UNHCR - Emergency food assistance to returnees, refugees, displaced persons and other war-affected population in Bosnia and Herzegovina: Joint WFP/UNHCR evaluation mission

UNHCR Evaluation Reports

Emergency food assistance to returnees, refugees, displaced persons and other war-affected population in Bosnia and Herzegovina: Joint WFP/UNHCR evaluation mission

INTRODUCTION

The Joint Evaluation Mission had as its task the evaluation of the food aid programme from its inception in 1992 until June 1997. It was able to interview many of the officials from the two agencies who had been involved in the design and implementation of the programme, in Geneva, Rome and Bosnia and Herzegovina (BiH). In BiH, the Mission traveled widely and held interviews with beneficiaries, with the staff of both agencies and with other multilateral and bilateral agencies, NGOs both national and international, and with the authorities at the various levels of government.

The humanitarian assistance operation in the Former Yugoslavia was one of the largest humanitarian initiatives ever undertaken by the international community. It was implemented in a political environment of great complexity and under conditions which, particularly during the war, were always difficult and were often dangerous. It is doubtful if a report of this length can do proper justice to the events of BiH over the past five years, even within the specialized area of emergency food aid. However, some general conclusions and recommendations stand out.

During the war, the food aid programme within the emergency relief operation had a critical impact on the survival of a very high proportion of the country’s population. It contributed to the survival of communities in siege, prevented famine from becoming a defining characteristic of BiH and signalled international solidarity at a time when other responses of the international community may have seemed less than adequate. In the unusually complex political and military environment in which the operation took place, and given international concern for such a crisis in Europe, UNHCR’s usual objectives related to refugees and displaced persons gave way to a concern with alleviating suffering and saving as many lives as possible. The food which was procured and delivered by WFP, and distributed by UNHCR took on an over-riding importance in the pursuit of these goals.

GENERAL CONCLUSIONS

War -time operation

1. Co-operation between UNHCR and WFP

Co-operation between the two agencies was good. During the war, no serious problems arose as a result of policy differences or failures of communication between the two agencies. However, frequent gaps in the pipe-line resulting from difficulties in mobilising and maintaining an adequate response from donors did occasionally give rise to some tension within and between the agencies.

2. Supply of assistance under war conditions

2.1 The important population displacements caused by the war meant that initially, no reliable needs assessments could be carried out. Nevertheless, it is to the credit of WFP and UNHCR that there was no under-supply. Estimates based on the available rudimentary data permitted coverage of the population’s basic needs for food. This occurred despite weaknesses in the pipe-line, including irregular and unpredictable supplies which were evidenced in the WFP’s and UNHCR’s urgent and repeated requests for donor response. It must be stressed that despite the lack of reliable data on differential food needs in different areas of Bosnia, the operation succeeded in getting food to the most needy areas, with only some inevitable over-supply in the initial stages. The operation ensured that there was no widespread hunger or malnutrition, although in the case of the isolated populations of the besieged cities, security-related difficulties of access made distribution problematic. When access by land was denied, distribution was by air transport, as in the difficult and relatively succesful airdrop operations in Srebrenica, Gorazde and Zepa. It seems that only in Bihac was it not possible to stave off hunger by such means in late 1994 and 1995 Difficulties of access also led to long-term air deliveries, as is illustrated by the Sarajevo air-lift, the longest running humanitarian air bridge in history, which lasted from 3 July 1992 to 9 January 1996.[1]

The humanitarian operation of the Former Yugoslavia provides an extreme example of the difficulties confronted by humanitarian agencies in emergency situations, where despite the absence of reliable population data, the survival of the victims must be ensured and supply cover basic needs. It is to the credit of WFP and UNHCR that the food distributed on the basis of initial calculations covered the most urgent needs. Later in this as in other emergencies, targeting became increasingly possible so that localised levels of under-supply or over-supply were gradually corrected.

2.2 Given the amounts of food supplied, the main cause of hunger was from the constraints of supplying communities under siege conditions Attention should be given to maximising the amounts of food delivered by providing a food basket constituted by food items procured in terms of their high nutritional value in relation to their volume.

2.3 The supply of large-scale humanitarian assistance during a conflict will probably have unintended consequences. Even if the distribution of assistance is rigorously restricted to non-combatants, it will result in freeing resources which can then be used for whatever the authorities prefer including military purposes.

Again, as access for the distribution of assistance is under the control of the authorities in situ, bargaining with them and agreeing to using the channels they control unavoidably must reinforce to some extent their authority. Also in some cases food levies are carried out by the military authorities to allow convoys to pass. In Former Yugoslavia, attempts to levy food taxes were systematically resisted by UNHCR field-staff who were given repeated and specific instructions to that effect. However, field-staff did occasionally undergo pressures to relinquish aid, and on exceptional occasions, food was seized, sometimes at gun-point. This was part of the reality of a war-time operation and illustrates the tense context in which staff operated and the difficult duties they had to perform

The conclusion is that under war conditions humanitarian agencies must select staff in terms of their capability to engage continually in stressful negotiations with difficult interlocutors, and to remain firm even under threat without jeopardising the continuity of the operation. This much was done by the humanitarian agencies operating in Bosnia during the war.

3. Management

The decisions of the two agencies to set up Special Administrative Units to handle the emergency was justified by the scale of the operation and the requirements to maintain a flow of information between the two agencies and with the donor community.

Likewise, it made sense to appoint a senior official to represent the lead agency in the country of intervention, with ample authority to respond on the spot to the continually evolving political and military events constraining the delivery of humanitarian assistance. UNHCR has found the appointment of a Special Envoy in the Former Yugoslavia and elsewhere a useful management formula in complex and large-scale operations.

GENERAL CONCLUSIONS

Post Dayton

1. Transfer of responsibility from UNHCR to WFP

Under Post Dayton conditions, responsibility for managing the food chain within BiH was progressively transferred from UNHCR to WFP, a process which was completed in January 1997. The transfer of management of logistics was justified by the parallel shift from the use of donor government supplied convoys to commercial transport, an appropriate component of the return to normal peace-time conditions and promotion of market economic arrangements.

The decision to set up a full-scale WFP programming and delivery field operation was more questionable. At that time, the operation should have been defined in terms of scaling down, with minimum negative impact on the beneficiaries. This is a different question from that which was addressed in practice, which was how to move from a war-time to a peace-time food aid programming. By asking the latter question, the programming shifted to assisting vulnerable categories rather than refugees, displaced persons and other war affected populations.

2. Scaling down of the food supply

With the cessation of hostilities, the decision was made to scale down the food aid programme. This decision was justified as the end of fighting made the revival of commercial transport possible. As the economy recovers, the problem of household vulnerability increasingly results from lack of household income, rather than from the dislocation of food supplies, and the lack of an adequate social welfare system to provide support for the vulnerable. In BiH, food aid should not be seen as an alternative to a social welfare system. As food aid is scaled down, the national authorities and donor agencies should take the steps required to create an effective social welfare system. This is now a matter of urgency.

RECOMMENDATIONS

Recommendation 1

The arrangement whereby WFP delivered to Extended Delivery Points outside BiH and UNHCR was responsible for management of the food chain within BiH made sense given the need to maintain a clear and unified chain of command and minimise security risks under war conditions. Moreover, it was important to avoid the costs connected with the setting up of two parallel command and administrative infrastructures in a large scale operation. However, given UNHCR’s difficulties in recruiting staff in general and logistics staff in particular. WFP and UNHCR should strive to reach agreement on the secondment of WFP specialised staff to fill key logistics posts in future similar operations.

Recommendation 2

The discrepancies between WFP’s calculations of tonnage delivered and UNHCR’s calculations of tonnage distributed in relation to reimbursement of ITSH costs to UNHCR, demanded labour intensive calculations and caused tension between the agencies. Evidently clearer procedures should have been established. Given the frequency with which the two agencies cooperate in joint operations, they should put some effort to make their systems of accounting and recording made mutually compatible and simple to operate.

Recommendation 3

In this joint UN operation, as the Lead Agency, UNHCR should have ensured media coverage for all participating agencies, both UN and non-UN. With respect to UN agencies, perhaps a greater equality of exposure could be achieved by more aggressive joint labeling of material than was the case (e.g. vehicles, commodity packaging) and, perhaps more importantly, by a conscious effort by the agencies, be they in the lead role or not, to emphasize the joint nature of the humanitarian operation in their public relations and information activities. Relations with the media are also a subject of inter-agency coordination.

Recommendation 4

During an operation taking place in war, the rapid field-level response capacity to cope with rapidly evolving conditions requires the establishment of a decentralised system in the country of operation. However, control of field activities must be permitted by effective communication between the field offices and the country central office. In addition, field offices should be regularly monitored, possibly by rotating country-office monitoring staff. In a war operation, decentralisation must be allowed to go as far as is compatible with the agencies’ minimal levels of control, while a maximum level of communication must be set up.

Recommendation 5

There are several unintended aspects of the distribution of aid, including food during a humanitarian operation: freeing of resources for whatever purposes the authorities prefer including the military; reinforcing the authority of those who approve or carry out distribution; and forceful appropriation of aid. These aspects of a humanitarian operation occurring in war must be taken into account in the recruitment of staff. Skills in negotiating with intransigent interlocutors in situations of danger are a critical requirement for humanitarian workers in the context of war.

Recommendation 6

WFP’s scaling-down exercise is being carried out more rapidly than the government can set up viable and sustainable social safety nets. Significant differences in this respect exist between the entities as also within them. WFP should be closely involved in assessing these differences as the reduction of the beneficiary lists takes place, in order that the authorities are aware of the destitute categories of the population.

Recommendation 7

Monitoring in the Post Dayton period has in common with pre-war monitoring the small ratio of staff to distribution points, and the even smaller ratio of monitors to beneficiaries. If the donors require that UNHCR and WFP carry out serious monitoring of aid distribution, including food, they should be aware of the cost implications with respect to staff and vehicles. At the same time, changed management approaches could maximise the use of staff dedicated to this activity.

