Why donors ignore the evidence on what works, and transparency and accountability projects are a dead end. David Booth’s Non-Farewell Lecture.

ODI is always innovating, and earlier this week organized a non-farewell lecture for one of its big thinkers, David Booth. As David Boothfar as I could work out, this was a celebration of them stopping paying him (aka ‘retirement’), while he continues to work for them for free as a visiting fellow. Interesting business model.

Anyway, for those that don’t know David’s work, he is an iconoclastic sociologist who, through big research projects like the Africa Power and Politics Programme, has spent decades pouring cold water on the half-baked buzzwords and unthinking fashions that flourish in the aid biz. We learned in the intro that he started on this course when he abandoned Trotskyism in the mid 80s – has anyone mapped the number and proclivities of lapsed trots in the aid sector? If not, why not?

David has been an influential voice in the critique of the failings of aid’s attempts in recent decades to implant Western style institutions in non-Western contexts. Phrases like ‘best practice’ and ‘good governance’ are useful alarm bells on this. Here he is (abbreviated) on what we have learned:

nb this is a Lant Pritchett slide, not David's
nb this is a Lant Pritchett slide, not one of David’s

‘If there is a single message that emerges from the best governance scholarship and research since the 1980s, it is that we know much less than we thought we did about the institutional arrangements that developing countries need….. We need a robust agnosticism about the forms of governance likely to contribute well to building economies and improving conditions of life in poor countries. The following things are clear:

  • The formal institutional arrangements that work reasonably well in OECD countries today are not exportable.
  • The institutional arrangements (formal and informal) that have helped countries break through into fast development are very diverse.
  • Historical path dependencies are really important, but, contra Acemoglu and Robinson, history contains more than just two typical long-term trajectories. There is much more to the global story than the persistence of ‘extractive institutions’ versus the spread of ‘inclusive institutions’.
  • Countries can and do make rapid and economic and social progress with institutions that, in overall terms, are seriously dysfunctional.

If there is a single takeaway from all of this, it is that ‘good’ institutions are neither easy to prescribe nor a precondition for successful developmental catch-up. Developmental catch-up invariably involves some borrowing ideas and technologies from abroad, but is never successfully led by borrowing. It is led by problem solving.’

If every serious scholar thinks this apart from Acemoglu and Robinson, (who he euphemistically categorized as ‘outliers’), have the donors done the decent, evidence-based thing? Ermm, not exactly.

‘That the immediate post-Cold War era should have featured a certain amount of liberal-democratic hubris is understandable. More striking is how durable much of the associated thinking has proven to be in the face of obvious disappointments and the build-up of the scholarly consensus against it.’

The discomfort zone for me and, probably, quite a few FP2P readers, is that David characterizes efforts to promote

Waste of time and effort?
Waste of time and effort?

transparency and accountability as a prime example of imposing irrelevant and inappropriate solutions. ‘Much of what lies behind the social accountability theme, like the content of most public sector capacity building, is a projection of solutions derived from experience in OECD countries onto the very different realities of developing countries. In other words, social accountability falls to the critique developed [that] it is profoundly solution-driven.’

I suspect he feels a bit that way (or at least ambivalent) about human rights too – we’ve crossed swords in the past about his apparent infatuation with Rwanda’s Paul Kagame. At least this time he moderated his usual dismissal of civil society organizations, acknowledging their role but adding ‘Civil Society Organizations are contributing to their own marginalization by doing these kinds of things that don’t lead anywhere (eg Transparency & Accountability – sorry about that, IBP).’

The way to fix aid is to really adopt ‘problem-driven iterative adaptation’ (PDIA), which means taking each letter in that clunky acronym seriously. In his view, donors have made more progress on the IA bit (Adaptive Management is all about iteration and adaptation), but have still not bought into problem-driven. The fudge has been ‘issue based’ programmes that say ‘we will use adaptive management techniques to fix governance/health/education’ – DFID’s vaunted SAVI governance programme in Nigeria would fit that bill. David thinks this falls short – what if you want to tackle poor quality education and find out that the real problem is bad roads, failing electricity or a low tax base? You need to be able to follow where the problem leads you.

Looks good, but is her placard in English just to please the donors?
Looks good, but is her placard in English just to please the donors?

Well fine, but then he’s basically saying aid donors should behave like governments, working across sectors and ministries, making trade-offs etc, which seems pretty impractical. Ironically, his weakest section was about why donors fail to act on this understanding – he puts it down to domestic donor politics and self-serving governance advisers who advocate institutional reform because that’s what they do. He even, true to his iconoclastic streak, advocated scrapping the unfortunately named GOSAC, the DFID governance team that funds lots of his work! But he didn’t really get into the broader realm of ideas and institutions, and in Q&A accepted that some of the most influential drivers of ‘best practice’ reforms are westernised leaders within developing country governments.

Conclusion? David makes people feel uncomfortable, sometimes goes too far, but who cares? In a sector full of groupthink, we need more iconoclasts like him, so I’m very glad he isn’t actually retiring.

Subscribe to our Newsletter

You can unsubscribe at any time by clicking the link in the footer of our emails. For information about our privacy practices, please see our .

We use MailChimp as our marketing platform. By subscribing, you acknowledge that your information will be transferred to MailChimp for processing. Learn more about MailChimp's privacy practices here.


7 Responses to “Why donors ignore the evidence on what works, and transparency and accountability projects are a dead end. David Booth’s Non-Farewell Lecture.”
  1. Diarmid

    The critique of the externally-driven “good governance” approach to development is a powerful one – I lived and worked inside this approach for years and agree its promise has been wildly overstated (including by the likes of me).

