Sustainability in a failed state

Updated: Oct 20, 2021





It is estimated that by 2030, 60 per cent of the world's poor will live in states affected by fragility, conflict and violence.


If we are to seriously address poverty, sustainably, we must take those living amidst protracted crisis into account. The recent collapse of much of the state in Afghanistan, and its concurrent change of regime, has re-focused debates about aid and humanitarianism. It has once again called into question the viability of state structures to govern and protect their populations, especially the most marginalised.


Meanwhile, the Covid-19 pandemic has tested the resolve and capacity of governments all over the world to make financial, ideological and protection choices that many have not faced for decades. This has exposed a range of fragilities at national, transnational and global levels. Every fragile state is fragile in its own way and, regardless of categorisation, every state has its failures. While one may lack authority, another may lack the capacity to supply basic public services And the situation often varies considerably between different parts of the same country.


Over the past six months Agora has worked on MSD programmes in Lebanon, Syria, and Afghanistan, among others. In these three countries the role of the state is either inadequate in ensuring positive developmental outcomes for its people or, in some cases, is actively working against these outcomes.


MSD is all about optimising markets to improve outcomes for disadvantaged people. To do this, we intervene temporarily to change the way systems work so that those with the appropriate capabilities, opportunities and motivations to contribute to improving outcomes can do so in the longer term. These roles, and who plays, them will differ greatly by context, but there are almost always roles for public, private, and civil society actors in every system.


So, if there is always a role for the state in ensuring positive developmental outcomes - but the state doesn’t exist or is actively impeding those outcomes - then how can any intervention focused on sustainability work?

Perhaps the clearest case of this quandary comes from the water sector. In no water sector in the world is there an entirely privately driven model. Governments perform roles ranging from full service provision, which is entirely free at the point of use, to supporting only the infrastructural cost, right down to only ensuring quality standards and providing licences. The water sector is riddled with market failures. Overcoming these failures requires an effective governance structure.


But if the government isn’t present, or effective, and there is no realistic prospect for private or self-governing models, the usual MSD playbook goes out of the window. Does that mean we shouldn’t try? We often talk about the humanitarian-development continuum, but most interpretation of this is simplistic; how do we stop handing out the relevant goods and services and get the private sector to do it instead?


Water service provision requires extraction, infrastructure provision (right from initial filtration or desalination through to households or water kiosks), maintenance of this infrastructure, retail and payment collection, regulation and quality assurance among a host of other functions. There are places where the state does not do any of these things in large regions of the country. In some states where the government has the intent to support service provision - and understands its own relative strengths and weaknesses - it may invite and legitimise effective private sector or NGO responses. However, in some instances state actors and structures actively mitigate against these functions – withholding access to materials for example. In a small cluster of states, protracted violent conflict has resulted from lack of access becoming entrenched over years or decades.


By anyone’s definition, water is seen as a basic need – an end in its own right - and so humanitarian agencies are obliged to do what they can to maximise distribution. That said, funding is fickle and if used inefficiently it will neither meet the scale of the challenge, nor have any lasting impact.

So what are humanitarian agencies to do? Here are four guiding principles derived from our recent work in areas of state failure, most particularly the water sector in Syria.

  1. Analyse the market system regardless of who is doing what. While this is always good practice, where direct delivery is the norm it can be tempting to just treat the service as a single function – in this case providing water. Still, as listed above, there are multiple functions embedded within that provision. Syria provides a very different context to, for example, WASH service delivery in Africa. In many cases, infrastructure had been at some point present and functional. Delivery models, and the skilled personnel to operate them, were too, at some point, present in Syria. So, the context of function (and dysfunction) is specific to this geography.

  2. Conduct a stakeholder analysis. Who is performing each of these functions and why? Again, even if the answer is “we are”, then it provides a framework to assess what might change at some point in the future. Situations might vary according to the exact geography and organisation. In some cases, humanitarian actors are doing it all – buying and maintaining pumps, checking quality, managing service delivery etc. In other areas, some NGOs are trying to utilise existing infrastructure and focusing on employing agents within water user committees. But in many cases, the existence of the first ‘aid does it all’ approach is undermining any efficiency or sustainability gains from the second more locally embedded approach.

