorities and the need for their technical strengthening are addressed. Some evidence confirming that such devolution of resources and technical strengthening are the keys to local-level inter-sector coordination is beginning to emerge. In Mozambique the empowerment of district administrators with discretional resources for development planning and the fielding and training of district planning officers have somehow reoriented the district-level line agents from the role of executing provincial programmes to preparing and implementing an integrated district-level plan. Moreover, they have increased the capacity of the district to negotiate with provincial-level line agencies—the content of the sector programmes that such agencies implement in the District. Similarly, in Vietnam the devolution of resources and the technical support to communes to implement the LPP has created an opportunity for a more holistic look at local development problems and for the integration of programmes managed by province and district-level agencies with locally financed and implemented activities.

While the impact of local-level financing and capacity building on horizontal, multi-sector coordination at the local level is to be expected, less intuitive is the impact on the coordination of local and supra-local agencies. More attention should be given in the LDF programme design to what such programme may accomplish in terms of fostering coordination and partnership arrangements between multiple levels of local authorities or state administrations.

Increase the speed and flexibility of decision-making and implementation

No comparative research has been carried out so far that could help determine whether decentralized planning procedures (LPP) introduced in LDF programmes lead to quicker decisions and implementation of local projects. Some observations based on evidence from Vietnam point in two different directions.

On one hand, the process of identification, selection and approval of local initiatives seems to be a quick one (a matter of a few weeks) even if it involves, in each commune participating in the LDF programme, a series of village-level participatory surveys and deliberations and requires an approval of each initiative (to check conformity with LDF guidelines) by provincial authorities. Given the discretion to allocate LDF resources and a "control of legality" of their choice by higher authorities, the commune authorities have experienced a shortening of the time between identification of a project and its approval.

On the other hand, once the project activities have been approved, management by commune authorities of the process of detailed formulation, technical design and procurement seems to be slow compared with the standards set by the LDF programme managers (but not compared with prevailing national-provincial standards). The reason for this seems to be the limited capacity of district-level technical agencies to provide timely and quality services to multiple commune authorities, while the private consulting industry is still underdeveloped and uninterested in servicing rural clients with small budgets or projects. This case indicates that, contrary to popular claims, the main delays in the overall cycle of locally managed projects are not associated with articulation of the demand (through participatory approaches), but rather with the inadequate supply response to the increased and diffuse request for technical services.

Generate additional resources and encourage more efficient use of existing resources

Additional resources have been generated by the implementation of LDF programmes because of the programme design, which requires matching community contributions of different levels (usually between 10% and 30% of the individual investment cost). In Vietnam, where the rural population already shoulders a heavy financial burden in the form of contributions for public programmes and services, the requirement of a matching contribution to the LDF has been accepted with trepidation. A few cases have been documented in which project proposals made by villages have been withdrawn because of the inability or unwillingness of villagers to pay the matching contribution. Yet the requirement for a matching contribution may play an important role in determining the actual local demand beyond the simple assessment of local needs. The lesson learned here is that a flexible co-financing matrix establishing matching requirements as a function of the type of project and the relative poverty of the demanding community would be more appropriate than a fixed percentage requirement across the board. No fully satisfactory solution to this problem has been implemented so far because of practical difficulties in determining an appropriate co-financing matrix and the fear that its introduction may produce as many distortions as it tries to correct. Considerable problems are associated with the form (cash or kind, labor and materials), which the additional resources mobilized within the LDF programmes take. Particularly when local contributions are available only in kind, their incorporation into the projects has consequences for the procurement modalities, and may lead to some inefficiencies and loss of quality in the final product. (15)

As to the effects of local-level planning on the efficiency of the use of decentralized resources, no in-depth investigation of this issue has been done across the ongoing LDF programmes. However, some reasonably strong evidence is available from a series of monitoring reports on the Cambodia LDF. In general, the commune and village-level management of the procurement process has resulted in cost-effective implementation of the selected projects. Costs have been kept within acceptable standards and, in many cases, they have been lower than those incurred by provincial or national agencies for similar projects. However, the choice of the projects and their appraisal has suffered from lack of local technical capacity and local political pressure. For example, in the early stage of LDF implementation in Cambodia the commune and village-level choice of some access road projects, had more to do with the creation of employment in the village than with the creation of a sustainable productive infrastructure. Those projects were close to "make work" projects. Anecdotal evidence from Palestine confirms that community participation in decision-making, project design and implementation resulted in cost-effective implementation of selected projects. In the village of Kufr-Rai an agricultural roads project proposed by the village council could not be funded with LDF resources because its cost exceeded the available allocation. A community-based committee was then set up to look into possible alternatives. The result was not only the redesign of a much cheaper, more cost-effective project, but also one that was more equitable and covered a larger number of beneficiaries—not just those in a section of the village who belonged to a single powerful local clan. Furthermore, once directly involved in monitoring the project implementation, the committee managed to extract from the contractor a number of small additional works, thus maximizing the actual work performed within the original contract’s price. (Top)

