Speakers:
Ariel Fiszbein - Chief Economist,  World Bank Human Development Department 
Dena Ringold - Senior Economist, Human Development Network, World Bank

Chair:
Andrew Norton - Director of Research, ODI

Discussants:
Dr Anuradha Joshi - Fellow, Governance and Public Policy, Institute of Development Studies
Marta Foresti - Head of Programme, Politics and Governance, ODI

This event was hosted by ODI to launch the World Bank Report ‘Citizens and service delivery: assessing the use of social accountability approaches in the human development sectors’ and was opened by the event chair Andy Norton, ODI Reserach Director.

Ariel Fiszbein Chief Economist for the Human Development Network at the World Bank, introduced the session by giving a brief background and overview of the study.

Background: Service delivery in most developing countries is characterised by failures and poor outcomes. Since the publication of the 2004 World Development Report Making Services Work for Poor People a key challenge has been to improve accountability of service providers and policy makers towards citizens and users. In recent years there has been an increasing interest in achieving this through social accountability mechanisms which involve the direct participation of users in their interaction with providers.

Overview: The report takes stock of World Bank interventions in the area of social accountability, looking at what is being done and the lessons learnt. It also looks at the growing body of evidence from impact evaluation studies and results coming from social accountability interventions. A key message coming out of the report is the critical need to continue supporting experimentation and the need to generate robust evidence about how social accountability mechanisms really operate.

Dena Ringold presented the findings from the research. Firstly she outlined some concepts and assumptions behind the research and then focused on the results of two types of interventions: 1) Information interventions and 2) Grievance redress.

Concepts and assumptions: Social accountability is about how citizens (service users) can use information to directly influence service providers (frontline providers as well as programme managers) and hold them accountable. This requires that people have access to information and the opportunity to use it to bring about change. Hence the importance of providing access to information also as a basis to make grievance redress possible.

The report acknowledges that a number of assumptions are made in relation to how social accountability is meant to improve outcomes. Firstly, it assumes that citizens have the ability to access and use information. Secondly, that citizens are willing to use redress channels to pressure providers and that, thirdly, providers will respond to citizen pressure. There are problems with these assumptions that we need to be aware of: firstly citizens have asymmetric access to information and may have different priorities which depict whether they use it or not; there might also be political or cultural factors which affect their willingness to use information and access available redress channels; and finally, provider behaviour and incentives are embedded in a complex set of contexts in which social accountability is just one factor.

Access to Information interventions

• Access to information laws (legal framework) have really been taking off, but they are largely general and it is not obvious the impact they have on specific sectors. There is little experience of evaluation and further work needs to be undertaken to examine who has access to this information and what the impact is.
• Information campaigns broadly look at the content of services (rights, standards, financing, organisation of services, channels for participation) and the information related to the quality of service provider performance. The results suggest that there is great diversity in the impact on provider behaviour. It is difficult to disentangle demand side effects from provider efforts. The pathways through which information campaigns flow are complex and difficult to evaluate.
• Scorecards and social audits, which facilitate discussion between citizens and providers, are popular interventions, particularly in the area of health. In India for example the percentage  of wage seekers who knew about the national rural employment act increased following the information campaign and social audit and in Uganda the use of health scorecards dramatically improved the infant mortality rate. However the body of evidence is limited and interventions have had mixed results.

Factors contributing to the success of access to information interventions include the way the message is framed (amount and form of information provided); the method of delivery (the tools and means used to provide information), and the clarity of the message (the way information is presented).

Grievance redress mechanisms

These are formal or informal channels through which citizens can express their (dis)satisfaction with service provision; they take place through three main agencies:  government, independent organisations and the courts. In the area of human development, grievance redress takes place in social development, health and education via different agencies and to varying extents. There is a growing interest in interventions in this area and more efforts are needed to better understand the effects of the different approaches.

Overall the findings show that:

• Over half of all World Bank human development projects used social accountability language, mostly in the area of social protection in Africa (health scorecards).
• Complaints mechanisms were mainly in social protection interventions where they are used to manage the targeting of cash transfers.
• However the incorporation of social accountability in project design was much less prevalent. This is critical as the design of social accountability interventions is a key determinant of their success.

Dena concluded with a tone of cautious optimism noting that there is a lot of innovation in social accountability. However, using citizens to influence providers is not necessarily straightforward:

• Social accountability tools can be effective but within a context of broader mechanisms to influence provider behaviour and change incentives.
• Intermediaries such as civil society and the media have an important role to play in facilitating access to information and redress.
• Design of information campaigns is especially important. Technology is an important aspect of the design but is not a panacea.

Going forwards there is a need to invest in evaluation and learning, to build an evidence base, to support implementation and to work with multiple partners.

Anuradha Joshi welcomed the report as timely research which makes a good contribution to the growing debate on social accountability. She drew attention for the need to take a more systematic approach to analysing the contextual factors which enable interventions to work in some contexts but not others as well as the processes by which they are expected to have impact. There is a need to better understand the conditions (politics and processes) under which social accountability actors take up the task of holding providers to account because intermediary impacts such as changes in provider and user behaviour could be equally important. Two useful areas for future research were flagged:

• A critical look at who gets included/excluded from both the process and the impact of social accountability interventions?
• The role of new technologies in social accountability interventions and how it feeds back into the management process.

Marta Foresti noted that the report was useful in the way in which it opens up about the assumptions which are made about social accountability interventions as well as its focus on building a quantitative evidence base, which is often missing from what is often an ideologically driven debate on these matters. She raised two key points.

• Firstly, the report shows that the impact of social accountability programmes is mixed but what is it that makes the difference? Clearly programme design plays a role but the overall theory of change and assumptions underlying these programmes also  matter. Recent ODI recent work looking at community scorecards in Malawi found that some interventions did contribute to some changes and better outcomes, although not necessarily as a result of access to access to information per se or holding providers to account. Rather, what was key was the the fact that the programme brought people together and enabled them to find solutions and collective ways of working with providers to overcome problems. By focusing on the nature of the demand side we risk selling short the potential impact of this work; there is more to the story than demand side.
• Secondly, there is a design bias in social accountability programming which means that programmes either work on the demand or the supply side. However, research shows that focusing on demand alone does not increase accountability and can be potentially damaging for local realities because they make assumptions about how demand works. There needs to be a greater focus on what is more likely to affect changes in incentives for providers and on identifying suitable entry points for programming which combine thinking around the incentives which can bring the demand and supply side together. The mechanisms which make things work, often overlooked,  tend to sit  in between demand and supply and are not either or so we need to start looking at demand and supply in a more integrated way.

Discussion

Following the presentation a number of comments and questions were put forward to the panellists. The discussion centred around providers incentives to respond to citizen demand and the mechanisms and tools by which social accountability functions. The key points raised were:

• Politics is an important contextual variable because provider and user behaviour is deeply embedded in politics.
• Understanding the complex incentives which motivate providers to respond to citizens demands is critical to social accountability work. It can enable us to identify how demand can play into existing incentive mechanisms.
• In the types of social accountability interventions analysed in the report there is a huge gap between access to information and formal grievance redress. At the local level access to formal grievance redress is limited. It would be interesting to look at how informal complaints are addressed.
• Types of grievance are more likely to be easily taken up at the community level than others. For example in the education system informing citizens about the quality of teaching is more difficult than questions such as absenteeism.
• There are fundamental knowledge gaps in social accountability work which need to be addressed including what the sustainability and costs of interventions are.