Recommendation 8

The response capability of agencies such as WFP and UNHCR in emergency situations ultimately depends on field-level response capability. Therefore, recruitment of a sufficient number of field-office level managers with previous experience in emergency operations, and knowledge of both the agencies’ principles and objectives is an important priority. Even in very large operations, it should be possible to deploy such staff for the relatively few key management field-office and central office posts. In UNHCR this could be ensured by the Division of Human Resources and Management in co-ordination with the Emergency Preparedness and Response Unit. In WFP the Human Resources Division and Operations Department would be responsible for achieving the same.

Recommendation 9

The post-war period coincides with the transition from emergency to rehabilitation and with the passage from a planned to a market economy. In this context WFP, along with other aid agencies, should be particularly vigilant with respect to the potential political manipulation of aid, including food. As part of the UN operation which is presently committed to the Dayton Peace Agreement, WFP should pay particular attention to avoiding too rapid a withdrawal which could lead to the kind of discontent and instability that can be manipulated by politicians, including those holding extreme ideologies.

Recommendation 10

It is outside WFP’s mandate to address the specialised issued involved in the reconstruction of the Social Welfare system. During the next phase of EMOP 5142 in BiH, the WFP country office could contribute to the planning of developments in this area, in particular because of the experience its staff have gained in helping the authorities to identify the socially vulnerable elsewhere.

I. Division of labour and cooperation between the two agencies

From the point of view of the cooperative arrangements between the two agencies, the history of the operation can be divided into three phases:

(1) The Initial Involvement of UNHCR (1991- late 1992) During this period, UNHCR became progressively more involved in Bosnia and Herzegovina, becoming the “lead agency” for humanitarian relief, but WFP was not yet involved;

(2) War- time Operations (late 1992-1995) During this period WFP became responsible for the delivery to Extended Delivery Points (EDPs), outside of Bosnia, of the bulk of the food which UNHCR distributed;

(3) Transition (1995-1997) During this period, the Dayton Peace Agreement was negotiated[2] and hostilities ended. The management of the food chain within Bosnia was handed over to WFP.

I.1 The Initial Involvement

When the Former Yugoslavia began to disintegrate in mid -1991, UNHCR’s office in Belgrade monitored the evolving situation and reported to UNHCR Headquarters on the population displacements which the hostilities had caused. By the October 1991 Session of UNHCR’s Executive Committee, the scale of events and the relevance of UNHCR’s mandate to population movements had led the then Yugoslav authorities to request the organization’s support to respond to the crisis This was followed by a joint UNHCR/UNICEF Special Assessment Mission, which visited key locations in Serbia, Croatia, Bosnia and Montenegro. The Mission recommended setting up a modest assistance programme in response to the displaced persons’ (DPs) needs, in conjunction with ICRC, which was to operate in the war areas, according to its mandate[3].

With the collapse of the Yugoslav Federal State, and the establishment of new borders, and as international recognition was accorded to some of the Former state’s constituent parts, so internally displaced populations became actual or potential refugees, reinforcing the perception by key actors of the international community that UNHCR possessed the special competence required by the crisis. Accordingly, on the 14th of November 1991, following a formal request by the Secretary General, UNHCR’s traditional mandate was extended to cover displaced persons in the Former Yugoslavia by application of the “Good Offices” concept[4]. Such a request was in line with past UNHCR interventions for internally displaced persons, not least that of 1990 in Northern Iraq, where donor states had insisted that UNHCR assist some 400 000 internally displaced people massed at the border between Turkey and Iraq. In 1991, the Iraqi operation constituted a relevant recent precedent for UNHCR intervention to assist a large internally displaced population.

In January 1992, in the brief period of international optimism which followed the establishment of the Vance plan, UNHCR’s presence was further expanded. In the Plan, UNHCR was given the responsibility for the registration and return of displaced populations. At about the same time, an internal debate within UNHCR considered the arguments for and against participating in humanitarian action in war.

Following the out-break of war in Bosnia in April 1992, UNHCR correctly assessed future funding requirements, and sharpened the international community’s sense of the scale of the crisis by requesting an unprecedented US$ 165 million to cover initial humanitarian needs in 1992. The positive response that followed indicated preoccupation with the evolving crisis among donors, their willingness to support extensive humanitarian action in the Former Yugoslavia, and their increasing acceptance of UNHCR’s involvement.

Inter-agency cooperation was evident early on, when the first Joint Appeal for the operation was launched jointly by UNHCR, UNICEF and WHO, on 3 December 1991, for some US$ 24,316,900. This enabled the distribution of basic items of assistance to the displaced population in Croatia, Slovenia and Serbia (370 000 in November 1991). Most of this population had received support from the general population, so that assistance took the form of complementatry family parcels, consisting primarily of food.[5] The programme had been introduced by ICRC and then taken over by UNHCR, which allowed ICRC to concentrate on its other activities, particularly in the combat zones.

This initial complementary distribution of food by parcels was the precursor of bulk distribution, which was introduced by WFP in November 1992, and thereafter maintained throughout the operation. But WFP did not participate in the initial period, until the scale of the required assistance became evident. This was mainly because it saw its mandate lying outside Europe; certainly per capita incomes in the Former Yugoslavia were well above the levels that could be considered a cut-off point for a country food aid recipient in normal circumstances.

The high-jacking of 12 UNHCR trucks on their way to Central Bosnia in the early summer of 1992, and the death of an ICRC delegate, when a mortar hit a convoy outside Sarajevo in the same period, caused consternation in ICRC and fueled the continual internal UNHCR debate whether to remain in the Former Yugoslavia. However, UNHCR was back operating soon after, and its long-term presence in the Former Yugoslavia was fully established in June, following the Secretary General’s request that the High Commissioner administer the Airlift Operation to Sarajevo in coordination with UNPROFOR.

Run by UNHCR through its Geneva-based Airlift Operations Cell, in close coordination with some 20 participating nations,[6] the airlift flew in a total of 160,677 metric tons of aid, including 144,827 metric tons of food and 15,850 metric tons of medicine, equipment and other supplies. In addition to aid delivery, the airlift helped with the medical evacuations of more than 1000 persons. Operating initially from Zagreb, Split, Frankfurt and Ancona, and from Ancona only as from early 1995, this air-bridge lasted from 3 July 1992 to 9 January 1996.

By entrusting the airlift to UNHCR, the Secretary General had renewed his support of the organisation as lead agency for humanitarian assistance and established a formal link between the civilian humanitarian and military aspects of the UN Operation. The military force was, in general terms, entrusted with the mandate of protecting the humanitarian activities under UNHCR (UN Security Council 771 of 13 August and 776 of 14 September). A variety of factors explain the occasional friction and misunderstandings between the two branches: the rift between the civilian, humanitarian culture of UNHCR and UNPROFOR’s military culture, the lack of prior experience in working together, and the fact that the roles of both organisations changed overtime. From supporting the humanitarian operation, the military took on a security advisory role, then a security command role and, under some circumstances, on UNHCR’s request, actually carried out food deliveries on UNHCR’s behalf.

Following the launching of the appeal for humanitarian funding in April 1992 and the beginning of the Sarajevo airlift in June of the same year, the seriousness of the crisis and the scale of the operation became increasingly evident, and with it the need to mobilize WFP capacity as a manager of large scale international food relief. WFP initiated its participation in July 1992 when, following consultations between the two agencies, the High Commissioner issued WFP an invitation “to join in the formidable task of providing humanitarian assistance to the continually increasing number of affected people”. WFP responded positively and participated in a UN inter-agency and donor mission in August 1992 which assessed the humanitarian assistance needed for the six republics of the Former Yugoslavia. The findings of this mission were the basis for a Joint Appeal in September 1992. In the Appeal it was stated that: ”WFP will mobilize resources from the international community and supervise the delivery of commodities to extended delivery points”. During the September- December 1992 period, UNHCR and WFP worked together towards increased WFP involvement until the latter took on full responsibility on 1 January 1993.

In considering the initial period of UNHCR presence, the main issue relevant to the subsequent joint operation is whether the WFP should have been involved earlier. Crucial operational aspects of the subsequent joint food programme were influenced by the fact that UNHCR had set up and staffed its operational capacity on the ground before WFP became involved. During 1992, UNHCR had found its role as a relief agency expand when it received ample international support for its responses to war-driven needs and population displacements, with an increase from 500,000 beneficiaries in December 1991 for the entire Former Yugoslavia through 650,000, 1,000,000, 2,700,000 to 3,055,000 by December 1992.

As has been mentioned, UNHCR was early to realize that it was embarking on a major operation when on 30 April 1992, the High Commissioner appealed to foreign ministers for US$165 million. But that action even took many UNHCR staff by surprise, and by no means represented a widely held appreciation within the international community or among UN agencies, including WFP, of the likely course of future events. WFP’s initial reticence seems to have been well grounded: when it was not clear what scale the operation would assume, there were strong reasons for not getting involved in what was a middle income European country well outside WFP’s usual focus on the poorest groups and food security crisis in low income countries. The decision regarding WFP involvement was therefore delayed so that first WFP shipments were not made until November 1992, when the international trucking fleets and the airlift were operational.

I.2 War-time operations

During the period of war-time operations, the division of labour between the two organizations was largely defined by two facts. First, WFP responsibility was concentrated on the delivery of food to EDPs (External Delivery Points), which were in all cases located outside Bosnia (initially in Croatia and Yugoslavia, followed by EDPs in Italy and Germany to service air deliveries). Secondly, the management of the food supply chain within Bosnia remained the responsibility of the UNHCR (see Logistics Annex). Therefore, during the war period, the joint operation involved a clear division of labour with a geographical dimension which meant that within Bosnia as such there was little “joint” activity since there was no WFP presence. Given the conditions on the ground within Bosnia, it made sense not to duplicate administrative structures.