 

Some conclusions for future LPP design (in progress)

• Efforts to introduce the LPP should be more context-sensitive and more explicitly designed as contributions to the improvement of the local-level component of the national-regional public sector planning system. Accordingly, they should also be conceptually and organizationally de-linked from the availability of financing under the LDF to avoid having the LPP be perceived as an externally driven procedure to be complied with only to access a specific donor-funded facility.

 

• A new and greater emphasis should be placed on those steps and procedures in the LPP that support an inter- and supra-local process of negotiation and the financing of local development activities.

 

• Appropriate platforms for interjurisdictional, intergovernmental and public-private negotiations should be promoted to foster links among local authorities, with higher-level deconcentrated administrations and with NGOs and the private sector.

 

• The focus on local choice, typical of LDF programmes, in which local authorities apply their own or transferred resources to the financing of local priorities, should be complemented by greater attention to contractual modalities of local development financing, including principal-agent arrangements with higher-level authorities or government agencies. UNCDF should support local authorities to enter into such contractual relations and participate in the planning and implementation of national-regional programmes of local relevance.

 

• Greater recognition should be given to the role of the central state in the successful implementation (or outright failure) of decentralization policies, including the development and adoption of local-level participatory planning procedures. This role should be understood and restructured as one of both support (financial and technical assistance) and supervision (upward accountability and legality control), and it should be reflected in an appropriate set of incentives and sanctions that affect local authorities’ performance.

 

• Appropriate support should be given in the frame of LDF programmes to deconcentrated state agents, enabling them to provide planning facilitation and technical services to emerging local authorities on a sustainable basis. The emergence and professional recognition of a new cadre of service-oriented local planners and technicians (often referred to as bureaucratic reorientation) should be recognized as being critical to successful decentralization. It should be interpreted as a dynamic complement to a demand-driven approach to local capacity building.

 

• The LPP should not be exclusively concerned with capital programmingand the capital development budget, but should more holistically aim at improved service delivery and therefore also affect the allocation of recurrent financing. This will require the establishment of a closer link between planning and budgeting, subtracting the budgeting to a purely accounting perspective and bringing it fully into the fold of the overall planning process. It will also require that local authorities engage in a more open and proactive search for inter-organizational (including public-private) partnerships for the co-financing of recurrent expenditures.

 

• Some pilot within the pilot experiences should be initiated immediately to develop and test institutional mechanisms that enable local authorities at appropriate jurisdictional or interjurisdictional levels to promote sustainable management of natural resources and protection of the environment and to foster local economic development. Such efforts should focus on the setup of effective and transparent institutional arrangements through which local public authorities can provide incentives to private and collective action with wider social and environmental benefits. (Top)

---

Notes:

(1) Jonathan Fox and Josefina Aranda, 1996, Decentralization and Rural Development in Mexico. Center for US-Mexican Studies at the University of California, San Diego, p.2 (return)

(2) James Manor, "The Political Economy of Decentralization"1997 Paper presented at the Technical Consultation on Decentralization, FAO, Rome, December 1997. Based on such assessment Manor seems to suggest that efforts to promote local-level participatory planning should be abandoned without even much regret since he concludes "Democratic decentralization can work quite effectively without effective planning from below". (return)

(3) As Manor himself recognizes "The main problem with planning from below, however, is that higher authorities who extol its virtues often have little interest in it in practice". Ibidem 1997. (return)

(4) See Diana Conyers, 1990 "Decentralization and Development Planning" in P. de Walk and K.H. Wekwete, eds., Decentralizing for Participatory Planning, Avebury: Brookfield. (return)

(5) In the words of David Jackson, a UNCDF-UNDP planner helping to develop the district-level planning process in the province of Nampula, the first obstacle to the introduction of such a process is the difficulty of "Getting away from the idea that it is a project, or something special, and introducing the idea that it is part and parcel of the [provincial] planning system. […] This point applies at all levels, from the community to the DPFF [The Provincial Planning and Finance Department]. We now call ourselves the "Gabinete de Planificacao Distrital" [District Planning Office, within the DPFF] and have banned the use of the word "project". (return)

