Humanitarian coordination team meet in Port-au-Prince to assess damage from tropical storm Emily Humanitarian coordination team meet in Port-au-Prince to assess damage from tropical storm Emily Photo credit: Leonard Doyle / IOM
Strength in numbers: catching up on a decade of NGO coordination
by Paul Currion and Kerren Hedlund February 2011

In the wake of the Haiti earthquake and Pakistan floods in 2010, the need for effective NGO coordination is clearer than ever. As part of a previously planned initiative, the International Council of Voluntary Agencies (ICVA) has carried out a review of NGO coordination in a range of humanitarian responses over the last ten years to contribute to the ongoing process of improving overall coordination. ICVA will shortly publish a series of case studies and develop basic guidelines as resources to help NGOs establish better coordination in the field.

This article presents an overview of some of the general findings and critical issues raised during the research. Based on the long-standing assumption that coordinating NGOs is like herding cats, we anticipated a struggle to find good examples of NGOs working together, but instead we found more than we could include. Our case studies cover Afghanistan, Haiti, Iraq, Kosovo, Myanmar, the Occupied Palestinian Territories, Pakistan and North and South Sudan, but there were many other potential examples from elsewhere, including Guatemala, Liberia and Somalia.

NGO coordination appears to be the norm rather than the exception. At the very least, NGO staff responding to emergencies maintain bilateral or multilateral relations to facilitate their work. This type of ‘coffee shop coordination’ is usually overlooked in most coordination studies – understandably, since it is almost impossible to measure – but it is nevertheless an essential element of a healthy coordination ecosystem.

Some documentation exists regarding more visible NGO coordination based around formally constituted coordination bodies. However, few of these documents identify lessons which might be replicated in other countries and, even if they did, there are no mechanisms for sharing those lessons. As a result, valuable lessons have been lost and NGO coordination has progressed little over the last ten years. This is a problem for the entire humanitarian sector; NGO coordination is a critical component of a well-functioning humanitarian system, providing a civil society counterpart (and sometimes counterweight) to coordination initiatives established by governments and UN agencies.

The post-cluster world

The introduction of the cluster approach following the 2005 UN Humanitarian Response Review marked a turning-point in the development of humanitarian coordination. However, the NGOs and Humanitarian Reform Project has found that humanitarian reform is still not working as intended, primarily due to lack of leadership at the field level. NGOs are unlikely to provide that sort of leadership given constraints on their resources (in particular staff time), but where UN agencies do not meet their cluster lead obligations, NGOs have convened their own sectoral meetings. One example came from South Sudan, where NGO leadership in the WASH sector contributed to the 2010 decision to include NGOs as official cluster leads and co-leads.

In general, however, most NGO staff believe that this is unsatisfactory, and most of our respondents expressed support for the cluster system – with two caveats. First, many NGOs would like to be more involved in improving the cluster system at field level, with some of the larger NGOs clearly prepared to take on leadership roles. This reflected a concern that active NGO involvement in sectors has been and should continue to be critical. In Afghanistan in the 1980s, sector working groups played critical roles in establishing basic standards, but as national government and UN institutions became stronger, these groups tended to be folded into centralised coordination structures, which seems to be a common pattern in many places.

Second, the cluster system is not mandated to deal with broader humanitarian issues: practical questions such as security procedures or NGO-related legislation, and questions of principle such as civil–military relations and humanitarian access. Such issues tend not to be addressed by the cluster system (or other institutional coordination mechanisms), and NGO coordination bodies usually aim to create space to discuss and resolve common positions on these issues.

The great weakness of NGO coordination is that it tends to be reactive, and NGO coordination bodies themselves are normally set up in response to one of three conditions:

  1. Resisting attempts to impose coordination by the UN or governments, although NGOs are usually committed to maintaining or improving links with such mechanisms.
  2. Filling gaps in existing coordination structures, for example in Iraq following the withdrawal of UN agencies after the Canal Hotel bombing in August 2003.
  3. Addressing NGO concerns that are not being dealt with by other actors, particularly where those concerns involve humanitarian principles, such as in Kosovo in 1999 or the Occupied Palestinian Territories since 2000.

The stated reasons for forming NGO coordination bodies are similar in every country, namely the recognition that collective action can achieve common goals more effectively than individual initiatives. The perception that coordinating NGOs is like herding cats appears to be based on a misunderstanding of what NGOs are trying to achieve through coordination.

The shape of coordination

While governmental and inter-governmental organisations attempt to impose top-down coordination structures (at least partly in an attempt to impose order on frequently chaotic situations), NGO coordination is usually built from the bottom up. As a result, externally imposed NGO coordination mechanisms fail to take root, while those generated and supported by NGOs themselves tend to show great resilience. This finding should not alarm the UN, governments or any other institutions responsible for coordination; while the UN’s humanitarian reform process has produced mixed results, the cases we looked at showed that overall humanitarian coordination tends to be seen as more successful (if more difficult) by all parties in those places where NGO coordination is more successful.