There were pros and cons to this arrangement. In its favour was the definition of a clear division of labour which left responsibility to a single agency to handle the difficult logistics of delivering food along with other humanitarian supplies within Bosnia, thus permitting, at least in principle, greater efficacy.

But, there is one aspect of the modality of cooperative field operations in crisis situations which deserves careful consideration by the agencies involved. On the whole when agencies involve themselves in field operations, they expect to create their own command structures and administrative infrastructure in the field, with joint activities involving horizontal cooperation between the two agency field offices. Apart from the financial costs involved, under war conditions in which there is an obvious need to minimize the number of field staff, such proliferation is undesirable.

However, the mobilization of the expertise and capacity of specialized agencies could also be achieved by the greater use of the secondment of the required specialized staff of other agencies to the lead agency for the duration of a field assignment. In this operation, in which UNHCR found it difficult to fill posts from its permanent staff, and found it necessary to recruit the large majority of field staff from outside the agency, secondment from other agencies would not even have had the disadvantage of blocking job opportunities for its own staff.

Recommendation 1

The arrangements whereby WFP delivered to Extended Delivery Points outside of BiH and UNHCR was responsible for management of the food chain within BiH made sense given the need to maintain a clear and unified chain of command and minimise security risks under war conditions. Moreover, it was important to avoid the costs connected with the setting up of two parallel command and administrative infrastructures in a large-scale operation.

However, given, UNHCR’s difficulties in recruiting staff in general and logistics staff in particular, WFP and UNHCR should strive to reach an agreement on the secondment of WFP specialized staff to fill key logistics posts in future similar operations.

Another issue related to the geographical division of labour became a source of difficulty between the agencies: the expenses incurred by UNHCR for internal, transport, storage and handling (ITSH) costs, and subsequent reimbursement of these costs to UNHCR by WFP. Lack of fit between the accounting systems of the agencies has given rise to disagreements between them. Disagreements on this count, however, have not impinged on the effectiveness of field operations in the difficult war period or later, though they have constituted a source of continued discussion, fortunately presently resolved, between the Headquarters of both agencies. (For an analysis of the issue, see Logistics Annex.)

Recommendation 2

The discrepancies between WFP’s calculations of tonnage delivered and UNHCR’s calculations of tonnage distributed in relation to reimbursement of ITSH costs to UNHCR demanded labour intensive calculations and caused tension between the agencies. Evidently clearer procedures should have been established. Given the frequency with which the two agencies cooperate in joint operations, they should put some effort to make their systems of accounting and recording mutually compatible and simple to operate

I.3 Transition of management within Bosnia

During 1995-1997 a change in the division of labour between the two agencies was agreed and implemented, culminating in WFP taking over the management of the whole food chain within BiH in January 1997. After Dayton, when there were no longer compelling reasons for UNHCR to continue management of the programming for BiH, the WFP was keen to implement a swift transfer. However, the impression has been left that cooperation between the two agencies met with some difficulties during this period. This partly derives from the inherent difficulties of transferring management of a large-scale operation in a particularly complex political and military context. It also possibly reflects the interests at stake of those who contributed to building the UNHCR managed system.

One change which would have occurred regardless of the transfer is the significant reduction of food flow into Bosnia during 1996, and 1997. The food reduction coincided with the transfer to WFP, which probably added to the agency’s difficulties in establishing its presence and gaining acceptance by the local authorities. Nevertheless, the transfer was completed without major difficulties between the agencies.

As is shown in the Logistics Annex, the transfer was followed by significant savings, the main element being WFP’s full shift to commercial transport, a shift which UNHCR had envisaged before the signature of the Dayton Peace Agreement in December 1995, but which had not been implemented because of war constraints. Apart from the significant savings it entailed for the humanitarian operation, the shift represented a very significant boost to the local economy.

There were, however, some costs which may be attributable to the transfer. The recruitment and reassignment of personnel, the setting up of administrative systems and the delineation of WFP operational areas different to those used to define UNHCR’s areas of responsibility, involved a learning process both on the part of WFP and the local authorities, which temporarily interrupted smooth food distribution at least in some areas. There were also direct financial costs involved. However, setting up the WFP administrative structure within BiH involved a net cost to the international community only during the brief period in which WFP had opened its offices and staffed them, while UNHCR had not yet withdrawn.

II. Inter-agency coordination

II.1 Coordination in war-time

Coordination humanitarian assistance in the Former Yugoslavia was one of UNHCR’s functions as Lead Agency. Coordination between WFP and UNHCR took place in that context.

The formal command structure of the UNHCR operation involved the UNHCR Special Envoy to the Former Yugoslavia, to whom the UNHCR offices in the six republics reported. The Special Envoy reported to the High Commissioner in Geneva, who in turn reported to the Secretary General in New York. The military component of the operation (UNPROFOR) was led by a commander who reported to the Under Secretary General in the DPKO (UN Department of Peace-Keeping Operations) and through him to the Secretary General, who thus constituted the highest coordination point of the UN presence in the Former Yugoslavia. Only in 1993 was a Special Representative of the Secretary General appointed to coordinate the military and humanitarian aspects of operations in the field. Another aspect of UNHCR’s coordinating role derived form UNHCR’s chairmanship, at the invitation of the Secretary General, of the Humanitarian Issues Working-Group, one of six working groups at the International Conference on Former Yugoslavia (ICFY) which was convened in London under the joint chairmanship of Cyrus Vance and Lord Owen.

These details are mentioned here not because they determined the nature of coordination of the food programme, but because they indicate the complexity of the institutional context of which the programme was a part.

From the point of view of the planning and implementation of the food programme as such, a number of critical connections were maintained by means of continual meetings being held in three key coordination points in the region: Zagreb, Belgrade and Split. This network permitted regular communication between WFP and UNHCR, and between them and the donors and agencies involved in humanitarian relief. More specifically, four types of meetings were held throughout the war to ensure communication on food issues.

- UNHCR weekly inter-agency meetings in Zagreb, Belgrade and Split, with the participation of WFP, UNICEF, WHO, ECHO, ICRC, IFRC and representatives of the bilateral donors.

- UNPROFOR fortnightly meetings which provided detailed information about military developments.

- WFP fortnightly food coordination meetings in Zagreb, involving UNHCR, ECHO and NGOs involved in food provision, to share information;

- As from May 1993, ICVA chaired NGO coordination meetings attended by the UN agencies.

Coordination between UNHCR and WFP regarding harmonization of policy and operations was carried out less in the formal meetings than through daily informal contacts between officials of both organizations, including the WFP Regional Manager and the UNHCR Regional Programme Coordinator. Regular telephone contacts and occasional meetings took place between WFP and UNHCR Headquarters staff within the Special Management Units created in both agencies to handle the Former Yugoslavia programmes.

In the field, coordination in the war period primarily involved creating links of mutual support among the different organisations, within and outside the UN. An example is provided below.

Click here for table

Written materials produced by both organizations were another important element in the information system. In Zagreb and Belgrade, the WFP collected the data on food distribution and the pipeline and produced a Situation Report, which was printed in Rome, and which regularly provided a global picture of the food situation for donors and other interested parties. The UNHCR Information Notes, produced by the office of the UNHCR Special Envoy in Zagreb and later in Sarajevo provided a useful overall data source. Finally, for purposes of reporting on and planning the strategy of the overall programme, the Joint Inter-Agency Appeals, made to the donor community semi-annually, was a key document.

These numerous channels ensured that on the whole, good communication was maintained between the two agencies. However, there seem to have been some dissatisfaction among the UNHCR field offices concerning communication about the recurring gaps in the food pipeline. This is hardly surprising, given the charged environment in which they were operating, and the fact that they had to confront local authorities’ and beneficiaries’ complaints when the food was not delivered. Thus it was felt that appeals from the field regarding the food situation were not passed on to the donors by WFP with sufficient urgency, though, in fact WFP officials continually provided information about the field situation to donors, both through the inter-agency coordination meetings in Zagreb which were attended by both agencies and by donor representatives, and through communications from WFP Headquarters in Rome and UNHCR Headquarters in Geneva. Again, some UNHCR field staff felt that they received insufficient advance information about the pipe-line gaps to alert the recipients of food and find alternative ways of supplying them.

If outside supplies were uncertain and the supply system was not sufficiently responsive to requests from the field, such difficulties mainly originated from the sources of food outside WFP control, and from the normal lags involved in information transfer within organizations, not from any difficulties in coordination between the two agencies at the level of field operations. Indeed, examination of the documentation from the period, and interviews with those who managed the operation in the field suggest that the relationship between the two agencies was usually harmonious.

Coordination with other actors loomed much larger as issues of concern to UNHCR. For instance, coordination with UNPROFOR (United Nations Protection Force), the multi-national military branch of the United Nations Operation in the Former Yugoslavia was sometimes problematic. UNPROFOR was created in principle to protect the Humanitarian Operation. But neither was its role clearly specified, nor the nature of its relationship to UNHCR. Differences in the culture of both organisations, lack of experience in working together and the shifting definition of UNPROFOR’s role occasionally gave rise to frictions and misunderstandings. Also, the presence of batallions of different nationalities, and the different initiatives taken by them, meant that the linkages between them and UNHCR were regio-specific. However, systematic efforts were made continually to maintain an on-going relationship of cooperation. For instance, Standard Operating Procedures were drawn up jointly in 1994, which helped define relationships at all levels and deal with differences which continued to appear.

This was an occasional source of friction and misunderstanding, which affected the efficiency of UNPROFOR’s support. Although UNPROFOR did contribute to ensure that food was delivered, especially when there were important security constraints, a clearer definition of its role in respect of the humanitarian operation could have resulted in a greater use of its transport capacity for humanitarian purposes. As it happened, UNPROFOR intervened according to the initiative of its local battalions, not on the basis of a clear mandatory definition of its involvement in humanitarian action.

