Lessons learnt from the Rwanda emergency

EC/1995/SC.2/CRP.21/Rev.1

I. INTRODUCTION

1. At the 16 January 1995 meeting of the Sub-Committee on Administrative and Financial Matters, a proposal was formally endorsed that UNHCR should, inter alia, conduct an internal "lessons learnt" exercise on the Rwanda emergency operation as well as a study by the Central Evaluation Section of recurring lessons from recent emergencies which would take into account the findings of the Rwanda exercise. The Sub-Committee would be kept informed of developments at its June 1995 meeting.

2. Both the internal lessons learnt exercise and the study of recurring lessons have now been undertaken under the aegis of the Inspection and Evaluation Service. The lessons learnt exercise has involved extensive consultations, both at Headquarters and in the field, with UNHCR staff who were involved in various capacities with the Rwanda emergency. A two- day regional workshop was held in Nairobi with field staff directly involved in the operation. More than 40 staff members, ranging from the most senior to junior field level, have been interviewed or consulted individually or in groups and given the opportunity to express their views orally or in writing. The principal conclusions on which there is general agreement are summarized in the second part of this paper and are now being reviewed and assessed by the substantive units concerned. They will also form part of the documentation for a comprehensive internal workshop on the emergency which it is hoped can be held once the situation in the field has stabilized to the point where key participants can be spared to attend.

3. The study on recurring lessons from recent emergencies has been carried out by a former UNHCR staff member with extensive emergency management experience. It has involved a close examination of earlier evaluation studies of emergency situations such as the Eastern Sudan (1985), the Persian Gulf (1990-1991) and the former Yugoslavia (1991-1992). This study has confirmed the key areas of concern in major emergencies and has helped to focus the analysis of the main lessons learnt from the Rwanda emergency. A summary of these follows.

II. LESSONS LEARNT

A. Early Warning/Emergency Preparedness/Contingency Planning

4. UNHCR needs to develop further its capacity to undertake discreet political analyses to warn of imminent refugee emergencies and examine possibilities for preventive action. UNHCR staff should be ready to strengthen liaison with the early warning systems of other United Nations agencies, with organizations such as the OAU, and with non-governmental organizations (NGOs). Every effort should be made to alert the international community to the need for political preventive action where this appears the only way to avoid a humanitarian disaster.

5. Sound contingency planning, centred in the field rather than at Headquarters, is an essential prerequisite of operational success. Ideally, the implementation team should be involved in the preparation of the contingency plan so that, in the words of an experienced Senior Emergency Officer, "they own their own plan".

6. Contingency planning could be greatly facilitated by the active participation and cooperation of potential first asylum countries. A major factor in eliciting such support would be the willingness of donors to fund necessary contingency measures, such as road building and infrastructural work on planned refugee reception sites.

7. A corollary to the above is that emergency preparedness and contingency planning, especially in technical sectors, is of minimal use unless preparedness action is actually embarked upon. Preparedness and contingency planning must involve some spending "upfront". Funding for such expenditures can result in significant savings later on.

8. Although UNHCR's emergency preparedness and response capacity has been greatly enhanced in recent years, it was severely strained by the rapid succession of emergency situations in the United Republic of Tanzania and Zaire in October 1993 and April and July 1994, respectively. Earlier pre-positioning in central or regional stockpiles of emergency relief items would have been warranted by the probable need. Donors should be requested to offer the requisite financial support to allow requirements to be determined through technically expert field-based assessments and permit the pre-positioning of supplies accordingly.

9. As an integral element of contingency planning, further steps should be taken to identify major NGOs, preferably local, if available, as potential operational partners in key sectors and to incorporate them in the planning process. The possibility of such NGOs assuming a coordinating or "umbrella" role vis-à-vis smaller NGOs should also be investigated and discussed with the NGOs concerned. NGOs selected as operational partners of UNHCR in the emergency should be prepared to make a minimum time commitment of six months duration and to participate in building the capacity of local NGOs. UNHCR should make its financial commitments to and expectations of selected NGOs explicit from the outset.

B. Protection

10. The profile of Rwandese arriving in the United Republic of Tanzania in April 1994 and Zaire in July 1994 was unique and reflected the genocide and conflict that preceded the exodus. This was not a typical refugee flight, but for the most part an orchestrated and organized mass population movement executed under coherent military and political control. From the nature of this movement, the following conclusions can be drawn:

(i) despite all the problems of identification and security involved, UNHCR must continue to encourage efforts by host Governments and the international community to ensure, under Article 1 (F) of the 1951 Convention, that persons whom there are serious grounds for considering as perpetrators of atrocities should be removed from refugee camps, excluded from refugee status and deprived of international protection and assistance. The international community should provide the necessary support and funds to assist host Governments at their request in removing criminal elements from refugee camps and in disarming armed militias;

(ii) the international community should also be prepared to provide the necessary support and funds to assist host Governments in moving the camps themselves away from border areas. Large, unmanageable camps should be broken up into smaller units of manageable size;

(iii) large-scale voluntary repatriation can become a feasible proposition only when the capability exists to ensure the safety of returnees in their country of origin and when action has been taken under (i) and (ii) above.

