Skip to content
Early recovery from disaster: the Pakistan earthquake

The earthquake killed at least 73,338 including 18,000 schoolchildren. Over 128,000 people were injured and 3.3 million displaced. Over 600,000 houses, 6,400 km of road network, 6,298 education facilities, 350 health facilities, 3,994 water supply systems and 949 government buildings were destroyed. The size of the IDP population and the number of houses destroyed were significantly greater than the Asian tsunami despite the death-toll having been lower.

The Pakistani military launched a massive response, supported by US, British, NATO and Australian military forces, amongst others. Coordination was an enormous challenge, as it always is for a host government in the aftermath of a disaster. This was particularly the case for a military lacking experience in working with NGOs and unfamiliar with the humanitarian principles they defend. It was thus necessary in Pakistan to use a model of ‘non-interfering coordination’ in which the military shared an open and honest assessment of needs with the humanitarian community and allowed NGOs to choose what operations they would undertake and where. In this model, gaps in humanitarian delivery are ‘back-filled’ by the army and government agencies.

As Pakistan had no designated national disaster agency at the time of the earthquake, the Federal Relief Commission (FRC), the ad hoc structure created to deal with the aftermath, decided to structure itself using the recently created cluster approach. This resulted in a series of key personal contacts between the national and international coordinators. Whilst there have been critics of the cluster approach in Pakistan, the facts speak for themselves:

  • A million tents, six million blankets and 400,000 emergency shelters were provided.
  • 350,000 IDPs were housed over winter, with 95% returning in the first year after the relief.
  • There was no second wave of deaths: medical statistics showed an improved rate of cold-related infections compared with normal years.
  • All schools and hospitals were restored to functionality.


Problems were overcome cooperatively as the cluster approach enabled a structure for overall engagement between national and international actors, humanitarians and the military.

Moving into recovery

Even whilst relief was continuing, early recovery planning had to begin. One of the first tasks for the Early Recovery Cluster was to convince decision makers, including the FRC’s successor body, the Earthquake Reconstruction and Rehabilitation Authority (ERRA[1]), that the matter was not quite that simple. Whilst the experience of most countries in post-disaster recovery shows a severe drop in momentum following the immediate relief period, Pakistan had to be convinced that planning was required so that no such loss of momentum would occur. The Heads of Clusters had to ensure that all interlocutors understood the Early Recovery concept whilst the Early Recovery Cluster ensured that all clusters did the detailed transitional planning. In essence, the Heads of Clusters also became a ‘network’ of early recovery planners for each of the clusters, as well as performing the advocacy tasks regarding early recovery on behalf of all clusters.

Guiding principles for recovery

The Early Recovery Framework is guided by a set of 10 rights-based principles to be applied during planning and implementation of early recovery interventions:

1. focus on the most vulnerable

2. restore capacities

3. rebuild people’s livelihoods

4. secure human development gains

5. reduce disaster risk

6. engage the private sector

7. independence and self-sufficiency

8. transparency and accountability

9. subsidiarity and decentralization

10. coordination


Once the government had bought into the concept of early recovery it instructed provincial and state governments to work with the clusters and coordinate proposals for the Early Recovery Plan (ERP)[2]. Provincial and state clusters fed draft plans into national-level clusters (that critically included donors and were chaired by UN and government representatives). The national-level clusters then drafted sectoral plans to be reviewed by the government and the Heads of Cluster forum to ensure suitability, coherence and ensure that cross-cutting issues such as gender issues were incorporated.

As clusters had never been tried before, no one knew what to do with them after a relief operation. A discussion paper on ‘Transitioning Clusters’ was circulated and went through 17 major revisions and significant changes in direction before it was finalised. The clusters were closed in Islamabad once the relief phase was declared ended and early recovery started but OCHA staff kept clusters operational at field level. Later, recognising this as an error, the Islamabad clusters were re-opened as ‘working groups’, effectively the ‘new’ clusters, a role they continue to this day to support ongoing reconstruction.

The final cluster transition paper probably left more divisions within the UN system than common decisions. This is not anyone’s fault but is demonstrative of how confusing it was to move out of relief when the structures used for relief were implemented in an ad hoc manner.

Funding success

Donors supported the ERP because:

  • Despite the passage of several months, the earthquake was still fresh in donors’ minds and HQs were still receptive to more funding.
  • Donors were treated as real and genuine partners: it is important not to be afraid to admit to mistakes/difficulties or ask them to be part of the decision-making process.
  • The cluster structure allowed nobody an escape clause or reason to argue that funding of early recovery should be left to somebody else.


The cluster approach, although sometimes difficult and hard to fathom, did improve both relief and early recovery. The timing of the earthquake meant that the clusters were tried in an ad hoc experiment, were ‘transitioned’ into recovery and are now coming to an end, also in an experimental way. Early recovery is harder to understand, harder to plan and harder to fund than relief. But whilst relief may keep people alive, it is early recovery that gives back their livelihoods and builds their futures. It is important that adequate thought and planning go into early recovery as soon as possible after a disaster – and preferably during disaster preparedness and risk reduction stages as well.


Andrew MacLeod ( was Chief of Operations for the UN Coordination Centre set up in response to the 2005 Pakistani earthquake. OCHA’s real-time evaluation of the cluster approach to the disaster is at:


This site is registered on as a development site. Switch to a production site key to remove this banner.