Other coordinating challenges faced by UNHCR were posed by the three warring factions and the shifting constellation of military and political forces. Against this complex and difficult background, and the UNPROFOR coordination issue, potential difficulties of communication between WFP and UNHCR were minimized.[7]

One important issue which was raised during the operation in the Former Yugoslavia was related to another UNHCR lead agency function: the coordination of humanitarian agencies’ exposure to the media. The perception by the authorities within Bosnia is that the WFP was not involved in food supply until the transfer of the field operation to WFP management. This has had one unfortunate consequence -- unwelcome changes in policy since the end of hostilities (e.g. the reduction in food supply and the consequent need for more selective targeting) are sometimes associated with the shift in management from the UNHCR to the WFP, which may increase the difficulty of the task of phasing down the programme.

Likewise, during the war UNHCR received most exposure in the international media, since in face of limited media coverage of the conflict particularly in the early days of the conflict, UNHCR considered it its responsibility to ensure a flow of information about the humanitarian situation outside Bosnia. This both alerted the international community to the needs of the population and proved to be a very effective strategy of fund-raising. Unfortunately in the process, the role of partnership among agencies operating alongside UNHCR in its lead agency role received insufficient publicity. Of course, if performance had been perceived by the media in negative terms, such a high profile would have been unwelcome. In practice, UNHCR received the lion’s share of the positive appreciation of the UN humanitarian effort. This is not of great operational significance and only resulted in the expression of minor irritation by WFP staff, but in future operations any such imbalance might become a source of inter-agency resentment, which in turn could erode the basis for cooperation.

Recommendation 3

II.2 Coordination Post Dayton

The Dayton Peace Agreement added one additional body to the coordination network operating from Sarajevo: the Office of the High Representative. He is the International Community’s highest level interlocutor with the Bosnian Authorities and therefore, in principle, the source of political support for all the non Bosnian agencies operating in Bosnia. The meeting of the Principals (OSCE; IPTF; SFOR; UNHCR; SRSG’s office) under the chairmanship of the High Representative provides a forum for coordination since 1996, though the dynamism within it seems to have suffered during the course of 1997.

The Dayton Agreement has maintained UNHCR in the lead role while it has re-focused its activities on refugees and displaced populations by making it responsible for the implementation of Chapter VII of the Agreement. UNDP is gradually expanding, according to the importance of rehabilitation and reconstruction activities, while the European Union, the World Bank and the European Bank for Reconstruction and Development continue to provide support to the government, specifically attempting to assist in the development of viable frameworks for the recovery of the economy and the social fabric.

Against the background of this proliferation of coordinating bodies, present coordination of the food programme has been carried out by WFP as follows. At the national level, the WFP Regional Manager in Sarajevo provides information about WFP activities in the country at weekly meetings chaired by the Special Representative of the Secretary General or his deputy. Information sharing takes place at monthly NGO meetings chaired by CRS (Catholic Relief Services) within the framework of ICVA. In Sarajevo also, WFP is regularly present at the monthly Agriculture Task Force meetings chaired by FAO.

At the operational level, WFP has established a wide network of links with the various organizations dealing with food. Each month, the heads of the 5 WFP sub-offices chair Regional Food meetings, which are attended by local agencies (municipalities, centres of social welfare, Red Crosses, the Commissariat for Refugees (in the Republika Srbska) and local NGOs); international NGOs (Caritas, Merhamet, ECHO-funded NGOs (MPDL, Equilibre, Premiere Urgence, AICF, Solidarite, InterSOS, German Red Cross, Australian Red Cross); and US-funded agencies (ADRA and IOCC). There is also the occasional participation of agencies such as UNHCR, OHR (Office if the High Representative) and SFOR.

WFP staff, the heads of sub-office particularly, obtain information about the over-all situation and share information about food issues at meetings organized by OHR, UNHCR, IMG and SFOR.

While information-sharing about the food programme has been a feature of this network of meetings in 1996, it is only towards mid-1997 that the meetings have operated as fora to coordinate programmes in the sense of establishing complementarity and avoiding duplication.

III. Response capability to beneficiary needs

III.1 Targeting

III.1.1 Targeting in the war

Access and the feasibility of delivery and distribution to specific groups were affected by various inter-linked factors including: the rapidity of the war build-up, the consequent massive movements of people, plus the complexity of the war itself: the several factions and parties involved and their different objectives, and the ever-changing patterns of conflict

Other evaluators have commended UNHCR for its enterprising adaptability in the Former Yugoslavia. In the first months of the operation, UNHCR focused on delivering large quantities of food to where it had access. Beneficiaries were initially described according to four categories: refugees, internally displaced persons (IDPs), returnees and the 'war affected'. But it soon became evident that there were strong operational as well as conceptual grounds for a maintaining a single category: casualties of war. And as one observer noted, "it (UNHCR) moved away from its usual statistical preoccupation with categorizing refugees and simply helped anyone who needed help.[8]

Distribution was in fact achieved through quite simple geographical targeting, guided by information on areas of conflict and movement of people, as well as crude estimates of populations based on pre-war figures. But a more category-based approach gradually evolved. As UNHCR developed its delivery systems, through international convoys, and its distribution systems, through local municipal authorities, the Commission for Refugees, local Red Cross Societies and international and national NGOs such as CARITAS and Merhamed, so it developed contacts with other relevant organizations as well as direct contacts with beneficiaries. To the extent the information base evolved and enlarged, so it became more feasible to ensure that the most needy beneficiaries were identified and targeted for food aid distribution.

In any case, the evidence suggests that the neediest (i.e. vulnerable, female-headed households, etc..), were looked after by those who distributed, in part because of the old communist Welfare State mentality. Also techniques were used to inform beneficiaries about the arrival of food, such as via radio announcements, so that failure to distribute was followed by beneficiaries reporting, blocking trucks, and generally making sufficient noise, so that eventually, they got the food.Programme beneficiaries: WFP food aid and logistics requirments

Click here for table

Source: Untied Nations Consolidated Inter-Agnecy Appeal (1992-1997)

Thus a system of prioritization was put in place to direct the distribution of food. Subject to the gaps in the pipeline and the obstacles to access, food was to be delivered to the following categories:

Category I - institutions (e.g., hospitals, orphanages)

Category II - extremely vulnerable individuals (e.g. persons in collective centres, mentally/physically handicapped, elderly persons without family support, minorities)

Category III - other beneficiaries who are in need of food assistance but who are not totally dependent on food aid.

Ensuring food supply to the ethnic minorities was a crucial objective. The minorities were served by agencies geared to two of the ethnic groups (Caritas and Merhamet) as far as possible, the local Red Cross organizations also being responsible for the minorities in some areas, after being evaluated for reliability and efficiency.

It should be noted however, that needs assessments continued to be based on estimates, rather than on in-depth studies. To a large extent initial estimates were provided by local authorities and local Red Cross societies. With increasing experience in the areas of intervention and the development of viable working relationships with local and international organisations, and with the acquisition of skills for operating in a dangerous environment, it was possible to identify populations and their needs with increasing accuracy. UNHCR field officers were able to visit institutions, collective centres, distribution points and Red Cross and municipal offices, as well as making household visits. Beneficiary lists were checked, distribution reports collected and limited spot checks carried out on the needs at the beneficiary level. In this way, a reasonable if somewhat limited overview was obtained.

Thus, UNHCR staff assessed the statements of local authorities about the status of beneficiary numbers and made an estimate of needs in their area for the following month. All UNHCR field offices transmitted these estimates to the UNHCR Programme Office in Zagreb. The WFP Coordinator then brought in information about the pipe-line, including food carry-overs from the previous period. WFP and UNHCR then jointly developed a chart of food distribution for each of the three principal EDPs (in Metkovic, Zagreb, and Belgrade).

However, the UNHCR managers of the EDPs in the field could not completely follow the charts, but distributed the food according to the evolving military and political situation, and to those areas where access was possible.

The formal system in which logistics was subordinated to programme became a flexible framework within which it was possible to respond to changing needs. However, despite the regular dispatch of information from the field about distribution to the Logistics Central Unit in Zagreb, there was little feedback and a lack of systematic monitoring visits by either the logistics or programme staff from Zagreb. The EDPs, therefore, operated as fairly independent centers of food distribution networks for specific parts of Bosnia (Zagreb for Western Bosnia, Metkovic for Central Bosnia and Belgrade for Eastern Bosnia). The effectiveness of these networks in getting the food to where it was possible remains undisputed, although undoubtedly there were communications problems with the central office in Zagreb and actual levels of monitoring and control remained weak.

Recommendation 4

Despite the cited difficulties in coordination, the Programme Unit was constantly appraising the evidence regarding beneficiary numbers, and finally decided to carry out an important cut in the supply of food at the end of 1994. The Unit had noted that despite a shortfall in deliveries, which had never met more than 70percent of targets anyway, evidence of extreme scarcity (e.g. malnutrition) only came from areas under siege, where access was constrained by transport rather than the overall availability of supplies. Following this observation, the Programme Unit carried out an evaluation of municipality claims about need, which together with the conclusions of the 1994 Joint FAO/UNHCR/WFP Assessment Mission and the 1994 CIET Survey, resulted in a quite significant reduction of the yearly tonnage supplied towards the end of the year (see Table above). Another programming issue which confronted the Programme Unit was the significant variations in food need according to geographical location. Distribution charts could not be adjusted to these differences because of the political considerations which were paramount in the logistical managers’ mode of distribution (i.e. one which required that the dynamics of power between the areas controlled by the Government and areas controlled by other parties be taken into account, as when a road passed through areas controlled by different forces). At times, and during the the three-sided war particularly, deliveries could not be carried out equally within Serb, Croat or Bosniak territories: while 90percent of deliveries were possible to the outer-rim Serbs, 60-70percent could reach Croats and only 40percent the Bosniaks during the worst months. Within these important constraints, the Programme Unit, the Field Offices and the Logistics Offices did try to ensure that more food was delivered to areas of greater concentration of displaced persons, despite the protestations of representatives of the warring parties from other areas with less beneficiaries.

