Duncan Green sets out how the International Monetary Fund can improve its engagement with fragile states.

This article is part of the #PublicAuthority blog series, part of the ESRC-funded Centre for Public Authority and International Development.

I took part in a really interesting discussion about the role of the IMF in fragile states in October 2017. Chatham House rules, so no names, no institutions. The Fund works in fragile states in three main ways – it lends money to governments, it trains officials and it tracks and reports on government economic performance (‘surveillance’). Although its lending is often not big compared to other aid flows, the IMF has disproportionate influence as a gatekeeper, acting as a kind of ratings agency in the eyes of other donors and foreign investors.

Twenty or thirty years ago, the Fund was the pantomime villain, a hit man imposing austerity through structural adjustment policies on governments hit by economic crisis. The 1998 photo of Fund boss Michel Camdessus standing, arms crossed, behind Indonesia’s president as he signed a painful bailout agreement came to symbolise its neocolonialist tendencies.

Times have changed, it seems. In Washington it is as likely to be talking about inequality or gender as balancing the books by cutting spending or even (gasp) raising taxes. In the Greek bailout, it was often the voice of moderation against the would-be grim reapers of Germany and the EC. It can be a source of stability, a mentor to governments emerging from chaos and conflict and desperately in need of help to get the machinery of the state working again. It often gets a better press these days.

But how well does its standard repertoire work in fragile states? In theory (and in its staff guidance notes) the Fund recognises that such places are messier, and require greater attention to understanding the political and social context. But the practice, according to many in the room much more familiar with the Fund than I am, is often pretty close to business as usual.

Which got me thinking. The goal of advocates and reformers is often to get a management statement, a new policy or a guidance note that captures their concerns and promises to do things differently. But that often has very little impact on what the institution actually does on the ground – where it is all about culture, rules of thumb and ‘mental models’. My impression is that the Fund’s mental model of a fragile state is of a normal state, only weaker, therefore requiring talking to the same people in the same ministries, advocating broadly similar policies and goals, but with more training programmes and maybe more patience about hitting targets.

But fragile states are often not like that – they can be qualitatively distinct from stable ones. Power (what my new LSE research outfit CPAID calls ‘public authority’) is more widely dispersed within society –  traditional authorities, diasporas, faith organisations, ‘non-state actors’ like armed groups may play central roles. In Myanmar I came across armed rebel groups running hydro dams that sold electricity to the government-run cities, and the government paid its bills. In DRC, I saw traditional leaders and state officials jointly collecting taxes (and it wasn’t that clear who was in charge).

Fragile states themselves are often unwieldy coalitions of those who want to use the state to improve the lives of citizens, those who want to grab as much as possible, and those who want to do a bit of both.

And the dynamics are chaotic – crises and critical junctures abound. There is little point in waiting for things to settle down – they won’t. Your three-year plan will be derailed within months.

Then there is violence: I’ve been struck in recent weeks by the way the aid business rejects and quarantines the whole subject. It doesn’t usually fit well with the mental model of homo economicus/ rational expectations that underpins so many aid debates; it’s not amenable to three-year project cycles and technocratic solutions; it ‘belongs’ to a different silo and different specialists (conflict resolution; peacekeeping; counterterrorism), with their own jargon and most of the time the rest of us just wish people weren’t disposed to hurt each other.

Any intervention into such a complex, unstable system is bound to have important consequences, strengthening the hands of some individuals, ministries or arguments and weakening others. Strengthening governments, whether benevolent or malign, relative to other players. Potentially stabilising or destabilising fragile political settlements.

So what should the Fund do? Should it try harder to understand the system in order to do the standards things better (infrastructure, macroeconomic stability, tax collection, encourage investment)? Or should it do different things (invest more in its role as catalyst, bringing together different parties to boost investment, maybe even get into peace-building).

My feeling was closer to the first – the Fund has a valuable role to play in helping governments do the basics, and if it pays more attention to the system, it should be able to play that role better. For that it needs to hire a wider range of expertise – political science and anthropology as well as economists.

One basic point is that it needs to be there – IMF staff are not allowed to even visit the most fragile states, let alone be based there. Safety concerns are entirely warranted – a senior official was killed in a suicide bomb attack on a Kabul restaurant in 2014 – but other agencies continue to function in such places, and you really can’t try and ‘dance with the system’ from a thousand miles away.

Read more about #PublicAuthority and visit our website.

This article was first published on the From Poverty to Power blog.

Duncan Green (@fp2p) is strategic adviser for Oxfam GB and author of ‘From Poverty to Power’ and a Professor in Practice at LSE.

The views expressed in this post are those of the authors and in no way reflect those of the Africa at LSE blog or the London School of Economics and Political Science.