    But I wonder whether people will look back at PDIA ten or twenty years from now, much as we are looking at “governance” now, and conclude that while a good idea in theory, which addressed some basic flaws in the orthodoxy that preceded it (just as the governance agenda did), it too often turned out in practice to mean a jumble of unconnected and unscalable projects driven by much the same kinds of donor and recipient-government incentives as before, only with iteration having replaced logframes, and with fly-in consultants with Powerpoints being replaced by resident stuck-in-the-muds who are steeped in local knowledge (as Rory Stewart advocates) but have become bogged down by it.

    I don’t know what the answer to this would look like. I’m hoping another commentator will suggest it, so that I don’t have to retrain as a plumber. Nothing wrong with plumbers but you can’t get the airmiles.

  2. I find the dismissal of transparency and accountability a bit over-blown.

    My take – see also comments and responses on David’s piece on “5 myths about governance and development” from 2015 https://www.odi.org/comment/9274-five-myths-about-governance-development – is that while urging countries to put into practice T/A principles on the grounds that they are good things in themselves may not be a smart or effective approach, there is value in exploring how T and A can support more effective cycles of learning and adaptation, which engage with the incentives and political dynamics around governance-related challenges.

    This is where we’ve been headed at Global Integrity, exploring the value of transparency, participation and accountability, not as ends in themselves but as means that can support the adaptive learning that’s needed to address complex and fundamentally political governance-related challenges. https://www.globalintegrity.org/2016/01/the-value-of-open-governance-adaptive-learning-and-development/

    I’d love to see more engagement between the opengov/T+A crowd and the doing development differently crowd. And reflection on the experience of countries where both types of approaches are being put into practice, to explore whether and how they are complementary? I think there’s much to learn in both directions. Maybe we – along with a combination of T/AI and ODI (the development one, not the data one, or both?!) – can help facilitate? A tale of two ODIs? 🙂

  3. Susan Watkins

    Good governance and good institutions are obviously important for development. But I think working for change at this high level overlooks the importance of country implementation. Two brief examples: (1) in a political economy study of learning in primary public schools in rural Malawi with funding from Dfid via CDG, we found that although improving the quality of education was of great interest to Dfid, it was of little interest to the ministry of education (or the parents, who only care that their child get a certificate that will permit them to get a job so that the child can eventually support the parents in their old age). The govt had to pay teachers’ salaries, but were reluctant to spend money on rural kids. A WB study found that the govt spent 3x as much per capita on tertiary ed than on primary ed (Vaikaathur et al 2016). The ministry’s logic was that tertiary ed helped drive development (though I think their main interest was in the government using donor money to subsidize their kids for tertiary ed ); in addition, the educated elites assume that the children of villagers could not learn anyway, the parents were “lazy”, so why spend money on them. (2) To implement government and donor projects requires a layer of brokers, from the cosmopolitan elites in the offices of the UN and INGOs in the capital to layers of govt and NGO brokers in in the districts. We spent a lot of time chatting with brokers, and learned that they had many obligations other than doing their jobs properly; in particular, they had to support their children in school and to support indigent relatives. It’s thus not surprising that implementation is often poorly done.

    The bottom line is that the donors have fantasies of changing poor people in far-off places, but at the local level there can be great resistance–e.g. often you can bring the horse to the water but you can’t make him/her drink.

  4. Mark Wentling

    The day you no longer see a clutter of NGO signs at every corner will be the day that marks progress being made. Local competency and good management are too often overlooked. A desire to change for the better among the people is needed to move to a higher stage of well-being. There is no substitute for hard work and discipline by most people. As long as women can’t get ahead, no country will move ahead. There is also no substitute for honest, inspired, competent leadership that puts the best interests of the people ahead of their own. Aid can’t work unless local institutions function well in the service of the people. In some cases, no aid may be the best aid.

  5. Steve Golub

    Good discussion here. A few thoughts, all along the lines of making matters a bit simpler, and somewhat confined to governance, democracy and the rule of law:

    1. Let’s get away from the failed nation-building narrative which has dominated much of development discourse, and focus on aid that benefits specific populations and policy/legal reforms.
    2. As a wise Ford Foundation representative once told me, her philosophy was to find good people and organizations with good ideas in a given country, and to support them. Such a foundation approach, I’d argue, relies on learning and adjusting such support (including restructuring, increasing or ending it) along the way in partnership with grantees to the maximum extent possible. This is anathema to the Rube Goldbergesque project cycle, logframes and other tools that unrealistically rely on A leading to B and on received “wisdom” from development agencies and consultants. It admittedly needs to be adjusted to country circumstances, but really is a more straightforward version of PDIA.
    3. We need robust support for longitudinal, retroactive learning, returning to the scene of a community, program, NGO aid or legal/policy reform years after the aid has ended, to ascertain what if any changes proved sustainable, as well as unanticipated impacts. Given that development aid is a field that aims for long-term change, it’s shameful how little long-term impact research goes on and how much more funding goes into superficial, often meaningless, short-term indicators.

    • Duncan Green

      Agree with all of that Steve, especially the unforgiveable lack of long term/longitudinal research and learning – the sector moves in an amnesiac short termist bubble

    • Steve, are you able to name to Ford Foundation rep? I’d like to speak with more program staff at foundations / aid agencies that take an organizational rather than strategic lens to their grantmaking.