The food basket under siege conditions

One issue special to the delivery of humanitarian aid in a war situation concerns the composition of the food basket to feed populations under siege conditions. One characteristic of the Bosnian conflict was the extended sieges of Sarajevo and a number of enclaves. In all these situations, supply by convoy or by air drop was uncertain, infrequent and potentially costly in terms of casualties.

In such a situation, in addition to choosing a basket which is most likely to contribute to nutrition, subject to cost and food availability, three other conditions must be taken into account:

the binding constraint on supply is transport space (defined in terms of tonnage and/or capacity);

if the siege is complete and lengthy, then assistance supplied may become the only source of outside supplies[10], and maintenance of morale is an objective, alongside defending nutritional status.

Consideration of these factors will result in a different optimum food basket than programming in other circumstances, where considerations of bulk only need to be taken into account insofar as transport costs are affected and where normally it can be assumed that, in non-famine conditions, the community will have other sources of supply available.

In light of these points, the complaints emanating from Gorazde that during their long siege, the food supplied was too heavily concentrated on flour sounded credible. It seems implausible, for example, that it was sensible to use the scarce space in air drops for such a bulky food item.

Also, under such conditions it is specially necessary to take account of the availability of ancillary requirements, such as fuel for cooking.

It could be useful to undertake a study of appropriate food baskets for siege situations, programming subject to the transport space constraint and in light of the likely environment under siege conditions.

It was inevitable that during implementation of such a huge relief programme, in a highly complex conflict situation, when the vast majority of the population was in need of food aid, some of the humanitarian assistance should have been channelled into the various militaries: in a country or region where nearly all adult males and many females are participating in the military, the food received by the women to some extent went to the men; also the civilians were keen to feed their military and would have willingly handed food over anyway. There is also the forced appropriation of food at gun-point, of which there were isolated but well-known examples. These facts were a subject of concern in Bosnia for the UNHCR staff of all levels throughout the war, not least the field officers and field assistants who continually had to confront the military and to exercise negotiating skills with very difficult interlocutors. To the cost of food aid on the international market, and delivery costs should be added the time and skills exercised in getting the food past political and military obstacles.

Recommendation 5

Some commentators have suggested that the provision of humanitarian assistance in Bosnia had contributed to the prolongation of the conflict.

Two lines of argument can be developed in support of the prolongation thesis. The first is the “fig leaf” argument - that by giving generous support to food aid, the donors were able to defend themselves against the charge of inaction in Bosnia, and by so doing they postponed the military intervention needed to end the conflict. This was apparently the view of some leaders in Sarajevo, who were opposed to the introduction of food aid at the early stages of the War. However, this a criticism of the priorities and political will of the international community, rather than of food aid as such. And it should be noted that food aid is only part of the package of humanitarian and rehabilitation aid. There is no evidence that the provision of aid, including food aid, deflected public opinion from support for military intervention, and as the implementation of the humanitarian assistance brought with it additional international media coverage, a precursor of the eventual decisive intervention, that hypothesis has little credibility.

The alternative line of argument is that humanitarian assistance prolonged the War by giving material assistance to the combatants. Of course, it was necessarily the case that humanitarian did support the military effort. Even if actual humanitarian supplies were not diverted to combatants, the provision of aid, including (but not exclusively) food aid, would have allowed the diversion to the War effort of resources which would otherwise have been needed to sustain the non-combatant population. Possibly, in the absence of food aid, the resulting suffering might have hastened the cessation of hostilities. But that would have been at the cost of greater non-combatant suffering (including the possible horrors of actual famine) and the resolution of the conflict could have been by the victory of one side. The conflict might have been shorter, but the suffering could have been greater and the outcome unsatisfactory: a world without Bosnia, since the neighbours would have divided it among themselves. There is no convincing case that the risks of increased suffering which would have occurred in the absence of humanitarian support would have been justified by an increase in the prospect of a swifter and satisfactory outcome of the conflict.

However, the provision of aid, including food aid did have some unavoidable negative political effects. As it had to be channeled through existing authorities or their approved agents, it contributed to the consolidation of bureaucratic structures linked to the governments, and consolidation of the authority of military leaders.

III.1.2 Targeting Post Dayton

Intensifying efforts to target more precisely became an issue almost immediately after the signing of the Dayton Peace Accord. The over-riding factor was donor pressure, which was inspired by a strong desire to reduce donations. The simple logic was that peace should bring stability and economic recovery, and that people would return to their places of origin; and hence large reductions in food aid would be both possible and necessary. Further justifications for reductions of humanitarian assistance were offered, including the usual arguments about avoiding the creation of food aid dependency and disincentives to agricultural production. In any case, what seemed clear was that a policy decision was made calling for a reduction in the quantities of food aid to be delivered, and hence the need for a reduction in the number of beneficiaries to be assisted.

The shift from a war-time relief operation and the resulting decision to reduce the availability of food coincided with the shift in responsibility for the management of the food chain within BiH from UNHCR to WFP. At that time, the targeting task should have been defined in terms of the objective of scaling down the war-time operation, with minimum negative impact on the beneficiaries. This is a subtly different question from that which was addressed in practice, which was how to move from a war-time to a peace-time food aid programme. By asking the latter question, the programme shifted to targeting in terms of criteria which were at best incidentally related to the specific objectives of the programme, to assist refugees, displaced persons and other war affected populations.

Since it was becoming increasingly responsible for the food aid chain, WFP initiated some attempts during 1996 to refine targeting in directions consistent with established WFP practice for non-emergency operations. The agency stopped defining its beneficiaries in terms of UNHCR categories and proceeded to identify the most needy in the total population. A first step was to reduce UNHCR categories to two: the "most vulnerable" and the "at risk." According to the WFP/UNHCR/FAO Food Aid needs Assessment mission to BiH in April 1996, the total beneficiary caseload - or beneficiary list - stood at 1.2 million according to WFP (while UNHCR was maintaining a planning figure of 2.7 million) - of a total population estimated at 3 million.[11] The mission recommended reducing the number of 'direct beneficiaries' (i.e. those eligible for general distribution) to an overall target of 600,000 people, or around 20 percent of the population.

Although in theory the criteria may have been accepted by local authorities, progress in their application has been slow. Several reasons have been given for the apparent difficulty encountered in applying the criteria. Firstly, various international agencies are also in the process of diminishing beneficiaries by the application of criteria, but they do not always co-ordinate well with each other, and propose different criteria of need or vulnerability for different commodities, using the same distribution channels. This has lead to confusion, at best, for the local partners. Secondly, despite considerable competence and capacity within the local authorities, the concept of targeting is apparently not very well accepted. There appears to be little motivation to put in place a targeting system. In fact given the present social tension and the fear, or risk of creating further instability, local authorities continue to regard targeting as an alien concept, and are more inclined to give lip-service to targeting while carrying on with a more or less general distribution. It seems that to give a little to nearly everyone, without regard to needs, has been preferable to an exclusive needs based targeted distribution.

The 1996 joint assessment mission also recommended the indirect targeting of some 420,000 families, or around 1.5 million people through the monetization of some 11,000 mt. of food commodities per month, representing nearly 80 percent of the overall food deliveries. The funds so raised were to be used for the subsidizing of various employment generation schemes for reconstruction. This programme proved to be quite unworkable and did not take off.

At present, i.e. during 1997, free distribution or 'direct targeting' continues and accounts for the major use of WFP-supplied food aid. Several agencies working in BiH, including WFP, believe that current increases in food production and availability mean that a more targeted approach is now more feasible and indeed necessary; while the continuing reduction in food aid commitments reinforces the need for better targeting. Most international agencies are trying to work with local government authorities; WFP and its partners continue to include the social welfare centres, offices of displaced persons and refugees, or the Commissioner for Refugees (COR). In the Republika Srbska the local Red Cross Societies have sole responsibility for food aid distribution.

The WFP Sub-offices have reported that progress has been made in verifying and regularizing beneficiary lists, and issuing of beneficiary cards, which has lead to some reduction in the numbers of food aid beneficiaries.

Based on recommendations contained in the report of the Joint Food Aid Need Assessment Mission, Former Yugoslavia, 1997, only one category of beneficiaries, i.e. the most vulnerable, should be assisted with WFP-supplied food aid in Bosnia. Furthermore this category should be reduced from a total of 1.48 million in May 1997 to some 600,000 beneficiaries by end September 1997. This will now account for between 16.5 and 18.2 percent of total population (estimates of total population now vary between 3,300,000 and 3,645,000 persons).

Criteria of vulnerability and food insecurity for selection of beneficiaries at the household level have been devised and are being gradually introduced to the local partners, under the guise of a re-categorisation of beneficiaries exercise. These criteria vary slightly in different areas of the country, but in essence rely on two types of criteria: individual monthly income; and vulnerability indicators, including the mentally and physically handicapped, single or female headed households, elderly (over 65 years of age), invalid, host family of displaced people, participant in home care or Especially Vulnerable Individual programme, and households with more than 3 dependents (children under 14 or elderly over 65) per income earner.

A person with a monthly income of DM 25, plus one vulnerability indicator, or with a monthly income of DM 75, plus more than one vulnerability indicator per household, should qualify for WFP assistance. The local authorities and the WFP sub-offices have drawn up letters of understanding, in which are described the modalities of the re-categorisation exercise and WFP's commitment to pay up to DM 2,500 to cover some of the costs incurred by the local authority. The letters of understanding also provide a ceiling figure for the beneficiary total - representing no more than 20 percent of the total population of the municipality.

