Innovation in think tanks

This report was commissioned by the RSA in partnership with the Think Tank Initiative to explore the changing role of developing country think tanks in influencing policy in today’s ever-changing and complex world.

‘Innovation in think tanks’ is also available to download as a PDF document

This think piece contributes to the far wider debate about how to ensure aid is more effective. It focuses on one small, but important, part of that wider discussion: how to support development of low and middle income countries in a politically realistic way. It argues that the messy complicated reality of how change happens requires, in most national settings, a set of institutions that are grounded in the importance of evidence, but are entrepreneurial, responsive and politically savvy in the use of that evidence. Its core contention is that think tanks have many of these attributes and therefore have the potential to play a vital role in supporting the development of locally-led, context-specific, policy change.

For this reason, we argue that donors and funders should pay think tanks more attention than they currently do. Think tanks are often better placed to influence policy than traditional civil society organisations (CSOs), while their research tends to be more politically informed than academic research. Yet CSOs and academic institutions tend to be more prominent on donor radars than are think tanks. This, we argue, needs to change. Donors should, ideally, look to build up a critical mass of think tanks in different countries.

This is not, however, to assert that think tanks are a silver bullet — ineffective or politically captured think tanks have little value. Nor are we saying that think tanks are the only — let alone most important — institution capable of helping achieve politically smart development. Clearly, in any given context other actors and interests will wield influence (often considerably more than think tanks). And think tanks tend to work best when other parts of the policy arena — government, parliament, the media, and civil society — are also strong (which is why they are more relevant in low- and middle-income countries).

But we are saying that locally-rooted and well networked think tanks, which effectively deliver on the core function of conducting policy-focused and politically savvy research, can and do influence their policymaking processes. However, given the many different ways in which politics happens in different developing contexts, from centralised undemocratic states to those which are more open and from those where politics operates largely through patronage networks to the more technocratic, we argue that think tanks need to experiment with a wider range of functions and approaches to policy influencing.

Indeed, another key argument in this paper is that while think tanks themselves are often very adept at developing strategies for achieving change that reflect their local context, they have been constrained in their ability to do so by funder pressures which can encourage (often inadvertently) a high degree of uniformity among think tanks. Without this experimentation, the potential of think tanks will remain untapped. Part of the answer to supporting more politically savvy development is both more think tanks but also greater innovation in how think tanks operate, to enable them to be more responsive to how politics works and how change happens in their respective countries.

One specific implication for donors is that there is a case for supporting the development of new think tanks, not just existing institutions. A small number of new organisations could be fully tailored to their local context, and creating such organisations — from recruiting the best skill-mix and forging the right culture — could, in many instances, be easier when creating a new organisation.

This paper develops the argument as follows: first it reviews current debates about the need to ‘bring politics’ into development, and explains why this agenda has only been partially successful; second, it makes the case for why think tanks are well placed to support politically savvy policy development; and third it argues that if think tanks are to have more impact they will need to innovate with their form and approach to policy influence if they are to better realise their potential.

1. Achieving change: challenges and frustrations

Recent debates on public policy in the developed world have struck an increasingly pessimistic tone. In the words of the Royal Society for the Encouragement of Arts, Manufactures and Commerce (RSA), Chief Executive, Matthew Taylor:

“… big policy is hard to get right. Very hard. From any perspective, the recent record of central government policy isn’t great. There are the disasters, like the poll tax, the Child Support Agency, and rail privatisation … then there is the underwhelming impact of 35 years of continuous reform of public services … despite all this policy activity, we are living with the failure to tackle major problems …”

Why policy fails and how it might succeed — Matthew Taylor, Annual RSA Lecture 2016

In the RSA’s critique this high failure rate has several causes. Chief among them is an inability among policymakers — and those seeking to influence policy — to fully consider the wider system in which social change happens. This leads to a scattergun approach of discrete policies in which impact is too often only possible with concerted and unusual focus and pressure — keeping the foot on the accelerator. Once this pressure is removed, progress stalls.

Thinking in the developing world has its own version of this pessimism. In some ways development thinkers there are ahead of those in the UK. Years of donor attempts to introduce ‘rational’ western-style approaches, from merit-based bureaucracies to transparent needs-based allocations and large scale reform efforts, resulted in numerous failures. As the Overseas Development Institute (ODI) put it:

“Despite vast amounts of support from the international assistance community, increased resourcing and improved policies and/or formal systems, many states and governments across the developing world have remained unable to provide adequately for the well-being of their populations at large.”

