Corporate Policy Papers
Taking Risks: Background Papers
Background Papers Index
Accountability in Decentralized Planning and Financing for Rural Services in Uganda
Laura Kullenberg and Doug Porter
- What is the District Development
Project?
- Uganda's decentralization
- Achievements and challenges
- The District Development Project - design features
- Accountability and project design
- Minimum conditions and performance measures
- Upward and downward accountability
- Information and communication campaigns
- Uganda's decentralization
[Published in Entwickling und Landlicher raum, (1998) 32 (3): 11-15.
Note: Laura Kullenberg is deputy director, UNCDF New York. Email: Laura.Kullenberg@undp.org.
Doug Porter is regional technical adviser, UNCDF, based in Kampala. Email: dporter@imul.com.
The authors wish to thank Annet Mpabulungi and Roger Shotton for helpful comments on an earlier draft.]
Many development agencies and governments are experimenting with decentralization programmes that transfer resources and authority from central governments to local bodies to promote rural development and poverty reduction. Here we deal with one specific type supported by UNCDF that transfers grant capital directly to local governments for planning and financing rural infrastructure and services. UNCDF supports these programmes, known as local development funds (LDFs), in a dozen countries. (1) One of the common challenges faced in all such programmes is how to address key concerns about accountability.
This paper discusses why various issues of accountability have become central in decentralization policy and practice and UNCDF's recent experiences in addressing these in Uganda through the District Development Project (DDP). We highlight efforts made to develop sanctions and incentives to encourage good local government performance. (Top)
What is the District Development Project?
The DDP is not a typical standalone donor programme. It tests mechanisms and procedures for decentralizing capital budgets to the lowest feasible level to provide services mandated in the Local Government Act of 1997. The project is fully integrated within the local government system so that government can draw lessons (both positive and negative) to inform national policy of the full-scale decentralization of development budgets scheduled to occur over the next three to five years.
In this project accountability is not concerned with the special arrangements donors make to ensure that local actors within a project are accountable for use of donor funds, or that physical outputs are delivered according to the approved project document. Rather, accountability is understood from the perspective of national policy-makers whose main challenge is to define sustainable checks and balances for national application to ensure accountability and thereby offset the political risks associated with the devolution of development funds for rural service delivery. (2) Why is accountability an issue in this context? What concerns must governments address when they decide to transfer development resources and authority from central to local governments? How do these shape decentralization policy and the ability of advocates to "sell" decentralization politically and maintain that support?
Worldwide, it is notoriously difficult to devise incentives and sanctions to ensure accountability in public sector investment programmes. In many LDC environments the challenges are greater. Decentralization policy is still evolving. Institutions and legal frameworks needed to promote accountability (such as an active press and effective judicial and accounting systems) are not strong. (Top)
Uganda's turbulent history since independence in 1963 is well known. Coups in 1966 and 1971 were followed by war with Tanzania in 1979. A protracted guerrilla struggle from 1981-85 led to the victory of the National Resistance Movement (NRM) in January 1986. At that stage the economy was shattered. Law and order had collapsed and the discredited civil service was unable to deliver services. On taking power, the NRM announced fundamental institutional, economic and political reform. Donors have provided generous support conditional on compliance with a standard package of structural adjustment measures. Private sector activity has since dramatically improved and, although Uganda still rates as one of the world's poorest countries, growth in GDP has averaged around 8% since 1990.
The decentralization policy, announced by President Museveni in 1992, has become the international hallmark of Uganda's political transformation. (3) This ambitious and far-reaching policy is guided by several principles:
• Political and administrative control over services shall be devolved to the lowest feasible level of government according to the principle of subsidiarity.
• The assignment of functions to key levels of local government should be clear and unambiguous.
• Control over resources (financing and authority) should accompany the assignment of functions.
• Local governments must be accountable to their constituents for carrying out mandated responsibilities and to the central government when they spend resources on its behalf as its agent.
All aspects of government have been affected by the devolution of functions, competency and resources to elected local government councils. Statutes introduced in 1993 and 1997 and the constitution of 1995 transfer real powers to local governments' discretionary powers. They can hire and fire staff, elect representatives, govern their affairs and approve their own budgets without reference to the centre. They have rights to retain tax income and can initiate and execute socio-economic development plans. Local governments relate to the centre under a unitary system. These governments are vertically related, according to the principle of non-subordination.