Ensuring coordination within the NGO community from outside that community is largely a matter of generating the interest, allowing the space and providing the tools to help NGOs form their own coordination bodies. Some of the most constructive experiences have come from simply providing additional human resources to existing coordination bodies, in the form of (for example) an Information Officer in post-conflict Kosovo or a Liaison Officer in post-cyclone Myanmar.

Despite the greater flexibility of NGOs compared to government bodies, all of our case studies uncovered a similar type of structure. This has enabled us to develop a basic model for NGO coordination bodies, which ICVA will pursue in the next stage of its project. While the recurrence of a particular model of NGO coordination shows its resilience, it is worth remembering that the cluster approach has demonstrated the weakness of a one-size-fits-all approach. We must recognise that coordination structures vary depending on local conditions, from crisis to crisis, country to country, and over time. The appropriate form of NGO coordination in the response to an ongoing conflict should – in theory – look very different to the appropriate form in response to a sudden-onset natural disaster.

NGO coordination does not always require the formation of an NGO coordination body. Before establishing a formal body, it is therefore worth pausing to consider whether there are other ways of achieving the same coordination goals; when managing such a body, it is worth continually asking how it needs to develop in order to stay responsive to changing conditions.

Coordination adding value

The persistence of NGO-led coordination bodies suggests that they have value for their members, and that NGOs are happy to participate in coordination mechanisms when the incentives are right. This does not necessarily mean financial incentives – institutional donors do not have to make participation in coordination a prerequisite for receiving funds, for example. Donors clearly have a role, but their influence should not be overstated; in the 2005 Indian Ocean tsunami and the Haiti earthquake, much of the relief funding was channelled through NGOs. Some observers suggested that this may have had the effect of removing one of the primary incentives for NGOs to coordinate, partly explaining the poor coordination overall, although this has not been confirmed by empirical study.

Our case studies suggest that lack of participation in coordination mechanisms (such as cluster meetings) does not necessarily indicate a reluctance to coordinate. Of the many barriers to effective coordination – different mandates, different sectoral interests, different target groups, different operating principles and so forth – the most basic obstacle is a lack of resources. Coordination has a cost – whether in time, money or other resources – and NGOs are willing to pay that cost only if they see the benefits, particularly so in the case of national NGOs.

We must bear in mind that in some cases the costs of coordination may not be worth the benefits, either to NGOs or to affected communities. A cost–benefit analysis of NGO coordination has never been attempted, but should be core to future evaluations of coordination. It is also notable that none of the coordination bodies covered has established clear measures for their success, something which also requires more attention in evaluations.

No matter what the details of a particular coordination mechanism are, the best way to ensure participation is to ensure that it delivers the benefits that stakeholders require and expect. Many coordination initiatives have adopted a clear service orientation, explicitly so in the case of initiatives such as the PakSafe security office currently being established in Pakistan (modelled on the successful Afghanistan NGO Safety Office). Every single case study in this review confirmed that the critical factor in keeping organisations engaged in coordination processes is surprisingly simple but seldom realised: maximising the benefits of coordination (particularly in terms of providing services), while minimising the costs of participation.


Our review of NGO coordination raises two critical policy questions. First, what do NGOs want from coordination? Given the weaknesses in the cluster system, for example, will becoming more active in taking the lead in sectoral coordination help us to achieve the impact we seek on the ground? Second, how can we use NGO coordination bodies to promote the broader changes we want to see in the sector around familiar issues such as humanitarian principles and new challenges such as urban disasters? Individual NGOs participating in coordination bodies are unlikely to be able to address either of these questions without guidance at the global level.

One particular issue that cut across all the case studies concerned partnership with national NGOs. Some international NGOs have a closer relationship with their national counterparts than other actors, and can act as entry points for national NGOs, yet coordination mechanisms instigated by international NGOs tend to focus on their own concerns. Different NGO coordination bodies have dealt with this question in different ways, but there are notably few models for engaging national NGOs in the context of coordination bodies. This has been highlighted in Haiti, where NGO coordination activities are struggling to include more than a handful of the huge number of local civil society organisations. This issue requires urgent attention.

We end on a note of caution, since the success of a particular model of NGO coordination in the past does not guarantee its success in the future. While we believe that NGO coordination is effective, it is effective in achieving a specific set of objectives, and must be part of a broader system of coordination if it is to be truly successful. NGOs in general have failed to address the growing complexity of the humanitarian endeavour, the increasing variety of humanitarian actors, the blurring of humanitarian principles and the incidence of new types of disaster, particularly related to climate change and urban risk. These are exactly the sort of challenges that require collective action by NGOs if they are to be addressed successfully; establishing and participating in field-based coordination mechanisms is now a requirement for NGOs, rather than an optional extra.


Paul Currion and Kerren Hedlund are consultants working on ICVA’s ‘Mapping and Learning from NGO Coordination Bodies’ project. This article was written on a personal basis, and does not reflect the official views of the ICVA Secretariat or ICA’s membership. The findings of the work are available at Any enquiries should be sent to Manisha Thomas at ICVA (