11. The composition of Emergency Response Teams (ERTs) should include a fully integrated protection element from the outset, and sufficient protection staff should be deployed for the duration of the emergency. The shortage of senior protection personnel in Goma between July and September 1994 has been characterized as the most serious deficiency in staff deployment, with consequent adverse consequences on the definition and implementation of protection strategy. The provision of "roving" protection officers covering a wide area was not an adequate solution. In Ngara, the assignment for a lengthy period of only a single protection officer was insufficient to deal with protection policy issues and to support other field personnel in their practical protection tasks.

12. Specific protection measures have been needed on behalf of Rwandese women and children in refugee camps, many of whom suffered physical abuse before leaving Rwanda and continue to be liable to intimidation, rape, forced prostitution, compulsory labour and other forms of abuse. A community-based approach, involving close cooperation between protection, medical and community services staff, both of UNHCR and NGOs, and a central role for the community itself, needs to be vigorously continued, along with the development of women's groups to facilitate the flow of information between refugees and humanitarian workers as well as to foster community solidarity and support. The development of a regional information network is essential, and of critical importance for the care of unaccompanied minors. The establishment in Kigali of a Regional Support Unit for Refugee Children has already proved of great value in this respect.

13. At the earliest opportunity, UNHCR should provide training in relevant aspects of refugee law for immigration, police and military officers of the host country and for the staff of involved NGOs.

14. Efforts should continue to ensure the integration of protection activities within the overall UNHCR response to emergencies.

C. Authority, operational management and coordination

15. The creation of a Special Unit (SURB), within the Regional Bureau for Africa, headed by a Coordinator to deal with the Rwanda and Burundi emergency was fully justified; the complex situation and exceptionally heavy additional workload could not have been adequately handled under traditional "desk" arrangements. The Unit has performed well under difficult circumstances. Nevertheless, there should have been greater clarity from the outset in defining the Unit's terms of reference and in establishing lines of authority, responsibility and the operational chain of command. Special units of this kind must operate under clear policy direction, with clear authority and provide a link for rapid communication between sections and services at Headquarters, and between Headquarters and the field. They also serve as an important focus for liaison and information sharing with donors and other Governments.

16. Where an emergency has regional implications, the emphasis, as had been recognized by the High Commissioner in responding to earlier situations, should be placed from the outset on a regional approach. The assignment of a Special Envoy can help considerably in this regard but, once emergency operations are under way, such appointees must fulfil both a political and an operational role and not act merely as a focal point. The task of Special Envoy must be clearly defined, particularly in terms of authority vis-à-vis the High Commissioner's Representatives in the field, and the Special Envoy must be given adequate means and resources to perform his/her role.

17. Special operational arrangements for emergency response (SURB, the Special Envoy, the Emergency Task Force) should not become institutionalized but should be phased out in favour of normal organizational structures as soon as the emergency crisis is over.

18. The early deployment of Emergency Response Teams to the field has been a crucial factor in improving UNHCR's emergency management capacity. The composition of such a team should be flexible according to need and adjusted in the light of the existing human resources available at Branch or Sub-Office level. In general, however, it should comprise a balance between programme/operational (including logistics), protection, sectoral specialist (including community services), public information and administrative personnel.

19. Experience in the Rwanda emergency has clearly demonstrated that technical staff, such as health, water and sanitation specialists as well as, where necessary, telecommunications experts, should be deployed from the very onset of an emergency as an integral part of the response team and should assume responsibility for the coordination of sectoral activities, particularly in situations where there is an early proliferation of voluntary efforts in technical fields. Technical staff may have to be directly operational until implementing partners can be found.

20. The environmental problems caused by massive and sudden refugee influxes caused particular concern, despite the efforts made by international organizations and NGOs to minimize their impact. Wherever possible, environmental issues should be taken into account whenever decisions are made on the location and design of refugee camps and related matters.

21. The early recognition of UNHCR's role as lead agency in first asylum countries, coupled with the establishment of good cooperating and coordinating relationships with other United Nations agencies, NGOs and governmental authorities have been critical elements in operational success. The readiness of agency representatives to put their own institutional interests aside, to act together as a team and to share the workload according to capability has set a commendable example.

22. The response to the Rwanda and Burundi emergencies has involved the proliferation of United Nations bodies and structures with a coordinating role but without proper clarification of the respective responsibilities of each, and without overall authority being assigned for humanitarian action. This should be avoided in future complex emergency situations.