Getting real about politics: From thinking politically to working differently — Overseas Development Institute, 2014

A plethora of critiques, emphasising the limitations of externally driven and ‘rational’ approaches to achieving change, have been published. An oft-quoted argument is that developing world institutions have taken on the appearance of reforms along western lines, but they have practiced ‘isomorphic mimicry’ which in biology describes different organisms that evolve to look alike without being related. As early as the 1990s thinkers were rejecting such attempts to achieve change in favour of more “adaptive”, locally tailored approaches. This emphasis on adaptive approaches is reflected in the RSA’s own argument about the need to achieve policy change in a more realistic and incremental way — looking for the short-term, small-win opportunities for change which cumulatively build up impact over time. Reflecting on this sense of frustration, Duncan Green, in his recently published book, How Change Happens, argued:

“Over the last 30 years, aid agencies and international financial institutions have devoted considerable attention to reforming states in developing countries. Their efforts to bring about ‘good governance’ have restructured budgets and ministries, rewritten laws, and even spawned new institutions, but by and large they made little change to the way states operate.”

2. Bringing in the politics: an incomplete revolution?

2.1 Bringing in politics and recognising complexity

The response to this challenge from some development policy thinkers has been an important exploration of how change happens.

One strand of thinking has been a focus on the wider ‘systems’ within which change happens. This emphasises how social and economic problems are multi-faceted, interrelated and often difficult to solve. As a result, often responses to policy problems need to be based more on trial and error; grounded in learning from local contexts more than ‘rational’ evidence from international best practice. The work of thinkers like Lant Pritchett has been particularly relevant here: he has emphasised that complex problems require an approach which is problem driven, but experimental and adaptive. He argues that “… answers cannot be pre-planned or developed in a passive or academic fashion by specialists applying knowledge from other contexts. Answers must be found within the change context through active engagement and learning.”

There are more than echoes here of the RSA’s argument that achieving change often requires organisations and approaches which can “think like a system, and act like an entrepreneur”. That is they need to understand the complexity of the system, but then act in a responsive, experimental way to explore different potential solutions. They need to recognise that most change happens when a series of ‘small-wins’, which are fully embedded and accepted by the wider system, build up to larger scale change. There will be moments when larger change is possible — in moments of major economic change, often on the back of crises, or political transitions for example — but mostly incrementalism is king.

Being able to identify the opportunities for lasting, if often piecemeal, change links to a second strand of debate amongst development thinkers — the need to have more politically informed approaches to achieving change.

The case for considering the politics — which is not taken to mean party politics in a narrow sense, but an understanding of power, who holds it, who influences it, how it is exercised and so on — around any policy challenge may seem uncontroversial. Any view that policy debates and social change happen in a rational way, based on objective evidence, is overly simplistic. Some policy challenges can be more technocratic, but in almost all cases normative, ethical and political aspects of change exist. In the case of complicated, or ‘wicked’, issues the messy reality of having to understand power dynamics and the formal and informal rules which explain how decisions are made is unavoidable. The growing body of development thinking and practice emphasising politics is therefore welcome.

2.2 An incomplete revolution

Despite, however, some development thinkers making the case for many years now, change in donors’ practice has been slow. Pockets of good practice have been reported and show what is possible but so far progress has been limited. Yet, overall progress has been slow. ODI sum it up:

“… the revolution is not yet complete. How to make the concept ‘politics matter’ operational remains hard. There is a growing acceptance of the importance of taking context into account, and a growing acknowledgement of the need to work in more iterative, adaptive and flexible ways. Yet making a jump from more technical approaches … to more politically aware programming, grounded in local realities, has proven considerably more challenging in practice.”

Getting real about politics: From thinking politically to working differently — Overseas Development Institute, 2014

So why the incomplete revolution? One obvious answer is that changing the culture of a system as complicated as a donor’s will take time. There are, however, reasons to believe the pace of change will remain very slow and the revolution incomplete.

• Domestic political pressures on donors mitigates against progress: Donor agencies will always have their own political context, domestic audiences and pressures. These may mitigate against the messy business of slow incremental change and thinking and working politically — with powerful incentives to focus on the measurable and tangible. This is likely to become an even more acute issue in the future.
• Donors reluctance to engage in politics: Donors can also feel that they shouldn’t drive politics in developing countries themselves — or at least that they feel that they can’t be seen to overtly do so. This can lead them to emphasise the technical or non-political.
• Barriers to working politically: Is it realistic to demand donor agencies to think politically? Many UK think tanks and parts of the civil service are fully focused on thinking politically — individuals have worked their whole careers in political cultures and are rewarded for working politically. Yet, even there many struggle. In a development context, the challenge is greater still given that expat donor staff have limited understanding of local political contexts, which differ significantly from the mature, interest-based politics associated with advanced democracies. High staff turnover, particularly of expats, further mitigate against building up local expertise.
• The complexity temptation: An additional risk is an overly simplistic equation of needing to understand complexity with a need to develop complicated (and hard to implement) responses. This could, for example, manifest itself in the creation of donor programmes which try to address a wide range of different parts of the ‘system’, but thinking that is possible to effectively intervene across so many parts of a complicated system risks overstretch and ultimately failure.