There are difficulties in the process of scaling down the programme. The reduction exercise is now being undertaken quite rapidly, allowing authorities little time to either make a reliable census and to establish structures to assist people who will no longer appear on the beneficiary lists. The 20 percent figure is arbitrary; in that data do not exist to predict the distribution of the population in terms of income level of other chosen criteria. It is left unclear what will happen if the application of the exercise yields beneficiary lists in excess of 20percent. The recategorisation exercise itself if carried out adequately would be quite complex and demanding, requiring detailed household surveys of food and nutritional needs as well as of non-food resources and income. It is unlikely that the local authorities will have sufficient human or financial resources to complete the survey properly even if they were motivated to do so.

The 1997 joint assessment mission has also re-introduced the notion of "indirect targeting" through providing food aid for "rehabilitation activities". This sub-programme primarily involves short-term, small-scale projects that aim: "to cushion the effects of a reduction of relief food aid; to revitalize the local economy; and to strengthen the spirit of entrepreneurship and self-reliance".(report of the Joint Food Aid Needs Assessment Mission, Former Yugoslavia, 1997). The mission recommended using some 10,000mt of commodities over the 12 month period (July 1997 to June 1998), i.e. some 8 percent of total projected food aid needs for BiH.

These proposals are considerably more modest than those of the previous joint assessment mission, but remain rather ambitious goals. The feasibility of such ventures is discussed elsewhere in this report. However the 1997 joint assessment report does suggest that "Wherever feasible such projects should target particularly vulnerable groups." Such targeting has proved quite difficult. The more successful projects, in terms of targeting, are those that have provided food aid to on-going programmes of reconstruction. But many of these 'support to rehabilitation' projects involve the subsidizing or topping-up of low wages through the provision of food aid commodities for closed circuit sales. However, people already employed are not usually the most poor and vulnerable, and in some cases this method of providing food aid merely contributes to the earnings of the small-scale entrepreneur-owner, through sustaining low wage payments. Support for small business may in itself have merit, but it is not usually the object of WFP targeting.

Recommendation 6

III.2 Monitoring

III.2.1 Monitoring in the war

UNHCR international field staff and local field assistants carried out food monitoring in a highly unstable environment in which attempts to organize activities in terms of regular routines were continually disrupted. Also, numerous obstacles to circulation lengthened the time required for each monitoring visit. Not least among these obstacles was the involvement of three ethnic groups in the fighting since in some cases, access to a given municipality required lengthy negotiations to ensure passage through territories under the control of other groups. Staff could never be sure that a given distribution point could be reached again by way of the same route, and knew that some routes represented considerable security risks despite all precautions.

Nevertheless each monitor planned monthly activities in their areas of responsibility in terms of intended visits, and covered as many distribution points as they could. Food monitoring during the war.

An example of the difficulties faced in monitoring food distribution during the war in provided by the experience of a field assistant in Central Bosnia responsible for food monitoring in 9 municipalities with some 100 distribution points in her area. Her other responsibilities included assessing the needs of new arrivals, assisting in the setting up of collective centres for the incoming population, monitoring established collective centres (32 out of 59 in September 1993), monitoring agencies operating in the collective centers.

Access to the 9 municipalities under the informant’s control in the April 1993-March 1994 period varied with the development of the Croat-Muslim conflict. Thus to reach Zavidovice, the 60 km Zenica-Zepce-Zavidovici route had to be abandoned for the 130 km. Zavidovici-Kakanj-Sutieska-Ponieri route. The route was dangerous because of its proximity to the front-line, a fact which made movements to Zavidovici less frequent. Conditions were sometime so risky that food distributions had to take place at night.

The choice to concentrate on particular distribution points was based on an on-going assessment which considered a variety of factors including ease of access, municipal attitudes, and security risks. The monitor’s schedule would be set up depending on the movement of the military conflict, closeness of the front-line, reliability of the municipal authorities or local Red Cross, and the number of collective centres and institutions in a given monitor’s area of responsibility. The scarcity of staff and the resulting large size of areas of responsibility meant that each monitor had to exclude many, in some cases most, of the final distribution points from their regular monitoring visits.

Staff report that in some areas significant differences were evident in the amount of food available between municipalities under the control of different ethnic groups. Reasons included differentials in stocking of food prior to the conflict, availability of alternative supplies of food by an NGO and differential possibilities of extracting a levy from convoys in transit. The consequence of this is that municipalities of the food rich municipalities could provide larger rations (According to one informant, “in December 1993, the reported food ration for K. town was 12 kg/person/month, but only 3 kg /person/month in Z.town for its 118000 beneficiaries”). Another factor resulting in differential food availability at the household level was the great variation in supplies in the open market. Items of consumption including food available on the open market could vary in price over a 20-fold range, depending on location and access ( e.g. 1 kg of flour:2DM or 20DM; 1 kg of coffee: 12 DM or 70 DM; 1 kg of sugar: 2 DM or 15-20DM).

The implications of such differences were widely discussed by programme staff, but the conclusion had to be reached that no differential distribution of food aid was possible, since in the complicated kaleidoscope of geographical areas under the three ethnic groups access to one area was dependent on the willingness of other ethnic groups to allow the convoys through, and differential supply of food based on need would have resulted in blockading of the convoys.

Within these very considerable constraints, the monitors regularly obtained distribution reports from the municipalities in their own areas. The degree to which the figures were inflated could never be determined, but monitors made some useful “guesstimates” of the amounts of food being deviated to alternative destinations in each of their municipalities of responsibility.

Although it was not possible to carry out spot checks with any frequency, when it was possible they did serve to indicate to some extent that the prioritized categories were being given the appropriate ration. The evidence suggests that deviations did not create hardship among the beneficiary population if those in charge of food distributions were of the same ethnic group as the beneficiaries. In the case of the minorities, distribution was largely in the hands of agencies with cultural-religious links to their beneficiaries, and these agencies had their own means to resist pressure including that of reporting to UNHCR

III.2.2 Monitoring Post Dayton

Current difficulties of monitoring in Mostar.

The reality is that even with the end of hostilities, detailed monitoring is not easy. For example, the WFP Mostar Area of Responsibility includes 28 municipalities and 368 final distribution points, serving 27000 families and 81000 beneficiaries. To cover this area there are 3 monitors with 1 four wheel drive vehicle and 2 skodas. About 220 final distribution points remain unvisited. The difficulty of monitoring at the final delivery point is exacerbated by the irregular distribution system by the municipalities, and difficult communications (e.g. by telephone).

The Food Aid Programme Strategy for BiH, drawn up by the WFP Country Office in early 1997, continues to emphasize the importance of needs assessments. It describes the current programme as reorienting from the war time "survival nutritional support" to "the changing food aid needs of the post-war period to provide a social safety net for the most vulnerable individuals and to contribute to revival of the economy." However, nutritional criteria were not used in the design of the earlier stages of the humanitarian operation - and nutritional monitoring was not a regular activity. The stated new orientation seems to be an attempt to claim a priori that food aid is synonymous with a social safety net and that food aid is an appropriate resource for the development of Bosnia's economy. These are both questionable assumptions.

Many of the criticisms of targeting apply also to beneficiary monitoring. Monitoring is, of course closely associated with targeting. To monitor whether the 'right' people are being included on beneficiary lists, and if so whether they are receiving the 'right' commodities, requires that there is an established understanding of the 'right' people and the 'right' commodities, and that there is an acknowledged methodology of how to reach the beneficiaries, i.e. that there is a recognized targeting system and known criteria for selection of beneficiaries and for defining needs. However monitoring should not be confused with targeting; although it should provide useful information for improving targeting and management, it neither replaces targeting nor the taking of management decisions. If the targeting plan has been well designed, monitoring should be capable of demonstrating whether that system is working to select the right type of beneficiary, and to point towards possible remedial measures if necessary. And while monitoring is a useful tool for management, there are limits to what can be expected from regular monitoring of the operation. The collection of lots of information does not necessarily mean that all management questions will be answered. The WFP country office has recently drafted a briefing paper on its proposed Monitoring Strategy. This is a commendable initiative. reflecting concern that as resources and funding decreases it becomes even more important that the food aid is used as efficiently as possible. The document is quite clear and explicit on the monitoring activities that should be undertaken - primarily by WFP monitoring staff of the Sub-offices, under the supervision and coordination of a trained monitoring officer in the regional office. The proposed activities, including visits to operational sites, liaising with other relevant IOs/NGOs/Municipal counterparts, collating, processing and analyzing information are quite appropriate, although possibly somewhat ambitious, given the few monitoring staff. The establishment of some priorities might be useful. The monitoring staff are also required to "identify and expand the number of WFP supported rehabilitation, reconstruction, employment and income generating activities." This is not a monitoring function, and given the importance of monitoring and the shortage of staff it does not seem appropriate to ask the monitors to also undertake these activities.

The draft monitoring strategy necessarily relates to and reflects the food aid programme strategy. And hence the overall goal of the proposed monitoring is stated in terms of ensuring "the progress of the implementation of the 1997 food aid programme strategy in the field" as well as providing "accountability by implementing partners in food aid distribution". This is what monitoring should of course do; however some problems arise when the objectives of the proposed monitoring are described. Potential weaknesses are revealed when, for example, monitoring seeks to provide data on whether "the needed (?) assistance ... (is) appropriately (?) used and programme objectives are achieved" - this of course rests on the assumption that food is an appropriate input and can be used efficiently to achieve the programme’s development - this is not immediately obvious. Similarly, monitoring should ensure that “food distribution be part of local capacity building and dependency reduction". And again this rests on the questionable assumption that food aid can contribute to a programme of local capacity building.

It is important, although sometimes difficult, that WFP avoid promising the impossible. Donors or other agencies may request or demand certain data, for example complete and current data on each and every beneficiary household. WFP cannot and therefore should not try to take on the task of providing such data, particularly when the exercise is largely mainly motivated by the wish to phase out the programme.

