Where have we got to on fragile states and what comes next?

June 22, 2010

     By Duncan Green     

Another week, another conference. This time it’s hosted by the UK development ministry, DFID, which among other things, has an impressive track record of funding research on development issues (declaration of interest – I worked for DFID for a year in 2004, and sometimes advise them on research issues). This week’s gabfest is called ‘The Politics of Poverty: Elites, Citizens and States’, and it is being held to pull together the findings of four large DFID-funded projects over the last ten years. So in terms of epistemic communities, we’re talking political scientists – the conversation so far has been pretty much an economics-free zone.

fragile statesThe synthesis paper of all that research is on DFID’s Research for Development website, along with links to the four research centres involved in the work.  It shows what can be achieved by a sustained research funding on a particular topic and  I’ll try and summarize/pick out highlights at some future date. For now, here are my notes on a great presentation from Tom Carothers, of the Carnegie Endowment for International Peace. Tom’s a guru on democracy and development. Sorry for the long post, but he’s worth it.

First, how did the debates stand ten years ago, when this research project got under way? Tom took note of the ‘huge movement towards ‘politics matters’’ in the development aid community, one that started in the 1990s and gathered considerable steam in this decade. The evolution of the governance debate in the 1990s meant that donors felt that they understood what policies were required to trigger development, but were frustrated by states’ incapacity to implement them. This led to a focus on state-building, but when the first efforts led to technocratic failures, aid donors began to think that political process must hold the answer, sometimes to the extreme that ‘politics became a metaphor for everything we didn’t know – everything that wouldn’t fit in the equation.’

Secondly, the rise in importance of post conflict situations meant that in countries like Mozambique, Angola, Liberia, and Cambodia, politics was ‘hitting them in the face’. Over the course of the decade, this concern with post-conflict morphed into the concept of fragile states. The focus on fragile states gained so much attention that it sometime overshadowed consideration of governance issues in non-fragile states.

Thirdly, there has been a big response to the question ‘Can we prove that good governance leads to better development outcomes?

How far have we got on these three points?

1. We have broken politics apart into more useful/researchable questions:

–         Understanding the incentives of elites and how they shape politics

–         Mapping the different forms of state-society relations

–         Understanding institutions in a political light, eg when do governments introduce progressive or substantial taxation systems?

–         Grasping the role of informal institutions

But Tom saw a deeper question: maybe having an ‘ideal’ for state-citizen evolution is neither necessary nor ideal. Are we getting to a world where we have no template, a ‘deeper anarchism’ in which we don’t know what the end of state-building should be, given that our ‘ideal forms’ seem unrealistic and many non-ideal forms are producing results in some areas?

2. Violence: three areas of progress

–         we have to understand ‘political settlements’ as the glue that keeps states together

–         there are certain core elements of the first stage of state-building: security, revenue collection etc

–         the importance of inclusion and inequality as drivers of conflict

3. He finds the question of the evidence about the value of good governance for development ‘surprisingly unsettled’, although he sees a ‘rich vein of argument and discussion’. It has been easier to prove impact at micro level, but proof at the national level has been very elusive. ‘It is impossible to explain why some regions do well without good governance, or that if good governance is introduced, it leads to growth’. That’s a pretty big hole.

Has all this research been translated into practice? Citing a recent article by Sue Unsworth in the Journal of International Development protestsas a rich source of analysis on this question, and drawing on his own experiences, Tom pointed to DFID’s research on ‘drivers of change’ in Nigeria, where what is now called ‘political economy analysis’ helped DFID make some basic changes in its work to reflect a new understanding of the importance of coalitions as potential agents of change in a context where the entrenched elites show no incentive for developmental reforms. So yes, it has had an influence, but only on some governments and projects. Why has it been so hard to introduce political thinking more generally into donor behaviour?

–         Aid organizations have a huge historical investment in technocratic approaches – that is very hard to change

–         This knowledge (on complex political interactions between states and citizens) is hard to know what to do with as an outsider. Research always stresses ‘there is no recipe’, because of complexity and context – not much help for policy makers.

–         It involves acceptance of much more limited goals for development actors – a very difficult lesson for the development industry.

More heretically, he thinks there may be a fundamental conflict between the Paris and Accra principles on aid effectiveness and taking politics seriously, which requires ‘infiltrating aid much more intensively in local political realities.’

Finally, he stood back and gave us the really big picture: these discussions are rooted in a post-Cold War interregnum characterized by a ‘certain benignness in international relations’. That assumed

–         we knew the direction of history and economic policy

–         the international community was involved in mopping up old Cold War conflicts

–         donors could get more involved in politics as governments around the world accepted the benefits of higher levels of pooled sovereignty.

But in the period that we’ve been trying to understand the political dimensions of development, these assumptions have become invalid, as part of a larger international shift away from the post-Cold War period into something different. There is much less confidence on what constitutes ‘sound economic policy’; it has become clear that conflicts are metastasizing and multiplying – Thailand, Mexico, Kyrgyzstan are not cold war hangovers. Finally, there is a ‘huge backlash’ against political intervention by outsiders, especially aid donors

So although we’ve moved far, the international context, and therefore the research questions we’re asking are changing underneath our feet. What new questions should we be asking?

And there frustratingly, hemmed in by his 20 minute limit, he left it. What are the equivalent questions on the politics of development for the next decade? Given what we now know, what political or other interventions in fragile states can an economically ailing West realistically undertake, with a decent chance of success? Answers, as ever, in the comments, please.

Postscript: see here for Foreign Policy magazine’s 2010 ‘fragile states index’, out this week, topped by Somalia, Chad and Sudan, with interactive map.

June 22, 2010
Duncan Green