Now that you have the background, what would you do if there was a major earthquake in country X tomorrow? Who would decide if an L3 system wide emergency should be declared, and what system would be implemented? How would this change if it was a conflict situation? What about if there was a major epidemic or possibly another pandemic such as Ebola? Let's therefore, look at different coordination models for large humanitarian crises. The cluster approach is one of the most well known coordination systems. Clusters are groups of humanitarian organizations, both UN and non-UN, in each of the main sectors of humanitarian action. They have clear responsibilities for coordination. As mentioned previously, for refugee settings, the United Nations High Commissioner for Refugees has the coordination role, and this system is called the refugee coordination model. Where neither system has been activated despite a major emergency, possibly because the government itself does not wish such a system to be implemented, then players respond according to the leadership of the government and existing coordination mechanisms or sectors are used as agreed with the government. Something relatively new following Ebola are when major health emergencies occur. These are called L-3 Plus. As we mentioned, L3 will be a system wide declaration, but the Plus refers to the major health component. And here, coordination at the global level can be expanded and the World Health Organization will have a privileged place at the country level in terms of technical guidance and strategy development. Where neither system has been activated for a variety of reasons, it may not meet the criteria or perhaps the government itself does not wish such a system to be implemented. Then players will respond, but it may not take the same role as the cluster approach or the refugee coordination model. This depends on the situation and the security, and its often very context specific. I want to come back to the cluster approach and the refugee coordination model in a bit more detail. The cluster approach was applied for the first time following the 2005 earthquake in Pakistan. Nine clusters were established within 24 hours of the earthquake. Since then, two major evaluations of the cluster approach have taken place, as there has been much criticism due to its lack of effectiveness. There was a need to refocus on strategic and operational gaps, assessment, and results. Currently, there are now 11 clusters at the global level. You can see this in the diagram above and I'll also read out the 11 clusters. One is Camp Coordination-camp management or CCCM, two is Early Recovery, three is Education, four, Food Security, five, Health, six, Logistics, seven, Nutrition, eight, Protection, nine, Shelter, ten, Telecommunications and number eleven, the last one, is Water Sanitation and Hygiene, or WASH. One of the core functions of the cluster is to support the delivery of services to the disaster affected people. Another major function is to develop a strategy on how best to address the needs. Each cluster has a so-called lead agency which in most cases is a U.N. agency. I will not now go into further details as it is quite complex, and it would take some time to explain, but in principle, lead agencies have substantial accountability of the delivery of humanitarian services. However, it's been my personal experience with the cluster approach, that there has been too much of an emphasis or focus on the process of coordination with a lack of emphasis on accountability regarding the humanitarian response and the results. Some have blamed it on a lack of funding and without funding, lead organizations, or at least some of them, have argued that they cannot be held sufficiently accountable. However, many people including myself believe that ultimate accountability can be implemented and that organizations that are responsible should be held to account. This remains a challenge particularly because many of the results of our humanitarian response are still reported in terms of process indicators and not outcome or impact indicators. So what do I mean by this? Well it's easier to report how many people have received health services or how many people have received hygiene kits, compared to outcomes of these interventions, such as how many people have not become ill because of the provision of hygiene kits for example. Ultimately, we would like to measure the outcomes and impact. But this is very complex in general and even more difficult in humanitarian settings. Yet, particularly with the aid of technology, outcome and impact indicators will increasingly be able to be measured and reported in the future. Primary responsibility to protect refugees rests with governments. The refugee coordination model is applicable in all refugee settings and throughout the duration of a refugee response. A refugee response does not include only UNHCR operations, but those of all actors involved in the response. A refugee response includes the needs of host communities, which is central to the preservation of asylum space and to the social cohesion necessary for a protective environment for refugees and their hosts. To achieve this, one must establish linkages with national partners and development partners, pursuing durable solutions and soliciting financial resources for refugees. As with any coordination system, coordination is only a means to an end to facilitate an effective and timely response. So if there is a major refugee crisis tomorrow, what would you do? First and foremost, what is the government's capacity to deal with the situation? In many situations, governments can handle a small refugee influx and UNHCR and its partners' main role will be to support the government in this response and perhaps to concentrate on key protection issues. In other settings, particularly with the large refugee influx, UNHCR and its partners may have a much more significant response and support to the government in many areas of assistance. Furthermore, issues of protection always remain paramount.