An initiative model for a plural think tank: a space of policy ideas

17 December 2014

Many think tanks are home to many voices. When encountering a new think tank, one of the first questions I ask, is whether the organisation has a single voice or many voices: when a researcher puts forward a proposal, is this proposal the researcher’s or the think tank’s?

More than often the answer is that the think tank is plural and open to the very different views of their researchers. So the proposals are made by the researchers … from the think tank, but not on behalf of all the researchers… it can get confusing.

Nonetheless, this can be explained by the origin of these organisations: university research centres, civic associations, research consultancies, etc. Their evolution has had to allow for the incorporation of different ideological strains (which may not constitute important variations) that tend to grow and shrink with the passing of years and the natural ebb and flow of political and ideological cycles.

A recent case in Peru reminded me of this issue and got me thinking about a model that could ‘work’ for think tanks that are and want to remain plural. But more importantly, that still want to have an impact.

When your view is not very popular -not even at home

Here is the case. A university based think tank produced a study on the national pension system (Peru has a combination of public and private providers but contributions are mandatory). I won’t go into the details of the proposal but the basic idea was that mandatory contributions should be abolished and instead a future universal basic pension should be paid for from general taxes and that if people wanted an additional pension then they should be free to choose whatever pension-savings product or strategy they wished.

Needless to say the private pension providers were not happy about this. And nor were many researchers and think tanks who hold an entirely different view. And many of these researchers were based at the think tank that published the report.

The presentation of the study was something to be seen -it is also one of the events that encouraged me to write about how to produce events. I couldn’t make it to the venue so I watched it online. Some of the fiercest reactions came from within the think tank, itself. Critique, however, missed the main ideas and focused on the assumptions, on the data and, even, on the researchers themselves. I had to Google logical fallacies just to keep up.

But the study reignited a debate that had been dormant. It had made the headlines a few months prior, after another think tank challenged the system, but it cooled off. Then it came alive again with this study. But it has cooled off again. Almost immediately.

Not much has been heard from the think tank in question about this issue, either. After the fierce reaction against their study, whatever ideas it may have had to communicate the findings and recommendations appear to have been put on hold. Who could blame them?

The thing is that something similar happened last year. Another study, on an entirely different topic (this time related to the country’s social policy), enjoyed an initial media boost followed by a wave of criticism -even from within. And in that case, too, the debate that had been generated quickly died off.

A few weeks ago, however, colleagues working in the social policy sector and deep within government confided that the study on social policy that was attacked so fiercely last year had been “sort of right” all along. It is a shame, they said, that it had been allowed to be obscured.

The point is that this kind of reaction is, and should be expected. It is good, in fact. We do not want everyone to agree all the time. But there is a need to find a way of taking advantage of that disagreement and turning it into something positive and constructive.

A new model for influence -even if it’s not your idea

This is a likely scenario, especially when dealing with issues of great political and economic interest. It is also not something that think tanks should avoid. Rocking the boat is one of the most important roles think tanks can play in a society. If they only went for safe bets we’d all be worst-off.

So what do to? What I propose is a modified initiative model (so CGD’s lessons learned paper should be a must read). Instead of developing an initiative to pursue the think tank’s solution to the policy problem, the modified initiative is developed to pursue a solution (any solution) to the problem. The difference is small but it has potentially big implications.

One of the important roles of think tanks is the formation and nurturing of spaces for debate. This is particularly relevant for think tanks that are plural in nature. They themselves are, already, spaces for debate. All they need to do now is make those spaces and debates public.

This happens already but I think few think tanks focus their attention and that of their audiences on key specific policy objectives.

The case of the pension reform study followed a rather traditional research to policy approach:

  1. researchers identify an issue of public interest,
  2. researchers undertake the study,
  3. publish it, and
  4. aim to change policy (or at least inform policy) based on their proposal.

The drivers of a process like this are usually the researchers -in this case, researchers with a very keen eye for the policy agenda as they surely anticipated it. This means, however, that if they move on to other things (or if they choose to take a step back from it all) the general aim of policy influence may be lost in a sort of no-man’s land.

An initiative, on the other hand, would have picked-up on the last step and placed it at its core. It would have worked backwards from the question of: ‘what do we need to do to change this?’ and possibly done lots more research, engaged in communications, capacity building, lobbying, etc. In an initiative, ownership is shared between various people with responsibility for research, management and communications, but the leadership is held by a policy entrepreneur. Someone who believes in the solution and wants to see it make a difference; now, in a few months, whenever.