The ambition to bring in politics and to consider complex systems is right, but changing development actors’ cultures sufficiently remains a big challenge. The next section explores why think tanks are well placed to support change in a more politically savvy, adaptive and responsive way.

3. Achieving change: the role of ‘core model’ think tanks

Think tanks in developing countries can make an important contribution to driving forward politically savvy reforms upon which effective development depends. Think tanks, which are embedded in their local polities and which exercise a degree of legitimacy outsiders lack, are well placed to influence change.

This is not, however, to be starry-eyed about the role think tanks can play. Developing country think tanks face many challenges which naturally constrain the contribution they can make (we discuss some of the main ones in section 3.3 below). Nor are we saying they are the only type of organisation that can support locally-led development. They are just one part of a more complex change-making process, which comprises a wide range of formal and informal institutions and interests, many of which wield more influence than think tanks. And in some contexts, particularly the poorest countries, think tanks may not exist and would have limited potential. However, in most countries organisations such as think tanks either already exist or could be strengthened with several of the ‘entrepreneurial’ attributes that theorists of social change, like Matthew Taylor and Lant Pritchett, place increasing emphasis on.

Below we make the case for why we believe think tanks can make a meaningful contribution to achieving change, and why donors need to pay them more attention than they currently do. First, however, we provide a brief overview of the developing world think tank landscape.

3.1 The emerging developing world think tank landscape and the core functions

Think tanks, while still in their infancy in developing countries, have nonetheless grown in number and importance. The University of Pennsylvania’s Think Tank and Civil Society Programme (TTCSP) estimates that the number of think tanks in Asia and sub-Saharan Africa increased significantly between 2007 and 2015, with the growth in Asia being dominated by China and India. This growth, on the TTCSP numbers at least, has leveled off since 2015. Nevertheless, this suggests that there is clear demand for establishing think tanks in developing countries.

We define the core functions of a think tank as an institution which undertakes policy-focused and politically savvy research with the ambition of using this to influence public policy. In other words, think tanks are in the business of trying to solve problems, not just diagnose them, and once they have a policy solution, they seek to try and get their recommendations implemented. There are of course plenty of organisations that describe themselves as a “think tank” which don’t credibly adhere to the definition set out above. Many of these are quasi-academic institutions which tend not be politically-engaged — characteristics which are often incentivised by the donors — or at the other end of the spectrum, act and behave more like NGOs.

The precise type of influencing strategy adopted will be context sensitive. Political systems in developing countries can be authoritarian and closed, democratic and open, or lie somewhere in between. The type of politics practiced runs from the patronage-based and more explicitly patrimonial models to the more technocratic, to those where politics operates largely through informal mechanisms networks to the more rule-based, all of which has a major bearing on the way think tanks engage policymakers. Notwithstanding these important contextual factors, in broad terms, it is possible to identify five distinct ways in which think tanks seek to exert influence:

(1) Direct policy influence: where a think tank advocates for the implementation of a specific proposal which is subsequently adopted by government.

(2) Indirect policy influence: where a think tank proposal shifts policy, but only as part of a messy and complicated process where a range of different interests all shape a change in policy direction.

(3) Influence the broader climate of ideas: this is often where they can have most impact, where they reframe a policy debate around new ideas.

(4) Informing public debate on key issues: through their communication and dissemination work they can play an important role informing public debate.

(5) Hold governments accountable: for example, by monitoring policy implementation or providing evidence to show policies are not achieving results.

In general think tanks rely on elite models of influence, which assume that targeting policy-makers in government is the best route to achieving change. Elite models of influence involve using a mix of ‘insider’ strategies (influencing policymakers through their networks) and ‘outsider’ strategies (using the media, and increasingly social media, to disseminate policy research). Which strategy is adopted will depend on both the nature of the political culture and the type of issue being researched.

In terms of political culture, the more closed it is the more a think tank will lean on ‘insider’ advocacy. The more open a political culture the greater the scope for using ‘outsider’ advocacy to raise the profile of an issue. And of course, the context can change. The closing down of political space in Bangladesh in recent years has forced think tanks to alter their approach. In contrast in Myanmar today think tanks are grappling with a potential opening of debate. Adapting to the prevailing political context was deemed, in a recent survey, the most important factor in determining the way developing country think tanks operate and work.