(6) The term "planning" in local planning process, does not exclusively cover the activity of planning, but is used as shorthand of the wider concept of "planned development". The LPP designates the process through which planned development takes place, of which the activity of planning is the centerpiece. The process of planned development is conventionally thought to include three sets of activities (the term "activity" is preferable to those of steps or phases, to avoid suggestions of linear sequencing). These are: analysis or knowledge-acquisition, planning and programmingand implementation-management activities. Because of its central position in the "knowledge-planning-actions" triad, a common and broad definition of planning is "the linkage between knowledge and action". For the same reason, the whole "planned development process" is often shorthanded as the "planning process". (return)

(7) This is true in a double sense. First, access to the financial resources of the LDF is conditional on the setup and implementation of a specific local planning procedure and machinery. In other words, access to LDF resources can be reduced or discontinued as a function of the local authorities' performance, including their commitment and ability to implement a participatory LPP. Second, the availability and the open public information on the amounts of financial resources from LDF programmes are meant to create an incentive for a more meaningful planning exercise, one that is potentially more open to popular scrutiny and participation. (return)

(8) In practice, the way in which the relation between the LDF and the LPP is understood by UNCDF local partners varies considerably from one country to another. As a general rule the establishment of a correct relation between the financial facility and the local planning process is facilitated by the existence of a wider decentralized planning programme or policy framework. For example, in Mozambique the district-level local planning process (LPP)-introduced in conjunction with the LDF facility-is meant to implement and further develop nationally mandated procedures, to access the LDF resources and those of provincial line agencies and locally active NGOs. In such case the capacity-building efforts made in the introduction and development of the LPP have a greater chance of leading to good and sustainable planning practices. By contrast, in the case of Vietnam, the LPP introduced in LDF-participating communes may still be perceived as only a tool to access LDF funds provided by a "special" project sponsored by two provincial administrations with UNCDF support. (return)

(9) This perception is strengthened when the LDF objects of financing are categorically limited to items in a specific menu (unless the menu itself and some associated rule for earmarking funds are used intentionally to widen the scope of LDF financing beyond the local public infrastructure. (return)

(10) Numerous studies of the decentralization process and decentralized planning policies in South Asia are available. See, for example, Habib Zafarullah, 1996, "The dilemmas of local-level planning in Bangladesh (1982-1990)", Contemporary South Asia 5(1): 47-65; Abdul Aziz, 1993, Decentralised Planning: The Karnataka Experiment, New Delhi: Sage Publications India; R.P. Misra, ed., 1990, District Planning: A Handbook, New Delhi: Concept Publishing Co. (return)

(11) For an analysis of constraints to participatory planning within the public sector planning system in developing countries, see Franklyn Lisk, ed., 1985, Popular Participation in Planning for Basic Needs. New York: St. Martin Press. Also P.de Walk and K.H.Wekwete, eds., 1990, Decentralizing for Participatory Planning, Avebury: Brookfield, and Bernd Jenssen, ed. 1992, Planning as a Dialogue, SPRING Centre, Dormundt University. (return)

(12) In the districts covered by the LDF programme in Mozambique, it was clearly logistically and financially unfeasible to conduct PRA exercises in all communities and villages within a yearly planning framework. A form of multi-year, intra-district targeting to select areas for more intensive annual consultation and participation exercises coupled with district-wide consultation through sub-district meetings is being tried. A similar process is being considered in Nicaragua, following the experience of the Social and Economic Investment Fund (FISE) there. By contrast, in Vietnam an intra-commune targeting approach, focusing on financial resources and conducting an intensive participatory rural appraisal exercise in a particularly "deserving" village, was initially chosen. Later, it had to be abandoned because the selection of the "deserving" village was not transparent and rather reflected the preselection of particular projects, which happened to be located in the selected village, based on priorities established by provincial planners. The subsequent adoption of Rapid Rural Appraisal techniques for consultation of all villages in a participating commune led to a better process of participatory project identification. (return)

(13) On this point, see Andrew Green, 1992, An Introduction to Health Planning in Developing Countries, Oxford University Press. Although primarily concerned with health sector planning, this volume provides a useful overview of the allocative planning problems common to most planning authorities in the public domain. (return)

(14) In UNCDF-supported LDF programmes capital programmingis considered a privileged "entry point" from which to start building local-level capacity in development planning and management. (return)

(15) For example, in a road construction project in Vietnam, commune authorities made separate use of the local contribution in kind (materials and labor) and the LDF cash resources. They dedicated the first to the self-help construction of the embankment and the second to contracting a local enterprise for construction of the pavement. When, during the period of maintenance of the finished pavement, some sections of it collapsed because of poor construction of the underlying embankment, the contractor could not be asked to make the necessary repairs since he could not be held responsible for the overall performance of the constructed facility under the terms of his contract. (return)