Fiscal decentralization quickly followed devolution of political and administrative functions from 1993. Recurrent budgets for decentralized services were devolved through unconditional block grants based on formulae governed by objective criteria and remitted directly to the local government. (4) Devolution of capital or development budgets is to start 1 July 1998. (Top)
Decentralization has recorded many positive results. (5) Local governments know and defend their rights. The budgeting and expenditure process, previously known to only a few, has become more transparent. No longer a bureaucratic, exclusive process, budgeting is now the responsibility of elected politicians. Whereas malfeasance and corruption went undetected at the central level, increased transparency in local government has greatly improved capacity to expose these practices and initiate corrective action.
Local governments have streamlined their operations. Many have retrenched staff while bolstering previously neglected capacities, such as planning and financial management. Some have introduced new methods of service delivery involving NGOs and private agencies.
Problems remain. Inconsistent central government actions can undermine their ability to call local governments to account. Central transfers fall short of requirements, unconditional grants have been arbitrarily cut, funds are sometimes withdrawn for certain services and abrupt changes are often made in modalities for fiscal transfers to local governments.
Many problems are caused by technical deficiencies in the way local governments plan and finance investments. New procedures and regulations are unfamiliar or unclear, legal provisions are sometimes contradictory and most local governments face major human capacity constraints. But more problems can be traced to a lack of political restraint to keep expenditure within approved budget limits and to observe rules and regulations for financial management. (6) Corruption and collusion disrupt revenue collection, contracting and the accountability of politicians.
Donors have supported government in many creative district-focused projects. However, donor policies also continue to undermine effective decentralization because many prefer sectoral financing and have been reluctant to integrate their funds into the local government planning and budgeting process. To ensure quick delivery of the projects they finance, many donors continue to create parallel ad hoc implementation structures and negotiate projects with line ministries without regard to the impact on fiscally strapped local governments. (7)
Nevertheless, the central government is committed to the devolution of development budgets-the next crucial stage of decentralization-although resource constraints dictate that achieving full devolution is still some way off. Donors and many government policy-makers are still concerned with two pre-conditions for decentralization: the need to improve the capacity of local councils to plan, finance and manage the delivery of services to their constituencies and a pressing need to develop a system of incentives and sanctions to promote accountability and establish a clear link between taxes and transfers received and services delivered. (Top)
The District Development Project - design features
The DPP focuses on these two pre-conditions. It was designed according to several basic assumptions:
• Decentralization fosters local-local dialogue, which is the essence of good governance. In turn, good governance is believed to result in improved services, which is expected to have a positive impact on poverty.
• Fiscal transfers of discretionary resources to local government improve the responsiveness of political leaders. While local governments are closer to the people, discretionary or horizontally integrated funds allow for full negotiation of investments within an elected political body that represents and is accountable to the local community.
• Planning and financing of local investments must be anchored in the national statutory and regulatory framework. Donor-created parallel structures jeopardize sustainability and can undermine accountability.
• Community participation is defined in terms of political representation for which clear, constitutionally mandated provision has been made in Uganda, from village upwards through five levels of local council.
• Successful decentralization requires a strong, capable central government to mentor, monitor and regulate. The central government has a crucial role to play in setting the framework of incentives and sanctions for local government action.
Central to the DDP is a LDF that is allocated to district and sub-county councils according to a transparent needs-based formula. Approximately $14 million will be allocated to five districts over three to four years. Districts receive block grants, 65% of which is passed down to sub-counties, which in turn pass 30% to lower local councils.
Advance knowledge of these entitlements enables communities and councils to rank their investments according to a known level of resources. This prior knowledge of resources (as well as information about costs and design standards) is a prerequisite for effective planning and local accountability. This varies from most community-based planning exercises where communities prepare "wish-lists", which they forward upward without any knowledge of the amount of funds available to complement their own resources.
DDP funds are used for investments mandated under the Local Government Act of 1997, which emphasizes basic services, including primary health care and education, agriculture, water and sanitation and selected classes of roads. Participatory planning involves direct community involvement and decision-making through the normal system of political representation.
The DDP was formulated through an intensive process of consultation with local governments, community and private sector agencies lasting more than one year. Accountability was a major concern: each level of government raised concerns about devolving resources to the level below it, while lower levels doubted the ability of higher-level governments to effectively manage resources on their behalf. People outside government tended to express blanket distrust of those inside government. Advocates of decentralization were, of course, concerned that misuse of DDP funds by local governments could adversely affect political commitment to decentralization. (Top)
Accountability and project design
How can central governments ensure accountability when transferring
power and resources to local governments? This question was discussed
extensively during the design of the DDP, resulting in a number
of innovative measures.