  3. Assess the potential for the transferral of functions to improve effectiveness, scale or sustainability of impact. It may be the case that there is no realistic prospect of the state funding infrastructure provision for the foreseeable future. But isolating infrastructure provision, for example, as a function performed by the humanitarian community allows you to examine whether there are more sustainable models incorporating other functions, delivered by other actors. Can local water user committees collect revenues to cover operation and maintenance costs? Can local service providers be used to source plumbing or distribution contracts? Can models be established to disincentivise wastage? Where a functioning system existed prior to conflict or other shocks to the system means that, at minimum, certain customer behaviour norms are in place – even if the service providers or hardware have long since gone. In such situations the potential exists for communities to run their own water service delivery models. Willingness to pay (something) is demonstrable, even if not at full cost recovery prices, which sets the right incentives to reduce wastage.

  4. Develop interventions focused on facilitating sustainable, large-scale impact in the areas where it’s feasible. We know this won’t be system-wide and we know that certain functions will continue to be both performed and paid for by humanitarian actors; the imperative of service provision remains, and without the NGO continuing to pay for water pumps, for example, people won’t get water. Water infrastructure can be considered both a merit good and a public good. It is beneficial to the population to have access to clean water and it is unrealistic for different private providers to utilise different service infrastructure. Infrastructure, therefore, is ripe for public investment which can be fully or partially recouped through temporary licencing models. However, in the absence of the state, investment in this infrastructure seems a valid entry point for coordinated humanitarian investment – unlike, for example, operation and maintenance. Leveraging market forces, humanitarian intervention can seek to encourage the development of models based on incentives and facilitate high quality service delivery. Even where commercial cost recovery is not immediately viable, the intervention can look to establish systems to make it so in the future. For example, by providing vouchers to households, redeemable in different ways exercising choice, or by putting contracts out to tender, rewarding those with reduced cost per user.

The difference between this “failed state” scenario and an ordinary MSD programme is that we’re not even contemplating a ‘no-aid’ future in the short to medium term. Rather, by following these steps there are some important and positive implications for the nature of the intervention’s impact:

  • We’re trying to establish models which might become transferable to local actors when feasible, or, at the very least, that don’t disincentivise others entering the market.

  • The intervention is likely to lead to a more efficient use of resources, meaning the money goes further and more people get the water they need. This can be achieved by encouraging actors to perform the functions which they are able to, using their own financial and human resources, and based on their own incentives.

  • Considering the system in this way is also likely to lead to a more coordinated and efficient use of resources amongst development actors. In humanitarian scenarios coordination is challenging and often leads to duplication of efforts and inefficient use of resources. A systemic analysis allows for better mapping of interventions by different actors - and hopefully better alignment behind a shared goal of a pathway to sustainability.

We have used water systems as an example of an area where there would typically be a role for the state, but where there are clear failures. However, it could easily have been another sector such as food and agriculture, energy, or healthcare. Furthermore, it is equally applicable in less urgent or obvious scenarios.


How often have you heard an NGO say that they’re delivering extension services to farmers because the state is ineffective? Here, we have attempted to go beyond the frequent, casual retort of the MSD community of ‘ just get the private sector to do it’. In many cases, particularly as the boundaries of MSD are pushed towards more fragile contexts, greater nuance is needed.


We wish to shift the tendency of designating fragile states as an excuse for bad development, while recognising the unrealistic nature and prescriptions of some decontextualised MSD approaches. As the humanitarian community considers the implications of change in Afghanistan for its future work there, and globally, we are providing a framework for humanitarian aid in exceptional circumstances, so that sustainable development might (eventually) be possible.


This blog was originally posted on The Beam Exchange on 8th October 2021