IPAR-Rwanda compared with Zambian think tanks

A comparison between think tanks in Rwanda and Zambia helps illustrate the way think tanks differentiate their strategies for influence according to the local context. In Rwanda, with constrained public debate and concentration of power, IPAR-Rwanda uses its networks and contacts with the government to pursue a largely insider approach. By doing so IPAR has built up credibility and trust with the government, which has enabled it to have real impact on policy. Think tanks in Zambia also carefully balance insider and more public approaches to achieving change, but in a country with considerably more open public debate and a stronger democratic tradition think tanks can adopt more ‘outsider’ and public approaches to achieving change. Think tanks such as ZIPAR and IAPRI have dedicated communications experts and put considerably more time and effort into making public arguments about the need for change than is possible in a country like Rwanda.

3.2 Thinking politically: the case for think tanks

At their best, think tanks have the technical research skills to develop solutions, but also the political insights to develop politically viable proposals for action. This makes them a strong contender to be a critical organisation in supporting more ‘politically-smart’ approaches to development. Earlier, we argued that the thinking and working politically revolution remains incomplete. Despite pockets of good practice, donors are not well placed to consistently deliver effective politically savvy development support. Think tanks offer the potential to help complete the revolution because they are embedded in their local polities. They understand how change happens in their country. By being locally rooted they often have the legitimacy and credibility that external actors lack. This leads to four advantages:

• Having an intuitive sense of how change happens: They have an intuitive sense of what will work politically and what will fall flat on its face; through their networks they learn where power really lies and who’s up and who’s down; and they know which interests will support change, and which will be determined to block it. Unlike outsiders they don’t need elaborate theories of change or Political Economy Analysis (PEAs) to understand how the internal power dynamics operate in their country. They have the local knowledge to understand how decisions are made, what formal and informal processes matter and how the patronage networks which may exist will help or hinder the prospects of achieving change. This kind of politically savvy thinking is — or is potentially — second nature to them. Indeed, often think tanks will have people who have worked in government and who understand the realities of policy making and politics.
• Ability to flex their research agenda to local context: It is not only influencing strategies which are shaped by political context but also research agendas. Relevance is key to a think tank’s influence: if they’re not trying to answer the questions at the top of the ministerial in-tray they will have less impact. Which isn’t to say think tanks shouldn’t be challenging what should be prioritised in the ministerial in-tray; but their chances of doing this will be greater if they have shown they can respond to ministerial priorities.
• Ability to develop policy solutions tailored to local circumstance: Development thinkers have emphasised for some time the importance of policies which are “best fit, not best practice”. The ability of think tanks to generate tailored policy is particularly important in developing countries where internal government policy capacity is weak: think tanks can help plug important gaps.
• Ability to recognise that political factors need to be weighed alongside evidence: Think tanks know they must combine rigour with influence. To take a hypothetical example, if the evidence says a subsidy should be removed but removing it will significantly undermine the interests of the core constituency of a governing party, then not only will it not happen, but the think tank would be advised not to propose it. Instead it should either develop short-medium term strategies that seek to achieve similar policy objectives but through different means or explore longer-term strategies to shift existing political positions and widen the scope of what is politically feasible. The key point here is that only a locally embedded organisation will be able to make informed judgements about which strategy to pursue.

Centre for Policy Dialogue in Bangladesh responding to local context

Following the Rana-Plaza tragedy in Bangladesh, the Centre for Policy Dialogue (CPD) initiated a monitoring exercise to track post-disaster activities, focusing on the delivery status of commitments made by different stakeholders. This was a spontaneous response to the Rana-Plaza disaster, facilitated by the core funding CPD received from the Think Tank Initiative.

3.3 Challenges and constraints facing developing world think tanks

While think tanks have the potential to contribute to politically savvy development, they are also constrained in several important respects. There are two fundamental issues.

First, there are few developing countries where there exists a ready-made set of institutions equipped with the core functions described above. Some countries will not have think tanks at all — this is particularly the case in the poorest or conflict effected states. Conversely, there are some settings in which the case for supporting think tanks is stronger: we would argue that they can make a more meaningful contribution to development in low/middle income countries than in the very poorest and fragile of states. For think tanks to be successful depends on a certain degree of economic and institutional development — they will prosper if other parts of the policy arena are strong. For example, good think tanks depend on a degree of technical competence in government to engage with and implement policy; and with there being effective higher education institutions to generate research they can apply in a policy context.

Second, think tanks as political actors can become captured by vested interests. When this happens, they can act to reinforce existing power structures, which help to block effective and inclusive development. The risk of capture is a challenge in highly patronage-based systems and in contexts where think tanks are overly dependent on narrow sources of funding.