A central point to emerge from field consultations is that clarity
about rules and procedures for decentralized planning and financing
is important. However, it is not sufficient. Transparency must
be linked with incentives that promote good performance of the
wide range of actors included in the process, and sanctions when
the various actors do not comply with agreements.
But first let us identify the specific actors or accountability relationships of most concern to the DDP:
• The accountability of local politicians to their constituents (honest, transparent use of development funds, according to agreed upon conditions).
• The accountability of technical staff to political authorities (to deliver services, including providing timely, reliable professional advice to support wise investment decisions).
• The mutual accountability of local governments and central governments.
At a minimum, local governments must not embezzle or misuse the funds entrusted to them on behalf of their constituents. They must also follow rules and agreements reached with the center governing the use of funds, just as the central government must provide funds on time and meet responsibilities to provide technical advice and supervision. Local governments must honour the outcome of the local planning process and investment decisions reached with local populations. Technical bodies must be accountable to local councils for technical oversight to ensure well-designed and appraised investments and that facilities are staffed by competent, responsive personnel. Local governments must also learn new roles as owners of investments in order to contract private service agents and ensure that they perform well.
To support these kinds of relationships, the UNCDF programme is testing two accountability measures that are new in Uganda: performance measures and public communication. (Top)
Minimum conditions and performance measures
All local governments are allocated development funds according to a set formula. However, they do not automatically get these funds, they must first meet minimum conditions. They must demonstrate basic standards of financial accounting and that functional committees are in place and are capable of steering investments through the design, appraisal and implementation process. Once they have demonstrated this minimum capacity, funds are released in guaranteed and predictable quarterly tranches. However, to maintain their quarterly payments, local governments must report on performance: accounts must be submitted for audit and they must show that periodic monitoring of the investments is taking place.
The above sanctions are matched by incentives. At the end of each financial year, higher level local governments are required to assess the performance of lower levels, again according to previously negotiated and agreed standards. If local governments perform well they will receive additional development funds, if they perform poorly funds will be reduced.
Performance is measured by questions such as: Were local plans honoured in practice? Were plans developed through an inclusive, participatory process recognizing the needs of different groups in the community? Did local councils invest development funds in services with a poverty focus? Did they meet their co-financing obligations? Was there adequate publicity about the rules of access, the amounts of money received, and the decisions of the council about approved projects? Did higher-level governments meet their obligations to support lower-level governments with technical advice and supervision of investments?
Some performance measures are based on objective, easily verifiable criteria. Others require subjective judgment and interpretation. It is easy to determine whether a council submitted the annual accounts for audit, or whether co-financing was provided on time. More difficult is to judge whether investments reflected community priorities, or whether higher-level governments provided adequate technical support. Thus the measures will need to be refined through practical experience.
Another challenge is that across the five pilot districts conditions vary greatly. (8) These differences, of course, translate into differing capacities to perform on service delivery. Should all local governments therefore be judged according to the same performance standards? If different standards are used will this be seen as fair, and will it prove technically and politically robust in the longer term?
Creating a workable system of performance measurement will take time. It is important to clarify roles and responsibilities for assessing performance (who judges who) and then agree on how they will be bound together to regularly negotiate over what good performance means in practice in specific situations. (Top)
Upward and downward accountability
However difficult, tying funding to performance is a crucial element in decentralized planning and financing. It encourages compliance with national standards-by providing tools and procedures for measuring local performance in financial management, planning and investment decisions. This is one necessary condition for upward accountability of local governments.
Further, performance measures that reward timely, high-quality service delivery create the conditions for downward accountability. Local politicians, more conscious of the need to deliver, put pressure on technical staff to provide timely, accurate information for their decisions. Politicians also become more responsive to pressure from below, which reinforces local democracy and new systems of representative politics. (Top)
Information and communication campaigns
Accountability requires that people are informed about the basic rules of the game. If communities are to engage meaningfully in local-local dialogue with their government officials about development budgets, they must know at least its size, terms of access and conditions of use. Given the radical changes in planning and financing that are being introduced and the extraordinary range of literacy and political awareness of people in these districts, development communication has become the central component of the DDP.
