Context, Politics and Institutions

A preliminary point that emerges is that policy makers are a heterogeneous group. They include in the cases, among others, government officials, legislators, civil servants, civil society, the judiciary and the media. (For example, the WAPDA case in Pakistan is interesting because it looks at the role of research in influencing and improving laws.) A related point is that groups of policymakers (e.g. government) are not a homogenous entity – significant differences of opinion exist (e.g. within government in the Tanzania PRSP case).

A clear finding is that the political context is very important in affecting the degree to which research will impact on policy.

• Political demand matters. Research appears to have a quicker and greater impact when it is policy driven, at least to start with, or has high-level political commitment. It is much more difficult to ignore the findings of any research in such circumstances. Among others, this can be seen in the cases of the Watershed approach in India, Development Planning Reform in Lithuania and Biodiversity Management in Saudi Arabia. Research commissioned by policymakers themselves is more likely to affect policy. Such demand often comes because of crises (e.g. DELIVERI).
• Ideology / perceptions matter. If findings are in accordance with the perceptions of key policymakers, it makes it more likely that they will be implemented (e.g. Self-managed schools in Argentina). Policymakers will even utilize incomplete research if it fits well with their interests and ideological viewpoints (e.g. the GALASA case).
• Interests matter. Research recommendations are unlikely to be implemented if there is no internal consensus on priorities (e.g. Kerala local research programme). However, research can help generate a consensus (e.g. PRSPs in Tanzania and Fiscal policy in Chile). A key finding was that research is unlikely to affect policy if reforms go against the interests of important political players (at least if there is no other incentive for reform, such as donor funds). This was a finding in the case of donor-driven research in the Kyrgyz Republic. In the Malawi case, the government adopted part of the research findings (about universal provision of seed packs) – where the evidence fits neatly with political imperatives – but ignored the other findings which raise serious questions about the desirability and viability of current agricultural policies.
• Timing matters. A number of cases highlighted how the incentive of entry into the European Union (and funds available to support research and policy change) encouraged reform in Eastern Europe. For example, the case of Lithuania concludes that the SIGMA review would have had much less impact if it had not been on the eve of the EC decision on whether to invite Lithuania to join the EU.

The cases indicate that changing the process is as important as changing policy. A local view of policy highlights policy implementation rather than formal policy (e.g. SPEECH case in India). Many cases highlight the importance of participation. The DELIVERI case in Indonesia addressed issues of decentralisation, privatisation and participation in livestock services. In the Chile case study, the government called for and followed advice from a panel of experts thus giving legitimacy to their policy. In the case of the CPI in Peru, the report stressed that a permanent review panel should be set up that would publish changes in approach in advance so the public and the experts could assess them. In short, we are often talking about issues of governance.

Of the three realms in the framework, the cases collected so far clearly indicate that the political context is probably the most important arena affecting the uptake of research. The case of PRSPs in Tanzania provides a good example of the issues here. As the case notes: “wide-scale confidence in research results is inadequate to ensure impact so long as policy decisions are fundamentally political.” This does not mean that work in this area should concentrate on political issues alone. As the Tanzania case also notes that “research results can shake up the balance of political forces and enable movement.”

But, it is clear that researchers and think tanks that aim to impact policy need to be aware of the environment in which they operate. Who are the policymakers? Where is the demand? What are the sources / strengths of resistance? Is government accountable and competent? In some cases, it seems more effective to remain close to policy-making. However the role of researchers and think tanks is less clear where the government is less democratic and / or there is no demand for policy research. For example, what can be done in situations like the case of slate mining in the Dhauladhar Himalayas, where the “entire research led effort, policies, action and administrative framework fails against the vested interests backed by political abettors.”

This leads to two sets of questions for further consideration: (i) What kind of specific policy contexts are there in the case studies? (ii) What kinds of policy context best draw on research evidence for development policymaking? The discussion in terms of bridging research and policy seem to revolve around the interaction of three sets of issues: openness, demand and contestation. It is also important to mention the issue of policymaker commitment to development here.

• First, the case findings reflect much of the literature and highlight that an open political system is important. This allows evidence to be freely gathered, assessed and communicated. While there is a great deal of diversity in the detail, the contexts of most of the case studies collected are basically open. There are very few cases from more closed systems so it is too soon to tell exactly how a closed political system affects evidence-based policymaking. In some of the closed-system cases there has been little impact of research on policy (e.g. Iran); in others there has been an impact (e.g. Ukraine).
• Second, the issue of policymaker demand for research is vital in many, if not most, of the studies. In some contexts, there will always be some level of demand (e.g. food security in India). In many of the collected cases, a new context means that policymakers try to find a solution to a problem (the case of regional development in Croatia provides a good example here). Where there is no initial policymaker demand, researchers can help “create” it by highlighting a problem and a way to address it (e.g. Rainwater harvesting in Tanzania and Remittances in Nepal).
• Third, the findings highlight that the degree of political contestation of any reform is vital. These cases suggest that the policy shifts more easily if there is consensus. In cases with a lack of policy shift, this is often a result of blocking by the special interests of groups or individuals (e.g. cases in Kyrgyzstan and Morocco). In some cases, it seems that there would not be a shift in policy even if the evidence was convincing.
• Finally, the cases highlight that the degree to which the policymaking community is committed to improving policy towards development goals is important. Policymakers committed to development are more likely to be interested in improving policies and learning from evidence. This is often implicit in the cases, although it is mentioned explicitly in the Philippines and Peru CPI cases. This cannot always be assumed (e.g. Peru and Indonesia before the change of regime) and may not exist at all (e.g. the Dhauladhar Himalayas case), but it does seem to significantly affect the nature of research uptake. This issue seems related to demand – i.e. there is a demand for research that might improve development performance. However, it is also worth looking further at whether it is the lack of funding to implement research findings rather than the lack of commitment that is the key issue.

It is instructive to look at how some of these factors played out in a few cases. In the DELIVERI case in Indonesia, the downfall of Suharto mid-way during the project’s lifetime meant that there was an increase in demand and a fall in contestation regarding the issues addressed in the project (i.e. decentralisation, privatisation and participation in livestock services provision). The change of context also saw an increase in openness. Given the action-oriented nature of the project, it was not surprising it had a substantial policy impact. In the contrasting case of Kyrgyzstan, there is high demand for public administration reform in a relatively closed context, but despite donor support for the research there is little progress due to a high level of contestation regarding the options.

In sum, it seems that the main issues in the context arena can broadly summed up as:

The likelihood of change = the degree of demand minus the degree of contestation.

The nature of the change is likely to be a function of the degree of openness and commitment (as well as the quality of the evidence and the nature of the links, which are covered in the next sections).