Even where think tanks exist and where they are not captured, they may still face challenges. In many instances, they are strong already, but many still require support. For example, some still want and need to build capacity to perform the core think tank functions effectively. In our experience, even relatively strong institutions would benefit from improving two specific areas. First, think tanks often need to strengthen their capacity to develop robust and credible policy options — to conduct policy development. The second concerns what might be called strategic advocacy, which involves going beyond a narrow focus on the media dissemination and elite networking, to thinking much more effectively about selecting the most appropriate forms of influencing strategy for specific issues, how to target different audiences with particular messages/arguments, and to do so in a timely fashion.

Capacity issues are to some degree a function of what in many countries is a small labour market for think tanks to draw on. Recruiting people with the relevant skills can be hard. Another challenge for think tanks is to hold on to good people who will often be poached by government, or international financial institutions or consultancies who can pay more.

There is also a tendency, in our experience, for relatively inexperienced think tanks to prioritise the wrong things. This is often the case for the think tanks which have — or aspire to have — strong academic cultures. Here a familiar problem is placing a premium on academic rigour (often encouraged by donor funding) over carrying out policy-oriented and politically savvy research. This often leads them to use inappropriate methods which don’t fit with skill-sets and the timelines needed to influence policymakers. Indeed, a desire to demonstrate academic credibility can mean that think tanks inadvertently neglect to use their local knowledge and political insights, in favour of carrying out academic research. Another example of prioritising the wrong things is that developing country think tanks tend to value economists over other potentially more relevant disciplines.

Funding is an obvious constraint on think tanks. Many have weak and precarious funding positions which make them dependent on government and donor money. Over-reliance on funding from government can compromise the independence of a think tank; while donor dependence can distort research priorities. Clearly in some countries it is unrealistic to think that domestic sources of funding might become available to support think tanks. Even in India, for example, the corporate sector is reluctant to give grants to think tanks for multiple reasons. It is also difficult to find core funding and often this is only available from donors, which raises the risk of think tanks being labelled as “foreign funded institutions” (which can provide the basis for governments to ignore them).

3.4 Why donors need to pay more attention to think tanks

Notwithstanding these challenges and constraints from a donor perspective, if change is to be locally-led and not externally imposed then think tanks have considerable potential. Indeed, they are exemplars of locally-led development.

Additionally, think tanks can help, in part, address the ‘complexity temptation’ described earlier. While the world is undoubtedly complex, this doesn’t necessarily mean donors have to respond with similar levels of complexity. And yet often they do. Witness the increasingly ambitious, and often unwieldy, programmes, they oversee. When it comes to influencing policy rather than trying to wrestle with every component of the policy arena, we would argue that donors might consider working with think tanks which provide a useful entry point into the policymaking process.

To some extent the donors already appreciate and understand the potential of think tanks; it is donor money that helps explain the rise in the number of developing world think tanks. Our argument is that there is a case for donors to place greater emphasis on think tanks. What are the options for donor support for think tanks? As countries develop, the ideal situation is that different think tanks emerge, which are funded from a mix of local sources of funding. This increases the prospects that the think tanks are focusing on genuine local priorities, rather than any external organisation’s interests. However, as we noted above even in middle-income countries such as India local sources of funding for think tanks are limited. So realistically, in many developing countries, governments, donors and foundations will remain important potential sources of funding. Donor support for think tanks can be divided into two broad categories.

• Build up a critical mass of think tanks: this is essentially a public good argument — if there is a critical mass of capable think tanks in any polity then they have the potential to provide the politically savvy policy solutions and play the kind of accountability role. A vibrant and healthy think tank community will also strengthen a culture of policy debate and effective policymaking more generally. From a donor perspective, this typically means providing more core funding, working on issues which are local priorities but also combining this with support for capacity building (see below).
• Co-opting think tanks as local partners who can help achieve change: in this case the relationship between the donor and the think tanks is more transactional with funding being on an issue-by-issue basis. Instead of aiming to build up a think tank as part of the wider policy ‘ecosystem’, this approach is based on more episodic and instrumental partnerships. For example, a donor doing some work on the education system with a policy influencing objective might look to partner up with a local think tank.

The Zambian Economic Advocacy Programme — enabling locally tailored policy influence

Between 2013 and 2016 the Zambian Economic Advocacy Programme (ZEAP) provided extensive mentoring to local economic policy think tanks in Lusaka. The focus of this mentoring was on increasing the capacity and expertise of local organisations across the core think tank functions, from conducting policy relevant research to effective communications and dissemination. However, decisions on which policy issues to work on and which advocacy strategies to adopt were made with the local think tanks in the lead. This reflected the fact that they have a deeper understanding of how power works in Zambia than any outsider. This combination of putting local politically savvy organisations in the lead, but combining this with intensive mentoring, has led to policy influence on important issues including reform of maize subsidies and government borrowing. More fundamentally, this approach has helped the development of a robust group of think tanks which will continue to generate policy relevant work which informs and influences policy.