Recommendation 7

III.3 Management and staffing under war-time conditions

Emergency operations being carried out in war of the scale of that in the Former Yugoslavia make special demands on humanitarian organizations such as to require the setting up of management structures and staffing levels that can respond flexibly to continual changes. At the Headquarters level, desk officers cannot accommodate the volume of processing nor the need for rapid and flexible response. This makes it necessary to create operation specific management units. At the Field Central Office level, such operations create the need for immediate high-level policy responses so frequently as to make it useful to appoint a senior official with the status to deal with the highest international and national authorities, and the authority to make important policy decisions whenever required. At the Field operational level, a war situation results in the relative isolation of the field offices, so that it becomes important that their managers possess the capacity to grasp the changing pattern of events, the implication of these events on the organization’s goals in the area of responsibility, and the ability to rapidly prioritize objectives according to available resources and military and security constraints. Special demands are also placed on the lines of communication and on the control and accountability mechanisms between the field offices and the central country office.

Some of these requirements were recognized early on by UNHCR in the Former Yugoslavia, as is shown by the early appointment of a Special Envoy in 1991 and by the creation in 1992 of a special unit, the Special Operation for Former Yugoslavia Unit (SOFY) outside the framework of the Europe Bureau.

UNHCR relied on external recruitment in order to fill many of the operation’s posts. Many of the “outsider” field managers had excellent management skills. But many had neither former experience in emergency operations; nor did they have any in-depth understanding of the agency’s goals and principles, which is important if effective guidance is to be given to subordinate staff operating in a continually changing situation. In the face of these facts, and despite the excellent performance of some among the outside recruited staff, it would have benefitted the operation if the combined efforts of the personnel and emergency sections of UNHCR (and of WFP, had some of its staff been seconded to UNHCR as suggested) had been able to fill the relatively few key managerial posts with skilled in-house personnel.

Recommendation 8

The response capability of agencies such as UNHCR and WFP in emergency situations depends ultimately on field-level response capability. Therefore, recruitment of a sufficient number of field-office level managers with previous experience in emergency operations and knowledge of both the agencies’ principles and objectives, is an important priority. Even in large operations, it should be possible to deploy such staff for the relatively few key management field-office and central country-office posts. The organisations’ Emergency Units should be capable of ensuring that this can be achieved.

IV. Food aid in the reconstruction period

IV.1 The transfer to WFP of management of the food chain within Bosnia.

Following the signature of the Washington Agreement in 1994 and the establishment of the Federation, and as the situation stabilized in Bosnia, a transition was made to a new division of labour in the food programme with WFP gradually moving into Bosnia, first by opening a warehouse in Zenitsa in November 1995. In March 1996 WFP moved its regional office to Sarajevo. Another important step was made by WFP in defining its strategy in mid-1996 following a joint WFP/UNHCR/FAO food aid needs assessment in April 1996. The process of transition culminated in January 1997 with the completion of the transfer of responsibility for management of all stages of the food chain to the WFP.

The transfer of management of the food chain between the two agencies was an understandable move. As the special circumstances of war no longer applied, it could be expected that the division of labour should correspond more closely to the global MOU as well the global mandates of the two agencies. However, although understandable, was it necessary? This question should be asked because of the financial and institutional costs of the transfer of responsibility between the two agencies, in particular within the context of a decreasing quantity of food supplied through the programme.

Whether the shift in responsibility was justified will depend in part on how long a large-scale food aid programme is to be maintained in Bosnia. Under one scenario, in which food aid could be phased out during 1998-99, the transfer is difficult to justify. It might have been sufficient for the declining quantity of food to be distributed by the UNHCR to the authorities in proportion to earlier allocations, leaving it to them to apportion the allocation of the diminished quantities.

The shift has been accompanied by the need from the WFP to create an administrative network, involving logistics, programme design and monitoring, through Bosnia, while the UNHCR has run down its capacity in those regards. The net costs may be mitigated, as the UNHCR has had to shift its attention to the complex issues of the return of DP’s and refugees. The shift has also involved changes in the choice of local partners, in logistic modalities and in policy with regard to targeting criteria.

Some of these changes were a necessary consequence of the end of hostilities, and may have been implemented even without the shift in agency responsibility, while in some respects the appearance of a new actor in the field may have made policy changes easier to implement. Nevertheless, the shift did involve a degree of cost in terms of confusion and misunderstanding.

There were two different components of the transition which have to be evaluated. One aspect was to shift in control over logistics within Bosnia; this is evaluated in the Logistics Annex, and is mainly a matter of cost effectiveness as the transport shifted from the official bilateral convoys appropriate for war-time siege conditions to commercial transport. That transfer can be viewed as separate from the other aspect of the transfer, which involved the decision that WFP should take over the management of relationships with local intermediaries and activities related to beneficiary targeting and monitoring.

IV.2 Relationship with the authorities

The relationship of the WFP with the authorities has been bedeviled by the fact that the changeover to WFP management coincided with the drastic scaling down of the programme, with the result that WFP staff had to develop relationships with local partners at the same time as carrying the message that resource availabilities were to be severely curtailed. This has been rendered more difficult by virtue of the fact that for some partners, food aid had become the main resource at their disposal. Feelings of resentment were compounded when the requirements of the programme suggested the need to shift emphasis away from partners from the war period.

Despite these difficulties, WFP have managed to establish good working level relationships with many partners and have in many instances succeeded in eliciting cooperation in implementing the often unpalatable steps required to trim the numbers of beneficiaries. This has, by and large, been possible to achieve through a process of dialogue, although there have been instances in which the sanction of withholding delivery has had to be used to ensure compliance with programme policies (e.g. re-categorizing of beneficiaries; the maintenance of the integrity of the distribution system).

However, difficulties of the relationships with local partners is a necessary and acceptable cost in the process of shifting responsibility for social welfare back to local institutions, a matter discussed further below.

The problem also remains that the choice of local partners has political implications, in peace as well as war. In the context of a highly charged and contested politics, the need to work with those in authority can result in an appearance of partisanship, as can a decision to work with one or other NGO. The only partial solution to this problem is continuing vigilance to ensure that food is not used as an explicit instrument in political contests.

Recommendation 9

IV.3 Rehabilitation

IV.3.1 UNHCR and WFP participation in rehabilitation projects

During the war both organisations tried to help people to retain and rehabilitate their assets. A notable example was UNHCR’s response to local requests for carrying out crop production activities: starting in 1993 a seed project was implemented with UNHCR funding through various NGOs (IRC, AICF-US, LWF) in 1993 and 1994. Through this project a total of 7,885 MT of seeds was distributed to some 435,000 families in Central Bosnia, Southern Bosnia, Banja Luka, Bihac and Sarajevo by road, airlift and airdrops. In 1994, the spring sub-component provided some 6076 MT of vegetable seeds (potatoes, beans, onions, peas, cabbage, cucumber, tomatoes, spinach and lettuce), while the winter component included some 1,796 grain (wheat) seeds. The programme provided a valuable supplement to the food aid being provided and contributed to cover the vulnerable categories in the selected areas.

Since late October 1996 WFP has been trying to support small-scale local level projects. According to the strategy, these projects should: "address the local level food deficits while contributing to the reconstruction of infrastructure and industry and rehabilitation of the economy." And they are also "aimed at the generation of employment ..., and economic rehabilitation and growth." The projects should "Address local food deficits and increase household purchasing power by supporting these local projects with food aid."

To-date there has not been sufficient time to monitor, let alone evaluate the effects of the projects. Many of the activities such as women's sewing workshops, are not directly related to increased food production, and their longer term sustainability is uncertain, they merit support to the extent that they contribute to a climate of reconstruction; and encourage participation in rehabilitation. Such criteria are probably more realistic than the standard longer term development parameters given in the country strategy document - such as "addressing local food deficits and increasing household purchasing power and food security...".

As the programme moves away from relief food aid, more attention should be given to finding appropriate partners for development type activities. There should however be a clear understanding that WFP inputs will, in general, be limited to relatively small, short-term, one-off grants, and as such could not be the main source of project funding. These factors should influence the choice of partners, as well as the type of activity.

The monitoring of rehabilitation activities to-date has been limited primarily to food management issues and quantitative information on project participants and, where relevant, production data. To the extent possible, future monitoring should include qualitative data, such as changes in the socio-economic status and vulnerability of beneficiaries, and the appropriateness and effectiveness of the implementing partners, while effects monitoring and evaluation should focus on the contribution that the projects make to people's attitudes towards overall reconstruction and rehabilitation.

IV.3.2 Food aid and social welfare assistance

An important rehabilitation issue relates to the framework for social assistance. There is a need for a government social welfare system to care for people unable to care for themselves. The weakness of the current arrangements results both from the economic disruption and the administrative and political uncertainties resulting from the war, and for the need to reform the tax and social security system as a result of the transition to a more market-oriented economy. At a minimum the foundations and framework of a system to cater for social welfare cases needs to be established as early as possible, and preferably before the international agencies completely withdraw their assistance.

In a situation in which fiscal arrangements remain in a confused and debilitated state, the responsible authorities have found food aid to be one of the few resources at their disposal to respond to the needs of their clients. However, food aid is not a viable alternative to a social welfare system, and its availability should not become an excuse for delaying the much needed, but difficult reforms.

Recommendation 10

As a first step more concerted efforts should be made to coordinate with other international agencies. In this context the Country Office should support the initiative of the World Bank to work, with government, other donors and NGOs, on a medium-term social assistance strategy. At a minimum, better coordination between agencies should lead to more coherent criteria for beneficiary selection. It is important that local authorities become more involved in identifying needs, hence in establishing criteria, as well as in building a new welfare system. The international agencies, including WFP will need to pay considerable attention to involving the authorities from the federal, canton and municipal levels, and assisting them in defining and understanding their roles and responsibilities in the establishment of their welfare system.

V. The future of food aid in BiH

This evaluation covers the period up to mid-1997. As such, it only covers the beginning of the reconstruction period; however, evaluation of the policies adopted towards the end of the period being studied required the team to take a view regarding the possible longer term of food aid in Bosnia, which should inform policies adopted in the transition from an emergency to a reconstruction mode of operation.

