Coordination in the midst of chaos: the refugee crisis in Albania

This article explores the coordination of the aid effort, the role of NATO and, finally, whether the response to the Kosovo crisis has strengthened or undermined the principles of universality that govern the global provision of humanitarian assistance.

The three months from April to June 1999 saw a massive emergency aid programme, as large and as high profile as any operation with refugees since the Great Lakes crisis of 1994. The stage may have been different but most of the actors were the same. UN agencies, NGOs, donors, foreign armies and the media all landed in force in Albania in early April and began to play their part in responding to the needs of the refugees. The resources available to aid agencies, via governments and through public appeals, were higher than they had ever been. The humanitarian programme was welcomed and facilitated at every level, by the government and population of Albania. There was an abundance of logistical support provided by foreign armies, principally from NATO countries. There were no particular security concerns. All the ingredients were apparently in place for an efficient and well-run emergency operation.

And yet, for many of us working in Tirana, the Albania emergency programme raised as many questions as it answered. Speculation and conjecture about what might have been done better will come to dominate the policy debates in the months and years to come.

Coordination of the aid effort

To discuss the coordination of humanitarian assistance in Albania, even at this early stage, runs the risk of repetitiveness. Public statements by some donor governments and NGOs about the quality of coordination in Albania were well publicised. UNHCR became more and more irritated with what they perceived as observations that were neither original nor constructive. Proof of poor coordination was not hard to come by. There were instances of several NGOs competing to work in the same camps, duplication of essential services, and, above all, a great variance in the standards in different camps. People looked to UNHCR to provide coordination and were quick to allocate blame where it was lacking. In fairness to UNHCR, there were aspects of the refugee crisis in Albania that, from the outset, made coordination extremely difficult, if not impossible. No analysis can afford to overlook these constraints.

The dispersal of the refugee population was a major factor. By its peak at the end of the first week of June, the refugee population in Albania was estimated at 460,000 people. Of these, some 270,000 refugees were living with host families.(1) The balance of refugees was spread out in camps and collective centres. Some of the camps held several thousand refugees, while some of the smaller collective centres held no more than a dozen. The number of these sites was estimated at between 700 and 900. However much one might be in favour ‘philosophically’ of refugees being so spread out and avoiding the creation of tented cities so vividly associated with previous refugee crises, this has to be balanced with an objective assessment of our own capacities and systems as relief agencies. Relief agencies are accustomed to responding to certain scenarios and were simply overwhelmed by having so many sites to assess and to service. NGOs were able to pick and choose where to concentrate their resources while the mandate of UNHCR compelled them to be everywhere at the same time as lack of resources exposed them to criticism for not doing so. In simple terms, they could not attain the coverage that decent coordination requires.

There is a saying that ‘you cannot DO coordination to people who do not want to be coordinated’, and nowhere was this saying more true than in Albania in the first three months of the refugee crisis. The simple truth was that, even if they had acquired the capacity, UNHCR would almost certainly have been unable to provide the coordination required due to other characteristics of the situation.

Albania saw arguably the highest ever proliferation of NGOs in a refugee crisis. At its peak, there were over 160 NGOs registered with the NGO Centre in Tirana. While many were motivated by altruistic considerations, there was also a high proportion of ‘briefcase NGOs’, there because they could not afford not to be there. As soon as they had found a site to work, their publicity machines back home would whirr into action and fundraise off the incredible sympathy that the plight of the Kosovans had evoked. Many of these organisations failed to attend general or sectoral coordination activities, actively resisted attempts at being coordinated and even refused to register their activities with the Government of Albania or UNHCR.

One can also better understand the particularities of coordination in Albania by comparison with past refugee crises. In Tanzania in 1994, for example, a courageous and visionary Field Coordinator for UNHCR in Ngara allocated each of the key sectors to a handful of respected NGOs and politely told the rest to pack their bags. The result was impressive. She did this because she was empowered to do so – the Government of Tanzania only allowed NGOs to operate in Tanzania if invited to do so by UNHCR. UNHCR could thus ‘DO coordination’ its way.

In Goma, a few months later, the UNHCR found itself with an even greater crisis to coordinate. After the chaos of the first few weeks, it was slowly able to exert its coordination role - one notable way in which it did this was though funding implementing partners to work in the three main camps. Many of the major donors (notably ECHO) empowered UNHCR to do this by channelling all of their funding direct to the refugee agency, rather than setting up a multitude of individual partnership agreements. It was then up to UNHCR to use its own funding of NGOs as a way to ‘DO coordination’. To no-one’s surprise, it worked.

To understand how much has changed, picture the scene in Tirana in early May 1999. At a meeting with NGOs, a senior UNHCR official referred to his organisation as technically bankrupt, having received no cash pledges in the first month of the refugee crisis. Senior political and diplomatic policy makers in the UK, USA and EU, disappointed at UNHCR reactions to the first wave of the crisis, were said to have placed a personal veto on cash funding to UNHCR.

