This paper discusses the steps required to build a robust evidence base for ‘thinking and working politically’ (TWP) in development. The ideas behind TWP are increasingly common in development discussions, and understanding which parts are necessary and sufficient conditions for success is crucial for moving it into mainstream development programming. After reviewing existing literature on TWP, public sector reform, and ‘pockets of effectiveness’ the paper suggests research questions, case study selection criteria, and a four-level analytical framework. It also calls for more focus on gender issues, and on different – and often more fragile – political contexts.
The current evidence base: Since TWP is a relatively recent arrival in the development debate, gaps in the literature are to be expected. Much of the existing research uses empirical examples to generate lessons and theories on politically informed development programming. This is an important first step but it is not in itself ‘evidence’. There are a number of methodological limitations that run through the literature: selection bias; limited range of contexts; insufficient testing of theories, follow-up and discussion of the change process. As such, claims that particular TWP approaches lead to more successful development outcomes cannot be justified by the existing literature. There is an urgent need for more systematic research and analysis if we are to understand which approaches can deliver better results. Addressing gaps in content on gender, diverse political contexts and looking beyond donor programmes will further enhance the evidence base.
Broad research questions to guide analysis include: why do politically informed programmes emerge in some contexts and not others? How do these programmes incorporate TWP?; how do different aspects of TWP affect the implementation and the outcomes of politically informed programmes?
A four-level framework enables the development of a broad approach that draws out the interaction and interdependencies between the different levels. This will help us to better understand how politically informed programmes emerge and succeed.
- Political. The political system, political and bureaucratic leadership and interaction, and the nature of the political settlement, as well as other types of power structures such as gender, religion, ethnicity, caste and rural-urban divides.
- Sectoral. The literature suggests that prospects for implementation will vary considerably according to a sector’s characteristics and political significance. What are the institutional characteristics of sectors associated with politically informed programming and programme success?
- Organisational. What organisational characteristics are associated with more politically informed approaches and successful programmes? Current suggestions in the literature include a problem-solving and iterative approach, flexible and strategic funding, and public organisations that have organisational autonomy and political support, but how might this look in different sectors and different contexts?
- Individual. A key question is whether the space to work politically despite organisational constraints is created by individuals. ‘Reform champions’ or ‘policy entrepreneurs’ are often seen as the source of innovation, but we need empirical evidence to understand who they are and how they work. If an individual renowned for TWP moves to a different organisation, do they take their TWP approach with them, and do they also leave it behind?
Case selection needs to allow for sufficient comparison. This means variation across and within levels of political contexts, programmes, sectors, and development actors implementing differing programmes. This should also help uncover whether there is one way to ‘think and work politically’ or whether there are multiple ways, each appropriate to the particular context.
To avoid selection bias, it is important to research both ongoing and completed programmes to learn more about what doesn’t work when it comes to approaches to TWP as well as what does.