In the traditional case, a think tank cannot possible host two ‘conflicting’ initiatives. You could not have an organisation that backed private pension only model and a public pension only model at the same time.

In the modified version of the initiative that I propose, however, this ‘conflict’ is possible; it is, in fact, desirable.

The objectives

In the modified version, the objectives become less specific. They are based on more fundamental ideas that everyone would be OK with. For instance, “Eliminate Child Poverty.” Who could be against this? Or, “Have a fair pension system.” Again, who could say, “No, I do not want a fair pension system”?

These objectives are sufficiently catch-all to get the process rolling. They also open a new and important (in fact, I think that this is fundamental) line of thinking: we need to stop assuming we all mean the same thing all the time. Eliminating child poverty may sound like something everyone will be in favour of but poverty is not defined by all in the same way. A fair system? For whom? Defined how?

The focus of the strategy: the space

Unlike the initiative or the typical research project, this model starts with and focuses on the space. What this initiative is doing is creating the space from where the best ideas will emerge. It may be an idea that the think tank hosting the space will develop and put forward. It may be an idea that takes bits from different ideas, own and borrowed. And it may even be some other think tank’s idea. Or even some other party’s idea: an NGO, a business leader, a trade union, intellectuals, the government, etc.

It does not matter. The idea will emerge from this initiative to find a solution to child poverty or to design a fairer pension system. That is what matters.

Open innovation

The concept of open innovation is not new -yet few think tanks have embraced it. Few think tanks would ever publish or feature the work of others on their websites. Few, too, would want to organise an event which featured a researcher from another think tank rather than their own. Many think tanks even have policies that demand that their researchers need to be the main authors of papers they publish -even if they didn’t even do they work themselves. This misses the point of a think tank as the source of ideas.

Some of the best think tanks in the world spend quite a bit of time looking for people and ideas to make their own. Take the case of FORMA, for instance. The Center for Global Development developed the idea but World Resources Institute has taken it forward as Global Forest Monitoring (GFW).

As much as we would like to think that we have the best ideas about the issues we are passionate and expert about, the chances are that there are others out there that could have better ideas. I spend more time than most thinking about think tanks but the best ideas on communications that I have published have come from people like Nick Scott or Jeff Knezovich, the best ideas on funding are from Goran Buldioski or more recently Gjergji Vurmo, and Hans Gutbrod‘s work on transparency is far better than anything I’ve come up with after years of looking into it. I draw from them in my own work. I’d be foolish not to.

Well, by embracing open innovation, think tanks can embrace the idea that it is possible that the solution to child poverty may be in another think tank, in an NGO, deep within the public system in some long forgotten public servant’s notebook (this is what Guerilla Policy assumed when it decided to work with the implementors of policy), or in the musings of a blogger. A fair pension system could just as well be designed by a post-doc on pension systems as by a graduate student struggling over his or her thesis.

Not one but many proposals to find what works

So the think tank needs to focus on what might work. This means planning not one but a few studies -quite possibly, allowing the different ‘factions’ within its plural research cadre to team up. It means inviting others to join -directly by joining their ‘competing’ research teams or inviting them to put forward their own.

But it also means that research must respond to the need to foster and nurture a public debate over a long period of time.

It cannot be: “go away for 6 months to do research and then come and let us know what you think.”

Instead, it has to be: “go and do research but keep updating us of your progress through frequent tweets, blogs, media articles or op-eds, research drafts, and by participating in the debates generated by the updates and presentations by others.”

This is particularly important in view of the findings of a recent study on Solution Aversion by Duke University that has shown that people are likely to challenge the existence of a problem if they disagree with the solutions presented to them. So it is much better for a think tank to come up with a few options from a few different ideological points of view than just one.


Communications professionals always tell us that think tanks have to work hard to be (and continue to be) a trusted source of information.

Journalists and policymakers echo this.

So it works well to be a trusted source of all ideas on a particular issue. Journalists and policymakers will find it easier to trust a think tank if they know that calling on them will give them access to all available views and not just one.