While this cash crisis ravaged UNHCR, the NGOs had never had it so good. The smallest of agencies found themselves awash with funds, whether from private or government sources, at a time when UNHCR were appealing for financial support live on CNN.(2) Meanwhile the Albanian government was neither willing nor able to use UNHCR as a ‘quality control’ tool to ensure that only the most established agencies operated on the ground. Agencies, therefore, had no incentive to be coordinated, and appeared unwilling or unable to coordinate themselves. In Albania, the inability to ‘DO coordination’ had never been more apparent.

The role of NATO

Coordination is linked to the second key issue, that of NATO’s role in the relief effort. Military disaster relief operations, and resultant intense debate, are not new. Nowhere have such operations been on the massive scale we have seen in the Kosovo crisis. The bilaterals, as the NATO armies came to be known, were hugely important actors in the relief effort, not only providing the logistical support traditionally associated with the role of the military but also setting up and managing refugee camps on behalf of the governments that they represented.(3)

At the onset of the crisis, the role played by the military was criticized by observers who noted the rather blurred line between NATO as warring party inside Kosovo and NATO as humanitarian actor just outside its borders. This defence of principle, important though it is, has to be seen against the reality that without such logistical and material intervention, the aid agencies would have found it hard or impossible to cope. This was particularly the case at the onset of the crisis. The middle ground that most people were happy to occupy was to recognize that NATO had a unique and important role to play in the provision of logistical and technical support, but that their support should be placed at the disposal, and under the coordination, of UNHCR. This, in theory, was fine.

In practice, however, this was not always the case. The bilaterals forged ahead with camp development, often without the input of experienced site planners. Some of the sites the military chose had earlier been rejected by aid agencies as entirely unsuitable for refugee camps. In the scramble to set up camps, NGO advice was ignored or went unheeded. Latrines were poorly sited and tents placed too close together, mistakes which could easily have been avoided had aid agencies assisted the armies at the planning stage.

The guiding principle for many donor governments came to be to pour as much money as possible into ‘their’ camp, and to try and attract the maximum number of refugees to live there. In Kukes, pictures of the major camps further south away from the Kosovan border were displayed on public notice boards, in order to attract refugees towards a particular camp. People could see the tents that waited for them, the kitchens where they would be served food, the spaces where their children could play. Competing camps were displayed next to each other in this makeshift estate agency for refugee camps.

Many coordination problems were exacerbated. There was often little or no incentive or mechanism for bilateral agencies to coordinate with UNHCR. Camps were often identified and prepared by armies, long before UNHCR was even aware of their existence. Even more galling to UNHCR was the fact that the same governments were often the most voluble in their criticism of UNHCR’s ‘lack’ of coordination.

A consequence of bilateralism was that NGOs were seemingly selected to work on a site on the basis of shared nationality. Broadly speaking, Spanish NGOs worked in camps prepared by the Spanish military, British NGOs worked in camps prepared by the British Army and so on. These arrangements were usually made via the aid department of the country concerned, who picked up the bills of both the army that had prepared the site and the aid agencies that would manage it. It goes without saying that it is difficult to ‘do coordination’ when agencies are selected and financed on this basis (and in total contrast to the positive examples from the Great Lakes in 1994). In brief, the lesson to be learnt from the Kosovo crisis is that ‘bilateralism’, almost by definition, undermines coordination.

Principles of universality

Another major concern has been the uneven standards set by the bilateral actors. In some camps, the standards were so high that people, only half jokingly, came to ask whether there should be maximum standards as well as minimum ones. In less fortunate adjacent camps standards were so low that refugee families preferred not to move in at all. One of the most striking examples was to be been in Shkodra, on the border with Montenegro. One camp had imported street lighting, hot showers and several television rooms while in another camp, only a few kilometres away, tents lay in a gravel pit without so much as a single communal space for the whole camp. Per capita expenditure on the former probably exceeded the latter by a factor of fifteen or twenty.

To raise questions about high standards in refugee camps is to enter a minefield. It is important, however, if one starts from the premise that all refugees, from Kosovo and wherever else in the world, have certain universal rights to assistance. One distinct characteristic about the aid effort in Albania over the first three months was the apparent absence of any sense of ‘absolute’ value of money, or any awareness that resources were finite. There was money to do almost anything and to do it almost anywhere. There was, on the face of it, no reason to hold back from installing electric showers and street lighting in a refugee camp if a donor was prepared to offer funds. At the same time, however, over half the refugee population was receiving no help whatsoever.

People living in host families or in private accommodation did not receive even a fraction of the attention or the assistance given to people in camps. For the first two months not a single food distribution reached the 270,000 refugees in private accommodation. By the end of June, financial assistance promised by UNHCR ($10 per person/month) had not reached host families. This failure to reach refugees living with host families will be looked upon (alongside the failure to complete the registration of refugees while they were in Albania) as among the gravest shortcomings of the aid effort in Albania.