The justification for the transfer of food aid administration depends in part on the changing role of food aid. During the war, UNHCR found itself managing a food distribution system which supplied food to the “war affected” as well as to refugees and DPs - this meant that as its peak, food aid was being supplied to a majority of the population in some areas, and issues of targeting and monitoring mainly related to ensuring that minority groups, institutions and collective centres were supplied with food. This, of course, extended the role of UNHCR well beyond its more usual mandate of aiding refugees. Following the end of hostilities, the food aid programme was inevitably to be scaled down. If the scaling down was to be done by narrowing the coverage of the programme by focusing on DPs and refugees, rather than feeding the general population, then it would have been sensible for UNHCR to continue to manage the field operation. If the shift was to be towards a poverty targeted programme including the general population, then WFP was the more appropriate agency.

The shift to WFP management in practice has been accompanied by the move to focus the programme on the needs of the vulnerable, including those whose vulnerability is not related to their refugee or DP status and, to a much lesser extent, to use food aid as an input into reconstruction projects. This involved the WFP shifting from a disaster relief mode of operation to a mode characteristic of its ongoing programmes in low income countries. This was an understandable institutional response - it was a plausible response to the question “if there is to be a food aid programme, what form should it take”. However, that question should be asked only after a positive answer has been made to the question “should there be a long-term food aid programme”.

The build-up of the food aid programme was fully justified by the war situation, and the continuance of the programme in the immediate post-war situation was justified by the overall food insecurity resulting from the dislocation of international and internal commerce and the disruption in agricultural production, all of which made it necessary to continue a large-scale food programme in the immediate aftermath of war.

By 1997, the international trading system had been restored and the commercial system within the entities was operative. Commercial transactions between the entities are recovering more slowly, but that recovery will be faster the more that households purchase commodities through open markets, rather than receiving commodity donations. Agricultural output has proved more resilient than industrial production.Agriculture in Bosnia

Previous to the War, 40percent or more of the population of Bosnia and Herzegovina was involved in agriculture on a full-time or part-time basis, in some 540,000 private small-holdings, which accounted for more than 90percent of the land in use. However, agriculture was based on relatively low labour productivity techniques, accounting for some 14percent of Gross Social Product; in addition forest products accounted for more than 10percent of GSP. Between one third and one half of food requirements were imported, mainly from other parts of the Former Yugoslavia, although the area now included in the RS entity was a grain surplus region. During the War the main cause of food insecurity was the disruption of transport and trade, rather than the collapse in domestic agricultural production.

During the War agriculture suffered from disruption of production, destruction of farm buildings, machinery and livestock[12] and the loss of access to about 15percent of farm land because of land-mines. On the other hand, there was a War-time effort to supplement food supply through gardening in the urban areas

In pre-war Former Yugoslavia many of the forests were undercut. During the war exploitation was haphazard, at best. Many forests were heavily cut for fuel-wood. The post-war period has given rise to quite large-scale logging operations; the risks of unsustainable management are significant.

The most critical requirements to get agriculture moving was the provision of seeds and other seasonal farm inputs and the replacement of livestock and farm equipment. Some support has been provided by the international community to the agricultural sector, although presentations to the evaluation team suggested that there was still a need for seed provision and improved technology.

The evidence available to the mission was fragmentary and it seems that there is a dearth of data on current agricultural and forestry production. But there has been some recovery in agricultural production in 1996-1997. Agriculture, being small-scale, relatively labour intensive and mainly based on private production could be expected to be more resilient than the much larger scale, publicly owned industrial sector. However, there will be some continuing output loss as a result of under-utilization of the land Formerly farmed by DPs. Given BiH’s comparative advantage, there is no economic reason for it to aspire to food self-sufficiency. The need to import substantial amounts of food by itself does not establish a case for food aid.

By now the source of food insecurity is the lack of income at the household level, i.e. the problem is one of household entitlement rather than food availability.

Some components of the household income deficiency are directly related to the war (most notably in the case of DPs), in other cases there is an indirect relationship (e.g. as a result of unemployment resulting from the War). However, significant parts of the household insecurity problem are either a manifestation of the sort of poverty to be found in all societies, or reflect the problems of transition from a planned to market economic system (including the difficulties in funding social entitlements and income losses resulting from the poor performance of public enterprises in transition). Social safety net issues and the need to reform the social security system are being confronted in all the transitional economies.

While it is understandable that the local authorities, faced with the difficult challenge of supporting the large numbers dependent on public support, should not wish to see the disappearance of food aid, one straw to grasp on to, there are a number of good reasons why food should not be seen as a long-term solution to the problems of vulnerable households.

Food aid is an effective instrument for intervention in very low income countries not only because of the moral justification of focusing support on the “poorest of the poor”, but also because at very low income levels food is a high proportion of the household budget, so that food aid can make a significant impact on welfare and can provide a plausible incentive (e.g. in Food for Work projects).

GDP and personal income - estimates and projections

Click here for table

Data from Bosnia and Herzegovina: From Recovery to Sustainable Growth, World Bank 1997, table 11, page 100. 1990-1996 figures are estimates; 1997-1999 figures are projections.

The figures in the table above include the available estimate of GDP and per capita income for Bosnia. This figures show that before the war, Bosnia was a middle income country (it had already suffered some economic decline and had been better off in the 1980s). The economy enjoyed middle income status, and as such would not have been a candidate for food aid. As a result of the war, output and incomes collapsed to one fifth pre-war levels. Recovery began in 1996 and continued into 1997; while estimate of GDP and income in 1997 were not available to the team, the continued revival in wage rates suggests that the World Bank projection for 1997 can be taken as a guide to current output levels. It is generally agreed, based on fragmentary data and direct observation, that there are considerable variations in the pace of recovery, resulting in substantial differences in the per capita incomes, with RS falling significantly below levels in the Federation, and within the Federation the cantons with Croat majorities being on average better off than those with Bosniak majorities.

The average income figures suggest that although the population is still suffering a severe decline in incomes compared with pre-war standards, average per capita income has now reached levels well above the levels which would normally justify a substantial food aid programme. Obviously, there will still be widespread perceptions of deprivation and vulnerability, with average income levels at around 40percent pre-war levels, and which justifies a substantial programme of international support for reconstruction to restore output to pre-war levels. But food aid is only an appropriate tool for groups with incomes far below current average incomes.[13] The need to focus the food aid programme on vulnerable groups with income levels considerably below the average was evident, along with the likelihood that if recovery matches the trajectory set out in World Bank projections, the programme would be phased out in the near future.

However, there was a dilemma in adopting poverty/food vulnerability criteria for food aid targeting and pushing the local authorities into elaborate exercises to identify beneficiaries on the basis of those criteria, in that it may be taken to imply both that food aid is an appropriate instrument to address social vulnerability in the Bosnia situation and that the WFP is accepting some responsibility for supplying food over the longer term for the vulnerable - both questionable conclusions.[14]

[1] Some 160,677 metric tons of aid (144,827 of food and 15,850 of medicine) were transported by the airlift, while more than 20,000 tons of aid were dropped by military cargo planes to besieged towns.

[2] Signed 14 December 1995.

[3] ICRC did not participate in the Mission, but was present at its initial and final meetings.

[4] The initial letter from the UN Secretary General to the UNHCR of 14 November 1991 was interpreted by UNHCR as inviting the agency to take the lead in coordinating humanitarian assistance to displaced persons. Subsequent UN documents in 1992 describe UNHCR variously as the “lead agency for humanitarian relief”, as “leading the international effort”, and as “lead agency for humanitarian activities in the Former Yugoslavia”. The mandate of UNHCR was extended as a Good Offices Activity at the request of the Secretary General rather than a Mandated Activity in response to a Security Council Resolution.

[5] The food content of the family food parcel consisted of: oil, cheese, sugar, tinned fish, corned beef, high protein biscuits and pasta.

[6] Among these, five flew regularly throughout the operation: Canada (1860 flights), France (2133 flights), Germany (1279 flights), the United Kingdom (1902 flights) and the United States (4597 flights). Italy was also a crucial participant throughout the airlift, providing planes and facilities at Ancona.

[7] The evaluation team was struck by the lack of complaints about inter-agency cooperation from those involved in field operations during the war period. Of course, when pressed, instances of difficulty were mentioned and are reflected elsewhere in this report, but these were few given the size and inherent stress involved in the operation.

[8] (Thomas G. Weiss and Amir Pasic, Reinventing UNHCR: Enterprising Humanitarians in the Former Yugoslavia, 1991-1995, Global Governance Vol. 3 No.1, Spring 1997).

[9] The average number of beneficiaries results from the amount of food supplied divided by the basic food ration. If, on average, beneficiaries received more or less than the basic ration, the number of beneficiaries covered would be fewer or greater.

[10] In Sarajevo, it was possible to supply the city from other sources to some degree through the tunnel, but by contrast Gorazde was exclusively dependent on humanitarian aid for outside supply.

[11] In neither case were these figures based on an actual survey of beneficiaries, but rather the difference resulted from a different way of interpreting and presenting the available data on the amount of food being shipped.

[12] It has been estimated that 70% of farm equipment and 60% of livestock were lost in the War, and that by 1995 fruit production has dropped by about half and wheat, maize and potato outputs were 60-70% 1990 levels. See Bosnia and Herzegovina: The Priority Reconstruction Programme: From Emergency to Sustainability, pages 103-115. Prepared by the European Commission, the EBRD, and the Central Europe Department of the World Bank, November 1996.

[13] One way of looking at the potential role of food aid is to assess what contribution a food aid basket could make to income. Roughly, the current WFP food basket could contribute around 8% to incomes at the average income level.

[14] It would not be surprising if there is some resentment if, after the promotion of an elaborate exercise to identify beneficiaries, there are no further benefits.