The media and policymakers are naturally sceptical of think tanks. They should be, after all, it is their job to question the advice they receive. It works well, then, when think tanks provide them with options. It certainly helps policymakers who need options to make more informed decisions.

But it is also easier, period.

Another lesson we have learnt already is that it is much easier to influence when the issue you are working on is ‘on the agenda’. So for many think tanks, ‘getting an issue onto the agenda‘ is as good as it gets. This is their main objective.

But getting an issue on the agenda is only half the problem. Once it is there you have to keep it there. And for that you need something new! Newspapers won’t pay too much attention to  your 5th call about a new repackaging of your first report. But they will be interested in the presentation of a 5th policy option; especially if it promises to challenge one or more of the previous ones.

You also have a better chance of keeping an issue on the agenda if the initiative produces lots of content: intermediate and final outputs. It is not enough to publish a single report every six months or a year. Research plans, updates, op-eds, events, blog posts, and other communication tools are all important.

All think tanks are under pressure to demonstrate their value. Whoever is asking, they have to be seen to be relevant. This is a great strategy to be relevant even when they do not have anything new to say or publish. Even if the policy solution that is eventually adopted came from another source, the think tank, through this kind of initiative, could still claim a share of the influence -a very significant share, in fact.


These are much larger efforts, of course, but not necessarily more expensive. They demand, first, a shift in the attention that hitherto forgotten aspects of a research project often get.

More resources will have to go into communications and the organisation of debates. More will need to be allocated to initial planning stages. These initiatives need leaders that are ‘beyond’ one or another policy option. These are initiative managers who can convene and nurture all policy options. Simon Maxwell’s policy entrepreneur description should be used to hire them.

All of this may be seen by researchers as an imposition and a loss of power; but should be interpreted as greater freedom to focus on research rather than management and communications.

In the end, researchers should have more time to generate more ideas and, crucially if you think this will be more expensive, more income for the organisation. So it should pay for itself.

There is also a special kind of cooperation that is necessary. Think tanks might find it hard to go at it alone -even if they are large and plural enough. After all, these are much larger and complex projects than traditional research projects or even traditional initiatives. They may have to partner with their ‘competitors’ to design and implement these initiatives. This kind of cooperation could be labeled: “cooperation in fundraising and competition on ideas.”

The usual approach that funders follow when confronted with broad objectives like “eradicating child poverty” (MDG-like objectives) is to fund one organisation to do research on how to do it in each country and then fund them (and others) to promote their recommendations. They think that funding two or more organisations to do the same research is overlapping and wasteful. But in not doing so they fail to recognise that public policy recommendations are never scientifically water-proof; nor are they politically neutral. (DFID Zambia has taken a different approach through the Zambian Economic advocacy Programme.)

They also miss the point of evidence based policy. The point is that policymakers can use evidence to help make up their minds about tackling different challenges and adopting different courses of action. To be informed properly, they need options. One piece of research should never be enough.

By funding efforts that purposely seek different options -and then bring them together to challenge each other on the merits of the ideas that underpin them- policy research funders would be supporting more transformative and sustainable outcomes. Not just policy change, but also change in the way that policy is done and in the context in which this happens.

Existing approaches

This is being done already. There are several models that could be adopted and adapted by plural think tanks keen to pursue this.

  • The What Works Centres in the UK, a network of centres of excellent, each focused on a specific policy issue, seek out evidence and knowledge from multiple sources. The centres may provide the solutions that policymakers are looking for but this does not mean that they will have emerged from within the centres themselves.
  • A programme I helped develop in Zambia, ZEAP, was set up by DFID to encourage more and better economic policy public debate. Rather than funding think tanks to work on entirely disconnected and unrelated issues, DFID would fund them to work on similar policy questions -from different perspectives.
  • The Battle of Ideas model in the UK is another interesting effort. The festival of debates that the Institute of Ideas organises every year hosts debates that refuse to acknowledge any one person or organisation as “The Expert”. Rather, it promotes an open debate on a range of fundamental issues. This is a kind of Anti-TED. In TED events, the speaker is the only one who is allowed to share his or her idea. In the Battle of Ideas model, everyone is assumed to have an idea (or an opinion) and there is space for them.
  • Also from the UK is the Policy Fight Club model, promoted by Policy Exchange. Here the think tank provides a space for ‘opposing views’.

These are approaches that could be incorporated into a modified initiative like the one I am proposing above.