Democratising Local Governance in Africa

Recent literature on decentralisation has recognized that if democratic (not simply decentralised) local governance is the ultimate objective, then just as much emphasis must be placed on mechanisms for genuine downward accountability to local residents as is placed on aspects such as the degree of autonomy from central government and capacity for service delivery.

This study took this proposition further by suggesting that in a context of transition from decades of top-down administration, clearly defined and guaranteed mechanisms for citizen participation, including structures for representation of marginalised groups, are an important part of overcoming the history of acceptance of privilege, resignation to patronage, and a tolerance for opacity in public affairs.

The empirical research for this study was conducted in Kenya between January and November 2005, on the back of significant legal-political events in the country: in March 2004 a national delegates conference issued a draft constitution for Kenya, which was revised by a parliamentary committee and re-drafted by the Attorney-General in August 2005, put to a national referendum and defeated in November 2005. The draft contained far-reaching proposals on re-organising local government to devolve some authority to the district level, with key provisions aimed at enhancing citizen participation in local government affairs.

Aims and Objectives

This study sought to understand how structural reforms intended to enhance citizen participation and downward accountability in local governance can be sustained and made effective in a context of transition from decades of centralised top-down administration.


The study combined desk-based and field research. The desk-based component primarily involved textual analysis of two versions of a proposed constitution for Kenya. The study linked analysis of the frameworks for citizen participation proposed in the drafts:

  • lessons learnt from relevant experience in Ghana, Tanzania and Uganda;
  • citizens’ current experience of interaction with local governance institutions;
  • citizens’ expectations of the proposed arrangements, as well as expectations of recent policies that have sought to enhance citizen participation in specific aspects of local governance.

The fieldwork was conducted in four districts in Kenya and it involved a survey with 516 respondents, 100 key informant interviews, eight focus group discussions and eight community-level feedback and analysis workshops.

Highlights from Findings

1. Structural reform must be accompanied by efforts toward incremental change in the culture of local governance. Rejection of the draft constitution in the November 2005 referendum marked the shelving of the first attempt at comprehensive reform of local government, creating a legal framework for citizen participation in management of local government affairs. In the absence of such a framework, efforts to enhance citizen participation have taken the form of piecemeal administrative reforms since the late 1990s. Expectations that structural reforms could deliver meaningful citizen participation and downward accountability have been dampened by the experience of these piecemeal administrative reforms.

Examples of such piecemeal reforms include a policy directive that conditions the release of 40 per cent of a local authority’s allocation under the Local Authority Transfer Fund (LATF) on the involvement of local residents in drawing up a Local Authority Service Delivery Action Plan (LASDAP).

Experience so far suggests that without a change in the culture of local governance, the participatory processes created through these structural reforms have been pigeon-holed: there is no indication that they have the potential to gradually transform mainstream local government toward a culture of transparency and answerability to local residents. The culture of opacity has not changed, as indicated by survey responses concerning public hearings, publicising of council financial information and voicing of complaints or suggestions to the councils.

2) Special Representation raises more conceptual and practical dilemmas than the literature suggests. A few writings on democratic local governance acknowledge the importance of special arrangements for the representation of marginalised groups if pro-poor outcomes are to be achieved. However, since these writings are set in contexts where such arrangements are already in place there is little discussion about the contentious political process of arriving at such arrangements.

The issues raised by Kenya’s recent experience include:

  • the process of arriving at a consensus on whom to include in the category;
  • the process of agreeing on what form the special arrangements will take; 
  • the social legitimacy of any special arrangements; 
  • the (perceived and real) effect of any such special arrangements on the democratic process.

As a result of this study, the research team learnt that social marginality is not necessarily equated with political marginality. There may be consensus about the marginality of a particular category of people in terms of their lack of access to services, or the fact that they are under-served by certain social institutions, for example, customary justice forums. However, the research team learnt that this is far from a consensus about the need to treat this category as a political category, let alone consensus on their eligibility for special representation in the political process.

This raises conceptual issues: how can we justify special representation arrangements as an important part of the political process in the face of public sentiment that is overwhelmingly against special representation, if at the same time we place significant weight on citizen participation in shaping political institutions? In the absence of broad social legitimacy or acceptance, where is the impetus for sustained political will or bureaucratic commitment to maintain the special arrangements and make them effective? In the absence of special measures, how is political voice to be assured for groups against whom there is comprehensive bias? If local understandings of marginality are so concrete and contextual is there a case to be made against lists of marginalised groups (for example, in constitutional frameworks or in laws on local representation), and in favour of broad statements of principle that allow for local flexibility? How is this local flexibility to be checked so that it does not end up being simply an articulation of bias?

Project details

start date
1 November 2004
end date
30 March 2006


About this project

Recent work