Skip to content

Independent thinking, but at What Price? – Brookings Institution

Peter Singer hits the nail on the head with an article on the ethics of think tanks and the threats that certain funding sources may create for think tanks’ independence.

On Saturday I blogged about the risks of foreign funding to think tanks in India.

Singer argues that:

Thinktankdom is a field that lacks any universal code of ethics, ombudsmen to hold people accountable, a professional association to regulate, etc. Even more, it is filled with people who are happy to speak about everything under the sun –that is, except their own field and the dirty little part that money plays in it.

And believes that:

thinktankers should not take on private consulting contracts with firms they might research and comment upon in their public work.

John Blundell, former chief of the IEA agrees with this assessment. And, I must say, that so do I.

However, this is not always possible. The reality of many think tanks in developing countries -and certainly in the poorest and most aid dependent countries- is one where the funders of research and influence are a few bilateral and multilateral donors; and more recently global foundations. Even in the developed world, international development think tanks (IDS, ODI, DIE, ECDPM, FRIDE, and other smaller non-for profit and for-profit outfits who portray themselves as source of independent sources of expertise) are almost entirely dependent of bilateral donor funds -even though their research and influence is also focused on these donors.

In the aid sector, funding for think tanks tends to come in the form of consultancy contracts rather than research grants or core funding. This creates, according to Singer’s assessment serious conflicts of interest and challenges the very essence of think tanks’ functions.

Singer’s recommendation, that think tanks and researchers should fully disclose the source of their funds -certainly more so when a particular study has been commissioned by a client- should be taken seriously in these contexts.

Transparency can only been a good thing.

2 Comments Post a comment
  1. chris #

    Regulating, creating codes of ethics, and establishing accountability mechanisms for ‘think tanks’ is not straightforward on account of the difficulties in even defining what they are. When Singer talks of ‘unaccountable’ think tanks, he actually means unaccountable sources of power and/or influence.

    This isn’t about ‘think tanks’; this is about consultancy firms (which the majority of think tanks are). I am not sure it is even a ‘dirty little secret’ as Singer suggests – but I could be wrong. The regulation issue is therefore a much wider one concerning who sells what to governments, and why.

    Thus far, the term ‘think tank’ is used as a bullet-proof vest to shield said-think tank from being criticised for their undeniable, inescapable political foundations and to shroud their proclamations in that most tantalising of fashion accesories – ‘non-partisanship’. The idea that by placing my brain in a see-through tank full of water – coming up for air only to be fed by the ambivalent hand that feeds me – is somehow tantamount to non-partisanship is indeed an interesting one. But not really one we can take seriously.


    December 15, 2010
  2. This is another excellent issue raised by Enrique. Chris is right in pointing that codes are hard to develop and to highlight the consultancy firms. However, his view is partial (probably makes sense in the Western world). I have been exploring the issue of ethics among the budding think tanks in central and eastern Europe. The fact that out of 52 organizations that constituted my sample, only one had a clear code of conduct publicly available on their web-site motivated me to write an article of what could constitute the basis for a code of conduct for think tanks. here is a taste of it (it is a bit academic, published in the Intyernational Journal of Not-for-profit Law :-):

    ‘Writing a code of ethics starts with looking at “the perennial debate—what it is about and what it is supposed to achieve” (van Wart 2003). Most of the literature on the subject recognizes three different levels and the code that reflects each of them: a) ideals—codes of ethics; b) norms—codes of conduct; and c) action—codes for rules and regulations.7 Given its brevity, this article will only provide content analysis and address the structure of the first two levels.

    One way to approach this study would be to analyze the codes of ethics in think tanks operating in the United States, where these organizations form a viable industry (Weaver 1989, McGann and Johnson 2005). However, considering that think tanks in CEE lag behind their US peers in development and policy influence, this article takes a different approach. As previously argued, think tanks are positioned between the spheres of NGOs, public service, and individual policy advisers. The first two fields have been the subject of regional comparative studies and benefit from the existence of codes of conduct (or “accountability charters,” as some NGOs call them).8 While there are no similar regional studies on the ethics of the region’s individual policy advisers, their positions could be easily compared with those of their peers working in other contexts. Comparison of all three, complemented by a list of accountability standards for international research organizations, will lead to enumeration of essential elements of a code of conduct for think tanks in CEE.
    See the full article here:
    Or summary here:


    January 19, 2011

Leave a Reply

Please log in using one of these methods to post your comment: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s


Get every new post delivered to your Inbox.

Join 7,065 other followers

%d bloggers like this: