Aid’s fragile state problem – why is it so hard to even think about?

I’ve spotted a recurring problem with the way the aid sector talks about fragile and conflict-affected states (FCAS). FCAS are characterized by states that are either absent or predatory – in terms of development, governments and officials are as likely to be part of the problem as part of the solution.

But the aid sector, especially the official world of bilateral and multilateral organizations, traditionally relies on the state as the main channel for spending aid dollars to provide education, health or the rule of law. Its instincts are statist. Aid wallahs can’t seem to cope when that recipe doesn’t work, and when confronted with FCAS, it often suffers some kind of intellectual loss of nerve where it ends up ‘assuming a state’ when discussing the role of aid. With a functioning (if imaginary) state, suddenly we can have all those comforting aid discussions about how it should best spend money, build schools etc etc. Most odd.

I mentioned this phenomenon in my review of a recent LSE-hosted ‘Commission on Fragility and Growth’, but it cropped up again in a recent ODI report on ‘financing the end of extreme poverty’.

Before it defaults back to a fictitious state, the report crunches some useful numbers. It projects poverty numbers out to 2030 and looks at what might ensure that the world ‘gets to zero’. It uses IMF and World Bank estimates of what states can reasonably be expected to raise in tax revenues and costs how much states need to spend to provide universal education and healthcare and on social protection to get everyone above $1.90 a day.

Key findings:

The proportion of people living in extreme poverty across the world is projected to fall from the latest estimate of 11% to 5% in 2030 as a result of economic growth. This would see 400 million people lifted out of extreme poverty ($1.90 a day), but leave another 400 million still in extreme poverty in 2030.

Of these 400 million people, 85% will be in fragile states.

Low Income Countries (LICs) will have 54% of the global total of extremely poor people (some fragile states are actually middle income, hence the difference between the numbers).

On average, LICs could only increase their revenues from 17% to 19% of gross domestic product. The rest is going to have to come from aid.

While additional tax revenue reduces their funding gaps, 48 of the poorest countries in the world still cannot afford to fully fund the three core social sectors needed to end extreme poverty even if they maximise their tax effort. They would still face an aggregate financing gap of $150 billion a year. Of these, there are 29 severely financially challenged countries (SFCCs) that cannot even afford half the costs.

The 48 under-resourced countries are predominantly low-income, least developed and fragile states. The concentration is even more pronounced in the 29 SFCCs: all bar one of the 29 is a LIC or a least developed country (LDC), and only three are not fragile states.

This is where alarm bells began to ring: 26/29 of the SFCCS are fragile and/or conflict affected, and the answer is – more aid. To whom? To do what? Isn’t the politics of these places rather important in determining how much of the aid ends up reducing poverty?

I’m a big fan of the ODI so sent this draft post over to Marcus Manuel, who wrote the report. He pushed back with two good points:

‘First I agree I don’t talk about how to spend the money and that is really important. But I think there are wide range of instruments that have been used and could be used. Failing all else, Social Protection could be done entirely by the UN and CSOs – i.e. we treat this as a humanitarian emergency. And you do see external money well spent in FCAS – the Afghanistan reconstruction trust fund, Liberia pooled health fund and Yemen social development fund to name but three. I agree these countries face deep governance problems – but there are hybrid ways of ensuring money for basic services is delivered and benefits the poorest.

Second the aim of my paper was to sound the alarm – yet again – that donors are switching their money away from LICS/LDCS (low income and least developed countries) to countries that serve their national interest. Do critique me but please don’t let the donors off the hook. We can discuss and debate how best to spend the money but without the money we will all be unable to help.’

This is really useful and links nicely back to the work I’m doing on Public Authority at the LSE – trying to understand how power and politics actually works in precisely these countries that ODI labels as ‘SFCCs’. So what I’d be interested in reading is anything on how effective aid spending in FCAS differs from aid spending elsewhere – bypassing the state is one option, as Marcus says, but what else have people seen to work?

And here’s Marcus presenting the report


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.


4 Responses to “Aid’s fragile state problem – why is it so hard to even think about?”
  1. Heather Marquette

    I wrote this paper on the challenges of combining a VFM approach with the increase in FCAS spending back in 2012 with Zoe Scott & Claire Mcloughlin, and I reckon a lot of the analysis still stands.

    I also reckon it’s a lot easier to write a paper, and a blog post, than it is to do something about it. As we say, ‘Given the prevailing climate of the global economic crisis, the age of austerity and the weakening of trust in government that has been demonstrated not only in the Middle East but also in the US and in Europe, explicit donor recognition of the possibility that public money spent on aid may be increasingly at risk would be unlikely to fly under the media radar and avoid the standard hyperbolic response. For this reason, there is unlikely to be a publicly accepted and advertised principle of ‘good enough’ aid effectiveness or value for money in fragile states. In both the international arena and the domestic political context, anything other than a fairly limited presentation of value for money is unlikely to be publicly palatable, and therefore not politically viable, any time soon.’

  2. Much aid has been–and is–implemented in organizational set-ups that are parallel to the State. When implemented in –for example– health facilites,within shouting distance of government health facilities, an automatic disparateness evolves that cannot be undone. It would be better, albeit it would likewise take more patience and time than most foreign governments and donors have, to ab initio work with government to gradually improve their own handling of issues and the speed and manner of their improvement.

  3. Daniel Shimmin

    Thank you for this blog post. From a primarily conflict sensitivity focussed lens we’ve been considering the same question in South Sudan. From our perspective bypassing the state is at best an incomplete option. While use of parallel institutions or delivery by UN, INGOs and others, may help limit fiduciary risks and sustain service delivery in an otherwise highly weak state delivery context, it also creates challenges to long term forging of an inclusive political settlement and – if accompanied by a fiction that we can compeltely bypass the state in delivering aid – will unwittingly play into political economies and conflict dynamics in ways that sustain violent conflict and state fragility.

    In South Sudan, state and society have adapted to an environment where aid is a major aspect of local economies but is largely delivered outside of state institutions, and has been adopting effective approaches to, among other things, co-opt and redirect to align to their own political strategies, and to take rents sometimes to finance violent conflict. Sometimes hamstrung by the unrealistic expectations of donors, aid actors can struggle to position themselves such that they understand, adapt and collectively respond smartly and effectively to such attempts.

    Some recent thinking on this issue is covered in a paper by Lauren Hutton on ‘Aid and Government: Conflict sensitivity and the state in South Sudan’ December 2018 ( Rather than exploring ways to bypass the state, this paper takes as given that international aid is an important part of South Sudan’s political economy and that it interacts with the government in a continuous and evolving manner. It therefore focuses on practical approaches by which aid actors can improve their ability to understand and manage risks of doing harm and be responsive to a continuously evolving government context.