Consensus is growing on the key role politics plays in development. Agreement on what donors should do in response remains elusive. Perhaps they should start with what not to do.
The forthcoming World Development Report 2017 (WDR17) on Governance and the Law shows just how much politics matters for reform in developing countries. As the supporting evidence builds, how should donors respond? At a recent ODI round table, Shanta Deverajan, Chief Economist of the World Bank’s Middle East and North Africa Region and a director of the WDR17, presented a paper suggesting one way forward.
His view is that donors’ traditional project approach - where finance and knowledge are bundled together - cannot tackle the government failures which plague developing countries. He advocates instead that donors offer their technical knowledge directly to citizens, in order “to build their capacity to select and sanction leaders who have the political will and legitimacy to deliver the public goods needed for development”. Meanwhile financing would be provided via budget support, with politically savvy conditions focused on transparency and promoting citizen engagement. This could be described as taking institutions seriously – acknowledging the importance of political will for good policies to succeed, and creating the conditions to support that will through citizens’ political engagement.
Oxfam's strategic adviser Duncan Green contrasted Deverajan’s view with the 'doing development differently' (DDD) movement, which aims to tackle the complex, political landscape of developing countries with programming that uses adaptive learning: experimenting with different approaches and responding rapidly with improvements, rather than sticking rigidly to an ex ante script. Green characterised DDD as doing more in the face of political realities, while he sees Deverajan’s approach as doing less. One might see complementarities between these two approaches. After all, information is vital to the feedback loops that are central to adaptive learning. And as many people at the ODI event pointed out, transparency on its own is rarely enough to change outcomes. So information seems necessary, but not sufficient, to shift from a bad political equilibrium to a good one. Other types of support would therefore be needed, perhaps using an adaptive approach.
But when questioned, Deverajan hinted that one reason some transparency measures had failed was because resources were still available via the old route of project financing. In fact, while DDD targets much better designed and implemented projects, Deverajan’s proposal differs radically from the project approach. His idea is that external actors provide information which enables citizens to demand policies that they believe will help them, thereby generating incentives for governments to deliver. Some external actors’ projects may not be welcomed if their chosen sectors or policy ideas do not align with citizens’ better-informed views - even if they are using adaptive programming.
Debate about the merits of the two approaches will no doubt continue. Acting Executive Director of ODI and chair of the event, Marta Foresti, said that even if a consensus exists that ‘politics matters’ for development, it has not yet changed many donor practices. So it may be useful to reflect on the similarities between the two views, of which there are many, as well as looking at the differences between them. One suggestion is that as well as trying to identify what external actors should start doing in light of political realities, researchers could look at what they should stop doing. They could ask ‘what never works?’ as well as ‘what works?’
So what might be on that list? One common donor practice which may not be politically smart is attaching financial disbursement conditions to the government’s passage of a particular law. This seems like a textbook case of targeting form over function, which can lead to isomorphic mimicry, where governments pay only lip service to reform. Research would help to identify more, like a recent DFID evidence paper which found that anti-corruption agencies are “rarely effective in curbing corruption” in developing countries, where they can lack sufficient independence from political influence, and suffer from weak institutional mandates and a lack of political will. And as Duncan Green suggested, too many projects omit plans for data collection, analysis and research. That prevents evidence from being gathered and lessons learnt about different approaches.
In the absence of agreement on what politically smart actions for external actors might look like, donors might helpfully demand more specific evidence on the politically naïve practices that should be phased out.