The first of these approaches is, in our view, most desirable in terms of achieving sustainable change. Because it is more focused on enabling locally led approaches and building up the capacity of a key set of institutions, it stands the best chance of both achieving positive policy change in the short and medium term, but also contributing to strengthening the approach to policymaking and debate in the longer term. However, given pressures on donors it is arguable that the second, more instrumental, approach is likely to be more realistic for think tanks in many countries.

The risk here, however, is that by using local organisations in this more instrumental way, their potential impact is undermined — if a think tank “chases donor money” and works on issues which are not local priorities, this risks undermining their own legitimacy and authority. Far more effective is to clearly place local actors in the lead and to build up their capacity: political change happens when locally led, not when the donors are in the driving seat.

This point about the risks of more instrumental short-termist approaches might also explain why donors have some reluctance to engage more thoroughly with think tanks. For all the talk of working politically they can sometimes recoil at the prospect of doing so: are they comfortable with striking a balance between what the evidence says and what is politically possible? Equally donors sometimes find it hard to really let go, for instance, when it comes to determining research priorities. The strength of a local think tank is that it knows which issues are salient, and where there is scope for making progress. But this can sometimes be at odds with the policy interests and priorities of the donors.

Perhaps this explains why donors have tended to prioritise other types of civil society organisations (NGOs via their grant-making programmes) and put money into more traditional academic and technical research. In opting to fund NGOs and academic research, it could be argued that donors have gone for the ‘safety first’ option. Yet often think tanks are better placed to influence policy than NGOs and their research is regularly more politically informed than that of the academy. Precisely because they have this greater potential for impact, think tanks, we argue, should be more of a priority of donors than they currently are. There is a need for donor support to be more balanced across think tanks, NGOs and academia — indeed greater balance would allow for more effective collaboration between the three, with each playing to their respective strengths (for think tanks to prosper they need to be able to draw on high quality academic research, and partner with campaign-minded NGOs).

In summary, there is a good case for donors building up a critical mass of ‘core model’ think tanks — as a public good. This isn’t to say think tanks are a silver bullet for politically smart aid. Nor are we arguing that think tanks should be privileged over other institutions, but rather that they should become more prominent than they currently are.

4. Beyond ‘core model’ think tanks: options for innovation

Think tanks have an important role to play in achieving ‘politically smart’ change in their respective national contexts. But this is not to be complacent. The best will be striving to improve and many others will be ambitious to build their capacity.

First and foremost, this means getting the basics right — strengthening the core functions of politically informed research, policy development and strategic advocacy. Perform these poorly and the credibility of a think tank will drain away. Most think tanks and support for think tanks should keep this focus. For many think tanks ensuring they are effective at these core functions should rightly be the limits of their ambition.

However, there is also a case for greater experimentation with new think tank forms and different approaches to achieving influence. For all the diversity of developing country think tanks it is striking how formulaic they are when it comes to their internal structures and basic approaches to policy influence. As you would expect there has been a high degree of emulation and borrowing from others resulting in the creation of institutions which look and feel very similar. One important reason for this is because developing world think tanks have leant heavily on replicating practice from the developed world.

While think tanks have proved adept at tailoring their strategies to local context they have on the whole been doing this within a basic think tank model inherited from the west. The problem with this, of course, is that developing countries operate in contexts vastly different from the advanced liberal democracies. The different ways in which politics happens in different developing contexts, from centralised undemocratic states to those which are more open and from those where politics is often contested along patronage and/or ethnic lines, require think tanks to experiment with a wider range of functions and approaches to policy influencing.

If anything, there is greater onus on think tanks to innovate precisely because of the relative weakness of other parts of the policy arena — to highlight some common features: parties are generally weak, there is little programmatic policymaking, interest groups are diffuse and poorly coordinated, and there is often little external accountability of the executive.

The relative lack of innovation limits the ability of think tanks to really exploit their local political knowledge and expertise. Indeed, while think tanks have considerable political insights and understanding of their own country, sometimes this can be latent potential, rather than real.

Moreover, the form and functions of think tanks incentivised by funders can sometimes make it harder to develop a strong and deep culture of thinking politically. Often funders will, through a more transactional relationship, ‘buy’ a particular output. Sometimes this can be too academic to be useful for local audiences. And significantly, much donor support tends to focus on existing institutions, not on the longer-term bet of creating new institutions which may be better suited to their context and more adept at influencing change.