When the DDP was being developed, local artists and communication specialists were hired to produce brochures, cartoons and illustrations to help communication between programme designers and local stakeholders about the project's goals and design. Stakeholders included local contractors, community organizations, councillors and local leaders involved in planning and delivering services. They all provided advice and ideas about how the programme should function.
The one-year investment made in communication created a high degree of ownership and commitment to the pilot project. Beneficiaries were no longer passive bystanders, and community representatives actively negotiated with designers about key design choices.
The experience provided a critical lesson: an ongoing public information and communication programme is crucial to providing conditions for accountability during implementation. But it also underscored the technical, political and logistical challenges of an effective communication campaign.
To hold their politicians accountable for the use of development funds for specific projects, people need independent and reliable sources of information about the level of resources being devolved to their local officials, conditions of use, investment options, local contribution required, basic costs and design standards and how they can access political deliberations over use of these funds. Even at a micro-project level, providing basic information about construction standards (that is, how much cement is required to build a clinic wall) gives villagers the ammunition to exercize strong informal supervision over local contractors. (9)
Providing this basic information has both technical and political implications. It is not just a matter of posting notices or holding public meetings. Information needs to be presented in a rapidly understandable form and conveyed through appropriate media. Information campaigns that are poorly conceived or executed can cause unrealistic expectations or add confusion to an already complex process. Many illustrations and techniques had to be discarded because they did not resonate with audiences.
Professional skills required for communication are often not appreciated or understood by local politicians and government policy-makers. Creating links and trust between local artists, NGOs, drama groups and sometimes skeptical local governments need to be carefully honed. Some local politicians are threatened by the wide release of information about their responsibilities and resources (particularly budgets) and can devise measures to distort and limit the impact of public information. Central ministries, which must play a key role in ensuring this communication, may be wary of charges that they are violating the independence of local governments. Donors may feel that they are overstepping the mark on sovereignty and interfering in local politics by supporting information campaigns.
However many the risks, the benefits have been well established in the DDP, as well as elsewhere. Information is fundamental to accountability. Without it systems of sanctions and political accountability are severely weakened. (Top)
Notes
(1) There are of course many different types of fiscal transfer instruments. The rationale for UNCDF's investment in LDFs is summarised in Romeo, L. 1996, Local Development Funds: Promoting Decentralized Participatory Planning and Financing, UNCDF, New York. (return)
(2) A large literature extols the virtues of decentralization. But within central ministries (and in some recent literature) decision-makers are cautious and see little incentive to decentralize. They cite poor local capacity, problems of accountability, potentially negative macroeconomic effects, and so on, many of which are problems equally applicable to centrally administered sector programmes. In Uganda many senior civil servants committed to decentralization are particularly concerned to limit adverse publicity, despite the reality that there will be downs as well as ups in performance, for fear that decentralization may be derailed. (return)
(3) See for example, Brett, E. 1994, "Rebuilding Organizational Capacity in Uganda Under the National Resistance Movement," Journal of Modern African Studies, 32 (1): 53-80. (return)
(4) Uganda's constitution provides for three different kinds of transfer instruments: block or unconditional grants, conditional grants and equalization grants. The government is clearly aware of the relative advantages, and complexities of each instrument and the potential conflicts among them. Conditional grants are received for universal primary education, maintenance of feeder roads and some health care programmes. Equalization grants have yet to be introduced. (return)
(5) A clear analysis of Uganda's decentralization policy and progress is provided by Lubanga, F. X. 1997, "Decentralization Programme in Uganda: Results and Problems," Paper to Seminar of Uganda Association of African Administrators and Public Managers, 23 June [city?]. (return)
(6) This is not unique to Uganda, of course, as has been observed across countries, different levels of local government and for different services. Schroeder, L. 1997, "Managing the Provision and Production of Rural Roads in Developing Countries," Development Policy Review, 15: 393-411. (return)
(7) For example, one donor invested heavily in building the capacity of the district council to plan and manage development budgets. In the same district, under a sectoral programme, another donor simultaneously funded an expensive health center whose recurrent cost obligations would absorb most or all of the district's annual budget. (return)
(8) The pressure to deliver services to a high density population of 850,000 presents different challenges than for a local authority servicing an agro-pastoral population of 200,000. Local authorities have widely differing fiscal capacities to service investments. At one extreme is a district with revenue of about $1.5 per person, compared with another capable of raising three times this amount. (return)
(9) In many UNCDF projects this type of village pressure has resulted in impressive improvements in contractor performance in terms of cost, quality and timeliness. (return)