Throughout this period some 190,000 refugees in camps and collective centres were tended to by over a dozen foreign armies, all the major UN agencies, and some 160 NGOs. Total expenditure will probably never be fully known. There are staggering comparisons to be made. The development costs alone of the US Army/OFDA site known as Camp Hope, near the town of Fier in the south of the country, have been estimated at some $50 million.(4) A maximum of 3,500 refugees lived in the camp at any one time. If the same sum had been allocated to UNHCR to enable it to give $10 per person per day to hosting families, all of the 270,000 refugees living in private accommodation would have been sustained for between four to five months. (More soberingly, expenditure on this one relatively small camp would almost fully fund this year’s UN consolidated appeal for Angola.)

Such comparisons may provoke unease among agencies. However, those who are familiar with the international assistance system know full well that there are only finite resources available for overseas aid budgets. Comparative judgements are more than merely useful – they are essential.

From the onset of the Kosovo refugee crisis, aid agencies working in other parts of the world noticed a drop in the resources available for their ongoing programmes. Initial commitments to rebuild Honduran and Nicaraguan livelihoods shattered by Hurricane Mitch were not acted upon. Proposals for emergency work in Sierra Leone and Angola lay on the desks of donors, unanswered and ultimately unfunded. NGOs, for their part, pulled many senior staff from programmes elsewhere in the world to staff their programmes in Albania and Macedonia. The international assistance system again showed itself to be like a searchlight, flitting from one corner of the world before moving onto the next. This lack of consistency is fundamentally at odds with the real needs of refugees and disaster victims and the prospects of achieving a global or universalist perspective.

People are already starting to ask to what extent the response to the Kosovo crisis creates precedents. Moral interventionism is again on the agenda and being discussed as a future foundation of Western foreign policy. Some urge caution and argue that NATO, for example, has a specific political and geographical sphere of operation, and that intervention to end a conflict in Europe fell directly within it. It would therefore be mistaken to assume or to advocate that the alliance will intervene in similar instances in Africa or Asia. This is a persuasive argument, however much one might wish that it were otherwise.

On the humanitarian level, however, no such argument exists. Our humanitarian charters are instead explicitly founded on global principles of universality – ‘the right to receive assistance, and to offer it, is a fundamental humanitarian principle which should be enjoyed by all citizens of all countries. As members of the international community, we recognize our obligation to provide humanitarian assistance wherever it is needed.’(5)

In recent times, these principles have been most commonly associated with initiatives (such as the Ground Rules in south Sudan or The Principles of Engagement in eastern Democratic Republic of Congo) that involve representatives of the international community trying to gain acceptance of these values by warring parties. How often, though, do we apply these principles to ourselves? Within this framework of universal rights, the quality and quantity of humanitarian assistance should be applied evenly to needs as they arise, wherever and however that may be. When the Kosovo refugee crisis is compared to other contemporary disaster zones such as Angola, the DRC or south Sudan, some of the costs of the refugee operation become impossible to justify. Aid personnel accustomed to the difficulties of securing donor funding for even the most life-threatening emergency in some of these forgotten areas found it hard to come to terms with the sheer quantity of resources available to refugees in the Balkans. The response to humanitarian need in Kosovo was, quite literally, beyond compare.

Why does it matter that so many resources were poured into this one refugee crisis? On what moral basis can one criticize this sum, or indeed any sum spent on a refugee population? The answers lie in the reality that aid budgets are finite. Money spent on streetlighting in Albania could have been spent on vaccines in Cuito. If we fail to attach a similar value to the preservation or amelioration of a human life in Angola as we do in Albania, then we have, in the most basic way, failed the very humanitarian principles that we are so keen for others to adopt. The figures of this crisis speak for themselves – the international community currently spends on an African refugee less than a tenth of the amount spent to assist a refugee in Europe. This gap has to be significantly and quickly narrowed, if our principles are not to appear hollow.


Toby Porter worked as Emergency Programme Coordinator with Oxfam GB in Albania. He wrote this in a personal capacity. Email: He is joining the RSP in October to do the Master of Studies in Forced Migration.


  1. In June, the term was changed to refugees living in ‘private accommodation’, in order to reflect the commercial nature of the arrangement in the majority of cases.
  2. This seismic shift in the relative positions of the UNHCR and the NGOs in the five years since Goma cannot be overstated – nor should it be welcomed. Only the most shortsighted NGOs would welcome a weakened UNHCR and the coordination vacuum that would ensue.
  3. Even the names of the refugee camps reflected this. A visitor to Kukes would arrive at ‘Italian 1’, and then proceed to ‘Greek Camp’ and ‘Italian 2’. These named after the armies that set up and managed the camp.
  4. Figure given by OFDA official during informal conversation
  5. Code of Conduct for the International Red Cross and Red Crescent Movement and Non-Governmental Organisations in Disaster Relief.


Opinions in FMR do not necessarily reflect the views of the Editors, the Refugee Studies Centre or the University of Oxford.
FMR is an Open Access publication. Users are free to read, download, copy, distribute, print or link to the full texts of articles published in FMR and on the FMR website, as long as the use is for non-commercial purposes and the author and FMR are attributed. Unless otherwise indicated, all articles published in FMR in print and online, and FMR itself, are licensed under a Creative Commons Attribution-NonCommercial-NoDerivs (CC BY-NC-ND) licence. Details at