To help unlock their potential and ensure think tanks go more fully with the grain of local context we argue that in addition to funder support for strengthening the core functions of think tanks described above, if think tanks are to really capitalise on the advantageous position they occupy in their policy arenas then more innovation around these core functions is needed.

In addition, we also argue that there is a case for establishing new institutions — fit for purpose for achieving change in their local context. Three directions for innovation could be considered:

• Political insights: explicitly developing the expertise and knowledge of local political context, for example through carrying out ‘political and power’ assessments on any given issue, and for instance developing an expertise in polling.
• Elite convening: developing a function which not only facilitates debate and discussion between key interest groups but which looks to identify collective interests and coordinate actions across these different stakeholders to help bring about policy change. This would also entail think tanks building partnerships with specific and potentially powerful interests, such as the church, which carry more weight than themselves.
• Campaigns and alliances: essentially this would mean engaging in more bottom-up approaches to policy influence, whereby think tanks leveraged the power of citizens and communities to press the case for reform.

Some think tanks, particularly the stronger institutions, will already be doing some of these. We highlight in the case studies below some examples of innovative practice. However, our argument is that few institutions — even the best funded and most established — have thought through how to develop one or other of these functions as a core part of their approach to achieving change. Think tanks (and donors funding them) do not systematically ask themselves what mix of these functions they should be aiming for. And funders tend not to think about how to lower the barriers to entry for new think tanks, or how to incentivise new innovative models of think tank.

Asking this question is important because naturally what mix of these different functions to innovate with, or adopt, will differ from context to context and from institution to institution. In some cases, anchoring think tanks more firmly to a campaigns and community partnership model will make sense as an approach to achieving change — in other contexts a bottom-up approach will be inappropriate, even naïve. And there will, for example, be contexts where great care would be needed with a focus on ‘political insights’.

Innovating by adopting some combination of the three functions outlined above would need to be done with care. There is a risk of trying to do too many things and doing them badly. There is, in particular, a risk of moving too far from the core function. Too much of a focus on either convening or community partnerships could crowd out a focus on the importance of objective evidence and research. Being too political can ‘pollute’ the reputation of think tanks and undermine the credibility of its research outputs. There are also potential trade-offs between focusing on convening and being more politically engaged: often convening power comes from having a reputation for acting as relatively neutral broker between different interests.

Below we set out in some more detail the types of innovation which could be considered. In each case the question for a think tank would be how to combine these with the existing core think tank function of robust policy analysis and research and policy influencing. The purpose is not to develop a prescriptive ‘blue print’ for an ideal think tank — a key argument of this paper is to avoid formulaic approaches. The argument is that these directions of travel provide useful potential focuses for further experimentation.

4.1 Think tanks as centres of political insight

While think tanks have considerable political insights and understanding of their own country, sometimes this can be latent potential, rather than real. Think tanks might therefore build expertise in areas of political analysis and political networking, including: polling and other attitudinal work and the skills of conducting political assessments of policy issues — to inform the think tank’s own work, but also market this ‘political intelligence’ to others (including donors but also government itself and other interest groups). This could have implications for the skills/culture of organisations and the structure — should there be a dedicated ‘political insights’ team or should researchers themselves build up expertise? It also has implications for the type of researcher recruited — suggesting there should be less emphasis on economists, and more willingness to bring in people with different backgrounds (political scientists, sociologists and historians may well be better placed to carry out this type of work).

4.2 Think tanks focusing on their convening power

A relatively untapped source of think tank influence comes from their convening power — bringing together and brokering deals between different interests to help drive change. Arguably this role is especially useful in those developing countries where interests are (relatively) disaggregated and disorganised. There is value in simply bringing together interests to identify collective interests. This isn’t to say think tanks simply assume the role of ‘honest broker’ — they have interests and are actors in their own right. But they are usually more disinterested than other groups eg networked businesses and ‘clients’ of the state.

To a large extent convening is best thought of as form of elite-level policy influencing (but of course it can also involve reaching out to grass roots organisations too) — but done well it has the potential to be more effective than traditional think tank advocacy approaches of policy dissemination. This is because it entails brokering support for policy recommendations among stakeholders who are often more influential than the think tank itself. To give an example: policy change in agriculture is more likely to happen if the farmers’ union, the private sector and the millers support change than if the think tank is advocating for it in isolation (see ‘ The role of IAPRI in achieving reform of agricultural subsidies in Zambia’ below) Convening involves think tanks developing skills in building partnerships and relationships with powerful interests, religious groups, who are well placed to champion reform.

Effective convening also depends on a clear assessment of the political and power relations underpinning a policy issue: think tank needs to know which actors are potential allies and which are opponents of change. Building expertise in convening and facilitation might help think tanks become more influential and improve their potential value as local partners for donors.