D. The procurement and delivery of material assistance

23. The physical layout and structure of refugee camps or settlements is crucial to easier access and the effective distribution of material assistance, especially food. The over- concentration of power in the hands of political or military leaders can be countered, in part, by the use of dispersed distribution points.

24. No guidelines or standards for food distribution were available at the outset of the Rwanda emergency, and initial delivery systems through the refugee leadership did not always ensure fair and universal coverage. The situation has since improved through registration, better organization, wider coordination and the targeting of particular community structures. UNHCR food distribution guidelines now in preparation should be finalized and issued in the near future.

25. When airlift operations are essential for the early delivery of relief, their temporary nature should be emphasized from the start, and surface supply and transportation systems identified and developed in parallel. The management of the airlift should take place in the field, in close proximity to the country of asylum and be adapted to the specific needs of the operation. Information on available supplies for transportation must be shared fully with the field.

26. The centralized commodity tracking system needs to be further adapted to use in emergencies. There should be close collaboration with other major distributing agencies, especially WFP, so as to ensure that their respective tracking systems are compatible with UNHCR's as a comprehensive and integrated supply information and management system.

27. UNHCR contractual procedures should make maximum use of arrangements for the award of annual or similar long-term contracts to suppliers of basic relief assistance items such as blankets, tents and kitchen utensils so as to ensure rapid delivery in emergencies. The range of commodities included should be expanded. Authority should be widely given for the establishment of local contracts committees with appropriate administrative support.

28. UNHCR should continue to encourage the provision by donor Governments of "service packages" for self-contained facilities and services to meet exceptional needs identified by UNHCR which cannot be met under normal operating arrangements. The specifications for, and operating principles governing the use of such packages should be agreed upon in advance.

29. Maximum use should be made of locally available resources in supply, logistics and management support.

E. Registration of refugees

30. Experience in the Rwanda emergency - especially at Ngara - demonstrated that (i) the registration of refugees should be done as early as possible in an emergency and (ii) registration is a very staff-intensive and time-consuming exercise which cannot be undertaken until minimum life support measures are in place and often not until the influx has stabilized. These two elements are difficult to reconcile; nevertheless, the importance of registration for the accurate determination of refugee numbers was illustrated at Ngara by the fact that the Ngara refugee population at the time of registration was counted as 230,000, well below all previous estimates and far below the figure of 350,000 derived from refugee leadership lists.

F. Staffing

31. The initial deployment of UNHCR emergency staff was, on the whole, rapid and efficient. At Goma, for example, within ten days from the beginning of the emergency most key staff were in place. However, identifying and deploying longer-term replacements proved more difficult and regular staffing needs were not always adequately met.

32. More rapid procedures need to be instituted for the creation and filling of longer- term posts for emergency operations. Procedures would be considerably accelerated by the establishment of standard models for the organization and staffing of Sub-Offices and Field Offices in emergencies with standardized organigrams and pre-classified job descriptions.

33. Wherever possible, greater use should be made of competent local personnel in meeting staffing needs. Authorization should also be given to Branch Offices to issue consultancy contracts to locally-identified and qualified technical specialists.

34. New types of posts have proved essential to effective emergency response, including such categories as base-camp managers and security officers.

35. Emergency staff have had to live and work under appalling office and accommodation conditions for extended periods. The situation should be remedied for future emergencies by the timely positioning of base-camp and emergency housing units. An emergency housing scheme should be implemented, to allow the immediate deployment with the first relief flights of accommodation providing acceptable working and living standards for up to one year, to be replaced thereafter with more permanent shelter if required. The scheme will involve the creation of a stockpile of two modules for living and office accommodation for staff and visitor housing. Each module will comprise fully furnished units, including linen, generators and office equipment and general medical supplies for the use of staff. Measures to address these needs are now in hand.

G. Staff security

36. It is essential that all field staff working in circumstances where there is a high security risk should receive appropriate training, particularly to develop cultural awareness and sensitivity to the local situation. The assignment of security officers is also necessary and should enhance, not substitute for, this training.

37. The system of establishing and implementing United Nations Security Plans should be reviewed to take into account the special circumstances in countries where no functioning national authority exists.

38. United Nations security arrangements should provide for the relocation of local staff to a safe area within the country (where one exists) or, in exceptional circumstances, the evacuation of local staff who have been identified by UNHCR as being in danger because of their status as UNHCR employees or as a result of the performance of their official duties.

39. Where the regular United Nations security mechanism is not in a position to make the appropriate determination, the UNHCR local security coordinator (normally the Representative or Emergency Team Leader) should have delegated authority to decide when to evacuate or relocate both international and local staff.

III. CONCLUSION

40. The lessons learnt from the Rwanda emergency reflect its size and special character. A number of lessons had been identified from earlier emergencies and had already contributed to a better emergency response. More can still be done, however, to improve UNHCR's capacity for effective action, even in situations as tragic and appalling as the Rwanda emergency has proved to be.