disaster management

Coordination, collaboration and the Cluster approach

We’ve recently been having a number of conversations about the role of the Cluster approach as a mechanism for as a collaborative decision-making in disaster response. Two aid workers with experience in Cluster coordination, Gregg MacDonald and Dave Hodgkin, had some comments.

__________________________________________________________________________________________________________

Kim Williamson: What is the purpose of the Cluster system in disaster response?

Gregg McDonald: One of the main goals of a Cluster is to have some sort of united approach across the disaster recovery. This helps to achieve equity across the response and provide technical assistance, information and direction to agencies that may not have enough resources. This is especially relevant for some of the smaller agencies with less overall capacity but some resources. Inequality across a disaster recovery response can be problematic in terms of the beneficiaries. For example, if you have many different agencies doing different designs, you could end up with many designs. The problem with that is that a few of them will be quite good but some may not be so good. The people getting the lower quality designs look at the other better designs and say, I want that. This was one of the problems with the aid response in post-tsunami Aceh. That’s one of the reasons I like the core shelter concept [giving everybody an adequate core house that is designed to be added to], because everybody gets essentially the same, and they can add to it depending on their own economic status and it can facilitate a return to the pre disaster community.

The bigger agencies are generally the ones with capacity and access to technical advisors and other experts. The first thing you have to do is get them involved in the Cluster process. It’s important to get them participating since they are the ones who will go and do large programs and have a significant impact on the disaster response. They might help out with the technical working groups (TWIGs and SAGs). While there will always be some voices of disagreement in a Cluster, many of the smaller agencies will be grateful for a good solution or design because they don’t have the capacity to do it themselves.

Kim Williamson: In a large-scale disaster response, there will be a multitude of agencies with as many ideas about how they could help. As a coordinator, what do you do if you have one organisation that wants to build domes, and another wants to build boxes, and a group of others want to build bamboo structures?

Dave Hodgkin: You try get as many of them as possible to come to a shared and common view, working together, and together in conjunction with the government and the communities. You’re not trying to make the decision for them. What tends to happen is that people coalesce around a common theme or idea, and work together, and then outsiders who come out with an idea that doesn’t fit with that coalesced vision go off on their own.

Gregg McDonald: Fundamentally, you’re trying to get people moving down the same path, in agreement. There will always be people out there doing their own thing.

Kim Williamson: When people talk about achieving global sustainability, they often talk about processes of social change. It almost seems like there’s a necessary process of change going on within a Cluster to try to get a group of people to achieve shared goals.

Dave Hodgkin: The thing that leaps to mind when I hear those words is that in a disaster our job isn’t to change the way people think. But lot of people come in thinking it is.

Kim Williamson: So as a Cluster, you might actually be trying to change the path of organisations to come to a point of agreement to not change communities’ own path of recovery?

Dave Hodgkin: Well it depends on the Cluster. One of the things that is great about the Cluster system is that it’s actually not a dictatorial system, it’s a participatory system. So people can choose to participate or not. And do choose to not to. It means you get diversity, and diversity by nature will mean that some of it is not so good and some of it is brilliant. What you’re trying to do is raise the bar of the average.

In fact what you tend to find in Clusters is that you have two or three actors who know what they’re doing and will volunteer to come in to put information into the Cluster, and support the Cluster process, to try to raise the bar of everyone else. And then you’ll have half a dozen actors who think they know what they’re doing, and be difficult to control, who will go a little bit against the flow and be argumentative, and then you’ll have another hundred actors who haven’t got a clue what they’re doing and are thankful that there is somebody there who does. They can be the best people in the Cluster, because they’ll come along, take part and be supportive, and then go away.

Gregg McDonald: It’s important to try to make the Cluster approach a collective process. It is critical for the agencies to have ownership of the process. It’s about including them all. Sometimes it doesn’t mean the best technical solution. Sometimes it doesn’t mean the most culturally appropriate solution. That’s what it should be, but if the quorum doesn’t have the capacity to do that, you do the best with what you’ve got. That’s the best you can do.

It’s not perfect, but it’s on the right track. Just try to make sure that it’s not something that is so out of left field that it’s just silly. For example, you wouldn’t go into downtown Melbourne after an earthquake putting up bamboo structures. Maybe tents would be more appropriate.

So the role of the coordinator is to support a collaborative process. You need someone who has technical knowledge, but they’re not necessarily doing technical work. The key role of the coordinator is ensuring that all the actors can talk, and have a sense of ownership with the process. If they don’t have ownership, then you don’t have anything.

Dave Hodgkin: We do now have a lot of donors in their requests for reporting, agencies are asking “how did your solution fit in with the general solutions of other agencies, or how did it fit in with the government (and with the Sultan, in the case of Jogyakarta). Ideally, with the Cluster approach, we come up with a solution, and that solution gets supported or verified by the government; then from a donor’s point of view, the government agrees, all the other agencies agree, we could fund that, that has success written all over it. It tends to narrow things down.

Kim Williamson: What can reduce the effectiveness of a Cluster? What can make a Cluster fail?

Gregg McDonald: Emergency response is essentially the same as your standard everyday life, it is a human/relationship process, therefore personalities affect greatly the effectiveness. So will poor coordination, lack of ownership from agencies which breeds lack of interest. Or a coordinator who thinks they know everything and feels the need to control the process. Constant changing of people in key positions within the coordination team, losing institutional knowledge - by that I mean knowledge about the current emergency and its context.

One of the major challenges for a Cluster is to prolong its value. At the start, agencies are interested in the Cluster because it is a source of valuable information. As time goes on, you need to keep adding value to the Cluster. But you can only give what you get which relates to the ownership / participation aspects. Agencies are giving you their information: it’s your job to collate it and disseminate that information.

But I would say that the biggest challenge for every Cluster is Cluster independence. The coordinator of a Cluster is usually employed by a major aid agency; if you cannot separate yourself from your agency’s mandate, you cannot coordinate effectively because coordination is not about the goals of the agency, but about taking into account everyone’s opinion, big or small.

This post has been reviewed by Gregg MacDonald and Dave Hodgkin and is published with their permission.

© Kim Williamson 2008