Advocates Coalition for Development and Environment (ACODE), Uganda — elite convening

Over the last 10 years ACODE has facilitated multi-stakeholder dialogues on contemporary, emerging, and sometimes controversial issues. Its ‘State of the Nation’ platform brings together individual leaders and professionals from the public and private sectors, civil society, the media and other interest groups. It draws from a TED Conference format and a Doha Debates model to incorporate formal debates, Q & A sessions, panel presentations, and personal storytelling. Issues such as budget policy, oil governance, regional security, public service delivery, and electoral reforms are discussed. The platform includes a local level segment that puts national policy debates into the local context and a student segment for nurturing a new generation of leaders.

The role of IAPRI in achieving reform of agricultural subsidies in Zambia — elite convening

Zambia’s agricultural subsidies are poorly targeted and drain scarce public resources. However, reform is notoriously difficult to achieve given the political interest supporting the status quo. IAPRI used its convening power to bring together key agricultural stakeholders to help make the case for practical and incremental change that was politically deliverable. Over time they helped deepen the constituency in favour of reform. Working with others they called for piloting an electronic voucher as a substitute for the traditional fertilizer subsidy programme. The initial pilot — which they evaluated — was deemed to work effectively and is set to be rolled out nation-wide.

4.3 Think tanks as campaigners and alliance builders

The core think tank approaches to policy influence are essentially elitist — based as they are on the idea of targeting government as the principal policymaker. In many contexts this makes sense: political change in many developing countries is largely elite driven. Any suggestion to move significantly away from elite models of influencing should be avoided. That said there is a case for some innovation with bottom-up approaches. Political cultures are not set in stone and it is at least reasonable to assume that elite systems will be prone to more bottom-up pressures, witness the growth of social media in many developing countries.

Thinking more politically about change also demands an account of how to leverage popular support for change. This could entail think tanks facilitating and supporting campaigns, and building alliances with other civil society and community organisations. It also demands think tanks think more creatively about how they reflect public opinion back to policymakers. It may well be that these activities can complement and strengthen the hand of the elite convening described above.

There is no doubt that this type of public engagement work is challenging, especially given the constraints on civil society in many contexts. But where there is the political space to experiment with these type of initiatives, think tanks should be encouraged to do so. This is partly because elite approaches to influence don’t necessarily result in the type of change that ‘sticks’ — it can often be insufficiently embedded or have insufficient political pressure behind it to ensure delivery (a particular challenge in many developing countries with weak government delivery capacity). Lasting change, with policy following through into tangible social change, needs to be driven by a wider alliance of interests in society.

Public Affairs Centre (PAC) India: influencing policy through empowered citizens

Since 1994 PAC has successfully used citizen feedback — through Citizen Report Cards — to improve public service accountability. PAC has generated a critical mass of grassroots evidence to prove the model works and is now working directly with some government departments to help design and implement evaluation systems that incorporate citizen voices. There is substantial evidence that feedback from the Citizen Report Card has improved government policy and service delivery in and beyond Bangalore. A recent report card on food security in Karnataka prompted the state government to formally acknowledge the issues that need to be addressed for the safety net to function effectively.

Centre of Analysis and Diffusion of the Paraguayan Economy (CADEP) and Investigación para el Desarrollo (ID), Paraguay: providing platforms and debate for citizens

In 2013 CADEP and ID, and five other civil society organisations developed the Paraguay Debate initiative, with the aim to promote political debates in the country around the April 2013 elections. Paraguay Debate reached out to politicians via policy notes, to journalists to better understand policy issues, and to the public to raise interest in policy debates. Debates with each candidate’s technical teams were broadcast by every terrestrial channel reaching an estimated 1.5 million households. The current National Development Plan was influenced by the analyses included in the policy notes for politicians.

Conclusion

There is a significant degree of consensus amongst development thinkers and practitioners that achieving change is an inherently political undertaking. Few cling to a technocratic view of change and many now agree that the demands of achieving lasting, sustainable change are high. Reducing the rate of policy failures and increasing aid effectiveness will require more adaptive, flexible and politically savvy approaches. While far from a silver bullet, in many contexts think tanks — operating as they do at the nexus between politics, evidence and policy change — can make a major contribution. Establishing a critical mass of ‘core model’ think tanks, focused on using policy relevant and politically savvy research to influence policy, will have important payoffs. But for think tanks to maximise their potential impact by embracing approaches to achieving change which go with the grain of their specific context, there also needs to be more innovation. That means increased innovation by some existing institutions, but also a generation of new think tanks.

For full references and bibliography please download the report ‘Innovation in think tanks’