What we do



Follow ODI

Options for Voice in Public Sector Management

Time (GMT +00) 13:00 14:00

Marian Barnes, University of Birmingham
Andrea Cornwall, Research Fellow - Institute of Development Studies

Simon Maxwell, Director - ODI

1. The fourth meeting of the series on Targets, Voice and Choice was held in the ODI Meeting Room on Wednesday 10 November. The speakers were Marian Barnes from the University of Birmingham, and Andrea Cornwall, Research Fellow at IDS-Sussex. The meeting was chaired by Simon Maxwell, Director of ODI

2. Marian Barnes, speaking to a PowerPoint presentation, started by saying that her views were mostly based on research and experiences gained in community care, health services and local government in the past twenty years. Her reflections focused on some of the problems that exist when introducing "voice" mechanisms in public service delivery. These stemmed from two commonly held views: on one hand, the view of public managers, professionals and politicians who question the validity and legitimacy of users' opinions. On the other hand, the view of citizens who feel that expressing their views will not lead to any significant changes. In her argument, there is a need to shift from an approach based on "voice" to one based on "dialogue".

3. The main problems with users' voices can be summarised as:

(a) Are they representative? Who selects them? Do they have legitimate mandate to speak on behalf of a whole category of individuals? Categories often include variables such as ethnicity, age and socio-economic status. However, other personal characteristics and political views are often overlooked;

(b) Are they informed? Often the knowledge necessary to express an informed opinion is quite technical, and therefore dialogue between experts and laymen can be difficult;

(c) Are they expressed in the right way? People' capacity to express their opinions will depend on the specific format of the voice mechanism, and also on their attitude and capacity to articulate their views. For example, anecdotal evidence may be underplayed, and strong grievances may be dismissed as inappropriate.

4. A range of other factors also influence the effectiveness of voice. Existing institutions, rules and norms strongly determine opportunities for voice. Professional power and defensive practice can discourage the use of voice. Often professionals feel threatened when more voice is given to users who take away their discretionary powers based on technical knowledge. Also, voice is often expressed outside the service delivery arena, through social movements, and in some cases the public-private divide can be blurred, with individuals playing different roles at the same time.

5. What is needed is a shift from approaches based on voice to others based on a more genuine dialogue between service providers and users. Dialogue, as opposed to voice, contributes to mutual understanding, allows for the construction of new identities and for different modes of expression, and ensures that participation will lead to change, given the dispersed location of power. A static approach which takes identities as given, and merely reflects an expression of grievances by those without power does not bear much hope. Conversations and informal discussions which address the attitudes and behaviours of those with power and include them in the identification of common problems and in the creation of different identities has much more potential for promoting improvements in public service management.

6. Andrea Cornwall's presentation, also in PowerPoint, started with two quotes from French thinkers Michael Foucault and Henri Lefebvre, introducing the concept of space and its importance in talking about participation and engagement.

7. The concept of participation has been widely used in recent development discourse: (a) as a way to improve efficiency, legitimacy and democracy; (b) to improve access and quality of care; (c) to enhance accountability; (d) to influence the type and range of services provided; (e) to build partnerships and influence policies; and (f) to encourage more active citizenship.

8. Spaces for engagement can be characterised in different ways. Current work being carried out at IDS has highlighted the usefulness of a two-by-two matrix which categorises examples of participation as either transient or regularised in time, and either as generated through invitation to official spaces or by popular/societal initiative. Examples included elections, public protests, citizens' fora, etc.

9. Individuals get involved in voice mechanisms with different roles, as beneficiaries, rights-holders and duty-bearers, either invited or selected, and through different spaces of engagement (associational spaces, formal politics, bureaucratic fora, etc.), and their inclusion or exclusion is often influenced by a number of factors, such as agendas and relationships, language and expertise, and the availability of new ways of structuring dialogue.

10. The future effectiveness of voice mechanisms and of participation more in general will depend on transforming the nature of the discourse, the rules of the game in order to improve voice opportunities. But also, it is important to build capacity to participate, by building the confidence and skills of individuals, give them access to relevant information, etc. The case of Zambia, where local farmers were not aware of wider political and institutional factors that were influencing their livelihoods, was brought as an example of this need.

11. A lively discussion followed the two presentations, focused mostly on the similarities that existed between debates around voice in developed and in developing countries. The importance of context and of the political environment in shaping opportunities for voice was noted. While in some cases participation and voice are largely dependent on political factors (as in Northern Brazil), in others, such as South Africa, the history and perceptions of disenfranchisement played a more important role.

12. Various participants questioned a view of participation and voice based on an assumption that decision-making power resides at the local level. Recent development discourse, mostly pushed by donor agencies, has idealised localisation and empowerment without spelling out some of the contradictions involved. The wider institutional framework determines which spaces are open for dialogue, and therefore will define some of the actions that citizens can take to engage with government, and the levels at which they will be most effective in bringing about change. The incentives faced by bureaucrats and professionals often are not conducive to dialogue, even if formal spaces exist for this purpose. However, it was also argued that local dialogue could provide a stepping stone for actions at national level, for example by giving people the skills and confidence needed to participate in citizen action or social movements.


Discussions on voice span a wide range of activities and actors from lobbies and protests to civil society consultation and joint civil society-government initiatives. Central themes include: differences between developed and developing countries vis-à-vis conditions for voice; perceived benefit/need for/value of voice; relative value-added of voice at different levels (policy setting, planning, budgeting and implementation, monitoring and evaluation); conditions conducive / threatening for voice; representation (whose voice?).