OECD’s ‘States of Fragility’ report: Some discomfort around the indicators
What's on this page
The OECD report on States of Fragility – Meeting post-2015 Ambitions proposed framework for assessing fragility misses out on a number of key risks and key building blocks for resilience in fragile states. I warmly welcome the OECD report on States of Fragility – Meeting post-2015 Ambitions. It moves away from a simple label of ‘fragile states’ and unpacks the concept into a number of different dimensions. This approach can help identify national and international priorities by shedding light on which countries are the most vulnerable to which risks, and can inform international financing allocations accordingly. It allows for the notion that a country can be vulnerable in one dimension and not in another, which breaks through the artificial juxtaposition of fragile countries on the one hand and non-fragile ones on the other. This shift in approach is timely, as it sets the scene for thinking about how to work towards achieving Goal 16 on peaceful and inclusive societies, assuming this goal will make it to the end of the negotiations. So far so good. But the deeper I dig into the framework, the more uncomfortable I become. This happens in particular when I take a good look at the indicators that are being used to assess the vulnerability in each dimension. Of course, when only a limited set of indicators can be used, there are choices that need to be made and trade-offs that need to be worked through. I am certain the authors made these choices after careful deliberation. But seeing them grouped together in this way, I can’t help but worry about possible unintended consequences. Click on the image to enlarge In essence the framework brings together a number of key outcome variables that directly show a country’s fragility (such as number of battle deaths), with variables that show a country’s exposure to external risks (the disaster risk a country faces), and a country’s internal mechanisms to cope with threats of violence and conflict (such as the extent of rule of law and the effectiveness of institutions). This logic makes sense. But to me, the framework seems to miss out on a number of key risks and key building blocks for resilience in fragile states. Rule of law and effective, accountable and transparent institutions are indeed key building blocks of resilience; they can help a country reduce its vulnerability to conflict and risk. But where are other key issues like inclusive political systems and decision-making processes? The security sector seems completely missing. And what about improved societal relations and reconciliation? Isn’t this the lesson we drew from South Sudan, that we should have focused more on the social contract and social cohesion? The choice of indicators used in the fourth dimension of economic foundations is not very clear to me; I would have thought they should refer to key vulnerabilities that are economic in nature, such as youth unemployment and economic, social and political exclusion. Yet, the indicators refer to the Doing Business Index, GDP growth and degree of education of the labour force (as a proxy for future economic potential). How do these relate to economic, social and political inclusion or youth employment? Are we saying that economic growth alone will reduce vulnerability to conflict and violence? And isn’t there an increased understanding that in fragile states more is needed to promote public sector development – in particular domestic private sector development – than scoring high on the doing business indicators? And even if we were simply talking about economic vulnerability, would we not need to look at economic diversification or over-reliance on commodity markets? The fifth dimension of resilience is also a bit of a mystery to me. First of all, to me resilience is what emerges when internal capacities are able to withstand and overcome vulnerabilities; as such it is an overall outcome, not a sub-dimension of fragility. But OK, fine, we can take it as an aggregate dimension of vulnerability to risk. What the dimension seems to want to cover is the exposure of a country to external shocks, as well as its adaptive capacity to mitigate shocks. But its disaster risk assessment seems to be based on natural disasters only. It doesn’t look at the risks so crucial to fragile states, including horizontal inequalities, societal fragmentation, economic or political grievances, youth unemployment, spill-over of conflict in the region, or radicalization. But most importantly, this framework does not sufficiently address the external stresses - or global drivers of conflict - that make it increasingly difficult for countries to overcome their internal fragility and build up their resilience. Organised crime, illicit trade and arms flows, and illicit financial flows are mentioned in the framework, but have no indicators associated with them. Yet, arguably, these are key drivers of fragility. Going forward, with the SDG agenda, let us not make the same mistake as we made with the New Deal, which focused only on countries’ own capacities to transition out of fragility. Let us look at what the international communities need to do to also address these external stresses. The report is fantastic, in the sense that it takes us one step closer to seeing fragility as a universal problem that all countries suffer from in a greater or lesser extent. It also presents very important facts and recommendations for financing for development that are very important and need to receive all the attention they can get. But the proposed framework for assessing fragility? I think it warrants a good discussion on the risks of choosing few indicators to assess where countries stand, especially if resource allocations are likely to be made accordingly. This discussion is an important one to have now, before the post-2015 agenda is finalized and monitoring and implementation of Goal 16 will begin. Therefore I warmly welcome this report, and I hope it can spark this so very necessary conversation. The views expressed here are those of the author, and not necessarily those of ECDPM.