Executive summary

On May 2 and 3, 2008, Myanmar was hit by a cyclone of unprecedented force, which had devastating effects on the Irrawaddy Delta and, to a lesser extent, Yangon. It is believed that 2.4 million people were severely affected by the cyclone. Given the scale of humanitarian needs created by Cyclone Nargis, the cluster approach was rolled-out rapidly in the first few days to ensure a coordinated response from the international community. Eleven clusters were activated by the end of June 2009, when they merged into a new coordination mechanism, namely Delta Recovery Groups.

The response to Cyclone Nargis was selected as one of the six country studies in the framework of the cluster approach phase II evaluation, The evaluation mission was conducted in September 2009 and met a wide range of actors involved in the humanitarian response (UN agencies, international and local NGOs, the Red Cross and Red Crescent Movement, the Humanitarian Coordinator, donors and representatives of the Myanmar government).

This report presents the results of the Myanmar country study. It analyzes the effects of activating the cluster approach following Cyclone Nargis and also takes into account the political nature of humanitarian aid in Myanmar. The report seeks to inform the global level and to distill lessons learned for other regions where the cluster approach is likely to be activated in the future.

First, it must be noted that the humanitarian response in Myanmar cannot be attributed alone to the international aid actors. Local efforts played an impressive role. In this context, the cluster approach in Myanmar proved to be a relevant and effective mechanism to coordinate the international response to cyclone Nargis, but to a lesser extent to coordinate with the national and local response. Despite an initial implementation run in an isolationist manner vis-à-vis the local actors, the cluster approach managed after one year both to better involve the Government of the Union of Myanmar (GoUM) and local actors in the response and strengthen the capacity to respond to further disasters at national and regional levels. It also helped to increase coverage by avoiding duplications and identifying gaps.

Nevertheless, the performance of different clusters varied greatly. The level and quality of global cluster support was uneven. Most lead organizations fulfilled their cluster leadership responsibilities though there were marked differences in their levels of commitment and effectiveness. Collaboration between cluster members was good but did not eliminate all uncoordinated approaches at village level. The cluster approach enhanced accountability among cluster members but had only a marginal effect on accountability towards the HC, and did little to nothing to encourage or standardize basic accountability practices at community level.

More globally, the cluster approach in Myanmar had a positive impact on relations between aid actors and government authorities in country and helped to demystify protection issues within the government. On the other hand, the cluster approach focused almost all resources and activities on the Delta, with the risk of neglecting long-lasting humanitarian needs in the rest of the country, where clusters were not activated. Finally, the Myanmar case shows that the "silo" frame of the cluster approach goes against an integrated approach, which is key in a recovery process.

The study identified a range of factors that contributed either positively or negatively to the functioning of the cluster approach in the response to the cyclone:

- Factors which strengthened the response: clear designation of leadership, existence of non-UN co-leads and other proactive NGO support to cluster leadership, role of clusters and cluster leads when dealing with the GoUM, backing of UN OCHA at field level and the strengthening of common planning mechanisms.

- Factors which impeded the work of clusters: high turn-over of staff, lack of training in special facilitation and coordination techniques, lack of clarity of UN agencies' roles with regards to food aid, food security, agriculture and early recovery resulting confusion with leadership, lack of clearly defined roles and reporting mechanisms between national (Yangon) and field clusters (in the Delta), lack of clear objectives in inter-cluster coordination limiting the ability of clusters to make cross-sectoral comparisons and determine priorities, vagueness of the concept of provider or advisor of last resort and lack of funding.